{
  "schema_version": 1,
  "cohorts": {
    "v1": {
      "categories": {
        "all": {
          "n_events": 790,
          "n_species": 79,
          "models": {
            "bioclip": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.40759493670886077,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.37744358138345485,
                    "high": 0.43668278056331855,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 1.0,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.9999999999999999,
                    "high": 1.0000000000000002,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.40759493670886077,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.3774435813834548,
                    "high": 0.4366827805633185,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.19377658997623626,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.17633500057238008,
                    "high": 0.20292895092335855,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": null,
                "wrong_species_rate_all_events": {
                  "value": 0.5924050632911393,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.5633172194366814,
                    "high": 0.6225564186165452,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.45378062209624936,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4129363937454471,
                    "high": 0.47521336608634596,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": null,
                "any_rank_conflict_rate": null
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.43356643356643354,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.38564615681848224,
                        "high": 0.4819290676218387,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.43356643356643354,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.38564615681848224,
                        "high": 0.4819290676218387,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5664335664335665,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.5180709323781613,
                        "high": 0.6143538431815176,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.22525144798559688,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.18581102096437715,
                        "high": 0.2327245798078507,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4650352474541355,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3836098497329077,
                        "high": 0.48046364863556285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3767313019390582,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3242506811989101,
                        "high": 0.429730288139379,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3767313019390582,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3242506811989101,
                        "high": 0.429730288139379,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6232686980609419,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5702697118606209,
                        "high": 0.6757493188010899,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.19503914425228208,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1471057016719108,
                        "high": 0.20126324698524756,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3768552956739009,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.28423813543386156,
                        "high": 0.3888815280731902,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5706666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5190208791964778,
                        "high": 0.6215506153351457,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5706666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5190208791964778,
                        "high": 0.6215506153351457,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.42933333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.3784493846648543,
                        "high": 0.4809791208035223,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.3569681825862559,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3021033359601287,
                        "high": 0.36228625823980637,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.576640910331644,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.48801308116636166,
                        "high": 0.5852316479258409,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2727272727272727,
                        "high": 0.3955579710144927,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2727272727272727,
                        "high": 0.3955579710144927,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6666666666666666,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.6044420289855073,
                        "high": 0.7272727272727273,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.19571272127354958,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13889177057499402,
                        "high": 0.20008868122703138,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3022398986756082,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.21449108873606673,
                        "high": 0.3089977102493396,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.15517241379310345,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.10285328185328185,
                        "high": 0.21142857142857144,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.15517241379310345,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.10285328185328185,
                        "high": 0.21142857142857144,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8448275862068966,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.7885714285714286,
                        "high": 0.8971467181467181,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.07969720820960491,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.04284947343218079,
                        "high": 0.08733185199203032,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.12523847004366484,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0673348868219984,
                        "high": 0.13723576741604762,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5918367346938775,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.529143220735149,
                        "high": 0.6536964980544747,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5918367346938775,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.529143220735149,
                        "high": 0.6536964980544747,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.40816326530612246,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.3463035019455253,
                        "high": 0.470856779264851,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.3709912983692708,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.29522804535874536,
                        "high": 0.37082372741107344,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5910708821476517,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.47036332650376383,
                        "high": 0.5908039046888289,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5307692307692308,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4428544165757906,
                        "high": 0.6194029850746269,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5307692307692308,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4428544165757906,
                        "high": 0.6194029850746269,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.46923076923076923,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.3805970149253731,
                        "high": 0.5571455834242094,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.3512735981966752,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24627254281666044,
                        "high": 0.35753108003108003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5189269064269064,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3638117109791575,
                        "high": 0.5281709136822773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3177570093457944,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.23075221238938057,
                        "high": 0.4056673827760134,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3177570093457944,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.23075221238938057,
                        "high": 0.4056673827760134,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6822429906542056,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.5943326172239866,
                        "high": 0.7692477876106193,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.19332045732045733,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11640574795574796,
                        "high": 0.2008696803196803,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2842947901771431,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1711849234643352,
                        "high": 0.2953965887054122,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3458646616541353,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.26153846153846155,
                        "high": 0.4296875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3458646616541353,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.26153846153846155,
                        "high": 0.4296875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6541353383458647,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5703125,
                        "high": 0.7384615384615385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.20422734764840028,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1273655459317224,
                        "high": 0.20972793816543817,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.30842497399962493,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19234796732545828,
                        "high": 0.3167319882498454,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.07894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.02531645569620253,
                        "high": 0.14492753623188406,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.07894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.02531645569620253,
                        "high": 0.14492753623188406,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.9210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.855072463768116,
                        "high": 0.9746835443037974,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.048546691403834254,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.011544011544011544,
                        "high": 0.0567099567099567,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.08126293995859213,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.01932367149758454,
                        "high": 0.09492753623188407,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.13541666666666666,
                        "high": 0.3,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.13541666666666666,
                        "high": 0.3,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7857142857142857,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.7,
                        "high": 0.8645833333333334,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.1114969135802469,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.058595792483660133,
                        "high": 0.12577215608465608,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.1745169082125604,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09171515345268542,
                        "high": 0.19686076604554864,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 790,
                  "population_events": 790
                }
              }
            },
            "speciesnet": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.2759493670886076,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.2565859544815241,
                    "high": 0.2972859471751877,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.48481012658227846,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.46199542897327717,
                    "high": 0.5091264133194513,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.5691906005221932,
                  "unit": "proportion",
                  "denominator": 383,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5307347404469154,
                    "high": 0.6091895497121522,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.21071676757461957,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.19353306451633928,
                    "high": 0.22214092042227154,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 0.6272151898734178,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.47349599459963393,
                    "high": 0.7919265404850383,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.2088607594936709,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.18530846439074286,
                    "high": 0.23283848886696987,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.30940689922349207,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.28417513270753625,
                    "high": 0.3261816046706772,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 1.9753164556962028,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 1.8936018240725518,
                    "high": 2.063675550451658,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.26962025316455696,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.24453574730473462,
                    "high": 0.29380076410930844,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.20279720279720279,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.16173021566427545,
                        "high": 0.24489910570969958,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3613053613053613,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.31279568657744744,
                        "high": 0.41067285382830626,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5612903225806452,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": {
                        "low": 0.47770700636942676,
                        "high": 0.6456733923884513,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1585081585081585,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.12284904521746627,
                        "high": 0.19529499188210314,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.5407925407925408,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 0.27845000459015645,
                        "high": 0.8170738306409038,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.5314685314685317,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 1.3355511875511876,
                        "high": 1.7330509169404138,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.24428938270401684,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.18414097253697911,
                        "high": 0.2530281419835335,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3230924093827319,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2435412862585853,
                        "high": 0.3346501232685443,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.19813519813519814,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.1592479431708823,
                        "high": 0.2375588235294117,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3628808864265928,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3103448275862069,
                        "high": 0.41690705003501666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.631578947368421,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5782122905027933,
                        "high": 0.6840695488721804,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5745614035087719,
                      "unit": "proportion",
                      "denominator": 228,
                      "interval": {
                        "low": 0.5064935064935064,
                        "high": 0.6431785765869863,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.26869806094182824,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.22126436781609196,
                        "high": 0.3172062048798318,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.7299168975069252,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 0.34613874345549744,
                        "high": 1.1223283565823887,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.5027700831024937,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.2835161339787478,
                        "high": 2.724413619378087,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.25608055726816586,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.21684344650954213,
                        "high": 0.2680882591219495,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.35590857111846774,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3013756375217365,
                        "high": 0.37259724149152307,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3545706371191136,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3019929208987381,
                        "high": 0.4069767441860465,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3253333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.2761394101876676,
                        "high": 0.3763164968873797,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.512,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.4576247968423497,
                        "high": 0.5646437994722955,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6354166666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": {
                        "low": 0.564591372125328,
                        "high": 0.7055214723926381,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18666666666666668,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.14758165436115628,
                        "high": 0.2281556816627923,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.0760000000000003,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 0.7533688205921247,
                        "high": 1.4008179660894136,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.2226666666666666,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 1.9975111750890218,
                        "high": 2.4471649484536084,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2765855604090898,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.22849024069612303,
                        "high": 0.28288112509819346,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3616888097657328,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.298794930141084,
                        "high": 0.36992147128225294,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.22933333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.18591420249244156,
                        "high": 0.27520435967302453,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2375,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.18375850340136055,
                        "high": 0.29338842975206614,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5125,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.44680633065922565,
                        "high": 0.5757575757575758,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4634146341463415,
                      "unit": "proportion",
                      "denominator": 123,
                      "interval": {
                        "low": 0.37189534118335404,
                        "high": 0.5564533660589059,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.275,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2155154892066162,
                        "high": 0.3359691699604743,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.13958333333333336,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": -0.26794018579888085,
                        "high": 0.5576962809917355,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.8687500000000001,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 1.6193415637860082,
                        "high": 2.123377314175187,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.18365873608592054,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13449501550229706,
                        "high": 0.19284751647425247,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2394537951499977,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.17535426071818475,
                        "high": 0.2514341037575697,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3458333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2833315450643777,
                        "high": 0.4085636948062933,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22413793103448276,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.15950920245398773,
                        "high": 0.29081864564007415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.39080459770114945,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3149171270718232,
                        "high": 0.4662585303646515,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5735294117647058,
                      "unit": "proportion",
                      "denominator": 68,
                      "interval": {
                        "low": 0.44926142697881827,
                        "high": 0.691381766381766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16666666666666666,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.11299435028248588,
                        "high": 0.2236024844720497,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.3333333333333333,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -0.14534986805211944,
                        "high": 0.7967142857142855,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.5977011494252873,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.2959708788898237,
                        "high": 1.9043779020183835,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.18976279650436953,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1136609270878934,
                        "high": 0.18685092593387373,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.21933621933621933,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13137431832237026,
                        "high": 0.21597055075473715,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.25287356321839083,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.18888364779874214,
                        "high": 0.31976744186046513,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.24897959183673468,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.19343028145745098,
                        "high": 0.3080168776371308,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.40816326530612246,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.34321723636621515,
                        "high": 0.47388097164778514,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.61,
                      "unit": "proportion",
                      "denominator": 100,
                      "interval": {
                        "low": 0.5094295946890287,
                        "high": 0.7096774193548387,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15918367346938775,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.11290165683444207,
                        "high": 0.20779348102877515,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.8816326530612246,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 0.5365200528139957,
                        "high": 1.2425657505910166,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.8,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 1.5376958020050129,
                        "high": 2.0705887965873373,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.27746545691751173,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19466994371321317,
                        "high": 0.2806427039474985,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3433047178809891,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24086281171295867,
                        "high": 0.34723588793504045,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1836734693877551,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.13461322869955156,
                        "high": 0.23577235772357724,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.46923076923076923,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.38297456781914896,
                        "high": 0.5581395348837209,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7076923076923077,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6260851891127608,
                        "high": 0.7867647058823529,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6630434782608695,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": {
                        "low": 0.5652173913043478,
                        "high": 0.7592681623931623,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.23846153846153847,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.16778523489932887,
                        "high": 0.31297709923664124,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.4423076923076925,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 0.8418699186991871,
                        "high": 2.059531440781441,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0192307692307696,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.6694182338777903,
                        "high": 3.375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.39531404268246373,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2764176065162907,
                        "high": 0.3891403382192857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5121113734750098,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.35808644480519486,
                        "high": 0.5041136199658927,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3153846153846154,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.23577235772357724,
                        "high": 0.3969465648854962,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1588785046728972,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.09174120795107034,
                        "high": 0.23232323232323232,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.32710280373831774,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.2389279045222051,
                        "high": 0.41818181818181815,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4857142857142857,
                      "unit": "proportion",
                      "denominator": 35,
                      "interval": {
                        "low": 0.3142857142857143,
                        "high": 0.6538461538461539,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16822429906542055,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.10084033613445378,
                        "high": 0.2413793103448276,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.2897196261682243,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": -0.20002136752136754,
                        "high": 0.7833376288659794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.3177570093457946,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 0.9826086956521739,
                        "high": 1.675,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.15526932084309134,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.07467896174863388,
                        "high": 0.16159250585480095,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.18571428571428572,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0893218954248366,
                        "high": 0.19327731092436973,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.205607476635514,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.13157894736842105,
                        "high": 0.2868870277975766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3007518796992481,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2222222222222222,
                        "high": 0.3821138211382114,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6616541353383458,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5781163194444444,
                        "high": 0.7416666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.45454545454545453,
                      "unit": "proportion",
                      "denominator": 88,
                      "interval": {
                        "low": 0.34615384615384615,
                        "high": 0.5643564356435643,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3609022556390977,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2773109243697479,
                        "high": 0.44696969696969696,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.018796992481203006,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": -0.5985568010659252,
                        "high": 0.6679961499550355,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.31203007518797,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.9830405484934088,
                        "high": 2.6500093984962407,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.21611721611721613,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.14600068232421173,
                        "high": 0.22378513831568128,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.28668610301263364,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19367437451170944,
                        "high": 0.2968578365412099,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.45864661654135336,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.3700787401574803,
                        "high": 0.5447761194029851,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.05063291139240506,
                        "high": 0.19480519480519481,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.2631578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.16666666666666666,
                        "high": 0.3625,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.45,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": {
                        "low": 0.22727272727272727,
                        "high": 0.6666666666666666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.07017543859649122,
                        "high": 0.22727272727272727,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.20394736842105263,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.7957815974796146,
                        "high": 0.391304347826087,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 0.9802631578947368,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6296296296296298,
                        "high": 1.3571669884169884,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.16049382716049382,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05555555555555555,
                        "high": 0.14814814814814814,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.18840579710144925,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.06521739130434782,
                        "high": 0.17391304347826086,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14544721407624633,
                        "high": 0.3333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.30612244897959184,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.21176470588235294,
                        "high": 0.40426129572077446,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.4897959183673469,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.38823529411764707,
                        "high": 0.59,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.625,
                      "unit": "proportion",
                      "denominator": 48,
                      "interval": {
                        "low": 0.4782608695652174,
                        "high": 0.761924939467312,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1836734693877551,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.11008944954128441,
                        "high": 0.2631578947368421,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.75,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 0.06382717980809345,
                        "high": 1.4224856149604412,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.076530612244898,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.6565476190476194,
                        "high": 2.509031308536259,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.24488536155202822,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1366547958214625,
                        "high": 0.24438029563029562,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2874741200828157,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.16042084726867334,
                        "high": 0.28688121660947746,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2653061224489796,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.18018018018018017,
                        "high": 0.35714285714285715,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 790,
                  "population_events": 790
                }
              }
            },
            "bioclip-country": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.43924050632911393,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.40883941743910096,
                    "high": 0.4685585871582707,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 1.0,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.9999999999999999,
                    "high": 1.0000000000000002,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.43924050632911393,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.4088394174391009,
                    "high": 0.4685585871582707,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.27682104545154496,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.25388708203886007,
                    "high": 0.28975491186793273,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": null,
                "wrong_species_rate_all_events": {
                  "value": 0.5607594936708861,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.5314414128417292,
                    "high": 0.591160582560899,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.47304862197415903,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4338576718385583,
                    "high": 0.49515079876165696,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": null,
                "any_rank_conflict_rate": null
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4592074592074592,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.4110552226027397,
                        "high": 0.5072824413923751,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4592074592074592,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.4110552226027397,
                        "high": 0.5072824413923751,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5407925407925408,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.4927175586076249,
                        "high": 0.5889447773972603,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.298983907576611,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2477162346035455,
                        "high": 0.30748940186481716,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.49669907226436993,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4115285832929869,
                        "high": 0.5108291676141319,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4155124653739612,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3628755309325946,
                        "high": 0.46978356528857956,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4155124653739612,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3628755309325946,
                        "high": 0.46978356528857956,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5844875346260388,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5302164347114204,
                        "high": 0.6371244690674053,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.27176729202217,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20811538906329702,
                        "high": 0.2800955122292008,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.39613537481197664,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3033546349058227,
                        "high": 0.4082748144357842,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6026666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5524296675191815,
                        "high": 0.6520547945205479,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6026666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5524296675191815,
                        "high": 0.6520547945205479,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3973333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.34794520547945207,
                        "high": 0.4475703324808184,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.44580929226526594,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.38067166726259244,
                        "high": 0.4522095006908864,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6001278934340117,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.512442629007336,
                        "high": 0.6087435586223469,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.31336257194280176,
                        "high": 0.4393305439330544,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.31336257194280176,
                        "high": 0.4393305439330544,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.625,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5606694560669456,
                        "high": 0.6866374280571982,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.2610571617924559,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19009548329839487,
                        "high": 0.2654312600239458,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3370611456054494,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24543973792957313,
                        "high": 0.34270871547395537,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1724137931034483,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.11666666666666667,
                        "high": 0.23115645431458368,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.1724137931034483,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.11666666666666667,
                        "high": 0.23115645431458368,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8275862068965517,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.7688435456854164,
                        "high": 0.8833333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.10125866241293545,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05421660362698098,
                        "high": 0.10816291556553269,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.1393950417632618,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.07463584395402577,
                        "high": 0.1488995980512528,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6204081632653061,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.5575202410092261,
                        "high": 0.6816326530612244,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6204081632653061,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.5575202410092261,
                        "high": 0.6816326530612244,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3795918367346939,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.3183673469387755,
                        "high": 0.44247975899077385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.4624122364563541,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3679518601190476,
                        "high": 0.46041481782727794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6269996426526834,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.49891777643260693,
                        "high": 0.6242912784098683,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5692307692307692,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4827586206896552,
                        "high": 0.656934306569343,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5692307692307692,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4827586206896552,
                        "high": 0.656934306569343,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4307692307692308,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.34306569343065696,
                        "high": 0.5172413793103449,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.4355060613989185,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.31235756802721093,
                        "high": 0.4413523158389229,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5542804417804418,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.39754599567099574,
                        "high": 0.5617211292495383,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34579439252336447,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.25742574257425743,
                        "high": 0.4347909060736806,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.34579439252336447,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.25742574257425743,
                        "high": 0.4347909060736806,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6542056074766355,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.5652090939263192,
                        "high": 0.7425742574257426,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.2116708908375575,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13088819745069744,
                        "high": 0.2191927208593875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.29882949294714,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.18478333757745521,
                        "high": 0.30944854709560593,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.39849624060150374,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.3115912330719886,
                        "high": 0.4838709677419355,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.39849624060150374,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.3115912330719886,
                        "high": 0.4838709677419355,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6015037593984962,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5161290322580645,
                        "high": 0.6884087669280113,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.29384673706707604,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19188352960598723,
                        "high": 0.29939263662522686,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.35381545891749977,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.23104343360720914,
                        "high": 0.3604931747120078,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.03125,
                        "high": 0.16216216216216217,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.09210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.03125,
                        "high": 0.16216216216216217,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.9078947368421053,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.8378378378378378,
                        "high": 0.96875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.05592020592020591,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.014414414414414415,
                        "high": 0.06516545391545392,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.08995859213250516,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.02318840579710145,
                        "high": 0.10483138238573021,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.23469387755102042,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.15217391304347827,
                        "high": 0.32222222222222224,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.23469387755102042,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.15217391304347827,
                        "high": 0.32222222222222224,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7653061224489796,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.6777777777777778,
                        "high": 0.8478260869565217,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.14702956424267902,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.07613816738816738,
                        "high": 0.16141058167697508,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.19497398736529173,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10096583066691761,
                        "high": 0.2140444670064235,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 790,
                  "population_events": 790
                }
              }
            },
            "speciesnet-country": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.28354430379746837,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.26437227593082024,
                    "high": 0.3049054286079602,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.34810126582278483,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.3270305678849983,
                    "high": 0.37001442430635473,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.8145454545454546,
                  "unit": "proportion",
                  "denominator": 275,
                  "population_events": 790,
                  "interval": {
                    "low": 0.7802622064846542,
                    "high": 0.8515058971710102,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.2996057929727042,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.27668206816782753,
                    "high": 0.31512404802767724,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 1.1753164556962028,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 1.0471834244342153,
                    "high": 1.3128998221363726,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.06455696202531645,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.05072203288975441,
                    "high": 0.07814962463696641,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.3185681849330019,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.29419359146958873,
                    "high": 0.33506860802942895,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 1.979113924050633,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 1.898304604590206,
                    "high": 2.067126918440209,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.16075949367088607,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.14020044126056783,
                    "high": 0.18097017785308925,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21212121212121213,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.17084282460136674,
                        "high": 0.2548555587283457,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.28438228438228436,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.2399976247030879,
                        "high": 0.33049834700954095,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7459016393442623,
                      "unit": "proportion",
                      "denominator": 122,
                      "interval": {
                        "low": 0.6607142857142857,
                        "high": 0.8260869565217391,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07226107226107226,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.04741379310344827,
                        "high": 0.09887640449438202,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.8752913752913752,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 0.620719965409601,
                        "high": 1.134664338013414,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.5396270396270397,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 1.3418575314700236,
                        "high": 1.744530675738047,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3155633928361201,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24246509857023862,
                        "high": 0.32642063674260646,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3359223214061924,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.25810800815541524,
                        "high": 0.34748003266148425,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13286713286713286,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.10048962341946124,
                        "high": 0.16625916870415647,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.31666666666666665,
                        "high": 0.4231807883120434,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.42382271468144045,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3693170332026464,
                        "high": 0.47863475177304965,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.869281045751634,
                      "unit": "proportion",
                      "denominator": 153,
                      "interval": {
                        "low": 0.8092105263157895,
                        "high": 0.9246575342465754,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.055401662049861494,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.03154506340784395,
                        "high": 0.08256948391112554,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.5318559556786706,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 1.1976325290878316,
                        "high": 1.8682243316686251,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.501385041551247,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.279096309641562,
                        "high": 2.7249301575931235,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.34954589419670884,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.29738036619515024,
                        "high": 0.36543287850737854,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.36139490755930914,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.30746105657464684,
                        "high": 0.3778204337110185,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.19390581717451524,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.15193282747045248,
                        "high": 0.23782234957020057,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.33866666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.28895026754800124,
                        "high": 0.39071126048295335,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.4106666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.35883700043230027,
                        "high": 0.46329422661315217,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8246753246753247,
                      "unit": "proportion",
                      "denominator": 154,
                      "interval": {
                        "low": 0.7579580513535032,
                        "high": 0.8875828912466842,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.072,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.04497354497354497,
                        "high": 0.10084033613445378,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.568,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 1.2716515448259325,
                        "high": 1.8679269227085529,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.2346666666666666,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 2.0076918158567776,
                        "high": 2.4622674357491987,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3580125159393452,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2972066855908319,
                        "high": 0.36557131269329507,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3763721321413629,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3124480540826695,
                        "high": 0.3843185594980794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13333333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.09755784865540963,
                        "high": 0.17187594582324453,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.24166666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.18749377490039842,
                        "high": 0.298702645901816,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.32083333333333336,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.25957446808510637,
                        "high": 0.38376795598313124,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7532467532467533,
                      "unit": "proportion",
                      "denominator": 77,
                      "interval": {
                        "low": 0.6385542168674698,
                        "high": 0.8589743589743589,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07916666666666666,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.042550682456844645,
                        "high": 0.12174187066130797,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.8291666666666668,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 0.4566050111504657,
                        "high": 1.2069019621988168,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.8708333333333333,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 1.624987449799197,
                        "high": 2.1270955504115228,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.23777415999638224,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.17657195428028763,
                        "high": 0.24916219068996848,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.24379375898363242,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1810421303380164,
                        "high": 0.25547009425173983,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.20833333333333334,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.15596155858652808,
                        "high": 0.26337448559670784,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22413793103448276,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.15950920245398773,
                        "high": 0.29081864564007415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.25287356321839083,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.18681318681318682,
                        "high": 0.32085891595695515,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8863636363636364,
                      "unit": "proportion",
                      "denominator": 44,
                      "interval": {
                        "low": 0.78,
                        "high": 0.9736842105263158,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.028735632183908046,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.005952380952380952,
                        "high": 0.056179775280898875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.8103448275862069,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.37878509358288776,
                        "high": 1.2380952380952381,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.5862068965517242,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.2837701140446807,
                        "high": 1.8932630318659056,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2171589310829817,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.129957805907173,
                        "high": 0.21377347160575008,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2227994227994228,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13333333333333333,
                        "high": 0.21932602930979556,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.15517241379310345,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.10344827586206896,
                        "high": 0.21142857142857144,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2653061224489796,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.20832908163265307,
                        "high": 0.3254014396456256,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3551020408163265,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.2923680860710051,
                        "high": 0.4184123458610583,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7471264367816092,
                      "unit": "proportion",
                      "denominator": 87,
                      "interval": {
                        "low": 0.6451612903225806,
                        "high": 0.845360824742268,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08979591836734693,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.052862988431646434,
                        "high": 0.13008130081300814,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.189795918367347,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 0.849994635193133,
                        "high": 1.5381638687101344,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.822448979591837,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 1.5576900921658987,
                        "high": 2.10148708230198,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.34928537071394217,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2458757112422526,
                        "high": 0.34844482237339375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3729657348301416,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.26254525098749004,
                        "high": 0.3720682001614205,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.12653061224489795,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.08366533864541832,
                        "high": 0.17269076305220885,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.47692307692307695,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.39090198863636366,
                        "high": 0.5666666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5153846153846153,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4271715210355987,
                        "high": 0.6043205733920326,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9253731343283582,
                      "unit": "proportion",
                      "denominator": 67,
                      "interval": {
                        "low": 0.8571428571428571,
                        "high": 0.9838709677419355,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.038461538461538464,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.008064516129032258,
                        "high": 0.07518796992481203,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2807692307692307,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 1.7434601449275364,
                        "high": 2.792614734299517,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0115384615384615,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.6638655462184873,
                        "high": 3.367660427807487,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5118422318422318,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3574269680936347,
                        "high": 0.5041812169312169,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5234750098386461,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.365550308277581,
                        "high": 0.5156398809523809,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.14615384615384616,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.08526719445818902,
                        "high": 0.21311475409836064,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1588785046728972,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.09174120795107034,
                        "high": 0.23232323232323232,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.21495327102803738,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.14,
                        "high": 0.29565882874889077,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7391304347826086,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": {
                        "low": 0.5416666666666666,
                        "high": 0.9090909090909091,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.056074766355140186,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.018018018018018018,
                        "high": 0.10281096444744184,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.6962616822429908,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 0.2340248226950355,
                        "high": 1.1434792673107892,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.3037383177570094,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 0.9714285714285714,
                        "high": 1.655970470183486,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.18214285714285713,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08760416666666666,
                        "high": 0.18956043956043958,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.18571428571428572,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0893218954248366,
                        "high": 0.19327731092436973,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.12149532710280374,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.06306306306306306,
                        "high": 0.18691588785046728,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3082706766917293,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.22949965846994536,
                        "high": 0.3902480788506515,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.40601503759398494,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.32025490196078427,
                        "high": 0.49296019170579025,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7592592592592593,
                      "unit": "proportion",
                      "denominator": 54,
                      "interval": {
                        "low": 0.6271123197571464,
                        "high": 0.8805970149253731,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09774436090225563,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.045454545454545456,
                        "high": 0.15909452808338154,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.9360902255639098,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 0.3780420054200542,
                        "high": 1.5069121078558463,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.327067669172932,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.9959652426019727,
                        "high": 2.6723097178060415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2878095238095238,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19799642857142855,
                        "high": 0.2964254604872252,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2936831875607386,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20203717201166183,
                        "high": 0.30247495968084204,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2781954887218045,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2,
                        "high": 0.358974358974359,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.05063291139240506,
                        "high": 0.19480519480519481,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.15789473684210525,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.08,
                        "high": 0.24324324324324326,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 12,
                      "interval": {
                        "low": 0.4666666666666667,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.039473684210526314,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.0,
                        "high": 0.08823529411764706,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.11842105263157894,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.45394736842105265,
                        "high": 0.6866686746987951,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 0.9736842105263158,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6216216216216216,
                        "high": 1.3513513513513513,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.18055555555555555,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0625,
                        "high": 0.16666666666666666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.18840579710144925,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.06521739130434782,
                        "high": 0.17391304347826086,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.17105263157894737,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.09090909090909091,
                        "high": 0.25758797653958926,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.30612244897959184,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.21176470588235294,
                        "high": 0.40426129572077446,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.32653061224489793,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.23076923076923078,
                        "high": 0.425531914893617,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": {
                        "low": 0.8387096774193549,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.02040816326530612,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.0,
                        "high": 0.05263157894736842,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.3469387755102042,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 0.7403750000000001,
                        "high": 1.9255417651694247,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.061224489795918,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.641282975085491,
                        "high": 2.495499412455934,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2932712215320911,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.16500862663906143,
                        "high": 0.2916230176828003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2932712215320911,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.16500862663906143,
                        "high": 0.2916230176828003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.07692307692307693,
                        "high": 0.21505376344086022,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 790,
                  "population_events": 790
                }
              }
            },
            "gemini-minimal": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4531645569620253,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.4263783860302847,
                    "high": 0.4795359848484848,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.610126582278481,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5833888957385793,
                    "high": 0.6359636372769284,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.7427385892116183,
                  "unit": "proportion",
                  "denominator": 482,
                  "population_events": 790,
                  "interval": {
                    "low": 0.7097350943188843,
                    "high": 0.7751527001066967,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.384620059106388,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.3583504031464412,
                    "high": 0.39855282226614575,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 1.8594936708860759,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 1.6622813978391506,
                    "high": 2.051830783053093,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1569620253164557,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.13510063336803843,
                    "high": 0.1789345991561181,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.5160724843705965,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.48082459156357926,
                    "high": 0.5347670779773601,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.09367088607595,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 2.9782408967614664,
                    "high": 3.1986251512200887,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.2468354430379747,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.2210091385407841,
                    "high": 0.27222614410905555,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5384615384615384,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.48853078890626983,
                        "high": 0.5871559633027523,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6806526806526807,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.6353699270862545,
                        "high": 0.7245667370567246,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.791095890410959,
                      "unit": "proportion",
                      "denominator": 292,
                      "interval": {
                        "low": 0.743201123618756,
                        "high": 0.8371215205803968,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14219114219114218,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.1100244498777506,
                        "high": 0.17592875577288777,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.312354312354312,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 1.9736765960038989,
                        "high": 2.6432040333213473,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.384615384615385,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.201609435682016,
                        "high": 3.5649213964452002,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4429138522252915,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37837072021640095,
                        "high": 0.45170270382921246,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5643579729967425,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4821175305983174,
                        "high": 0.57555667100819,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21445221445221446,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.1755536858974359,
                        "high": 0.25547772577002387,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3518005540166205,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.30057762875884914,
                        "high": 0.40298507462686567,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5263157894736842,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.47246156617823687,
                        "high": 0.5795454545454546,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6684210526315789,
                      "unit": "proportion",
                      "denominator": 190,
                      "interval": {
                        "low": 0.599009900990099,
                        "high": 0.736559795436664,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1745152354570637,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.1352112676056338,
                        "high": 0.21487603305785125,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.3213296398891967,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 0.9622798503929553,
                        "high": 1.6893758066829196,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.7479224376731306,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.5482899912587413,
                        "high": 2.9529590313669583,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.28475413397310384,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.21815943209231833,
                        "high": 0.29202158641477943,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3716282765411694,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.28471654696794085,
                        "high": 0.3811129178633562,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2853185595567867,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.2371967654986523,
                        "high": 0.33246897163120537,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6293333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5785090462363189,
                        "high": 0.680628272251309,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7893333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.745945945945946,
                        "high": 0.8310258742367175,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7972972972972973,
                      "unit": "proportion",
                      "denominator": 296,
                      "interval": {
                        "low": 0.7507984700819382,
                        "high": 0.8421094139133322,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.12403049641038227,
                        "high": 0.19726027397260273,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.8226666666666667,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 2.4706623913927337,
                        "high": 3.1644222445862926,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.916,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.7564086294416246,
                        "high": 4.072605273273694,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5094660158435669,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4416274509778075,
                        "high": 0.5124802425132557,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6400983275983275,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.554865258920835,
                        "high": 0.64388543290127,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21866666666666668,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.17595123712403662,
                        "high": 0.26285787923416787,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4125,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.35019455252918286,
                        "high": 0.4756583180715735,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5791666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5145203279604212,
                        "high": 0.6428571428571429,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7122302158273381,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": {
                        "low": 0.6381459763539283,
                        "high": 0.7862595419847328,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16666666666666666,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.12083333333333333,
                        "high": 0.21428571428571427,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.666666666666667,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 1.2301170657883025,
                        "high": 2.0983800876201246,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.020833333333334,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.786035558863949,
                        "high": 3.256637168141593,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4520707900018244,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.32700141493244944,
                        "high": 0.4337621462190427,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.49785011050833833,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.36011548226738105,
                        "high": 0.47768742684881915,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2708333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.21518987341772153,
                        "high": 0.32644628099173556,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.12643678160919541,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.07975460122699386,
                        "high": 0.1783783783783784,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.26436781609195403,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.2010575192276223,
                        "high": 0.3333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4782608695652174,
                      "unit": "proportion",
                      "denominator": 46,
                      "interval": {
                        "low": 0.3333333333333333,
                        "high": 0.6222222222222222,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13793103448275862,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.08928571428571429,
                        "high": 0.19047619047619047,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.031609195402298854,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -0.40238122171945706,
                        "high": 0.46428571428571436,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.4109195402298853,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.1697477650063859,
                        "high": 1.6630449314112998,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.12717480931766648,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.06375661375661375,
                        "high": 0.12991307634164775,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.1387361556192725,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.06955266955266956,
                        "high": 0.1417233560090703,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.27586206896551724,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.21052631578947367,
                        "high": 0.3423913043478261,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.689795918367347,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.6282051282051282,
                        "high": 0.75,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8448979591836735,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.7972350230414746,
                        "high": 0.8906882591093117,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8164251207729468,
                      "unit": "proportion",
                      "denominator": 207,
                      "interval": {
                        "low": 0.7625539241655945,
                        "high": 0.8679245283018868,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15510204081632653,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.1111111111111111,
                        "high": 0.2008201008299609,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.148979591836735,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 2.7312179691163667,
                        "high": 3.5514410856935372,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.108163265306122,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.923828125,
                        "high": 4.286766130870954,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5430831128747795,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.44114724877745715,
                        "high": 0.5395518275864477,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.662745493677697,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5383491849487613,
                        "high": 0.6584361285800718,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.19183673469387755,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.1405605894652992,
                        "high": 0.2446351931330472,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5153846153846153,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4295774647887324,
                        "high": 0.6050420168067226,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6846153846153846,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6027397260273972,
                        "high": 0.7647058823529411,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7528089887640449,
                      "unit": "proportion",
                      "denominator": 89,
                      "interval": {
                        "low": 0.6623332748332749,
                        "high": 0.8421052631578947,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16923076923076924,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.10686240770867227,
                        "high": 0.23448803659394785,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2076923076923083,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 1.5952193475815524,
                        "high": 2.8177974313493626,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.5538461538461537,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.2703502415458936,
                        "high": 3.8396281692216983,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.42257495590828925,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.30799402757736094,
                        "high": 0.4257711372989151,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5186147186147186,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37799267020857935,
                        "high": 0.5225373048668502,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2692307692307692,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.19117412888722368,
                        "high": 0.34814814814814815,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5046728971962616,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.4090909090909091,
                        "high": 0.5979381443298969,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6355140186915887,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.5416666666666666,
                        "high": 0.7254901960784313,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7941176470588235,
                      "unit": "proportion",
                      "denominator": 68,
                      "interval": {
                        "low": 0.6935483870967742,
                        "high": 0.8852459016393442,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1308411214953271,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.07079646017699115,
                        "high": 0.2,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.032710280373832,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 1.3715448357502222,
                        "high": 2.6698160377358495,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.2009345794392523,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.8246252291175704,
                        "high": 3.559833333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5196428571428571,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3259757653061225,
                        "high": 0.4816921768707482,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5705882352941177,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3579341736694677,
                        "high": 0.5289169000933706,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2336448598130841,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.15730337078651685,
                        "high": 0.315219839404622,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3383458646616541,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.25862068965517243,
                        "high": 0.4198473282442748,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5338345864661654,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.4473684210526316,
                        "high": 0.6197183098591549,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6338028169014085,
                      "unit": "proportion",
                      "denominator": 71,
                      "interval": {
                        "low": 0.5185185185185185,
                        "high": 0.746031746031746,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19548872180451127,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.12799548872180452,
                        "high": 0.26515151515151514,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.3721804511278197,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 0.7933870523415978,
                        "high": 1.943609022556391,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.8759398496240607,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 2.577512289657781,
                        "high": 3.1796953125,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3334750566893424,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20898224914965985,
                        "high": 0.3308555581323438,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.38111435050210557,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.23883685617103984,
                        "high": 0.3781206378655358,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3007518796992481,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2222222222222222,
                        "high": 0.3800041322314049,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.03076923076923077,
                        "high": 0.16304347826086957,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.21052631578947367,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.12194297956493079,
                        "high": 0.3055555555555556,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4375,
                      "unit": "proportion",
                      "denominator": 16,
                      "interval": {
                        "low": 0.19047619047619047,
                        "high": 0.6923076923076923,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.05063291139240506,
                        "high": 0.1969750320102432,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.02631578947368421,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.685513440860215,
                        "high": 0.6214333976833977,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.2894736842105265,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.9615259740259741,
                        "high": 1.6353046218487393,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.10690943043884221,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.027450980392156862,
                        "high": 0.1134453781512605,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.11853002070393373,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.03043478260869565,
                        "high": 0.12577639751552794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2631578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.16666666666666666,
                        "high": 0.36619718309859156,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.15306122448979592,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.08602150537634409,
                        "high": 0.22916666666666666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.30612244897959184,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.21739130434782608,
                        "high": 0.40404040404040403,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 30,
                      "interval": {
                        "low": 0.32,
                        "high": 0.6785714285714286,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15306122448979592,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.08510136491368929,
                        "high": 0.2289156626506024,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.07653061224489797,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -0.5284131493506494,
                        "high": 0.678954816285998,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.5051020408163265,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.161454133064516,
                        "high": 1.8722327044025153,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.12275641025641025,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.06046932234432234,
                        "high": 0.14026556776556773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.13876811594202898,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.06835662525879917,
                        "high": 0.1585610766045548,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.2,
                        "high": 0.37777777777777777,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "saved_event_telemetry",
                "coverage": {
                  "observed_events": 779,
                  "estimated_events": 11,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "779 saved telemetry estimates plus corrected selected-response result tokens at retained prices; no provider invoice. Earlier replaced calls are separate.",
                "value": 0.0017474293114182412
              }
            },
            "gemini-low": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.46582278481012657,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.4380837982341147,
                    "high": 0.49218202595576016,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.6278481012658228,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6001950563501196,
                    "high": 0.6533056386300059,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.7419354838709677,
                  "unit": "proportion",
                  "denominator": 496,
                  "population_events": 790,
                  "interval": {
                    "low": 0.7082568164755375,
                    "high": 0.7739269137725746,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.38942685947078304,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.36217189284084783,
                    "high": 0.40367993488144427,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 1.8443037974683545,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 1.6361009507370263,
                    "high": 2.0410591856280145,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1620253164556962,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.14017642255775167,
                    "high": 0.18462976281805393,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.52745156915663,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4905366143540598,
                    "high": 0.5467563674976524,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.1544303797468354,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.0410697618134326,
                    "high": 3.2579031385281385,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.2620253164556962,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.23668882102426408,
                    "high": 0.2884082057183323,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5501165501165501,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.5,
                        "high": 0.598086689967989,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6993006993006993,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.6536547924689774,
                        "high": 0.743878874905425,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7866666666666666,
                      "unit": "proportion",
                      "denominator": 300,
                      "interval": {
                        "low": 0.738093336376084,
                        "high": 0.8327759197324415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14918414918414918,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.11566265060240964,
                        "high": 0.1848341232227488,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.3578088578088576,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 2.015617629716981,
                        "high": 2.6878038069687973,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.453379953379953,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.2727143300890456,
                        "high": 3.631114173228346,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.41656594752569986,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3575201952296675,
                        "high": 0.42504995181813016,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5643796708412706,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.48438219998858173,
                        "high": 0.5758741282697247,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2191142191142191,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.18013747359434346,
                        "high": 0.26054884574224124,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3656509695290859,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3131868131868132,
                        "high": 0.418283937486606,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5429362880886427,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.4888165452130524,
                        "high": 0.5966862119318952,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.673469387755102,
                      "unit": "proportion",
                      "denominator": 196,
                      "interval": {
                        "low": 0.6038624555646706,
                        "high": 0.7417582417582418,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1772853185595568,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.13662790697674418,
                        "high": 0.21846448087431689,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.234072022160665,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 0.854408613445378,
                        "high": 1.6187522644927537,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.799168975069252,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.592685761102194,
                        "high": 3.001535218413762,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.29717896697986373,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.22896800683550192,
                        "high": 0.3051930029136122,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3727329416357613,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.28718021196317195,
                        "high": 0.38278444433232717,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.31301939058171746,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.2631542898341745,
                        "high": 0.36193029490616624,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6426666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5922077922077922,
                        "high": 0.6926832340111814,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7973333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.7542833200319234,
                        "high": 0.8388906926406926,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8060200668896321,
                      "unit": "proportion",
                      "denominator": 299,
                      "interval": {
                        "low": 0.7593220338983051,
                        "high": 0.8491803278688524,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15466666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.11968028277868704,
                        "high": 0.19200263157894734,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.7933333333333334,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 2.437844279661017,
                        "high": 3.140320568835269,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.94,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.7804500125989784,
                        "high": 4.0960092178770955,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5238805312334723,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.454377981352297,
                        "high": 0.5283396577396728,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6514924555082925,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5650597973227283,
                        "high": 0.6570377794967727,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.22933333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.18701245908562983,
                        "high": 0.2727327624066754,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4041666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.34156378600823045,
                        "high": 0.4669421487603306,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5377299528301887,
                        "high": 0.6625514403292181,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6736111111111112,
                      "unit": "proportion",
                      "denominator": 144,
                      "interval": {
                        "low": 0.5961506622516556,
                        "high": 0.75,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19583333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.14601532530669606,
                        "high": 0.24778857252789532,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.5291666666666668,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 1.0787335369513789,
                        "high": 1.982075475874281,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0708333333333333,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.8427887206169284,
                        "high": 3.304624328272081,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.41970945720945724,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.30556971910665087,
                        "high": 0.40520989269568813,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4675244586636991,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.34038145925804153,
                        "high": 0.4513730450281083,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.30833333333333335,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.25,
                        "high": 0.36681222707423583,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.16666666666666666,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.11309271589991929,
                        "high": 0.2251655629139073,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.2988505747126437,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.23214285714285715,
                        "high": 0.3697954299132948,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5576923076923077,
                      "unit": "proportion",
                      "denominator": 52,
                      "interval": {
                        "low": 0.42371251629726203,
                        "high": 0.693556616194865,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13218390804597702,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.08333333333333333,
                        "high": 0.18391157182145118,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.21551724137931033,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -0.24450844854070658,
                        "high": 0.6833477011494251,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.5660919540229887,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.3060077145612343,
                        "high": 1.836436225551394,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1611013129531648,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08885768038545817,
                        "high": 0.16342025530751023,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.16947021232735518,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09347366378210534,
                        "high": 0.17190961921958867,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.27011494252873564,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.20555555555555555,
                        "high": 0.33689984887235525,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7061224489795919,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.6459123702983139,
                        "high": 0.7647058823529411,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8408163265306122,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.7890295358649789,
                        "high": 0.8870967741935484,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8398058252427184,
                      "unit": "proportion",
                      "denominator": 206,
                      "interval": {
                        "low": 0.7887323943661971,
                        "high": 0.8883495145631068,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1346938775510204,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.0931174089068826,
                        "high": 0.17857142857142858,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.2306122448979595,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 2.8220330508474576,
                        "high": 3.620264076822348,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.148979591836735,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.9650643790321136,
                        "high": 4.326447437399778,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5688163971718239,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4616772500720305,
                        "high": 0.5615022397967091,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6845078677830421,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5555777077137993,
                        "high": 0.6757060851790907,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1836734693877551,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.13692946058091288,
                        "high": 0.23423423423423423,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5230769230769231,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4351145038167939,
                        "high": 0.6120689655172413,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7153846153846154,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6349206349206349,
                        "high": 0.7948717948717948,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7311827956989247,
                      "unit": "proportion",
                      "denominator": 93,
                      "interval": {
                        "low": 0.6363636363636364,
                        "high": 0.8192771084337349,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.12666455696202533,
                        "high": 0.2631578947368421,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.9692307692307693,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 1.3157793522267207,
                        "high": 2.6260523128170186,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.546153846153846,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.2584710451977403,
                        "high": 3.8387130098640365,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4262057387057387,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3120593295593296,
                        "high": 0.43502816627816626,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5036976911976911,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.36879738947920765,
                        "high": 0.5141241965105601,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3153846153846154,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.23529411764705882,
                        "high": 0.3969465648854962,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4953271028037383,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.40186915887850466,
                        "high": 0.5887850467289719,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7009345794392523,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.6105263157894737,
                        "high": 0.7857142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7066666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": {
                        "low": 0.6025641025641025,
                        "high": 0.8076923076923077,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.205607476635514,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.13043478260869565,
                        "high": 0.2833333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.925233644859813,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 1.2473184697855753,
                        "high": 2.5840718302869905,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.3271028037383177,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.9770075031928482,
                        "high": 3.6584158415841586,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.46988304093567246,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2997446267183109,
                        "high": 0.4413140490114174,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5251633986928105,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.335008700449877,
                        "high": 0.4932333488951136,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2803738317757009,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.19801980198019803,
                        "high": 0.36666666666666664,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3308270676691729,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.25,
                        "high": 0.4146341463414634,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.518796992481203,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.4297520661157025,
                        "high": 0.6083347902097901,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6376811594202898,
                      "unit": "proportion",
                      "denominator": 69,
                      "interval": {
                        "low": 0.5147058823529411,
                        "high": 0.7534246575342466,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18796992481203006,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.12194297956493079,
                        "high": 0.2597402597402597,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.2105263157894737,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5952193475815524,
                        "high": 1.8134343165126747,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.8646616541353382,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 2.563380281690141,
                        "high": 3.1614017125837677,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3162954217499672,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20259188523894406,
                        "high": 0.3204784244276222,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3550254733928203,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.22739905486003925,
                        "high": 0.3597206804799842,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3308270676691729,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2517482517482518,
                        "high": 0.4132231404958678,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.05061708860759495,
                        "high": 0.19444444444444445,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14492276887871855,
                        "high": 0.33766233766233766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 18,
                      "interval": {
                        "low": 0.2631578947368421,
                        "high": 0.7333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.05128205128205128,
                        "high": 0.19753086419753085,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.11842105263157894,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.5571635610766046,
                        "high": 0.76,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.368421052631579,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.0208261986301368,
                        "high": 1.7333544303797468,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.13647798742138365,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0440251572327044,
                        "high": 0.13836477987421383,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.15724637681159417,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05072463768115942,
                        "high": 0.15942028985507245,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.15714285714285714,
                        "high": 0.35366322701688546,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.20408163265306123,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.12675610625779996,
                        "high": 0.2882882882882883,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3469387755102041,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.25263157894736843,
                        "high": 0.44554455445544555,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5882352941176471,
                      "unit": "proportion",
                      "denominator": 34,
                      "interval": {
                        "low": 0.42105263157894735,
                        "high": 0.75,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.07692307692307693,
                        "high": 0.2127708825847123,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.29081632653061223,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -0.36046791144901613,
                        "high": 0.9495527522935778,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.7193877551020411,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.3490541470125788,
                        "high": 2.1087010331467932,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1511175898931001,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08830911152339725,
                        "high": 0.17303759166004062,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.16097308488612835,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09406840140535792,
                        "high": 0.18432265198569542,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.19801980198019803,
                        "high": 0.375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 790,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.0022434227848101267
              }
            },
            "gemini-medium": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5050632911392405,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.47745408072939727,
                    "high": 0.53048139624089,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.689873417721519,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6645643596909419,
                    "high": 0.7147793372239575,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.7321100917431193,
                  "unit": "proportion",
                  "denominator": 545,
                  "population_events": 790,
                  "interval": {
                    "low": 0.7008204108738859,
                    "high": 0.7612962443950506,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.39410302408254455,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.369012504140656,
                    "high": 0.40790367039280523,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 1.9759493670886077,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 1.7623803135672547,
                    "high": 2.16761077881331,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1848101265822785,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.16303734108718287,
                    "high": 0.208183674623548,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.5587283379398098,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5231569678956136,
                    "high": 0.5782938111897998,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.311392405063291,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.1984617193636176,
                    "high": 3.4113239195755023,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.2670886075949367,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.2428527291976659,
                    "high": 0.2938827847609493,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6013986013986014,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.5514012835827887,
                        "high": 0.6498830935251798,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7808857808857809,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7410698227132579,
                        "high": 0.8198198198198198,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7701492537313432,
                      "unit": "proportion",
                      "denominator": 335,
                      "interval": {
                        "low": 0.7226858731297396,
                        "high": 0.8164585049809003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1794871794871795,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.14285714285714285,
                        "high": 0.21695819413729825,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.375291375291375,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 2.019091143288857,
                        "high": 2.728666281755196,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.6107226107226107,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.4311163895486936,
                        "high": 3.788964060133415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4654919191330071,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3976692922665775,
                        "high": 0.4713525415265455,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6306664710834289,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5387777508127825,
                        "high": 0.6386066691649969,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.24708624708624707,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.205377349019052,
                        "high": 0.2885106086214535,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.39058171745152354,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.338623297087864,
                        "high": 0.44385026737967914,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5817174515235457,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5283018867924528,
                        "high": 0.6332475813544415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6714285714285714,
                      "unit": "proportion",
                      "denominator": 210,
                      "interval": {
                        "low": 0.6048780487804878,
                        "high": 0.7355795704845816,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19113573407202217,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.15056818181818182,
                        "high": 0.23351648351648352,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.5013850415512469,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 1.118985631200821,
                        "high": 1.8689475743348982,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.9556786703601112,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.7528397187658715,
                        "high": 3.154173259493671,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2888468366428893,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.22631365892797076,
                        "high": 0.29702985247806746,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.39165672765137527,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3068659782074179,
                        "high": 0.4027523423431422,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.29085872576177285,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.24331550802139038,
                        "high": 0.3408450704225352,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6826666666666666,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.6322713627765442,
                        "high": 0.7336818168918152,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.856,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.8179271708683473,
                        "high": 0.8922662558436039,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7975077881619937,
                      "unit": "proportion",
                      "denominator": 321,
                      "interval": {
                        "low": 0.7507148093841642,
                        "high": 0.8433734939759037,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17333333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.13350785340314136,
                        "high": 0.21448467966573817,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.9173333333333336,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 2.559053099131839,
                        "high": 3.2723582434034064,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.0906666666666665,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.9360546552904845,
                        "high": 4.243351063829787,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5561552774710669,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.49417766617328757,
                        "high": 0.5620633601445499,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6773686071762993,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6018830549546451,
                        "high": 0.6845643488940031,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.23466666666666666,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.18994413407821228,
                        "high": 0.27979274611398963,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.44583333333333336,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.38235294117647056,
                        "high": 0.5084033613445378,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6291666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5670961850649351,
                        "high": 0.6901978136387298,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7086092715231788,
                      "unit": "proportion",
                      "denominator": 151,
                      "interval": {
                        "low": 0.6350259685569904,
                        "high": 0.781021897810219,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18333333333333332,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.13565891472868216,
                        "high": 0.23236742010628228,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.8687500000000001,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 1.4132825552628197,
                        "high": 2.3114820468573707,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.2229166666666673,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.997965326581232,
                        "high": 3.4480593270366002,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.428970606687998,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.31851245222169133,
                        "high": 0.41735012252595366,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.49956070652273177,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3709258937265267,
                        "high": 0.4860279907897182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2708333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.21338772663877267,
                        "high": 0.3292181069958848,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.20114942528735633,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.143646408839779,
                        "high": 0.2648648648648649,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.41379310344827586,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3413173652694611,
                        "high": 0.4913294797687861,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4861111111111111,
                      "unit": "proportion",
                      "denominator": 72,
                      "interval": {
                        "low": 0.3723267866884888,
                        "high": 0.6027397260273972,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21264367816091953,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.15300546448087432,
                        "high": 0.2752808988764045,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.07758620689655173,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -0.4342857142857143,
                        "high": 0.5925171604046238,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.7442528735632186,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.4666620111731845,
                        "high": 2.0287397847319113,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1834154351395731,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10519978106185005,
                        "high": 0.18124816514902722,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.20723562152133582,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1188620902906617,
                        "high": 0.2047868878956541,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.26282051282051283,
                        "high": 0.4037274370862858,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7428571428571429,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.6818139257745999,
                        "high": 0.803347280334728,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9102040816326531,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.8710905997983871,
                        "high": 0.9444444444444444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8161434977578476,
                      "unit": "proportion",
                      "denominator": 223,
                      "interval": {
                        "low": 0.7592592592592593,
                        "high": 0.870590770791075,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1673469387755102,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.11764705882352941,
                        "high": 0.21900924832743013,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.1938775510204085,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 2.7397476408596226,
                        "high": 3.63292194092827,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.255102040816326,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.064514336917562,
                        "high": 4.435743893003205,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.605582264957265,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4907888170970279,
                        "high": 0.593548859619793,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7390156453715775,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5989287259489153,
                        "high": 0.7243308117394084,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21224489795918366,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.1582966207966208,
                        "high": 0.2681992337164751,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5692307692307692,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.48333333333333334,
                        "high": 0.6571526418786692,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7538461538461538,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6744186046511628,
                        "high": 0.8321180644173345,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7551020408163265,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.6666666666666666,
                        "high": 0.8387096774193549,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18461538461538463,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.12,
                        "high": 0.2540983606557377,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.396153846153846,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 1.7829299161525076,
                        "high": 2.9803149606299217,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.7807692307692307,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.5090791536050165,
                        "high": 4.050789091012773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.45083943833943835,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.34779137529137527,
                        "high": 0.45883793290043284,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5328102453102453,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.411026170798898,
                        "high": 0.5422630116096026,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.27692307692307694,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.2,
                        "high": 0.3577981651376147,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5700934579439252,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.47572019735795,
                        "high": 0.66087324981577,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7476635514018691,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.6630434782608695,
                        "high": 0.8285714285714286,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7625,
                      "unit": "proportion",
                      "denominator": 80,
                      "interval": {
                        "low": 0.6626506024096386,
                        "high": 0.8536653877774731,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17757009345794392,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.10678578806031812,
                        "high": 0.25555764411027565,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.1588785046728973,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 1.4380803571428573,
                        "high": 2.8539425373794565,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.5140186915887854,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 3.1666666666666665,
                        "high": 3.8482245249017035,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5385204081632653,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.34609539270253564,
                        "high": 0.502640306122449,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5913165266106443,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.38002631355572525,
                        "high": 0.5519187675070027,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.27102803738317754,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.18867924528301888,
                        "high": 0.35714285714285715,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3458646616541353,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2635556915544676,
                        "high": 0.42857142857142855,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5338345864661654,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.4485294117647059,
                        "high": 0.6166666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.647887323943662,
                      "unit": "proportion",
                      "denominator": 71,
                      "interval": {
                        "low": 0.5342465753424658,
                        "high": 0.7567567567567568,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18796992481203006,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.125,
                        "high": 0.2565833926031294,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.635338345864662,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.0679806722689076,
                        "high": 2.184435138987883,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.9887218045112784,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 2.6951219512195124,
                        "high": 3.277372262773723,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.31372074253430193,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20401790300095382,
                        "high": 0.31341736467517023,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.37774538386783285,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2456542097358424,
                        "high": 0.3773800921598988,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2706766917293233,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.19548872180451127,
                        "high": 0.34965909090909075,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18421052631578946,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.10168877425693934,
                        "high": 0.2753623188405797,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.40789473684210525,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2987012987012987,
                        "high": 0.5185185185185185,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.45161290322580644,
                      "unit": "proportion",
                      "denominator": 31,
                      "interval": {
                        "low": 0.27586206896551724,
                        "high": 0.6296296296296297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.13333333333333333,
                        "high": 0.32,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.006578947368421052,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.7619360902255639,
                        "high": 0.7800312499999996,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.6513157894736843,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.25,
                        "high": 2.0641233766233764,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1812121212121212,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.07454329004329004,
                        "high": 0.1777215007215007,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.21666666666666667,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08912784679089027,
                        "high": 0.21249309868875085,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.32894736842105265,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2261904761904762,
                        "high": 0.43478260869565216,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.13592233009708737,
                        "high": 0.3020893895348836,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.41836734693877553,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.31868131868131866,
                        "high": 0.5252525252525253,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5121951219512195,
                      "unit": "proportion",
                      "denominator": 41,
                      "interval": {
                        "low": 0.358974358974359,
                        "high": 0.6666666666666666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.20408163265306123,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.125,
                        "high": 0.28846153846153844,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.1326530612244898,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -0.5543589934484813,
                        "high": 0.8275909961685823,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.816326530612245,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.4471153846153848,
                        "high": 2.2209302325581395,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.15296495956873316,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08513426039369436,
                        "high": 0.17103250915750914,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.17624223602484473,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09808947393186523,
                        "high": 0.19705919533365185,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.336734693877551,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.24443798449612403,
                        "high": 0.4326923076923077,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 790,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.003248393670886076
              }
            },
            "gemini-high": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5139240506329114,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.48570456969191145,
                    "high": 0.5386299776173193,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.7316455696202532,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.7062302566456048,
                    "high": 0.7552009716724906,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.7024221453287197,
                  "unit": "proportion",
                  "denominator": 578,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6708708952491154,
                    "high": 0.7324712048388243,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.3771972779892321,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.3529444026478563,
                    "high": 0.38958075995382013,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 1.8740506329113924,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 1.6584591735620213,
                    "high": 2.0725644858833143,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.21772151898734177,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.19377996739547376,
                    "high": 0.24249152482538558,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.5681832415280837,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5316504293049987,
                    "high": 0.5868368409430961,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.3424050632911397,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.229050540981337,
                    "high": 3.4410532211283797,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.2936708860759494,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.26745416942410616,
                    "high": 0.32023772992127425,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6060606060606061,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.5576459745188398,
                        "high": 0.654320987654321,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8135198135198135,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7748799387214119,
                        "high": 0.8495151249550521,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7449856733524355,
                      "unit": "proportion",
                      "denominator": 349,
                      "interval": {
                        "low": 0.6974063400576369,
                        "high": 0.7916666666666666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.20745920745920746,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.16904761904761906,
                        "high": 0.24708624708624707,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.358974358974359,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 1.9977703682602166,
                        "high": 2.7055291931293306,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.641025641025641,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.461084547009136,
                        "high": 3.8148159067085956,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4296899965310341,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37659909514081835,
                        "high": 0.44071373632432786,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.602952091906451,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5284535689879226,
                        "high": 0.6184208880680085,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2564102564102564,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.21461852125483066,
                        "high": 0.29904306220095694,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.40443213296398894,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.352112676056338,
                        "high": 0.4568245125348189,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6343490304709142,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5826233626609914,
                        "high": 0.685016980080085,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6375545851528385,
                      "unit": "proportion",
                      "denominator": 229,
                      "interval": {
                        "low": 0.5714285714285714,
                        "high": 0.7021276595744681,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2299168975069252,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.18579234972677597,
                        "high": 0.2755681818181818,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.2977839335180055,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 0.8977095651638335,
                        "high": 1.6852459287700254,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.987534626038781,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.7849311242060155,
                        "high": 3.186638303536293,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2868774052367834,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2262036091715442,
                        "high": 0.2938424611300952,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4084356277947424,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3220525961086392,
                        "high": 0.4183519785581017,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3379501385041551,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.28653216194242503,
                        "high": 0.38875035521454954,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6906666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.6416181349327075,
                        "high": 0.7393787036321076,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8746666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.8404255319148937,
                        "high": 0.9076527298558961,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7896341463414634,
                      "unit": "proportion",
                      "denominator": 328,
                      "interval": {
                        "low": 0.743202416918429,
                        "high": 0.833846394984326,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.184,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.14477123606603642,
                        "high": 0.22493224932249323,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.9493333333333336,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 2.6005215453290926,
                        "high": 3.29264715531882,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.1226666666666665,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.9755128232212273,
                        "high": 4.269241192411925,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5346079846079846,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4735553460474888,
                        "high": 0.5418435870400683,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6785409035409036,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6010510161371972,
                        "high": 0.687724552781625,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.23466666666666666,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.19137466307277629,
                        "high": 0.27823691460055094,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.475,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.4126984126984127,
                        "high": 0.5381526104417671,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7041666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.643936799501868,
                        "high": 0.7615070807853235,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6745562130177515,
                      "unit": "proportion",
                      "denominator": 169,
                      "interval": {
                        "low": 0.6031746031746031,
                        "high": 0.7430205757548005,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.22916666666666666,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.17721394398485,
                        "high": 0.28270042194092826,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.7041666666666666,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 1.2284370689655173,
                        "high": 2.173439161298393,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.2875,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 3.0572033898305087,
                        "high": 3.5150230779191953,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.452971341681019,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3382683254267695,
                        "high": 0.43860563197256747,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5332447440042375,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3982146109454374,
                        "high": 0.5163332123221365,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.31666666666666665,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2562999166608792,
                        "high": 0.3782612766650382,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1839080459770115,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.12727272727272726,
                        "high": 0.2441893498942917,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.45977011494252873,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3864995817066369,
                        "high": 0.534893980848153,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 80,
                      "interval": {
                        "low": 0.29333333333333333,
                        "high": 0.5068493150684932,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.27586206896551724,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.21176470588235294,
                        "high": 0.34355828220858897,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.22701149425287356,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -0.7452244321595962,
                        "high": 0.29697601010101005,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.7270114942528736,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.4580775196594764,
                        "high": 2.006373589775613,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.16573100812231245,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09024741034931252,
                        "high": 0.16307320383687074,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.19801626944484085,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10782807470307469,
                        "high": 0.19484071107781958,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.39080459770114945,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3172043010752688,
                        "high": 0.46703296703296704,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7551020408163265,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.6956521739130435,
                        "high": 0.8122275198288923,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9306122448979591,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.8966942148760331,
                        "high": 0.9610894941634242,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8114035087719298,
                      "unit": "proportion",
                      "denominator": 228,
                      "interval": {
                        "low": 0.7567567567567568,
                        "high": 0.863854398319743,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17551020408163265,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.12652940227474388,
                        "high": 0.22608695652173913,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.3000000000000003,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 2.872339468850452,
                        "high": 3.7107553790253633,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.3,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.121716212527863,
                        "high": 4.469433595752096,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5714178464178464,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.47556384306384314,
                        "high": 0.5685184446581505,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7166935700834006,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5964699048597354,
                        "high": 0.713057032283104,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.15017857142857144,
                        "high": 0.252,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5692307692307692,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.48360655737704916,
                        "high": 0.6549356483729966,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7692307692307693,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6949152542372882,
                        "high": 0.8409090909090909,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.74,
                      "unit": "proportion",
                      "denominator": 100,
                      "interval": {
                        "low": 0.6511627906976745,
                        "high": 0.8240766178266178,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.13380281690140844,
                        "high": 0.271189725397451,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2884615384615383,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 1.6850615993595448,
                        "high": 2.8949600082034457,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.7884615384615388,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.5269078947368424,
                        "high": 4.046524050884516,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4380217519106408,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3333241809630698,
                        "high": 0.4464432209040053,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5375721500721501,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.40907967663649475,
                        "high": 0.5479075892912791,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.22033898305084745,
                        "high": 0.3787878787878788,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5794392523364486,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.48514851485148514,
                        "high": 0.6698113207547169,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7850467289719626,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.7064220183486238,
                        "high": 0.8571428571428571,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7380952380952381,
                      "unit": "proportion",
                      "denominator": 84,
                      "interval": {
                        "low": 0.6395187338501293,
                        "high": 0.8300127118644065,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.205607476635514,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.1308411214953271,
                        "high": 0.2857142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.107476635514019,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 1.3818157380254155,
                        "high": 2.8136385918003564,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.55607476635514,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 3.211108562691132,
                        "high": 3.890476190476191,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5098466505246166,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3329161163695062,
                        "high": 0.4811956122973072,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5898225957049487,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3851382522706052,
                        "high": 0.5566772769713946,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2897196261682243,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.20407667921537548,
                        "high": 0.3786407766990291,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.39097744360902253,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.30767716535433076,
                        "high": 0.47580784204671855,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6390977443609023,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5555555555555556,
                        "high": 0.7196969696969697,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.611764705882353,
                      "unit": "proportion",
                      "denominator": 85,
                      "interval": {
                        "low": 0.5046428571428573,
                        "high": 0.7142857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.24812030075187969,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.17518248175182483,
                        "high": 0.32608695652173914,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.3796992481203008,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 0.7605633802816901,
                        "high": 1.9843780281007755,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0714285714285716,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 2.762403483184626,
                        "high": 3.371212121212121,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.35121693121693115,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2299664953102453,
                        "high": 0.34595628908128906,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4300615484288954,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2815916269105045,
                        "high": 0.4236199458138234,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3383458646616541,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.2569444444444444,
                        "high": 0.4227655917498772,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.15789473684210525,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.078125,
                        "high": 0.24358974358974358,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.47368421052631576,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.36,
                        "high": 0.5853658536585366,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 36,
                      "interval": {
                        "low": 0.18181818181818182,
                        "high": 0.48717948717948717,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3157894736842105,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.21621621621621623,
                        "high": 0.4225352112676056,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.35526315789473684,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -1.081085304054054,
                        "high": 0.39103915662650585,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.6184210526315792,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.2272727272727275,
                        "high": 2.019743421052632,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1596045197740113,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.057908931934355666,
                        "high": 0.1556383447061413,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.20471014492753623,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.07427449965493443,
                        "high": 0.1996230942970073,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.39473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2857142857142857,
                        "high": 0.506333223738287,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.20408163265306123,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.12631578947368421,
                        "high": 0.2894798327594687,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.4489795918367347,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.3490566037735849,
                        "high": 0.550561797752809,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.45454545454545453,
                      "unit": "proportion",
                      "denominator": 44,
                      "interval": {
                        "low": 0.3055555555555556,
                        "high": 0.6052631578947368,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.24489795918367346,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.1619047619047619,
                        "high": 0.3333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.12755102040816327,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -0.8724489795918369,
                        "high": 0.6044327731092431,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.8112244897959184,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.4444314641744551,
                        "high": 2.2070801096067054,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.14779183081069872,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08113157282968603,
                        "high": 0.16378640135299402,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.17028189202102245,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09347768173855131,
                        "high": 0.18871041895018875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3877551020408163,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.28888888888888886,
                        "high": 0.49,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 790,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.004065537974683544
              }
            },
            "gemini-medium37": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5569620253164557,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5289582736180838,
                    "high": 0.5833146679270096,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.669620253164557,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6436389208541109,
                    "high": 0.6939144382340586,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.831758034026465,
                  "unit": "proportion",
                  "denominator": 529,
                  "population_events": 790,
                  "interval": {
                    "low": 0.8035102363854232,
                    "high": 0.8593842856522088,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.5158212687832463,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.485934472060884,
                    "high": 0.5317084254079105,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 2.7278481012658227,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 2.535525864782194,
                    "high": 2.9083559706819213,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.11265822784810127,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.0935086880419159,
                    "high": 0.13236580929618905,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.6268207823188815,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5905026495929728,
                    "high": 0.6461266941665748,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.5189873417721524,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.4080927322501697,
                    "high": 3.623406529300516,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.15822784810126583,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.1360821597530458,
                    "high": 0.18104793413337716,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6503496503496503,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.6017073642589098,
                        "high": 0.6977912871355494,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7482517482517482,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7034266356168403,
                        "high": 0.7906976744186046,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8691588785046729,
                      "unit": "proportion",
                      "denominator": 321,
                      "interval": {
                        "low": 0.8292008880447617,
                        "high": 0.9060791360474888,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0979020979020979,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.06994773361353672,
                        "high": 0.12809276865456637,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.1573426573426575,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 2.851803304457175,
                        "high": 3.4441782769806046,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.7983682983682985,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.623852675126468,
                        "high": 3.96946045310016,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5693708656290797,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4885538818334084,
                        "high": 0.5733599868573677,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7071218815070828,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6067524016318138,
                        "high": 0.7120761127099566,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1282051282051282,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.09579187720788802,
                        "high": 0.16312056737588654,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.44598337950138506,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.39322749732905987,
                        "high": 0.498567335243553,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5761772853185596,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5220994475138122,
                        "high": 0.6293103448275862,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7740384615384616,
                      "unit": "proportion",
                      "denominator": 208,
                      "interval": {
                        "low": 0.7162790697674418,
                        "high": 0.8287068318318318,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13019390581717452,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.09710886208476019,
                        "high": 0.1657612328528249,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2174515235457064,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 1.8729848545894077,
                        "high": 2.553152482500938,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.186980609418283,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.9888252463749923,
                        "high": 3.3839804083459364,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3922731053912909,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.31622015203762194,
                        "high": 0.4006330134517594,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.458760072406764,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3698167879762019,
                        "high": 0.4685369140368034,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.19390581717451524,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.15363881401617252,
                        "high": 0.23697916666666666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7386666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.6916395027521984,
                        "high": 0.7849506798187149,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8453333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.8061776354695136,
                        "high": 0.8841322275825756,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8738170347003155,
                      "unit": "proportion",
                      "denominator": 317,
                      "interval": {
                        "low": 0.8360128617363344,
                        "high": 0.909375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10666666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.07646722047597666,
                        "high": 0.13920544447640965,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.6106666666666665,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.3203246099556925,
                        "high": 3.883692024575624,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.264,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 4.122714801904145,
                        "high": 4.401832460732984,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.6917675788264023,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6092684192932306,
                        "high": 0.6934404355584225,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7538492846185153,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6639463543580076,
                        "high": 0.7556722695187938,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13066666666666665,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.09714210723659543,
                        "high": 0.1671018276762402,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5083333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.4435146443514644,
                        "high": 0.572,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6083333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5450368050977807,
                        "high": 0.6709429824561403,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8356164383561644,
                      "unit": "proportion",
                      "denominator": 146,
                      "interval": {
                        "low": 0.7741880247459125,
                        "high": 0.891724048889654,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.06493506493506493,
                        "high": 0.13924310022826308,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.583333333333334,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.1633674461626033,
                        "high": 2.9716656098076517,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.458333333333334,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 3.232063695125236,
                        "high": 3.677928882272103,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5349987267634326,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.403278626602156,
                        "high": 0.5148592693144843,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5756315414543262,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4339073830529527,
                        "high": 0.5539625049586224,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.175,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.12704918032786885,
                        "high": 0.22666801619433194,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22988505747126436,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.16867294760437096,
                        "high": 0.2947383632491633,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3735632183908046,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3016759776536313,
                        "high": 0.449438202247191,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6153846153846154,
                      "unit": "proportion",
                      "denominator": 65,
                      "interval": {
                        "low": 0.5,
                        "high": 0.734375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14367816091954022,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.09259259259259259,
                        "high": 0.19753086419753085,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.0114942528735633,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.5570423302720298,
                        "high": 1.4634932896379524,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.988505747126437,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.7085710466004587,
                        "high": 2.27457044400686,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.23909191583610187,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13719663411523877,
                        "high": 0.22759589247961343,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2670377241805813,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.15323260433650043,
                        "high": 0.2541980097824254,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.19540229885057472,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.13736263736263737,
                        "high": 0.2561599445812808,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.746720363028998,
                        "high": 0.8512412055669913,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8938775510204081,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.8493123505109806,
                        "high": 0.9356223175965666,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8949771689497716,
                      "unit": "proportion",
                      "denominator": 219,
                      "interval": {
                        "low": 0.8530612244897959,
                        "high": 0.9339622641509434,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09387755102040816,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.05859375,
                        "high": 0.13158165475416933,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.8877551020408165,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.555752551020408,
                        "high": 4.199579831932773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.438775510204081,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.283329646017699,
                        "high": 4.586618240385241,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7142061642061642,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5822732846372024,
                        "high": 0.695857211341837,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7989424887729972,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6513565556958536,
                        "high": 0.778416541501038,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.11020408163265306,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.07199924242424242,
                        "high": 0.15126050420168066,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6230769230769231,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.5371862565044383,
                        "high": 0.7086614173228346,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7538461538461538,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6759243295019157,
                        "high": 0.8297872340425532,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.826530612244898,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.75,
                        "high": 0.8977272727272727,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13076923076923078,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.07627118644067797,
                        "high": 0.1896551724137931,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.088461538461539,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.546589281451421,
                        "high": 3.5990611047097794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.934615384615385,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.6607142857142856,
                        "high": 4.203547481279783,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5803902116402117,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.43799546807359313,
                        "high": 0.5790610166514771,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6331529581529581,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4778132378984651,
                        "high": 0.6317029272561568,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.16923076923076924,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.10526315789473684,
                        "high": 0.23741744309470444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5981308411214953,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.5042735042735043,
                        "high": 0.6885245901639344,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6822429906542056,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.5904761904761905,
                        "high": 0.7663551401869159,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8767123287671232,
                      "unit": "proportion",
                      "denominator": 73,
                      "interval": {
                        "low": 0.7966101694915254,
                        "high": 0.9473684210526315,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08411214953271028,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.03571428571428571,
                        "high": 0.14130434782608695,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.9392523364485985,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.361319933378757,
                        "high": 3.4854403939507095,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.640186915887851,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 3.2889858826320792,
                        "high": 3.961166094743891,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5605442176870749,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.36844954648526074,
                        "high": 0.5252251317193544,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6154995331465919,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.40457205104263927,
                        "high": 0.5767177916918402,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.14018691588785046,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.07563025210084033,
                        "high": 0.2087987012987012,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.43609022556390975,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.3515625,
                        "high": 0.5220604502863092,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5488721804511278,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.46218487394957986,
                        "high": 0.6349206349206349,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7945205479452054,
                      "unit": "proportion",
                      "denominator": 73,
                      "interval": {
                        "low": 0.6973684210526315,
                        "high": 0.8815982972136219,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11278195488721804,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.06299074940807224,
                        "high": 0.1680016806722689,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2969924812030076,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.7519056510516782,
                        "high": 2.8104869257087,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.3120300751879705,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 3.0147031987977675,
                        "high": 3.597744360902256,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.46521578421578424,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.30725842490842487,
                        "high": 0.4542333638583639,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.47470998389365743,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3135290050085969,
                        "high": 0.46350343250853454,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.20300751879699247,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.1366906474820144,
                        "high": 0.2765957446808511,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14473684210526316,
                        "high": 0.33783783783783783,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2602739726027397,
                        "high": 0.4794520547945205,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6428571428571429,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": {
                        "low": 0.4583333333333333,
                        "high": 0.8181818181818182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13157894736842105,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.060596933187294644,
                        "high": 0.2112676056338028,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.0855263157894737,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.4090909090909091,
                        "high": 1.7711254107338439,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.9407894736842106,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.5249922839506174,
                        "high": 2.3852666528325375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2759259259259259,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11724046015712682,
                        "high": 0.25044091710758376,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3239130434782609,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1376301054018445,
                        "high": 0.2939958592132505,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.17105263157894737,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.09090909090909091,
                        "high": 0.25925925925925924,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22448979591836735,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.14457496653279786,
                        "high": 0.31314616755793206,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.37755102040816324,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.2803738317757009,
                        "high": 0.47959183673469385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5945945945945946,
                      "unit": "proportion",
                      "denominator": 37,
                      "interval": {
                        "low": 0.4358974358974359,
                        "high": 0.75,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15306122448979592,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.08602150537634409,
                        "high": 0.22772277227722773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.9540816326530612,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 0.33683941997851774,
                        "high": 1.5693069306930694,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.0255102040816326,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.6553279920955408,
                        "high": 2.421081871345029,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.18898225957049486,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10469771241830067,
                        "high": 0.20046724245253655,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.20952380952380947,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11607789855072465,
                        "high": 0.22225716011042096,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.1346153846153846,
                        "high": 0.2990887850467286,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "saved_event_telemetry",
                "coverage": {
                  "observed_events": 790,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "Saved provider-token cost estimate for each selected response; not an invoice.",
                "value": 0.0032213569260759493
              }
            },
            "gemini-medium38": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5670886075949367,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5389654798436445,
                    "high": 0.593053275838086,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.660759493670886,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6348663748347293,
                    "high": 0.6849344765642235,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.8582375478927203,
                  "unit": "proportion",
                  "denominator": 522,
                  "population_events": 790,
                  "interval": {
                    "low": 0.8315573625250614,
                    "high": 0.883581261189724,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.5717006119944849,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5400427581382938,
                    "high": 0.5877039707227594,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 2.8784810126582276,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 2.695389768020306,
                    "high": 3.0551292879061864,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.09367088607594937,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.07666521681553327,
                    "high": 0.1117573029993916,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.6368310614622108,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.6015666166603778,
                    "high": 0.6546575876405419,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.568354430379747,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.4558900259374945,
                    "high": 3.66975747529545,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.1379746835443038,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.11743710491337073,
                    "high": 0.15924142225566273,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6783216783216783,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.631336405529954,
                        "high": 0.7246393861892583,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7575757575757576,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7136038186157518,
                        "high": 0.8,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8953846153846153,
                      "unit": "proportion",
                      "denominator": 325,
                      "interval": {
                        "low": 0.8610271903323263,
                        "high": 0.926984126984127,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07925407925407925,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.055023325358851675,
                        "high": 0.10552989667438312,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.371794871794872,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.0942477422003285,
                        "high": 3.6412313254587243,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.8962703962703964,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.7297603165396094,
                        "high": 4.059288663908255,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.6164839736143914,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5357398042384097,
                        "high": 0.6193322491825886,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7159168725844545,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6221494500833143,
                        "high": 0.7192245474378447,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1048951048951049,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.07586206896551724,
                        "high": 0.1358590827513136,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.43490304709141275,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3808187961364721,
                        "high": 0.488890423572744,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5457063711911357,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.4913282195187525,
                        "high": 0.5988712048137431,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7969543147208121,
                      "unit": "proportion",
                      "denominator": 197,
                      "interval": {
                        "low": 0.7357512953367875,
                        "high": 0.8523829039812646,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11080332409972299,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.07926605750661612,
                        "high": 0.14641059061551626,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2922437673130194,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 1.952243848778736,
                        "high": 2.621520630144567,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.1786703601108037,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.9796682209469156,
                        "high": 3.3761107563178534,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.41804116678889525,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3363835465957599,
                        "high": 0.4298271630594536,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4534683843133779,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3648906268157397,
                        "high": 0.4662531938272039,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1772853185595568,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.13829630788485608,
                        "high": 0.21957671957671956,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7626666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.7164128680919726,
                        "high": 0.8079118644067796,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8426666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.802138331945875,
                        "high": 0.8820375335120644,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9050632911392406,
                      "unit": "proportion",
                      "denominator": 316,
                      "interval": {
                        "low": 0.8706624605678234,
                        "high": 0.9367088607594937,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.05316225868043278,
                        "high": 0.109375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.88,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.6189169604386997,
                        "high": 4.124301675977654,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.346666666666667,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 4.210327206029347,
                        "high": 4.477721032449445,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7303325857542724,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6490451096466334,
                        "high": 0.7319890034241242,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7771487771487771,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6906505653932123,
                        "high": 0.7789113754384911,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.09333333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.06417112299465241,
                        "high": 0.12430939226519337,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.49583333333333335,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.4297872340425532,
                        "high": 0.5603448275862069,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6083333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5447154471544715,
                        "high": 0.6708860759493671,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.815068493150685,
                      "unit": "proportion",
                      "denominator": 146,
                      "interval": {
                        "low": 0.7481481481481481,
                        "high": 0.875,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1125,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.07441764366711513,
                        "high": 0.15451282444308187,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.54375,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.1189447486506845,
                        "high": 2.938818565400844,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.439583333333333,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 3.2120573105585972,
                        "high": 3.6611603572380695,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5312174100409395,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.39787418300653593,
                        "high": 0.5096462758997816,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5715630361199983,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4280924753867792,
                        "high": 0.5483535879934358,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.17916666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.13022877670637173,
                        "high": 0.23333333333333334,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2413793103448276,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.17901234567901234,
                        "high": 0.3072639324158605,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3390804597701149,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.269836811128946,
                        "high": 0.4114285714285714,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.711864406779661,
                      "unit": "proportion",
                      "denominator": 59,
                      "interval": {
                        "low": 0.5909090909090909,
                        "high": 0.8245614035087719,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09770114942528736,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.055865921787709494,
                        "high": 0.14450867052023122,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.1695402298850575,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.7028532818532819,
                        "high": 1.6114369747899158,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.060344827586207,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.776532842916132,
                        "high": 2.3493589743589745,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.294006919006919,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1657100075850076,
                        "high": 0.2729249835132188,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2978251906823335,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.16786208560559213,
                        "high": 0.2764694638185852,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1781609195402299,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.12209302325581395,
                        "high": 0.23837209302325582,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8448979591836735,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.7940074906367042,
                        "high": 0.8933345588235294,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9061224489795918,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.8619399081995879,
                        "high": 0.9458333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9324324324324325,
                      "unit": "proportion",
                      "denominator": 222,
                      "interval": {
                        "low": 0.8976715825297787,
                        "high": 0.9641255605381166,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.061224489795918366,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.032520325203252036,
                        "high": 0.09282700421940929,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 4.259183673469388,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.9860382059800674,
                        "high": 4.512346954392409,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.585714285714285,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.44924659208998,
                        "high": 4.71401380597015,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7884168741311597,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6433944250002739,
                        "high": 0.764112325064706,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.841868865597679,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6870143860172415,
                        "high": 0.8159165504928217,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.0653061224489796,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.03539562727745966,
                        "high": 0.09767616716209125,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6076923076923076,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.5217313664596274,
                        "high": 0.695659518213866,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7230769230769231,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6408395686619719,
                        "high": 0.8017260795278036,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8404255319148937,
                      "unit": "proportion",
                      "denominator": 94,
                      "interval": {
                        "low": 0.7582417582417582,
                        "high": 0.9146341463414634,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.06086809635722679,
                        "high": 0.1760014084507042,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.1653846153846157,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.649604459489293,
                        "high": 3.653508771929825,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.896153846153846,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.6222154471544714,
                        "high": 4.171433901918976,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5672949735449735,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4270411532911533,
                        "high": 0.5654266826923077,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6188672438672438,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.46586307631762175,
                        "high": 0.6168291083916084,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.14615384615384616,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.08546511627906977,
                        "high": 0.21088568239795918,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5794392523364486,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.48484848484848486,
                        "high": 0.6704614745011085,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7009345794392523,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.6131999391357273,
                        "high": 0.7840909090909091,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8266666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": {
                        "low": 0.734375,
                        "high": 0.9090909090909091,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12149532710280374,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.0625,
                        "high": 0.18811881188118812,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.7570093457943927,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.1176395558223287,
                        "high": 3.3617042899152403,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.5981308411214954,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 3.252332385209264,
                        "high": 3.9272880244755246,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5518796992481203,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3559276663881927,
                        "high": 0.5115302144249513,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6168067226890757,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3978015094926859,
                        "high": 0.5717102396514161,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.16822429906542055,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.09900990099009901,
                        "high": 0.24347826086956523,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.42857142857142855,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.34285714285714286,
                        "high": 0.514290890269151,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5338345864661654,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.44881166049096805,
                        "high": 0.6201550387596899,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8028169014084507,
                      "unit": "proportion",
                      "denominator": 71,
                      "interval": {
                        "low": 0.704225352112676,
                        "high": 0.8902480788506515,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10526315789473684,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.05714285714285714,
                        "high": 0.16058394160583941,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.3721804511278197,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.8257386363636365,
                        "high": 2.871570642201835,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.3120300751879705,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 3.02082902892562,
                        "high": 3.595238095238096,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4678095238095238,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3112497308247308,
                        "high": 0.4589404952890247,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.47735665694849366,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.31760176614768454,
                        "high": 0.46830662784594357,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.18796992481203006,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.12295081967213115,
                        "high": 0.25925925925925924,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.27631578947368424,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.17855919765166342,
                        "high": 0.38095238095238093,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.35526315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.25,
                        "high": 0.4657534246575342,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7777777777777778,
                      "unit": "proportion",
                      "denominator": 27,
                      "interval": {
                        "low": 0.6071428571428571,
                        "high": 0.9259259259259259,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.025,
                        "high": 0.14516129032258066,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.3552631578947372,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6551621510673235,
                        "high": 2.0357250430292595,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.0789473684210527,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.6428571428571428,
                        "high": 2.5294117647058822,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.35763888888888884,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1581345390720391,
                        "high": 0.3180654761904761,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3731884057971015,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1650099538143016,
                        "high": 0.33189440993788805,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.14473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.07042253521126761,
                        "high": 0.22857142857142856,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.13541666666666666,
                        "high": 0.3023332313341492,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.32653061224489793,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.23333333333333334,
                        "high": 0.4226903251387786,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.65625,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": {
                        "low": 0.4827586206896552,
                        "high": 0.8181818181818182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11224489795918367,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.05319148936170213,
                        "high": 0.1794943820224718,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.0255102040816326,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 0.40525657894736844,
                        "high": 1.6443298969072164,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.045918367346939,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.6768277285045574,
                        "high": 2.4350000000000005,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.19630192502532925,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10437310030395137,
                        "high": 0.2128021092382794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.20056935817805382,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10664208074534162,
                        "high": 0.2174282420478072,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.20408163265306123,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.125,
                        "high": 0.28846153846153844,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "verified_selected_result_token_reconstruction",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 790,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "Selected-response tokens repriced with the run's retained list prices and GBP buffers; no provider invoice. Earlier failed/uncertain attempts are separate.",
                "value": 0.0075642658227848105
              }
            },
            "gemini38-low": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.569620253164557,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5417234620161835,
                    "high": 0.5961324408713649,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.6670886075949367,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6418739268818383,
                    "high": 0.6918905716294957,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.8538899430740038,
                  "unit": "proportion",
                  "denominator": 527,
                  "population_events": 790,
                  "interval": {
                    "low": 0.8265536540508284,
                    "high": 0.8789170821302109,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.5374228127338921,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5076246402887419,
                    "high": 0.5529720878074124,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 2.867721518987342,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 2.6761928204531054,
                    "high": 3.0395548118372484,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.09746835443037975,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.08026443229766014,
                    "high": 0.11624285700235067,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.6394651189491882,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.604009065660022,
                    "high": 0.6579667880240097,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.5702531645569624,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.4587884623288105,
                    "high": 3.6734464502621154,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.14050632911392405,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.12054258680840958,
                    "high": 0.16327144638062357,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6876456876456877,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.6402873628698851,
                        "high": 0.7333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7645687645687645,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7213086424416408,
                        "high": 0.8069309615343313,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.899390243902439,
                      "unit": "proportion",
                      "denominator": 328,
                      "interval": {
                        "low": 0.8660428501618715,
                        "high": 0.9300911854103343,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07692307692307693,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.052995391705069124,
                        "high": 0.10267884430667132,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.4149184149184153,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.1325187542623323,
                        "high": 3.6757502194426155,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.9160839160839163,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.746691479009378,
                        "high": 4.0813108796543265,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5889353821931136,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5126891394994124,
                        "high": 0.5943557395231674,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7314197488527379,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6367268345395928,
                        "high": 0.7381514829561918,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.10023310023310024,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.07242990654205607,
                        "high": 0.13013908989936385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4293628808864266,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.3771917293233083,
                        "high": 0.48265915221579964,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5512465373961218,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.49865210848873986,
                        "high": 0.6034493507610285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7788944723618091,
                      "unit": "proportion",
                      "denominator": 199,
                      "interval": {
                        "low": 0.7184452070081464,
                        "high": 0.835,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12188365650969529,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.08963376395334254,
                        "high": 0.15691764111130643,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2174515235457064,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 1.8790728689931353,
                        "high": 2.540733877016808,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.1592797783933517,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.964001595474383,
                        "high": 3.357340720221607,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4066446751628938,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.32916461914704476,
                        "high": 0.4191355405619619,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4479983709421711,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3626389871958967,
                        "high": 0.46175949383944964,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1883656509695291,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.14763002876843692,
                        "high": 0.2305393379906853,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.768,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.7222222222222222,
                        "high": 0.8128491620111732,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8426666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.8016991232373748,
                        "high": 0.8810810810810811,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9113924050632911,
                      "unit": "proportion",
                      "denominator": 316,
                      "interval": {
                        "low": 0.8786861677054417,
                        "high": 0.9409838611139395,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07466666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.049479166666666664,
                        "high": 0.10236220472440945,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.9133333333333336,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.6608060845819916,
                        "high": 4.152521610448182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.366666666666666,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 4.233070701079844,
                        "high": 4.494318984334874,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7129652754652754,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6376155725448074,
                        "high": 0.7178825557336174,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7678087581933735,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6866629242790233,
                        "high": 0.7731042907900493,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.09066666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.06284117243155093,
                        "high": 0.12096774193548387,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5208333333333334,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.4570106490872211,
                        "high": 0.5833333333333334,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6166666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5542168674698795,
                        "high": 0.6785714285714286,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8445945945945946,
                      "unit": "proportion",
                      "denominator": 148,
                      "interval": {
                        "low": 0.7832167832167832,
                        "high": 0.9,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09583333333333334,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.060603005865102644,
                        "high": 0.13524821274497717,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.7604166666666665,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.3496228654262774,
                        "high": 3.135757412721186,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.5104166666666665,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 3.2865711554361305,
                        "high": 3.727277384500745,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5414579606440071,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.40863227857413903,
                        "high": 0.5186205636978279,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5894352482960077,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.44484020199210067,
                        "high": 0.5645742845318128,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.15,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.10384615384615385,
                        "high": 0.2,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.20689655172413793,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.14917127071823205,
                        "high": 0.26993865030674846,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3563218390804598,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.2857142857142857,
                        "high": 0.4301675977653631,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5806451612903226,
                      "unit": "proportion",
                      "denominator": 62,
                      "interval": {
                        "low": 0.45651220442410373,
                        "high": 0.704225352112676,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14942528735632185,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.09815950920245399,
                        "high": 0.2057142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.75,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.2813984865154173,
                        "high": 1.216388192703982,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.9281609195402298,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 1.6583261494252879,
                        "high": 2.2088302139037435,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.24082892416225748,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13113870697204033,
                        "high": 0.22645930201485756,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2533395176252319,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13795110733422425,
                        "high": 0.23822342160004498,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.23563218390804597,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.1728395061728395,
                        "high": 0.30120481927710846,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8448979591836735,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.7934262600395763,
                        "high": 0.8931623931623932,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8938775510204081,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.847925829230257,
                        "high": 0.9354838709677419,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9452054794520548,
                      "unit": "proportion",
                      "denominator": 219,
                      "interval": {
                        "low": 0.91324200913242,
                        "high": 0.9734513274336283,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.04897959183673469,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.023622047244094488,
                        "high": 0.07751937984496124,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 4.295918367346939,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.025087889703338,
                        "high": 4.535417076771654,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.581632653061225,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.441803392658509,
                        "high": 4.7126193843253,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7860630551106741,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6448308171493854,
                        "high": 0.765472030563067,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.8393554656266521,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6885481606849368,
                        "high": 0.8173684394148003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.05714285714285714,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.029661016949152543,
                        "high": 0.08823529411764706,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6230769230769231,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.5352112676056338,
                        "high": 0.7096835754740212,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7461538461538462,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6666666666666666,
                        "high": 0.8222222222222222,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8350515463917526,
                      "unit": "proportion",
                      "denominator": 97,
                      "interval": {
                        "low": 0.7582330827067669,
                        "high": 0.9052667140825036,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12307692307692308,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.06956521739130435,
                        "high": 0.1810344827586207,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.192307692307693,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.6790911513859275,
                        "high": 3.672947865531598,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.9615384615384617,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.690826164976644,
                        "high": 4.217062689585441,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5785545423843296,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4417746231310061,
                        "high": 0.58267972275951,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.618001443001443,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4718956201626657,
                        "high": 0.6224078856749311,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.15384615384615385,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.0948905109489051,
                        "high": 0.21768844634628043,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6448598130841121,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.5517241379310345,
                        "high": 0.7327586206896551,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7383177570093458,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.6526315789473685,
                        "high": 0.8198243243243242,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8734177215189873,
                      "unit": "proportion",
                      "denominator": 79,
                      "interval": {
                        "low": 0.7948717948717948,
                        "high": 0.9431818181818182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09345794392523364,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.041237113402061855,
                        "high": 0.15315315315315314,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.116822429906542,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.510989010989011,
                        "high": 3.659309734513274,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.771028037383178,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 3.4215557275541792,
                        "high": 4.087962962962963,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5935139573070608,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3889114668856049,
                        "high": 0.549341133004926,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6749766573295986,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4422914721444134,
                        "high": 0.6247408963585432,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1308411214953271,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.0707070707070707,
                        "high": 0.2,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.33793103448275863,
                        "high": 0.5038167938931297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.518796992481203,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.4338235294117647,
                        "high": 0.6033057851239669,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8115942028985508,
                      "unit": "proportion",
                      "denominator": 69,
                      "interval": {
                        "low": 0.7160493827160493,
                        "high": 0.8983050847457628,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09774436090225563,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.05042016806722689,
                        "high": 0.15151515151515152,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.473684210526316,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.9599814814814818,
                        "high": 2.9500088652482273,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.300751879699248,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 3.012597293015567,
                        "high": 3.57936507936508,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4639014047177312,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3063629737609329,
                        "high": 0.4534023980405559,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4639014047177312,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3063629737609329,
                        "high": 0.4534023980405559,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.16541353383458646,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.10317460317460317,
                        "high": 0.23484975594874924,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14492753623188406,
                        "high": 0.33707865168539325,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3815789473684211,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2714285714285714,
                        "high": 0.49230769230769234,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6206896551724138,
                      "unit": "proportion",
                      "denominator": 29,
                      "interval": {
                        "low": 0.43586956521739134,
                        "high": 0.8,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.07042253521126761,
                        "high": 0.22917520491803264,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.9736842105263158,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.25609393147502907,
                        "high": 1.6667134831460668,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.9605263157894737,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.5460376794258375,
                        "high": 2.3947368421052633,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.29084967320261434,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1222222222222222,
                        "high": 0.2619047619047619,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.322463768115942,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1355072463768116,
                        "high": 0.2903726708074534,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.19736842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.1111111111111111,
                        "high": 0.2916666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1836734693877551,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.1111111111111111,
                        "high": 0.26732673267326734,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.336734693877551,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.24489795918367346,
                        "high": 0.4380952380952381,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5454545454545454,
                      "unit": "proportion",
                      "denominator": 33,
                      "interval": {
                        "low": 0.37140211640211646,
                        "high": 0.7142857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15306122448979592,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.08571428571428572,
                        "high": 0.23,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.576530612244898,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -0.05617977528089888,
                        "high": 1.2143049155145926,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.903061224489796,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.5494419642857142,
                        "high": 2.284727822580645,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.15170068027210884,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08027210884353742,
                        "high": 0.17067077480342782,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.1615942028985507,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0855072463768116,
                        "high": 0.18180147750799922,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2653061224489796,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.17755792445482865,
                        "high": 0.3592233009708738,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_usage_ledger_plus_selected_canary",
                "coverage": {
                  "observed_events": 789,
                  "estimated_events": 1,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "789 saved reconciled selected-response costs plus one prior canary cost reconstructed from its hash-bound result tokens and retained price/FX buffer; earlier failed or uncertain attempts remain separate. Not an invoice.",
                "value": 0.006292627848101266
              }
            },
            "gemini38-high": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.640506329113924,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.6124181760574166,
                    "high": 0.6668709305813737,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.7734177215189874,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.7483349430105759,
                    "high": 0.7955571450349932,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.8281505728314239,
                  "unit": "proportion",
                  "denominator": 611,
                  "population_events": 790,
                  "interval": {
                    "low": 0.8019380435410597,
                    "high": 0.8543367724636198,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.5700173160569032,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.5417594480316914,
                    "high": 0.5854840138287108,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 2.945569620253165,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 2.7489428758653442,
                    "high": 3.1343632299486726,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.13291139240506328,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.1122995894752224,
                    "high": 0.15348890121833156,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.6998946792091089,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.6651983096085324,
                    "high": 0.7188854347010752,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 3.8063291139240505,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 3.6953067172806087,
                    "high": 3.907151573373883,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.17215189873417722,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.14883463143747955,
                    "high": 0.19550607708044415,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7319347319347319,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.6877931616850425,
                        "high": 0.7739745180962888,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8531468531468531,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.8173076923076923,
                        "high": 0.8868384938549561,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8579234972677595,
                      "unit": "proportion",
                      "denominator": 366,
                      "interval": {
                        "low": 0.8205128205128205,
                        "high": 0.8931512946628186,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12121212121212122,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.09070294784580499,
                        "high": 0.15328510191433686,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.4055944055944054,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.1085524416555246,
                        "high": 3.6841507177033495,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.093240093240094,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 3.9378910168718657,
                        "high": 4.241252165841584,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.6111803656842256,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5328345661557514,
                        "high": 0.6144362888196777,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7787620788557069,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6789343665532962,
                        "high": 0.7829107551089441,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13752913752913754,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.1047613110081938,
                        "high": 0.1721870147976805,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5318559556786704,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.47780523897699717,
                        "high": 0.5849596220689877,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6786703601108033,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.6287243104526535,
                        "high": 0.7277936962750716,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7836734693877551,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.7288135593220338,
                        "high": 0.8347835791925465,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14681440443213298,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.11080332409972299,
                        "high": 0.18571623731459794,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.398891966759003,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.0397243781287595,
                        "high": 2.75220023020918,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.465373961218837,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 3.266196099674973,
                        "high": 3.661727993332388,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4679993612043887,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.38971527884421053,
                        "high": 0.47986115462039364,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5314569017066787,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.44255802851800186,
                        "high": 0.5449270738909553,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21329639889196675,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.17109047508150907,
                        "high": 0.25806451612903225,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8133333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.7716535433070866,
                        "high": 0.8532345068137572,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9066666666666666,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.8743311855293534,
                        "high": 0.9375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8970588235294118,
                      "unit": "proportion",
                      "denominator": 340,
                      "interval": {
                        "low": 0.8643067846607669,
                        "high": 0.927377255312326,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09333333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.06557377049180328,
                        "high": 0.12335958005249344,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.9280000000000004,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.6716412868091974,
                        "high": 4.169231266149871,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.488,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 4.3692523209549075,
                        "high": 4.60340784183275,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7348382989559459,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6601597557648675,
                        "high": 0.7394745897377476,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.8007853257853257,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.7194048620514581,
                        "high": 0.8058376939449814,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.112,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.082010582010582,
                        "high": 0.1447730335926398,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6083333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5439983263598327,
                        "high": 0.6707818930041153,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7416666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.6848739495798319,
                        "high": 0.7974151571366216,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8202247191011236,
                      "unit": "proportion",
                      "denominator": 178,
                      "interval": {
                        "low": 0.7597765363128491,
                        "high": 0.8757070504462943,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13333333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.09130434782608696,
                        "high": 0.1799163179916318,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.8625000000000003,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.4336278595821246,
                        "high": 3.2680866077556625,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.7791666666666672,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 3.5635946937014666,
                        "high": 3.991562067865121,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.604353086249638,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4608287727316632,
                        "high": 0.5789137339352856,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6655533987812469,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5074949775652493,
                        "high": 0.637537909523669,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.18333333333333332,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.13274336283185842,
                        "high": 0.2384937238493724,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3103448275862069,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.2422360248447205,
                        "high": 0.3801232375417069,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5287356321839081,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.45454545454545453,
                        "high": 0.603448275862069,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5869565217391305,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": {
                        "low": 0.4842105263157895,
                        "high": 0.6888888888888889,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21839080459770116,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.15789473684210525,
                        "high": 0.28160919540229884,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.9310344827586207,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.4084854631507775,
                        "high": 1.4542986279257464,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.367816091954023,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 2.0602826198410664,
                        "high": 2.6705202312138727,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3339663643235071,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19873754950698228,
                        "high": 0.30663355989248847,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3643269428983714,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.21680459946216246,
                        "high": 0.3345093380645328,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.28735632183908044,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.22042782941498087,
                        "high": 0.35542168674698793,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8653061224489796,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.8189646303391515,
                        "high": 0.9082578822963824,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9510204081632653,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.9182128092552236,
                        "high": 0.9795081967213115,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9098712446351931,
                      "unit": "proportion",
                      "denominator": 233,
                      "interval": {
                        "low": 0.8709677419354839,
                        "high": 0.9444444444444444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08571428571428572,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.05263157894736842,
                        "high": 0.12244897959183673,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 4.183673469387755,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.8799105553160786,
                        "high": 4.456528448738593,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.63265306122449,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 4.5042359584426634,
                        "high": 4.752233678735256,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.7473849382940291,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.622920330821473,
                        "high": 0.7324318183425803,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.8360577275831513,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6968261327833427,
                        "high": 0.8193305086544119,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.08979591836734693,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.05622489959839357,
                        "high": 0.12766191770550187,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7153846153846154,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.6363636363636364,
                        "high": 0.7950868077965664,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.823076923076923,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.7534223744292238,
                        "high": 0.8888888888888888,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8691588785046729,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.8021978021978022,
                        "high": 0.9304347826086956,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1076923076923077,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.05737704918032787,
                        "high": 0.16393442622950818,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.446153846153847,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.9541227470882947,
                        "high": 3.9178540851272015,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.2153846153846155,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 3.976557221283784,
                        "high": 4.446979665071771,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.6379794973544973,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4953498151154401,
                        "high": 0.638349511721222,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6959776334776334,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5403816164895711,
                        "high": 0.6963812855140604,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.15384615384615385,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.09401709401709402,
                        "high": 0.21804511278195488,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6915887850467289,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.603448275862069,
                        "high": 0.7764705882352941,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.822429906542056,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.7474720893141946,
                        "high": 0.8910891089108911,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8409090909090909,
                      "unit": "proportion",
                      "denominator": 88,
                      "interval": {
                        "low": 0.759493670886076,
                        "high": 0.9139784946236559,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1308411214953271,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.06930693069306931,
                        "high": 0.19820013979496737,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.2242990654205608,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.617297880690738,
                        "high": 3.7857142857142856,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.9719626168224305,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 3.656514945652174,
                        "high": 4.268915643180349,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.6417898193760262,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4235570607553366,
                        "high": 0.590068076081007,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7298786181139122,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.48169234360410823,
                        "high": 0.6710578120136943,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.14953271028037382,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.0847457627118644,
                        "high": 0.22,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5413533834586466,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.45384615384615384,
                        "high": 0.6283249051833122,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6766917293233082,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5948261775974758,
                        "high": 0.7575757575757576,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8,
                      "unit": "proportion",
                      "denominator": 90,
                      "interval": {
                        "low": 0.7093023255813954,
                        "high": 0.8829787234042553,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13533834586466165,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.0782608695652174,
                        "high": 0.2,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.5714285714285716,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 1.9797007722007725,
                        "high": 3.1229524460057245,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.62406015037594,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 3.326441152335737,
                        "high": 3.9029879671410392,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5442528060175119,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37526196842373316,
                        "high": 0.5285456547572916,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5664672062631246,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3905787834614366,
                        "high": 0.5501189467882014,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21052631578947367,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.1399965034965035,
                        "high": 0.2868429436126152,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35526315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.25,
                        "high": 0.4642857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5789473684210527,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.46875,
                        "high": 0.6896673142302087,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6136363636363636,
                      "unit": "proportion",
                      "denominator": 44,
                      "interval": {
                        "low": 0.4666666666666667,
                        "high": 0.7560975609756098,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.13414634146341464,
                        "high": 0.3230769230769231,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.131578947368421,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.3209853510226645,
                        "high": 1.925931186868687,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.5131578947368425,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.057142857142857,
                        "high": 2.975,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3902964959568733,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.18658243486073675,
                        "high": 0.34380091344713987,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4496894409937888,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.21497541407867496,
                        "high": 0.3961184437543133,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.27631578947368424,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.17857142857142858,
                        "high": 0.38095238095238093,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2755102040816326,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.18811669628501312,
                        "high": 0.3695652173913043,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.4897959183673469,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.38823529411764707,
                        "high": 0.5904761904761905,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5625,
                      "unit": "proportion",
                      "denominator": 48,
                      "interval": {
                        "low": 0.4166666666666667,
                        "high": 0.7045454545454546,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.13592233009708737,
                        "high": 0.29896907216494845,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.7755102040816326,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 0.07605978260869568,
                        "high": 1.474375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.2551020408163267,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.8520408163265305,
                        "high": 2.6770917338709683,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.23257142857142857,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13551338383838382,
                        "high": 0.2434307692307692,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.25279503105590057,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.14729715634606938,
                        "high": 0.26459866220735784,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.29591836734693877,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.20731432652164358,
                        "high": 0.38636363636363635,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_usage_ledger_plus_selected_canary",
                "coverage": {
                  "observed_events": 789,
                  "estimated_events": 1,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "789 saved reconciled selected-response costs plus one prior canary cost reconstructed from its hash-bound result tokens and retained price/FX buffer; earlier failed or uncertain attempts remain separate. Not an invoice.",
                "value": 0.013143683544303798
              }
            },
            "qwen-repaired": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4050632911392405,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.37932481460171324,
                    "high": 0.42949030839062485,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.8607594936708861,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.8362663136792884,
                    "high": 0.8822261686976877,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.47058823529411764,
                  "unit": "proportion",
                  "denominator": 680,
                  "population_events": 790,
                  "interval": {
                    "low": 0.4443503349531054,
                    "high": 0.4971062481667023,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.2028763468703074,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.1880233773912816,
                    "high": 0.21117139483899539,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 0.4044303797468355,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.20845360774949376,
                    "high": 0.59785657204961,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.45569620253164556,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.4297494839899903,
                    "high": 0.48085025997209546,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.4365693540247121,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4046072678040237,
                    "high": 0.4544194572484709,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 2.809493670886076,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 2.6910758979944105,
                    "high": 2.9168447357846095,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.4810126582278481,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.4549480003119244,
                    "high": 0.5062528926769433,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.48951048951048953,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.4397313015972721,
                        "high": 0.5395348837209303,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8741258741258742,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.8426952033308618,
                        "high": 0.9047619047619048,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.56,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5062034739454094,
                        "high": 0.6129055039183524,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.38461538461538464,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.33647058823529413,
                        "high": 0.43387470997679817,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.1002331002331003,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7157678752693073,
                        "high": 1.4771219614096984,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.174825174825175,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 2.979155585106383,
                        "high": 3.3698033785759662,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.27320079576369943,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.23530682696226463,
                        "high": 0.28262942481471504,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.46708523146696995,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.40229876867742026,
                        "high": 0.4832051456509644,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4149184149184149,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.36658157397767716,
                        "high": 0.46454902957786093,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3047091412742382,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.2553191489361702,
                        "high": 0.355747044580544,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8448753462603878,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.8033696193470073,
                        "high": 0.8834701820029564,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.36065573770491804,
                      "unit": "proportion",
                      "denominator": 305,
                      "interval": {
                        "low": 0.30475875118259227,
                        "high": 0.41818478906714196,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5401662049861495,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.48633879781420764,
                        "high": 0.5945223991114402,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.422437673130194,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": -0.8128314781896204,
                        "high": -0.007040778188685206,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.3753462603878117,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.166655027932961,
                        "high": 2.589780907198335,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.147244129838632,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11118732279369133,
                        "high": 0.15303127974848377,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.30946223898288755,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.23368183095623263,
                        "high": 0.3216250625222371,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5595567867036011,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5051175548589342,
                        "high": 0.6135709975864843,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5893333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5376344086021505,
                        "high": 0.6413043478260869,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9893333333333333,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.9778393351800554,
                        "high": 0.9974620895714195,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5956873315363881,
                      "unit": "proportion",
                      "denominator": 371,
                      "interval": {
                        "low": 0.5437113413209914,
                        "high": 0.6472222222222223,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.34908136482939633,
                        "high": 0.4520547945205479,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.5613333333333335,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 1.1409053184281845,
                        "high": 1.9761252675227394,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.708,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.527214921811124,
                        "high": 3.883602531102531,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3341092241092241,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.28877395047234,
                        "high": 0.3394712910434869,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5568487068487068,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.48128991745389993,
                        "high": 0.5657854850724782,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.42933333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.3775201932053416,
                        "high": 0.48128591136648075,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32083333333333336,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2616033755274262,
                        "high": 0.38257658102766795,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8916666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.8508754244482173,
                        "high": 0.9297520661157025,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3598130841121495,
                      "unit": "proportion",
                      "denominator": 214,
                      "interval": {
                        "low": 0.2949293985584308,
                        "high": 0.42660550458715596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5708333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5061728395061729,
                        "high": 0.6347860593512767,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.3875,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": -0.867782395432465,
                        "high": 0.11163256795211454,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.591666666666667,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.3473420599803343,
                        "high": 2.8394313604833297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2151797535797536,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.15615385336885335,
                        "high": 0.21379193192772142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3404742936388506,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24707888191274266,
                        "high": 0.33827837330335664,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5958333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5321078320603787,
                        "high": 0.6600790513833992,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1206896551724138,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.07262569832402235,
                        "high": 0.17318435754189945,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5402298850574713,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.4634146341463415,
                        "high": 0.6170243161094224,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.22340425531914893,
                      "unit": "proportion",
                      "denominator": 94,
                      "interval": {
                        "low": 0.13829787234042554,
                        "high": 0.3113207547169811,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.41954022988505746,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3465909090909091,
                        "high": 0.49404761904761907,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -1.0229885057471264,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -1.4401252891943386,
                        "high": -0.5952255639097747,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.1609195402298853,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.90625,
                        "high": 1.4273352129448025,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.07115812536373284,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.03657070538495772,
                        "high": 0.08156063303259564,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.09888207031064172,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05081903215831787,
                        "high": 0.11333750304529523,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4367816091954023,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3619631901840491,
                        "high": 0.5126590310351474,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6857142857142857,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.6234309623430963,
                        "high": 0.7449412874816523,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9918367346938776,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.9789029535864979,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.691358024691358,
                      "unit": "proportion",
                      "denominator": 243,
                      "interval": {
                        "low": 0.6290322580645161,
                        "high": 0.7510377820963986,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.30612244897959184,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.24686192468619247,
                        "high": 0.368,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.369387755102041,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 1.8681965593940053,
                        "high": 2.84439949285385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.083673469387755,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.882128891155832,
                        "high": 4.27451258912656,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4293459552495697,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3572486670288907,
                        "high": 0.43041251452611723,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6039951573849879,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5025701587016598,
                        "high": 0.6054955712825039,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.34285714285714286,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.2813658507604563,
                        "high": 0.40613326149425283,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4076923076923077,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.32231404958677684,
                        "high": 0.4957282913165266,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9846153846153847,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.96,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4140625,
                      "unit": "proportion",
                      "denominator": 128,
                      "interval": {
                        "low": 0.3274296057924377,
                        "high": 0.5035460992907801,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5769230769230769,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.48836751510712323,
                        "high": 0.6635535259133389,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.038461538461538464,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": -0.6760057377049181,
                        "high": 0.7698660714285711,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.6773041586073507,
                        "high": 3.3248220802919715,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2205654761904762,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.15291540230602732,
                        "high": 0.229375744047619,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4010281385281385,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.27802800419277696,
                        "high": 0.4170468073593073,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5923076923076923,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.5038744896147701,
                        "high": 0.6779661016949152,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3364485981308411,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.24770642201834864,
                        "high": 0.42609369733737773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.897196261682243,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.8365384615384616,
                        "high": 0.9516206575682381,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": {
                        "low": 0.2792740414833438,
                        "high": 0.4717034131863472,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5607476635514018,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.46609963797926607,
                        "high": 0.6551724137931034,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.17289719626168223,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": -0.8701967592592592,
                        "high": 0.5244118009361909,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.7242990654205608,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.375,
                        "high": 3.0809634551495018,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2309206349206349,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.14162012987012987,
                        "high": 0.22733333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.33958916900093367,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20826489686783803,
                        "high": 0.3343137254901961,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5794392523364486,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.4891304347826087,
                        "high": 0.6728971962616822,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3082706766917293,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.22880750605326877,
                        "high": 0.3944954128440367,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8872180451127819,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.8309859154929577,
                        "high": 0.9396551724137931,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3474576271186441,
                      "unit": "proportion",
                      "denominator": 118,
                      "interval": {
                        "low": 0.2601626016260163,
                        "high": 0.44118309485956536,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5789473684210527,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.4895067698259188,
                        "high": 0.6642335766423357,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.5601503759398496,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": -1.1965811965811965,
                        "high": 0.11539115646258497,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.484962406015038,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 2.1547579887218045,
                        "high": 2.828,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.19558103076621594,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.12479618571938833,
                        "high": 0.19909069631528667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3233074182053774,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20629573557694805,
                        "high": 0.3291091102354738,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.6090225563909775,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5220588235294118,
                        "high": 0.6936936936936937,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.06578947368421052,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.013513513513513514,
                        "high": 0.12698412698412698,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.4605263157894737,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.34882840028188866,
                        "high": 0.5731747967479673,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 35,
                      "interval": {
                        "low": 0.03125,
                        "high": 0.2702702702702703,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.39473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.287496408045977,
                        "high": 0.5061728395061729,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -1.25,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -1.7608954451345757,
                        "high": -0.7465584998877164,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 0.8552631578947368,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5416666666666666,
                        "high": 1.2025316455696204,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.048205128205128206,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.010256410256410256,
                        "high": 0.05846153846153846,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.06811594202898551,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.014492753623188404,
                        "high": 0.08260869565217391,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.3125,
                        "high": 0.5316455696202531,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.16326530612244897,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.09090909090909091,
                        "high": 0.24691721132897596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6020408163265306,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.5,
                        "high": 0.7040816326530612,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2711864406779661,
                      "unit": "proportion",
                      "denominator": 59,
                      "interval": {
                        "low": 0.15623263888888891,
                        "high": 0.39285714285714285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4387755102040816,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.33962264150943394,
                        "high": 0.5421686746987951,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.8469387755102041,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -1.470977628918099,
                        "high": -0.21126369327073558,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.3979591836734695,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.0287315270935962,
                        "high": 1.7973116733755032,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.08222912353347137,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.039195709684840124,
                        "high": 0.09879227053140097,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.12334368530020705,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.058793564527260186,
                        "high": 0.14818840579710144,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4489795918367347,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.3474553969669938,
                        "high": 0.5505732630130702,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "provider_usage_or_saved_generation_metadata",
                "coverage": {
                  "observed_events": 790,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "Selected evaluated response attempt only: 786 first-pass outputs, four successful repair outputs. Four superseded first-pass length charges are excluded from this comparison and retained in the audit below.",
                "value": 0.0019741087746835443
              }
            },
            "qwen-first-pass": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.40379746835443037,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.3781239382979889,
                    "high": 0.42818381600818317,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "answer_coverage": {
                  "value": 0.8556962025316456,
                  "unit": "proportion",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.8312612492992241,
                    "high": 0.8774901250296819,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "accuracy_among_answers": {
                  "value": 0.4718934911242604,
                  "unit": "proportion",
                  "denominator": 676,
                  "population_events": 790,
                  "interval": {
                    "low": 0.445274581992246,
                    "high": 0.49831855704556355,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "proportion",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "macro_f1": {
                  "value": 0.20385000853147844,
                  "unit": "f1",
                  "denominator": null,
                  "population_events": 790,
                  "interval": {
                    "low": 0.18882257406561237,
                    "high": 0.2121418537359752,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "f1",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "adjusted_tis": {
                  "value": 0.4151898734177215,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "population_events": 790,
                  "interval": {
                    "low": 0.2174646349606793,
                    "high": 0.6079605744993402,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "unit": "points_per_event",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.4518987341772152,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.4261005757603859,
                    "high": 0.47705611723016794,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "reference_species_macro_f1": {
                  "value": 0.4360841954660741,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.40393689895048723,
                    "high": 0.45382244659974436,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "raw_tis": {
                  "value": 2.80126582278481,
                  "unit": "points_per_event",
                  "denominator": 790,
                  "interval": {
                    "low": 2.6832318911749287,
                    "high": 2.90948593556584,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.4772151898734177,
                  "unit": "proportion",
                  "denominator": 790,
                  "interval": {
                    "low": 0.4512202951759915,
                    "high": 0.5025796215036722,
                    "level": 0.95,
                    "method": "species_balanced_source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_headline_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 429,
                  "n_species": 62,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.48951048951048953,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.4397313015972721,
                        "high": 0.5395348837209303,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8717948717948718,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.8400954653937948,
                        "high": 0.9026144159139113,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5614973262032086,
                      "unit": "proportion",
                      "denominator": 374,
                      "interval": {
                        "low": 0.5079787234042553,
                        "high": 0.6143617021276596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3822843822843823,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.3348309624056404,
                        "high": 0.43137254901960786,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.1118881118881119,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 0.7304206372051449,
                        "high": 1.4860597783674105,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.174825174825175,
                      "unit": "points_per_event",
                      "denominator": 429,
                      "interval": {
                        "low": 2.979155585106383,
                        "high": 3.3698033785759662,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.27580270810430607,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.23754784436190526,
                        "high": 0.2853211336224742,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.46708523146696995,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.40229876867742026,
                        "high": 0.4832051456509644,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4125874125874126,
                      "unit": "proportion",
                      "denominator": 429,
                      "interval": {
                        "low": 0.3642852999668545,
                        "high": 0.46153846153846156,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 361,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.30193905817174516,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.2532964485624485,
                        "high": 0.35278514588859416,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8365650969529086,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.7946416505791505,
                        "high": 0.8761329305135952,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3609271523178808,
                      "unit": "proportion",
                      "denominator": 302,
                      "interval": {
                        "low": 0.30491803278688523,
                        "high": 0.41883116883116883,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5346260387811634,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.4807692307692308,
                        "high": 0.5894751341162824,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.4127423822714682,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": -0.7982748540217572,
                        "high": 0.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.357340720221607,
                      "unit": "points_per_event",
                      "denominator": 361,
                      "interval": {
                        "low": 2.1464444503964,
                        "high": 2.5723926468229417,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1464895552300676,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11064952326137786,
                        "high": 0.15225922164258712,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3078763533648877,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2325515404137433,
                        "high": 0.3200024319267933,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.554016620498615,
                      "unit": "proportion",
                      "denominator": 361,
                      "interval": {
                        "low": 0.5,
                        "high": 0.6073328432769768,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 375,
                  "n_species": 78,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5866666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.5345744680851063,
                        "high": 0.6381782945736434,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9786666666666667,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.9629629629629629,
                        "high": 0.9920429666119321,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5994550408719346,
                      "unit": "proportion",
                      "denominator": 367,
                      "interval": {
                        "low": 0.5474254742547425,
                        "high": 0.6511643748811863,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.392,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.3410841510955774,
                        "high": 0.4432724478415798,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.584,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 1.1693301829268294,
                        "high": 1.9973200959205475,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.690666666666667,
                      "unit": "points_per_event",
                      "denominator": 375,
                      "interval": {
                        "low": 3.5081075617396373,
                        "high": 3.8676101996068355,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.33875402284119394,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2927426374648299,
                        "high": 0.3443524435912997,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5559040374829848,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4803981743012593,
                        "high": 0.5650911894831584,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.42133333333333334,
                      "unit": "proportion",
                      "denominator": 375,
                      "interval": {
                        "low": 0.36923076923076925,
                        "high": 0.47342071859243323,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium": {
                  "n_events": 240,
                  "n_species": 79,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32083333333333336,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.2616033755274262,
                        "high": 0.38257658102766795,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8916666666666667,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.8508754244482173,
                        "high": 0.9297520661157025,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3598130841121495,
                      "unit": "proportion",
                      "denominator": 214,
                      "interval": {
                        "low": 0.2949293985584308,
                        "high": 0.42660550458715596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5708333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5061728395061729,
                        "high": 0.6347860593512767,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.3875,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": -0.867782395432465,
                        "high": 0.11163256795211454,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.591666666666667,
                      "unit": "points_per_event",
                      "denominator": 240,
                      "interval": {
                        "low": 2.3473420599803343,
                        "high": 2.8394313604833297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2151797535797536,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.15615385336885335,
                        "high": 0.21379193192772142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3404742936388506,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24707888191274266,
                        "high": 0.33827837330335664,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5958333333333333,
                      "unit": "proportion",
                      "denominator": 240,
                      "interval": {
                        "low": 0.5321078320603787,
                        "high": 0.6600790513833992,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard": {
                  "n_events": 174,
                  "n_species": 77,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1206896551724138,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.07262569832402235,
                        "high": 0.17318435754189945,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5402298850574713,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.4634146341463415,
                        "high": 0.6170243161094224,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.22340425531914893,
                      "unit": "proportion",
                      "denominator": 94,
                      "interval": {
                        "low": 0.13829787234042554,
                        "high": 0.3113207547169811,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.41954022988505746,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3465909090909091,
                        "high": 0.49404761904761907,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -1.0229885057471264,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": -1.4401252891943386,
                        "high": -0.5952255639097747,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.1609195402298853,
                      "unit": "points_per_event",
                      "denominator": 174,
                      "interval": {
                        "low": 0.90625,
                        "high": 1.4273352129448025,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.07115812536373284,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.03657070538495772,
                        "high": 0.08156063303259564,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.09888207031064172,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05081903215831787,
                        "high": 0.11333750304529523,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4367816091954023,
                      "unit": "proportion",
                      "denominator": 174,
                      "interval": {
                        "low": 0.3619631901840491,
                        "high": 0.5126590310351474,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 245,
                  "n_species": 59,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6857142857142857,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.6234309623430963,
                        "high": 0.7449412874816523,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9877551020408163,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.97265625,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6942148760330579,
                      "unit": "proportion",
                      "denominator": 242,
                      "interval": {
                        "low": 0.6322314049586777,
                        "high": 0.7543103448275862,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3020408163265306,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.24267782426778242,
                        "high": 0.3632657047193878,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.3897959183673474,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 1.8948178982022512,
                        "high": 2.8605840032726237,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.083673469387755,
                      "unit": "points_per_event",
                      "denominator": 245,
                      "interval": {
                        "low": 3.882128891155832,
                        "high": 4.27451258912656,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.43458188153310107,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.36160535809021865,
                        "high": 0.4356614476300943,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6039951573849879,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5025701587016598,
                        "high": 0.6054955712825039,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.33877551020408164,
                      "unit": "proportion",
                      "denominator": 245,
                      "interval": {
                        "low": 0.2767830674760023,
                        "high": 0.4016745342059149,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 130,
                  "n_species": 44,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.3153846153846154,
                        "high": 0.48717948717948717,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.9615384615384616,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.9252336448598131,
                        "high": 0.9920634920634921,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.416,
                      "unit": "proportion",
                      "denominator": 125,
                      "interval": {
                        "low": 0.3283582089552239,
                        "high": 0.5042735042735043,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5615384615384615,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.4732824427480916,
                        "high": 0.648854961832061,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.06538461538461539,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": -0.6318213383838385,
                        "high": 0.7879231974921623,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.95,
                      "unit": "points_per_event",
                      "denominator": 130,
                      "interval": {
                        "low": 2.6277971370734527,
                        "high": 3.2810105929424114,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.22110709262607994,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.15301787084223792,
                        "high": 0.2298774362065501,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3969877344877345,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2747366317394726,
                        "high": 0.4127344877344877,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5769230769230769,
                      "unit": "proportion",
                      "denominator": 130,
                      "interval": {
                        "low": 0.48836751510712323,
                        "high": 0.6623608534322817,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 107,
                  "n_species": 51,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3364485981308411,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.24770642201834864,
                        "high": 0.42609369733737773,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.897196261682243,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.8365384615384616,
                        "high": 0.9516206575682381,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": {
                        "low": 0.2792740414833438,
                        "high": 0.4717034131863472,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5607476635514018,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.46609963797926607,
                        "high": 0.6551724137931034,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.17289719626168223,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": -0.8701967592592592,
                        "high": 0.5244118009361909,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.7242990654205608,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": {
                        "low": 2.375,
                        "high": 3.0809634551495018,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2309206349206349,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.14162012987012987,
                        "high": 0.22733333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.33958916900093367,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20826489686783803,
                        "high": 0.3343137254901961,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5794392523364486,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": {
                        "low": 0.4891304347826087,
                        "high": 0.6728971962616822,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 133,
                  "n_species": 49,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3082706766917293,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.22880750605326877,
                        "high": 0.3944954128440367,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8872180451127819,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.8309859154929577,
                        "high": 0.9396551724137931,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3474576271186441,
                      "unit": "proportion",
                      "denominator": 118,
                      "interval": {
                        "low": 0.2601626016260163,
                        "high": 0.44118309485956536,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5789473684210527,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.4895067698259188,
                        "high": 0.6642335766423357,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.5601503759398496,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": -1.1965811965811965,
                        "high": 0.11539115646258497,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.484962406015038,
                      "unit": "points_per_event",
                      "denominator": 133,
                      "interval": {
                        "low": 2.1547579887218045,
                        "high": 2.828,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.19558103076621594,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.12479618571938833,
                        "high": 0.19909069631528667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3233074182053774,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20629573557694805,
                        "high": 0.3291091102354738,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.6090225563909775,
                      "unit": "proportion",
                      "denominator": 133,
                      "interval": {
                        "low": 0.5220588235294118,
                        "high": 0.6936936936936937,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 76,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.06578947368421052,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.013513513513513514,
                        "high": 0.12698412698412698,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.4605263157894737,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.34882840028188866,
                        "high": 0.5731747967479673,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 35,
                      "interval": {
                        "low": 0.03125,
                        "high": 0.2702702702702703,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.39473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.287496408045977,
                        "high": 0.5061728395061729,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -1.25,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -1.7608954451345757,
                        "high": -0.7465584998877164,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 0.8552631578947368,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5416666666666666,
                        "high": 1.2025316455696204,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.048205128205128206,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.010256410256410256,
                        "high": 0.05846153846153846,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.06811594202898551,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.014492753623188404,
                        "high": 0.08260869565217391,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.3125,
                        "high": 0.5316455696202531,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 98,
                  "n_species": 46,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.16326530612244897,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.09090909090909091,
                        "high": 0.24691721132897596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6020408163265306,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.5,
                        "high": 0.7040816326530612,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2711864406779661,
                      "unit": "proportion",
                      "denominator": 59,
                      "interval": {
                        "low": 0.15623263888888891,
                        "high": 0.39285714285714285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4387755102040816,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.33962264150943394,
                        "high": 0.5421686746987951,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.8469387755102041,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": -1.470977628918099,
                        "high": -0.21126369327073558,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.3979591836734695,
                      "unit": "points_per_event",
                      "denominator": 98,
                      "interval": {
                        "low": 1.0287315270935962,
                        "high": 1.7973116733755032,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.08222912353347137,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.039195709684840124,
                        "high": 0.09879227053140097,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.12334368530020705,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.058793564527260186,
                        "high": 0.14818840579710144,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4489795918367347,
                      "unit": "proportion",
                      "denominator": 98,
                      "interval": {
                        "low": 0.3474553969669938,
                        "high": 0.5505732630130702,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 790,
                "basis": "selected_response_attempts",
                "source_kind": "provider_usage_or_saved_generation_metadata",
                "coverage": {
                  "observed_events": 790,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 790,
                  "population_attempts": 790
                },
                "note": "First pass includes 786 model outputs and four charged HTTP 200 length nonresponses. 785 charges came from result usage and five exact charges from hash-bound provider generation metadata.",
                "value": 0.0019925871011392403
              }
            }
          }
        },
        "birds": {
          "n_events": 400,
          "n_species": 40,
          "models": {
            "bioclip": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.3875,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.33886158349595763,
                    "high": 0.43596059113300495,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 1.0,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 1.0,
                    "high": 1.0,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.3875,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.33886158349595763,
                    "high": 0.43596059113300495,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.6125,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.5640394088669951,
                    "high": 0.6611384165040424,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.15161317024192555,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.13137522302392565,
                    "high": 0.16129822046561862,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.4889524740302099,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4236850942521602,
                    "high": 0.52018676100162,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": null,
                "raw_tis": null,
                "any_rank_conflict_rate": null
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.41975308641975306,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.36569402305825244,
                        "high": 0.4741648691048142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.41975308641975306,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.36569402305825244,
                        "high": 0.4741648691048142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5802469135802469,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.5258351308951859,
                        "high": 0.6343059769417475,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.1783428019138341,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.15185341092813434,
                        "high": 0.1896237204681453,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.49748255270701086,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.42359109364163783,
                        "high": 0.5289503781479845,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.15,
                        "high": 0.3563218390804598,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.15,
                        "high": 0.3563218390804598,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6436781609195402,
                        "high": 0.85,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.12016272418058131,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05414349489795918,
                        "high": 0.13376677713246035,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.33645562770562776,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1516017857142857,
                        "high": 0.37454697597088904,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5409090909090909,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5409090909090909,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4590909090909091,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25742574257425743,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.25742574257425743,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7425742574257426,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8846153846153846,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5833333333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5833333333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4166666666666667,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.24,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.24,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.76,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3076923076923077,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3076923076923077,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6923076923076923,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.9107142857142857,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.18181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 400,
                  "population_events": 400
                }
              }
            },
            "speciesnet": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.12,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.08644859813084112,
                    "high": 0.15625,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.2675,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.2210796915167095,
                    "high": 0.31368040140584535,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.4485981308411215,
                  "unit": "proportion",
                  "denominator": 107,
                  "interval": {
                    "low": 0.3461448598130841,
                    "high": 0.5545454545454546,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1475,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.11056378603548415,
                    "high": 0.18518518518518517,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.10598496391563618,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.08181920280091934,
                    "high": 0.12321767899743628,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.14837894948189065,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.11454688392128708,
                    "high": 0.1725047505964108,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 0.11375,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": -0.11410714285714285,
                    "high": 0.3510692853246044,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 1.13875,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 0.9665684624924107,
                    "high": 1.3190677147034429,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.205,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.1628498727735369,
                    "high": 0.24880382775119617,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.10493827160493827,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.0707395498392283,
                        "high": 0.14285714285714285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.24074074074074073,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.19314534321624235,
                        "high": 0.29070026223173,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4358974358974359,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": {
                        "low": 0.3188405797101449,
                        "high": 0.5569620253164557,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13580246913580246,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.09771940169980156,
                        "high": 0.1760797342192691,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.12808641975308643,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": -0.10691869342349221,
                        "high": 0.37199057859344353,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.007716049382716,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 0.831817480359147,
                        "high": 1.197641925501764,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.12773670773670773,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08511994134176797,
                        "high": 0.1475071928071928,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.16807461544303648,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11199992281811574,
                        "high": 0.19408841158841156,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.17592592592592593,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.13354037267080746,
                        "high": 0.22115384615384615,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18421052631578946,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.09722222222222222,
                        "high": 0.2835820895522388,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.3815789473684211,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.26582278481012656,
                        "high": 0.5,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4827586206896552,
                      "unit": "proportion",
                      "denominator": 29,
                      "interval": {
                        "low": 0.2857142857142857,
                        "high": 0.68,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19736842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.10588235294117647,
                        "high": 0.2962962962962963,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.05263157894736842,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.5915492957746479,
                        "high": 0.7105448480355816,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.6973684210526316,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.235612591431557,
                        "high": 2.1538732394366193,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.12527472527472527,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.0699538602423218,
                        "high": 0.15511197136197138,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.16285714285714287,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09094001831501831,
                        "high": 0.2016455627705628,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.32894736842105265,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.21516726943942135,
                        "high": 0.44871794871794873,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.17272727272727273,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3409090909090909,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5066666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16818181818181818,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.4772727272727273,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.5,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.0594059405940594,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.2079207920792079,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 21,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1485148514851485,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.2722772277227723,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.7673267326732673,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.05128205128205128,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.14102564102564102,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.36363636363636365,
                      "unit": "proportion",
                      "denominator": 11,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08974358974358974,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.4166666666666667,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.6089743589743589,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.15104166666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3020833333333333,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 58,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15104166666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.4244791666666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.3098958333333333,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6071428571428571,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5294117647058824,
                      "unit": "proportion",
                      "denominator": 17,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.8392857142857143,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.8035714285714284,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.04,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.17333333333333334,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.23076923076923078,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13333333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.14666666666666667,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.6533333333333333,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3076923076923077,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6346153846153846,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.0961538461538463,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.03571428571428571,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.125,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 7,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.5267857142857143,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.45535714285714285,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.18181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 4,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.13636363636363635,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.0,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 400,
                  "population_events": 400
                }
              }
            },
            "bioclip-country": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4175,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.3686746987951807,
                    "high": 0.46594142718893056,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 1.0,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 1.0,
                    "high": 1.0,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.4175,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.3686746987951807,
                    "high": 0.46594142718893056,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.5825,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.5340585728110694,
                    "high": 0.6313253012048192,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.2078819215356367,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.18175755511458577,
                    "high": 0.2205133767698909,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5197048038390918,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4543938877864644,
                    "high": 0.5512834419247273,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": null,
                "raw_tis": null,
                "any_rank_conflict_rate": null
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4444444444444444,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.39062127976190475,
                        "high": 0.49847140996147343,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4444444444444444,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.39062127976190475,
                        "high": 0.49847140996147343,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5555555555555556,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.5015285900385266,
                        "high": 0.6093787202380951,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.2387020889204657,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2043271972044104,
                        "high": 0.2530410745582378,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5276572491926084,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4516706464518546,
                        "high": 0.559353954286631,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3026315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.19480519480519481,
                        "high": 0.4142857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3026315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.19480519480519481,
                        "high": 0.4142857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6973684210526315,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5857142857142857,
                        "high": 0.8051948051948052,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.16010989756790428,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.07980530090824207,
                        "high": 0.17705178686561662,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3762582592845751,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.18754245713436893,
                        "high": 0.41607169913419906,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5727272727272728,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5727272727272728,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.42727272727272725,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2871287128712871,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2871287128712871,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7128712871287128,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.14102564102564102,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.14102564102564102,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8589743589743589,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.609375,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.609375,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.390625,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6785714285714286,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.26666666666666666,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.26666666666666666,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7333333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.10714285714285714,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.10714285714285714,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22727272727272727,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.22727272727272727,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7727272727272727,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 400,
                  "population_events": 400
                }
              }
            },
            "speciesnet-country": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.1225,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.08844842347228982,
                    "high": 0.15931545818841672,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.2025,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.16058394160583941,
                    "high": 0.24611498605021917,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.6049382716049383,
                  "unit": "proportion",
                  "denominator": 81,
                  "interval": {
                    "low": 0.4861111111111111,
                    "high": 0.7222222222222222,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.08,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.05223880597014925,
                    "high": 0.11027659075746919,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.13656107706909845,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.10622364253569397,
                    "high": 0.1583469879089808,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.15021718477600832,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.11684600678926339,
                    "high": 0.17418168669987888,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 0.315,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 0.09007572808365832,
                    "high": 0.5449877720025769,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 1.1025,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 0.9310322779786578,
                    "high": 1.2804919585299395,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.1575,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.11893171083705065,
                    "high": 0.19845360824742267,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.10802469135802469,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.0732484076433121,
                        "high": 0.14679158243922663,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.18209876543209877,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.1387057909055577,
                        "high": 0.22769306930693067,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5932203389830508,
                      "unit": "proportion",
                      "denominator": 59,
                      "interval": {
                        "low": 0.4528301886792453,
                        "high": 0.7288135593220338,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07407407407407407,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.045317220543806644,
                        "high": 0.10632183908045977,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.3179012345679013,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 0.08882289438637557,
                        "high": 0.5504439248304542,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 0.9814814814814815,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 0.8074624548067117,
                        "high": 1.1682246664315525,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.15916400672498232,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10907171247539746,
                        "high": 0.18310944622715564,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.17172958620327042,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.11768263714450776,
                        "high": 0.19756545513982582,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13271604938271606,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.09523444347063979,
                        "high": 0.17365269461077845,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18421052631578946,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.09722222222222222,
                        "high": 0.2835820895522388,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.2894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.18643082380764683,
                        "high": 0.4025974025974026,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6363636363636364,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": {
                        "low": 0.4166666666666667,
                        "high": 0.8421052631578947,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10526315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.04054054054054054,
                        "high": 0.17978909559771578,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.3026315789473684,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.3382939045183291,
                        "high": 0.9500176056338026,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 1.6184210526315792,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.174223098330241,
                        "high": 2.0696403102157244,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.14805194805194805,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.08267274392274392,
                        "high": 0.18331414797323886,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.16285714285714287,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.09094001831501831,
                        "high": 0.2016455627705628,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2631578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.15625,
                        "high": 0.3780487804878049,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.17727272727272728,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.2818181818181818,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6290322580645161,
                      "unit": "proportion",
                      "denominator": 62,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10454545454545454,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.6818181818181818,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.4545454545454546,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.0594059405940594,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.10891089108910891,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5454545454545454,
                      "unit": "proportion",
                      "denominator": 11,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.04950495049504951,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.01485148514851485,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.7277227722772277,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.05128205128205128,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.10256410256410256,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05128205128205128,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.2948717948717949,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.6025641025641025,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.15625,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.2552083333333333,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6122448979591837,
                      "unit": "proportion",
                      "denominator": 49,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09895833333333333,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5807291666666666,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.2838541666666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4642857142857143,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6923076923076923,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.375,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.625,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.04,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.08,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 6,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.04,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.14666666666666667,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.6133333333333333,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6,
                      "unit": "proportion",
                      "denominator": 5,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.4807692307692308,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.0576923076923077,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.03571428571428571,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.07142857142857142,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 4,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.03571428571428571,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.35714285714285715,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.44642857142857145,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.18181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 4,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.13636363636363635,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.0,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 400,
                  "population_events": 400
                }
              }
            },
            "gemini-minimal": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.44801980198019803,
                    "high": 0.551219512195122,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.665,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.6159600997506235,
                    "high": 0.7113924050632912,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7518796992481203,
                  "unit": "proportion",
                  "denominator": 266,
                  "interval": {
                    "low": 0.6966280130605974,
                    "high": 0.8042704626334519,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.165,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.1287122076818942,
                    "high": 0.2034313725490196,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.3868238089527153,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.3507570427397413,
                    "high": 0.4062130086997508,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.560894522981437,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5085977119726248,
                    "high": 0.5890088626146385,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.9987500000000002,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 1.6232044158201147,
                    "high": 2.360165420536899,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.22375,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.023929471032746,
                    "high": 3.4178276909722225,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.245,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.20191587193460492,
                    "high": 0.28967372113279505,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5493827160493827,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.4939015038156516,
                        "high": 0.6046511627906976,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6975308641975309,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.6470588235294118,
                        "high": 0.7461785249036835,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7876106194690266,
                      "unit": "proportion",
                      "denominator": 226,
                      "interval": {
                        "low": 0.7324561403508771,
                        "high": 0.8407079646017699,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14814814814814814,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.11038961038961038,
                        "high": 0.18688601195036003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.3425925925925926,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 1.9472764788459676,
                        "high": 2.721859417608103,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.3919753086419755,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.176378373346573,
                        "high": 3.6021732440598306,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4111462295345362,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37239530279133876,
                        "high": 0.4320253382883027,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5626211562051548,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5095935722407794,
                        "high": 0.5911925681839932,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.20987654320987653,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.1651376146788991,
                        "high": 0.256797583081571,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.18292682926829268,
                        "high": 0.4024429838454228,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5263157894736842,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.4074074074074074,
                        "high": 0.6406350160256409,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.55,
                      "unit": "proportion",
                      "denominator": 40,
                      "interval": {
                        "low": 0.3870967741935484,
                        "high": 0.7058823529411765,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14285714285714285,
                        "high": 0.33784259611724393,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.5328947368421053,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.3354498191681736,
                        "high": 1.4109790491539078,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.5065789473684212,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.097697645079899,
                        "high": 2.9262355351976854,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.21617186059376728,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10451746529332735,
                        "high": 0.24018622182415286,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.31344919786096254,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1515503246753247,
                        "high": 0.34827002164502163,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.39473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2816791373239437,
                        "high": 0.5087884806355509,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6409090909090909,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8136363636363636,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7877094972067039,
                      "unit": "proportion",
                      "denominator": 179,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17272727272727273,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.9045454545454548,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.9727272727272727,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.504950495049505,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6435643564356436,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7846153846153846,
                      "unit": "proportion",
                      "denominator": 65,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13861386138613863,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.9702970297029703,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.1584158415841586,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08974358974358974,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.2692307692307692,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 21,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1794871794871795,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.5576923076923077,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.1730769230769231,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8385416666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8198757763975155,
                      "unit": "proportion",
                      "denominator": 161,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15104166666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.2005208333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.111979166666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6428571428571429,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 18,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.875,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.017857142857143,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5333333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.68,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7843137254901961,
                      "unit": "proportion",
                      "denominator": 51,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14666666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.14,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2066666666666666,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4230769230769231,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5384615384615384,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7857142857142857,
                      "unit": "proportion",
                      "denominator": 14,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.4807692307692308,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.019230769230769,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.23214285714285715,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.38461538461538464,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.375,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.1428571428571428,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.36363636363636365,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2727272727272727,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.0227272727272727,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.25,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "saved_event_telemetry",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 10,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "779 saved telemetry estimates plus corrected selected-response result tokens at retained prices; no provider invoice. Earlier replaced calls are separate.",
                "value": 0.0017722417305069645
              }
            },
            "gemini-low": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5075,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.4556054298813513,
                    "high": 0.5578425920861516,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.6825,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.6346141975308641,
                    "high": 0.728,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7435897435897436,
                  "unit": "proportion",
                  "denominator": 273,
                  "interval": {
                    "low": 0.6885783204684589,
                    "high": 0.7952231866921071,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.175,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.13793103448275862,
                    "high": 0.21445783132530122,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.37121324271570644,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.3358122897889391,
                    "high": 0.3900793684897925,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5661001951414523,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5121137419281322,
                    "high": 0.5948710369469336,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.9912500000000002,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 1.620342585646412,
                    "high": 2.3522058396093573,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.2662500000000003,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.07001959168418,
                    "high": 3.455266093383064,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.255,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.21148711403816428,
                    "high": 0.30025445292620867,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5555555555555556,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.4984423676012461,
                        "high": 0.6117656192911948,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7160493827160493,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.6655390377031837,
                        "high": 0.7645262161297205,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7758620689655172,
                      "unit": "proportion",
                      "denominator": 232,
                      "interval": {
                        "low": 0.7194547779586142,
                        "high": 0.8305785123966942,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16049382716049382,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.12084334926529047,
                        "high": 0.20327868852459016,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.358024691358025,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 1.966961977082823,
                        "high": 2.7375443480579142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.4382716049382718,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.222213541666667,
                        "high": 3.647846747571789,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3806293538084315,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3422577334462904,
                        "high": 0.3995077819102162,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5709440307126472,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5133866001694357,
                        "high": 0.5992616728653243,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.21604938271604937,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.1710914454277286,
                        "high": 0.2625080128205127,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3026315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.19696969696969696,
                        "high": 0.41333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5394736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.4230769230769231,
                        "high": 0.6538461538461539,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5609756097560976,
                      "unit": "proportion",
                      "denominator": 41,
                      "interval": {
                        "low": 0.4,
                        "high": 0.717948717948718,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14285714285714285,
                        "high": 0.33766233766233766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.4276315789473684,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.4294117647058824,
                        "high": 1.2945293291183704,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.5328947368421058,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.114940587138863,
                        "high": 2.9629697712418297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2020428861199652,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.10126468333304835,
                        "high": 0.22335180145524974,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.29296218487394954,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1468337908329201,
                        "high": 0.3238601121101121,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.3076923076923077,
                        "high": 0.5333333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6545454545454545,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8181818181818182,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8,
                      "unit": "proportion",
                      "denominator": 180,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16363636363636364,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.8136363636363635,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.9727272727272727,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4752475247524752,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6633663366336634,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7164179104477612,
                      "unit": "proportion",
                      "denominator": 67,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18811881188118812,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.8663366336633664,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.202970297029703,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1282051282051282,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.32051282051282054,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 25,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.20512820512820512,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.3333333333333333,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.703125,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8489583333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8282208588957055,
                      "unit": "proportion",
                      "denominator": 163,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14583333333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.2005208333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.138020833333333,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6071428571428571,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5294117647058824,
                      "unit": "proportion",
                      "denominator": 17,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.16071428571428573,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.8392857142857144,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.49333333333333335,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7066666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6981132075471698,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21333333333333335,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.9933333333333334,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.26,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4230769230769231,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5384615384615384,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7857142857142857,
                      "unit": "proportion",
                      "denominator": 14,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.5,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.0384615384615383,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.125,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.26785714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4666666666666667,
                      "unit": "proportion",
                      "denominator": 15,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.08928571428571429,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.25,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.13636363636363635,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.45454545454545453,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3,
                      "unit": "proportion",
                      "denominator": 10,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.5,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.5454545454545454,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 400,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.0022678025
              }
            },
            "gemini-medium": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.55,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.4987593052109181,
                    "high": 0.601467446561282,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.7675,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.7243086776218898,
                    "high": 0.8108761853987947,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7166123778501629,
                  "unit": "proportion",
                  "denominator": 307,
                  "interval": {
                    "low": 0.6643829691425321,
                    "high": 0.7682119205298014,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.2175,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.17585301837270342,
                    "high": 0.2595641551071879,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.37299546410126566,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.3405976362492402,
                    "high": 0.3908888539057059,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5967927425620252,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5449562179987842,
                    "high": 0.6254221662491294,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.04625,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 1.6640897097625331,
                    "high": 2.425840659404469,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.44625,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.255233578021267,
                    "high": 3.6349624035989714,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.28,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.2347412100967664,
                    "high": 0.32673267326732675,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5987654320987654,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.541795665634675,
                        "high": 0.6547619047619048,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7901234567901234,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.7460802607354332,
                        "high": 0.8333333333333334,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7578125,
                      "unit": "proportion",
                      "denominator": 256,
                      "interval": {
                        "low": 0.7017543859649122,
                        "high": 0.8125,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19135802469135801,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.14733475716429706,
                        "high": 0.2367601246105919,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.3796296296296298,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 1.9649620472040952,
                        "high": 2.7861490835135414,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.5987654320987654,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.3887239225229786,
                        "high": 3.80791864538731,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.414646816983449,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.372733505192438,
                        "high": 0.4361010420009951,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6110584671335039,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5492914813362246,
                        "high": 0.6426752197909402,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.24382716049382716,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.1962025316455696,
                        "high": 0.2920661216293746,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34210526315789475,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.22972972972972974,
                        "high": 0.4597837985769018,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6710526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.548376049491825,
                        "high": 0.7875094696969696,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5098039215686274,
                      "unit": "proportion",
                      "denominator": 51,
                      "interval": {
                        "low": 0.36363636363636365,
                        "high": 0.6491228070175439,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.32894736842105265,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.22093023255813954,
                        "high": 0.4444444444444444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.625,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.29733437705998683,
                        "high": 1.5367715141612202,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.7960526315789473,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.3714125200642058,
                        "high": 3.2228962001853576,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.19913419913419914,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1034880018970928,
                        "high": 0.21906589080573632,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3285714285714286,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.17075520313020315,
                        "high": 0.361458719829465,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4342105263157895,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.3176470588235294,
                        "high": 0.5584415584415584,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6909090909090909,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7676767676767676,
                      "unit": "proportion",
                      "denominator": 198,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.20909090909090908,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.8727272727272726,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.122727272727273,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5346534653465347,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7128712871287128,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 72,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1782178217821782,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.198019801980198,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.4356435643564356,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.16666666666666666,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.46153846153846156,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3611111111111111,
                      "unit": "proportion",
                      "denominator": 36,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2948717948717949,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.5192307692307693,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.5320512820512822,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7395833333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.921875,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8022598870056498,
                      "unit": "proportion",
                      "denominator": 177,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18229166666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.1927083333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.260416666666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35714285714285715,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.47619047619047616,
                      "unit": "proportion",
                      "denominator": 21,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.39285714285714285,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.6785714285714286,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.1785714285714284,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.56,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7333333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7636363636363637,
                      "unit": "proportion",
                      "denominator": 55,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17333333333333334,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.2733333333333334,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.473333333333333,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.46153846153846156,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7058823529411765,
                      "unit": "proportion",
                      "denominator": 17,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.9807692307692308,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.326923076923077,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.16071428571428573,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4107142857142857,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.391304347826087,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.3125,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.4732142857142858,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5909090909090909,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3076923076923077,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.0454545454545454,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.6818181818181819,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 400,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.0032434425
              }
            },
            "gemini-high": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.56,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.5099502487562189,
                    "high": 0.6092575112467866,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.8075,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.7692307692307693,
                    "high": 0.8445040214477212,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.6934984520123839,
                  "unit": "proportion",
                  "denominator": 323,
                  "interval": {
                    "low": 0.6426424072975797,
                    "high": 0.7435897435897436,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.2475,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.206266318537859,
                    "high": 0.2889541525518862,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.36069291156813826,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.32860588250455736,
                    "high": 0.3773509533405161,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.6131779496658349,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5586300002577476,
                    "high": 0.6414966206788774,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.9475000000000002,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 1.5561621431314814,
                    "high": 2.3280870179689077,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.4725,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.2828629605918334,
                    "high": 3.6583794557289324,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.305,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.25925925925925924,
                    "high": 0.35189980951210514,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6018518518518519,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.5471698113207547,
                        "high": 0.6570605187319885,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8240740740740741,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.7835341579928539,
                        "high": 0.8626229405500219,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7303370786516854,
                      "unit": "proportion",
                      "denominator": 267,
                      "interval": {
                        "low": 0.67578125,
                        "high": 0.7844548105375047,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2222222222222222,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.17717717717717718,
                        "high": 0.26726797552269255,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.2978395061728394,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 1.8764981353987378,
                        "high": 2.7043360690850324,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.625,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.4187080344117824,
                        "high": 3.8276935975609754,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.39320527755125284,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.35413933447537865,
                        "high": 0.4131148520925158,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6208504382388202,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.559167370224282,
                        "high": 0.6522866085671303,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2654320987654321,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.21751105379513633,
                        "high": 0.31446540880503143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3815789473684211,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.2682926829268293,
                        "high": 0.5,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7368421052631579,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6363636363636364,
                        "high": 0.8333333333333334,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5178571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": {
                        "low": 0.38,
                        "high": 0.6538461538461539,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.35526315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.24675324675324675,
                        "high": 0.4677572478562676,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.45394736842105265,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -0.46380182002022247,
                        "high": 1.382820109577922,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.822368421052632,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.3903933319903303,
                        "high": 3.2605969148222664,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.22140615948665487,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.12458978311251037,
                        "high": 0.23829817755676638,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3653201631529805,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20557314213564212,
                        "high": 0.39319199296866447,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.47368421052631576,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.35896739130434785,
                        "high": 0.5869651483781918,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7136363636363636,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9181818181818182,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7772277227722773,
                      "unit": "proportion",
                      "denominator": 202,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.20454545454545456,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.9136363636363636,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.163636363636364,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5544554455445545,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7623762376237624,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7272727272727273,
                      "unit": "proportion",
                      "denominator": 77,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2079207920792079,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.108910891089109,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.495049504950495,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1282051282051282,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5512820512820513,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.23255813953488372,
                      "unit": "proportion",
                      "denominator": 43,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4230769230769231,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.0256410256410255,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.4743589743589745,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7552083333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9322916666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8100558659217877,
                      "unit": "proportion",
                      "denominator": 179,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17708333333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.2786458333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.294270833333333,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.42857142857142855,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8214285714285714,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5217391304347826,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.39285714285714285,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.4107142857142857,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.267857142857143,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.56,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7733333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7241379310344828,
                      "unit": "proportion",
                      "denominator": 58,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21333333333333335,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.1466666666666665,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.48,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5384615384615384,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7307692307692307,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7368421052631579,
                      "unit": "proportion",
                      "denominator": 19,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.0,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.5384615384615383,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.125,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5178571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2413793103448276,
                      "unit": "proportion",
                      "denominator": 29,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.39285714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.9107142857142857,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.5,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.13636363636363635,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6363636363636364,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 14,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.3181818181818181,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.4090909090909092,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 400,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.0040265475
              }
            },
            "gemini-medium37": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.625,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.5748651987247069,
                    "high": 0.6742424242424242,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.7425,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.6974342105263157,
                    "high": 0.7862237156180149,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8417508417508418,
                  "unit": "proportion",
                  "denominator": 297,
                  "interval": {
                    "low": 0.7979443105341033,
                    "high": 0.8819708783153405,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1175,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.08730109250682816,
                    "high": 0.15,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.5250346685202693,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4828035179866669,
                    "high": 0.5469169879771919,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.6956709357893567,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.6397146613323337,
                    "high": 0.7246650090697794,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.93375,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 2.6071387043189374,
                    "high": 3.2452200988333395,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.68375,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.4959038639978797,
                    "high": 3.864797642029785,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.15,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.11491442542787286,
                    "high": 0.1876433031047829,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6728395061728395,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.6191948214414286,
                        "high": 0.7256944444444444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7654320987654321,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.717391304347826,
                        "high": 0.8117293028322441,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8790322580645161,
                      "unit": "proportion",
                      "denominator": 248,
                      "interval": {
                        "low": 0.8352919145186165,
                        "high": 0.9181826630618453,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09259259259259259,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.0625,
                        "high": 0.1261003672293995,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.2623456790123457,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 2.920601379981677,
                        "high": 3.588061856371637,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.8487654320987654,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.6464968152866244,
                        "high": 4.045180722891566,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5778212733813252,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5153238895506652,
                        "high": 0.6001641279245447,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7298795032185161,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6509354394324193,
                        "high": 0.7581020563257408,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.11728395061728394,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.08231707317073171,
                        "high": 0.15540746539222144,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.3,
                        "high": 0.5405516171909613,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6447368421052632,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5277701465201465,
                        "high": 0.7571428571428571,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6530612244897959,
                      "unit": "proportion",
                      "denominator": 49,
                      "interval": {
                        "low": 0.5094253859348199,
                        "high": 0.7857142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.13333333333333333,
                        "high": 0.32222222222222224,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.5328947368421053,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6217350746268657,
                        "high": 2.3904583772391983,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.9802631578947367,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.513513513513514,
                        "high": 3.426670731707317,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2977832512315271,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.17351796909986567,
                        "high": 0.3246398716226302,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.43178571428571433,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2516010551948052,
                        "high": 0.47072781385281376,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.18518518518518517,
                        "high": 0.40229885057471265,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7772727272727272,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8909090909090909,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8724489795918368,
                      "unit": "proportion",
                      "denominator": 196,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11363636363636363,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.6977272727272728,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.3568181818181815,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6237623762376238,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7029702970297029,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8873239436619719,
                      "unit": "proportion",
                      "denominator": 71,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07920792079207921,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.0693069306930694,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.712871287128713,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3717948717948718,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5172413793103449,
                      "unit": "proportion",
                      "denominator": 29,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1794871794871795,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5769230769230769,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.7307692307692308,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8177083333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9010416666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9075144508670521,
                      "unit": "proportion",
                      "denominator": 173,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08333333333333333,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.9973958333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.4921875,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8214285714285714,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6086956521739131,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.6428571428571428,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.4285714285714284,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.64,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.72,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8888888888888888,
                      "unit": "proportion",
                      "denominator": 54,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.2066666666666666,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.74,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5769230769230769,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8823529411764706,
                      "unit": "proportion",
                      "denominator": 17,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.673076923076923,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.6346153846153846,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.35714285714285715,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.7857142857142857,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.7678571428571428,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.13636363636363635,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 9,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2727272727272727,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.045454545454545456,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.6363636363636365,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "saved_event_telemetry",
                "coverage": {
                  "observed_events": 400,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "Saved provider-token cost estimate for each selected response; not an invoice.",
                "value": 0.0032643324735
              }
            },
            "gemini-medium38": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.66,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.6116745108668868,
                    "high": 0.7079213704080688,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.76,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.7164547515586625,
                    "high": 0.8026370483810477,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.868421052631579,
                  "unit": "proportion",
                  "denominator": 304,
                  "interval": {
                    "low": 0.8278145695364238,
                    "high": 0.9057239057239057,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.07105263157894737,
                    "high": 0.13101604278074866,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.6194118479870407,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5731193928964311,
                    "high": 0.6434837705102915,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.7278089213847727,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.6734152866533064,
                    "high": 0.7560934303495924,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 3.16125,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 2.8486029573438874,
                    "high": 3.457500606796117,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.79875,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.6180435231023105,
                    "high": 3.975128621597893,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.1275,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.09571788413098237,
                    "high": 0.1618798955613577,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7098765432098766,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.6594410313467491,
                        "high": 0.7603557385106906,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7870370370370371,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.7405243200354283,
                        "high": 0.8317167226204518,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9019607843137255,
                      "unit": "proportion",
                      "denominator": 255,
                      "interval": {
                        "low": 0.8634682642007142,
                        "high": 0.9367088607594937,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07716049382716049,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.04985314404217027,
                        "high": 0.10714285714285714,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.483024691358025,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.161762504402959,
                        "high": 3.7821341190348763,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.9614197530864197,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.7663339149219013,
                        "high": 4.150772865853659,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.6462027392780747,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5791572090196322,
                        "high": 0.6688190750504688,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7652400859871937,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6858440633127222,
                        "high": 0.7920225888755551,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.09567901234567901,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.06432748538011696,
                        "high": 0.1298741506171846,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.32529367469879517,
                        "high": 0.5692307692307692,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6447368421052632,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5238095238095238,
                        "high": 0.7600140845070421,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6938775510204082,
                      "unit": "proportion",
                      "denominator": 49,
                      "interval": {
                        "low": 0.5476078167115903,
                        "high": 0.8292682926829268,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19736842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.10666666666666667,
                        "high": 0.3,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.7894736842105265,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.8818716397849464,
                        "high": 2.6323573706591072,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.1052631578947367,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.6384246575342467,
                        "high": 3.5507432864450124,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.34795198781266895,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.20830328272351295,
                        "high": 0.3856887374321584,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4175423853752028,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.24996393926821556,
                        "high": 0.46282648491859013,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2631578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.16216216216216217,
                        "high": 0.3736607142857138,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8181818181818182,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9090909090909091,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9,
                      "unit": "proportion",
                      "denominator": 200,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.038636363636364,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.493181818181818,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6237623762376238,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7326732673267327,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8513513513513513,
                      "unit": "proportion",
                      "denominator": 74,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10891089108910891,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.9356435643564356,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.6782178217821784,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2564102564102564,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3717948717948718,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6896551724137931,
                      "unit": "proportion",
                      "denominator": 29,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11538461538461539,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.9551282051282052,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.9807692307692308,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8697916666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9270833333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9382022471910112,
                      "unit": "proportion",
                      "denominator": 178,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.057291666666666664,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.359375,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.645833333333333,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4642857142857143,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7857142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5909090909090909,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.32142857142857145,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.8392857142857142,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.4464285714285716,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6266666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7466666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8392857142857143,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.0,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.6666666666666665,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6153846153846154,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6923076923076923,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8888888888888888,
                      "unit": "proportion",
                      "denominator": 18,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.75,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.7115384615384617,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.26785714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.35714285714285715,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.0982142857142858,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.9910714285714286,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22727272727272727,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5555555555555556,
                      "unit": "proportion",
                      "denominator": 9,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5909090909090909,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.9545454545454546,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "verified_selected_result_token_reconstruction",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 400,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "Selected-response tokens repriced with the run's retained list prices and GBP buffers; no provider invoice. Earlier failed/uncertain attempts are separate.",
                "value": 0.0076179075
              }
            },
            "gemini38-low": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.66,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.6120481927710844,
                    "high": 0.707808564231738,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.765,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.7220750394465348,
                    "high": 0.8070175438596491,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8627450980392157,
                  "unit": "proportion",
                  "denominator": 306,
                  "interval": {
                    "low": 0.8228218725054168,
                    "high": 0.8993294038176589,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.105,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.07673267326732673,
                    "high": 0.1358974358974359,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.5410575647262954,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5022617742238907,
                    "high": 0.5616383998065115,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.7304277123804986,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.6780533952022526,
                    "high": 0.7582118397387907,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 3.145,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 2.8400485131048385,
                    "high": 3.4411764705882355,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.7825000000000006,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.604438241350774,
                    "high": 3.9582335176085177,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.1275,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.09660574412532637,
                    "high": 0.1605287019361876,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7129629629629629,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.6615372670807453,
                        "high": 0.7628578792341679,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7870370370370371,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.7412082962342219,
                        "high": 0.8317460317460318,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9058823529411765,
                      "unit": "proportion",
                      "denominator": 255,
                      "interval": {
                        "low": 0.8686411135120535,
                        "high": 0.9395161290322581,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07407407407407407,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.04745461839334069,
                        "high": 0.10377440099236142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.4984567901234573,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.180839280904929,
                        "high": 3.7940342874044415,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.9614197530864197,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.766561383625463,
                        "high": 4.1507468427095295,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5983094983387549,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5386304688811944,
                        "high": 0.6184960786156588,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.771504353121026,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.6945498151362771,
                        "high": 0.7975344171622969,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.09259259259259259,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.0626956908248725,
                        "high": 0.12538226299694188,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4342105263157895,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.313953488372093,
                        "high": 0.5555555555555556,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6710526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5569620253164557,
                        "high": 0.7812589798850573,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6470588235294118,
                      "unit": "proportion",
                      "denominator": 51,
                      "interval": {
                        "low": 0.5068661971830988,
                        "high": 0.7777777777777778,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.14544721407624633,
                        "high": 0.3384615384615385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.6381578947368423,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 0.7714199655765922,
                        "high": 2.4687690548780488,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0197368421052637,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.5579640468227427,
                        "high": 3.466688311688312,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3142857142857143,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19118086797008363,
                        "high": 0.34836276120998344,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.42428571428571427,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2580941717596129,
                        "high": 0.4702897276334776,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.27631578947368424,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.1794871794871795,
                        "high": 0.3823620188816266,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8227272727272728,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9045454545454545,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9095477386934674,
                      "unit": "proportion",
                      "denominator": 199,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08181818181818182,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.056818181818182,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.488636363636363,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6633663366336634,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7425742574257426,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8933333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07920792079207921,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.3069306930693068,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.801980198019802,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3974358974358974,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4838709677419355,
                      "unit": "proportion",
                      "denominator": 31,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.20512820512820512,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.33974358974358976,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.75,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8645833333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9166666666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9431818181818182,
                      "unit": "proportion",
                      "denominator": 176,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.052083333333333336,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.333333333333333,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.619791666666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5357142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8214285714285714,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6521739130434783,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.1607142857142856,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.5892857142857144,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.68,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.76,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8947368421052632,
                      "unit": "proportion",
                      "denominator": 57,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.36,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.8266666666666667,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6153846153846154,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6923076923076923,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8888888888888888,
                      "unit": "proportion",
                      "denominator": 18,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.1538461538461537,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.730769230769231,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.23214285714285715,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6190476190476191,
                      "unit": "proportion",
                      "denominator": 21,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.7946428571428571,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.8660714285714286,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.45454545454545453,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2,
                      "unit": "proportion",
                      "denominator": 10,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.36363636363636365,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.8181818181818182,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.4545454545454546,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_usage_ledger_plus_selected_canary",
                "coverage": {
                  "observed_events": 399,
                  "estimated_events": 1,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "789 saved reconciled selected-response costs plus one prior canary cost reconstructed from its hash-bound result tokens and retained price/FX buffer; earlier failed or uncertain attempts remain separate. Not an invoice.",
                "value": 0.00642868
              }
            },
            "gemini38-high": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.7,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.6542546575632476,
                    "high": 0.744899247577819,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.8475,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.8102564102564103,
                    "high": 0.8836633663366337,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8259587020648967,
                  "unit": "proportion",
                  "denominator": 339,
                  "interval": {
                    "low": 0.7840218423551757,
                    "high": 0.8647906357061286,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1475,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.11369509043927649,
                    "high": 0.18357607796597977,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.5473418155271824,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5095120118424594,
                    "high": 0.56817361341227,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.7525949963498759,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.7005790162833817,
                    "high": 0.7812387184418713,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 3.1887500000000006,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 2.8758883770375623,
                    "high": 3.4824826208607362,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.97625,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 3.8129860786695318,
                    "high": 4.135311963058419,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.1575,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.12339067507131035,
                    "high": 0.19444444444444445,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7376543209876543,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.6881347757370975,
                        "high": 0.7850746268656716,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8611111111111112,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.8207540563395169,
                        "high": 0.8993906949054697,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8566308243727598,
                      "unit": "proportion",
                      "denominator": 279,
                      "interval": {
                        "low": 0.8136185529433206,
                        "high": 0.896797153024911,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12345679012345678,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.0884124701854274,
                        "high": 0.16088418724391745,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.4413580246913584,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.10647095959596,
                        "high": 3.749222217627436,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.104938271604938,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.9262788461538465,
                        "high": 4.274697185856551,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5835279762023525,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.5304215943957684,
                        "high": 0.6049760598081002,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7831559680610518,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.7118816135311629,
                        "high": 0.8119415539529766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13271604938271606,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.09687477742165243,
                        "high": 0.17251485342749845,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5394736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.41975308641975306,
                        "high": 0.654320987654321,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.6933333333333334,
                        "high": 0.8809523809523809,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6833333333333333,
                      "unit": "proportion",
                      "denominator": 60,
                      "interval": {
                        "low": 0.5535714285714286,
                        "high": 0.8028169014084507,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.15384615384615385,
                        "high": 0.35714285714285715,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.111842105263158,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.2222222222222223,
                        "high": 2.91783176369863,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.4276315789473686,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 2.983328651685394,
                        "high": 3.845085637280473,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3525867136978248,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.23130200817700816,
                        "high": 0.3825755571588904,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.47599206349206347,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.312257711038961,
                        "high": 0.5164770021645022,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2631578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.1684925093632959,
                        "high": 0.36709506587445095,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.85,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9363636363636364,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9077669902912622,
                      "unit": "proportion",
                      "denominator": 206,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08636363636363636,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.156818181818182,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.611363636363636,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7128712871287128,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8415841584158416,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8470588235294118,
                      "unit": "proportion",
                      "denominator": 85,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12871287128712872,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.366336633663366,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.00990099009901,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2564102564102564,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6025641025641025,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.425531914893617,
                      "unit": "proportion",
                      "denominator": 47,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.20512820512820512,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.128205128205128,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8802083333333334,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9479166666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9285714285714286,
                      "unit": "proportion",
                      "denominator": 182,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.06770833333333333,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.3515625,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.690104166666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6428571428571429,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8571428571428571,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 24,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.8214285714285716,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.071428571428571,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6933333333333334,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.84,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8253968253968254,
                      "unit": "proportion",
                      "denominator": 63,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14666666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.2,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.933333333333333,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8461538461538461,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.8461538461538463,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.230769230769231,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.30357142857142855,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5892857142857143,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5151515151515151,
                      "unit": "proportion",
                      "denominator": 33,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2857142857142857,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.6160714285714286,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.3125,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.13636363636363635,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6363636363636364,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 14,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.8409090909090909,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.6590909090909092,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_usage_ledger_plus_selected_canary",
                "coverage": {
                  "observed_events": 399,
                  "estimated_events": 1,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "789 saved reconciled selected-response costs plus one prior canary cost reconstructed from its hash-bound result tokens and retained price/FX buffer; earlier failed or uncertain attempts remain separate. Not an invoice.",
                "value": 0.012317345
              }
            },
            "qwen-repaired": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.455,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.40540540540540543,
                    "high": 0.50495111078272,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.875,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.8414322250639387,
                    "high": 0.9066994942608523,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.52,
                  "unit": "proportion",
                  "denominator": 350,
                  "interval": {
                    "low": 0.4668587896253602,
                    "high": 0.5738185781760929,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.42,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.36939313984168864,
                    "high": 0.469334693877551,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.2042805257464443,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.1858802338328159,
                    "high": 0.21486251008167706,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.4953802749351276,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4507595670445786,
                    "high": 0.5210415869480669,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 0.7425000000000002,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 0.3414286390077459,
                    "high": 1.1350294478970948,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.005,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 2.7986644549763033,
                    "high": 3.2057441653481007,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.4525,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.40245962672855623,
                    "high": 0.5024878731343284,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5092592592592593,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.45364238410596025,
                        "high": 0.5641891891891891,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8734567901234568,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.8368564371968766,
                        "high": 0.9093569459903535,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5830388692579506,
                      "unit": "proportion",
                      "denominator": 283,
                      "interval": {
                        "low": 0.5243055555555556,
                        "high": 0.6411149825783972,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.36419753086419754,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.3111111111111111,
                        "high": 0.4171789298338276,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.2098765432098766,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 0.7786121836338644,
                        "high": 1.6238953927492443,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.231481481481482,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.0063086522627622,
                        "high": 3.4471302330599913,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2564240868637153,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2322738089112476,
                        "high": 0.2707035669158167,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5061001714415434,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4584351491669362,
                        "high": 0.5342833557549014,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.404320987654321,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.3501577287066246,
                        "high": 0.4591564188757925,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.1323529411764706,
                        "high": 0.32142857142857145,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.881578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.8028169014084507,
                        "high": 0.95,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2537313432835821,
                      "unit": "proportion",
                      "denominator": 67,
                      "interval": {
                        "low": 0.1506849315068493,
                        "high": 0.36363636363636365,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6578947368421053,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5443037974683544,
                        "high": 0.7666666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -1.25,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -2.073346590909091,
                        "high": -0.3856728778467915,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.039473684210526,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.6249663978494624,
                        "high": 2.4743589743589745,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.10329004329004329,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05166590909090908,
                        "high": 0.11227152406417111,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2582251082251082,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.12916477272727273,
                        "high": 0.28067881016042784,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.6578947368421053,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5443037974683544,
                        "high": 0.7666666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6272727272727273,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9954545454545455,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6301369863013698,
                      "unit": "proportion",
                      "denominator": 219,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.36818181818181817,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.7954545454545454,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.840909090909091,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3564356435643564,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.900990099009901,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3956043956043956,
                      "unit": "proportion",
                      "denominator": 91,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5445544554455446,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.07920792079207921,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.742574257425743,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08974358974358974,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.1794871794871795,
                      "unit": "proportion",
                      "denominator": 39,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.41025641025641024,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.2179487179487178,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.9615384615384616,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9947916666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6910994764397905,
                      "unit": "proportion",
                      "denominator": 191,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3072916666666667,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.3020833333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.072916666666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7857142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.6785714285714286,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.25,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.36,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8666666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4153846153846154,
                      "unit": "proportion",
                      "denominator": 65,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5066666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.12,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.7866666666666666,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6538461538461539,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.6153846153846154,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4642857142857143,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.1428571428571428,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.9107142857142857,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5909090909090909,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.15384615384615385,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.4090909090909092,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.0909090909090908,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "provider_usage_or_saved_generation_metadata",
                "coverage": {
                  "observed_events": 400,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "Selected evaluated response attempt only: 786 first-pass outputs, four successful repair outputs. Four superseded first-pass length charges are excluded from this comparison and retained in the audit below.",
                "value": 0.0021851117630625
              }
            },
            "qwen-first-pass": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.455,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.40540540540540543,
                    "high": 0.50495111078272,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.875,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.8414322250639387,
                    "high": 0.9066994942608523,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.52,
                  "unit": "proportion",
                  "denominator": 350,
                  "interval": {
                    "low": 0.4668587896253602,
                    "high": 0.5738185781760929,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.42,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.36939313984168864,
                    "high": 0.469334693877551,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.2042805257464443,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.1858802338328159,
                    "high": 0.21486251008167706,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.4953802749351276,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4507595670445786,
                    "high": 0.5210415869480669,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 0.7425000000000002,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 0.3414286390077459,
                    "high": 1.1350294478970948,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.005,
                  "unit": "points_per_event",
                  "denominator": 400,
                  "interval": {
                    "low": 2.7986644549763033,
                    "high": 3.2057441653481007,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.4525,
                  "unit": "proportion",
                  "denominator": 400,
                  "interval": {
                    "low": 0.40245962672855623,
                    "high": 0.5024878731343284,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 324,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5092592592592593,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.45364238410596025,
                        "high": 0.5641891891891891,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8734567901234568,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.8368564371968766,
                        "high": 0.9093569459903535,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5830388692579506,
                      "unit": "proportion",
                      "denominator": 283,
                      "interval": {
                        "low": 0.5243055555555556,
                        "high": 0.6411149825783972,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.36419753086419754,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.3111111111111111,
                        "high": 0.4171789298338276,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.2098765432098766,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 0.7786121836338644,
                        "high": 1.6238953927492443,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.231481481481482,
                      "unit": "points_per_event",
                      "denominator": 324,
                      "interval": {
                        "low": 3.0063086522627622,
                        "high": 3.4471302330599913,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2564240868637153,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2322738089112476,
                        "high": 0.2707035669158167,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5061001714415434,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4584351491669362,
                        "high": 0.5342833557549014,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.404320987654321,
                      "unit": "proportion",
                      "denominator": 324,
                      "interval": {
                        "low": 0.3501577287066246,
                        "high": 0.4591564188757925,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 76,
                  "n_species": 20,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.1323529411764706,
                        "high": 0.32142857142857145,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.881578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.8028169014084507,
                        "high": 0.95,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2537313432835821,
                      "unit": "proportion",
                      "denominator": 67,
                      "interval": {
                        "low": 0.1506849315068493,
                        "high": 0.36363636363636365,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6578947368421053,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5443037974683544,
                        "high": 0.7666666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -1.25,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": -2.073346590909091,
                        "high": -0.3856728778467915,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.039473684210526,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": {
                        "low": 1.6249663978494624,
                        "high": 2.4743589743589745,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.10329004329004329,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.05166590909090908,
                        "high": 0.11227152406417111,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.2582251082251082,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.12916477272727273,
                        "high": 0.28067881016042784,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.6578947368421053,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": {
                        "low": 0.5443037974683544,
                        "high": 0.7666666666666667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 220,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6272727272727273,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9954545454545455,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6301369863013698,
                      "unit": "proportion",
                      "denominator": 219,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.36818181818181817,
                      "unit": "proportion",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.7954545454545454,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.840909090909091,
                      "unit": "points_per_event",
                      "denominator": 220,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 101,
                  "n_species": 40,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3564356435643564,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.900990099009901,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3956043956043956,
                      "unit": "proportion",
                      "denominator": 91,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5445544554455446,
                      "unit": "proportion",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.07920792079207921,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.742574257425743,
                      "unit": "points_per_event",
                      "denominator": 101,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 78,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08974358974358974,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.1794871794871795,
                      "unit": "proportion",
                      "denominator": 39,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.41025641025641024,
                      "unit": "proportion",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.2179487179487178,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.9615384615384616,
                      "unit": "points_per_event",
                      "denominator": 78,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 192,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9947916666666666,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6910994764397905,
                      "unit": "proportion",
                      "denominator": 191,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3072916666666667,
                      "unit": "proportion",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.3020833333333335,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.072916666666667,
                      "unit": "points_per_event",
                      "denominator": 192,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 28,
                  "n_species": 11,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.21428571428571427,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7857142857142857,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.6785714285714286,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.25,
                      "unit": "points_per_event",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 75,
                  "n_species": 34,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.36,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8666666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4153846153846154,
                      "unit": "proportion",
                      "denominator": 65,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5066666666666667,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.12,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.7866666666666666,
                      "unit": "points_per_event",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 26,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.34615384615384615,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6538461538461539,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6538461538461539,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.6153846153846154,
                      "unit": "points_per_event",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 56,
                  "n_species": 31,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.08928571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4642857142857143,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.19230769230769232,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.1428571428571428,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.9107142857142857,
                      "unit": "points_per_event",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 22,
                  "n_species": 14,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.09090909090909091,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5909090909090909,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.15384615384615385,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.4090909090909092,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.0909090909090908,
                      "unit": "points_per_event",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 400,
                "basis": "selected_response_attempts",
                "source_kind": "provider_usage_or_saved_generation_metadata",
                "coverage": {
                  "observed_events": 400,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 400,
                  "population_attempts": 400
                },
                "note": "First pass includes 786 model outputs and four charged HTTP 200 length nonresponses. 785 charges came from result usage and five exact charges from hash-bound provider generation metadata.",
                "value": 0.0021851117630625
              }
            }
          }
        },
        "mammals": {
          "n_events": 390,
          "n_species": 39,
          "models": {
            "bioclip": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4282051282051282,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3753351206434316,
                    "high": 0.48052038239538236,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 1.0,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 1.0,
                    "high": 1.0,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.4282051282051282,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3753351206434316,
                    "high": 0.48052038239538236,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.5717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5194796176046176,
                    "high": 0.6246648793565683,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.21273344644357361,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.18372988850400115,
                    "high": 0.22799181345311462,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.45274041166196427,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.39101489091877173,
                    "high": 0.4852133465797054,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": null,
                "raw_tis": null,
                "any_rank_conflict_rate": null
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.47619047619047616,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.3734820233314209,
                        "high": 0.5798319327731093,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.47619047619047616,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.3734820233314209,
                        "high": 0.5798319327731093,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5238095238095238,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.42016806722689076,
                        "high": 0.6265179766685789,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.29517698525795694,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1912797510223981,
                        "high": 0.3149832404437667,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.47966260104417996,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.31082959541139693,
                        "high": 0.511847765721121,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4105263157894737,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.35116993030202454,
                        "high": 0.4709946780817955,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4105263157894737,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.35116993030202454,
                        "high": 0.4709946780817955,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5894736842105263,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.5290053219182044,
                        "high": 0.6488300696979754,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.22266362223470063,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.17962296574783346,
                        "high": 0.23753944284421868,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.41107130258713964,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3316116290729233,
                        "high": 0.43853435602009605,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6129032258064516,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6129032258064516,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3870967741935484,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.38848920863309355,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.38848920863309355,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6115107913669064,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1875,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.1875,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8125,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6226415094339622,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6226415094339622,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.37735849056603776,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6078431372549019,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6078431372549019,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.39215686274509803,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35514018691588783,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.35514018691588783,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6448598130841121,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.95,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7763157894736842,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 390,
                  "population_events": 390
                }
              }
            },
            "speciesnet": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4358974358974359,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3842239185750636,
                    "high": 0.48894365862597905,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.7076923076923077,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.6601458407736052,
                    "high": 0.753695500330858,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.6159420289855072,
                  "unit": "proportion",
                  "denominator": 276,
                  "interval": {
                    "low": 0.5546875,
                    "high": 0.6772911898505269,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.2717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.22538860103626943,
                    "high": 0.318769888140068,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.2985159203125846,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.2703997825889327,
                    "high": 0.3164523007682581,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.4745637707533396,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.429866321038816,
                    "high": 0.5030780166059488,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.1538461538461537,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 0.7627304656491979,
                    "high": 1.5520725201349208,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 2.833333333333334,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.622361909755108,
                    "high": 3.0442858573717952,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.33589743589743587,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.28690591868022713,
                    "high": 0.38536646802236285,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5047619047619047,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.4077669902912621,
                        "high": 0.6018540992448759,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7333333333333333,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.6463287601626017,
                        "high": 0.8151283339422725,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6883116883116883,
                      "unit": "proportion",
                      "denominator": 77,
                      "interval": {
                        "low": 0.5769230769230769,
                        "high": 0.7957044107965766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.22857142857142856,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.14678305915849416,
                        "high": 0.3148148148148148,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.8142857142857143,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 1.068133899233297,
                        "high": 2.5385135135135126,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.1476190476190475,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.7521195652173915,
                        "high": 3.5309803650442477,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4264029373404373,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3022962184873949,
                        "high": 0.4499582747350258,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5685372497872497,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4030616246498599,
                        "high": 0.5999443663133679,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.26666666666666666,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.18,
                        "high": 0.354846246608381,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4105263157894737,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.35018050541516244,
                        "high": 0.4725274725274725,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6982456140350877,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.6420620574448987,
                        "high": 0.7526501766784452,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5879396984924623,
                      "unit": "proportion",
                      "denominator": 199,
                      "interval": {
                        "low": 0.5149962871287129,
                        "high": 0.6614583333333334,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.28771929824561404,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.23371548761701624,
                        "high": 0.3448303142732431,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.9105263157894739,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 0.4514911951036187,
                        "high": 1.3721405388087147,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.7175438596491226,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.476114513814917,
                        "high": 2.966105898574119,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.31125373401485507,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.26469666469774217,
                        "high": 0.32501670601740834,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.45490930356017284,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3868643560967001,
                        "high": 0.47502441648698135,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.36140350877192984,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.3028161999887773,
                        "high": 0.4204946996466431,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5419354838709678,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7548387096774194,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.717948717948718,
                      "unit": "proportion",
                      "denominator": 117,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2129032258064516,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.9258064516129032,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2483870967741937,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3669064748201439,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7338129496402878,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3669064748201439,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.43884892086330934,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.6690647482014387,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3645833333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.59375,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6140350877192983,
                      "unit": "proportion",
                      "denominator": 57,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.22916666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.9427083333333334,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.4010416666666665,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6037735849056604,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7924528301886793,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7619047619047619,
                      "unit": "proportion",
                      "denominator": 42,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18867924528301888,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.5377358490566038,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.5754716981132075,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5098039215686274,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7352941176470589,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6933333333333334,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.22549019607843138,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.607843137254902,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.0784313725490198,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6363636363636364,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.3125,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.875,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34579439252336447,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7476635514018691,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4625,
                      "unit": "proportion",
                      "denominator": 80,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.40186915887850466,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.17757009345794392,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.607476635514019,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.65,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5384615384615384,
                      "unit": "proportion",
                      "denominator": 13,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.7,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.45,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5789473684210527,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6363636363636364,
                      "unit": "proportion",
                      "denominator": 44,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21052631578947367,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.006578947368421,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.388157894736842,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 390,
                  "population_events": 390
                }
              }
            },
            "bioclip-country": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.46153846153846156,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4083742576132689,
                    "high": 0.5144375999833795,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 1.0,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 1.0,
                    "high": 1.0,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.46153846153846156,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4083742576132689,
                    "high": 0.5144375999833795,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.5384615384615384,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4855624000166206,
                    "high": 0.591625742386731,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.2874866981663751,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.2507405535490119,
                    "high": 0.3077505863215063,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.4644015893456829,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.40504243265609624,
                    "high": 0.4971355625193564,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": null,
                "raw_tis": null,
                "any_rank_conflict_rate": null
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5047619047619047,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.4019607843137255,
                        "high": 0.6068376068376068,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5047619047619047,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.4019607843137255,
                        "high": 0.6068376068376068,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.49523809523809526,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.39316239316239315,
                        "high": 0.5980392156862745,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.3514661528947243,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.22981918280132563,
                        "high": 0.37307821741136643,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5125548063048063,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.33515297491859997,
                        "high": 0.544072400391576,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4456140350877193,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.38571428571428573,
                        "high": 0.5076335877862596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 1.0,
                        "high": 1.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4456140350877193,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.38571428571428573,
                        "high": 0.5076335877862596,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5543859649122806,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.49236641221374045,
                        "high": 0.6142857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "macro_f1": {
                      "value": 0.3042640295820698,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2486817262463597,
                        "high": 0.32480684543414234,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4212886563444043,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.344328544033421,
                        "high": 0.4497325552165048,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6451612903225806,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6451612903225806,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3548387096774194,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.43884892086330934,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.43884892086330934,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5611510791366906,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.19791666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.19791666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.8020833333333334,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.660377358490566,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.660377358490566,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.33962264150943394,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6372549019607843,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6372549019607843,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3627450980392157,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.53125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.53125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.46875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.411214953271028,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.411214953271028,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5887850467289719,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.95,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.23684210526315788,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.7631578947368421,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 390,
                  "population_events": 390
                }
              }
            },
            "speciesnet-country": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.44871794871794873,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3969839656296751,
                    "high": 0.502591010902418,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.49743589743589745,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.44472953216374267,
                    "high": 0.551076441959617,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.9020618556701031,
                  "unit": "proportion",
                  "denominator": 194,
                  "interval": {
                    "low": 0.8502355072463769,
                    "high": 0.9476792406039898,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.04871794871794872,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.02564102564102564,
                    "high": 0.07653061224489796,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.4789549804666704,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.43602614790514854,
                    "high": 0.5059200301212662,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.4912358774017132,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4472063055437421,
                    "high": 0.518892338585914,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.0576923076923075,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 1.7278996532573665,
                    "high": 2.3867222222222213,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 2.878205128205128,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.6675257731958766,
                    "high": 3.0911420094936712,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.1641025641025641,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.12698412698412698,
                    "high": 0.20308483290488433,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5333333333333333,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.43636363636363634,
                        "high": 0.6296296296296297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5081967213114754,
                        "high": 0.6907263355201498,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8888888888888888,
                      "unit": "proportion",
                      "denominator": 63,
                      "interval": {
                        "low": 0.8059597844112769,
                        "high": 0.9622641509433962,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.06666666666666667,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.021505376344086023,
                        "high": 0.11764705882352941,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.5952380952380953,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 1.968412653975364,
                        "high": 3.178579619921363,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.2619047619047623,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.8734004602991945,
                        "high": 3.6416789215686274,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5720583860583861,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4137630670771644,
                        "high": 0.6001817644229408,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5958941521441522,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4310031948720463,
                        "high": 0.6251893379405634,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13333333333333333,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.07272727272727272,
                        "high": 0.2,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.41754385964912283,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.35815602836879434,
                        "high": 0.48,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.45964912280701753,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.39855072463768115,
                        "high": 0.5226480836236934,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9083969465648855,
                      "unit": "proportion",
                      "denominator": 131,
                      "interval": {
                        "low": 0.847995652173913,
                        "high": 0.9600066225165561,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.042105263157894736,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.01818181818181818,
                        "high": 0.07063197026022305,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.8596491228070176,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 1.477178362177449,
                        "high": 2.2422693606840016,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.736842105263158,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.49119640546459,
                        "high": 2.989547038327526,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4632091458681123,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.39594358076506175,
                        "high": 0.48295268414449993,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4632091458681123,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.39594358076506175,
                        "high": 0.48295268414449993,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.17543859649122806,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.1310344827586207,
                        "high": 0.22183098591549297,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.567741935483871,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5935483870967742,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9565217391304348,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.025806451612903226,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.825806451612903,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.3419354838709676,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.37410071942446044,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4748201438848921,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7878787878787878,
                      "unit": "proportion",
                      "denominator": 66,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10071942446043165,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.4424460431654675,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.7014388489208634,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3645833333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9722222222222222,
                      "unit": "proportion",
                      "denominator": 36,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.010416666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.7083333333333333,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.3854166666666665,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.660377358490566,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7169811320754716,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9210526315789473,
                      "unit": "proportion",
                      "denominator": 38,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05660377358490566,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.3962264150943398,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.7735849056603774,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5196078431372549,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5294117647058824,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9814814814814815,
                      "unit": "proportion",
                      "denominator": 54,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.00980392156862745,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.5294117647058822,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.1176470588235294,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.53125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8235294117647058,
                      "unit": "proportion",
                      "denominator": 17,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.984375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.921875,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35514018691588783,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.45794392523364486,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7755102040816326,
                      "unit": "proportion",
                      "denominator": 49,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.102803738317757,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.280373831775701,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.635514018691589,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.875,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.45,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.45,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 1.0,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.7763157894736843,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.3684210526315788,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "unavailable",
                "reason": "Local compute cost was not measured; zero is not assumed",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 0,
                  "missing_events": 390,
                  "population_events": 390
                }
              }
            },
            "gemini-minimal": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.40512820512820513,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.35426873953098825,
                    "high": 0.45641439205955325,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.5538461538461539,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5013546502757252,
                    "high": 0.6058201058201058,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7314814814814815,
                  "unit": "proportion",
                  "denominator": 216,
                  "interval": {
                    "low": 0.6698113207547169,
                    "high": 0.7903950925348304,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.14871794871794872,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.11398963730569948,
                    "high": 0.18414322250639387,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.37517134480551423,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.33174762079806003,
                    "high": 0.39839433383733197,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.47136912552487686,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4168111133103831,
                    "high": 0.5005467271289555,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.7166666666666666,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 1.3637564484126985,
                    "high": 2.060849787069299,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 2.9602564102564104,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.766413892865665,
                    "high": 3.1554586862609466,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.24871794871794872,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.2040289735203078,
                    "high": 0.2948717948717949,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5047619047619047,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.39795644064077246,
                        "high": 0.6100097402597401,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6285714285714286,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5306122448979592,
                        "high": 0.7265151515151512,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.803030303030303,
                      "unit": "proportion",
                      "denominator": 66,
                      "interval": {
                        "low": 0.6956521739130435,
                        "high": 0.8947368421052632,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12380952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.0660377358490566,
                        "high": 0.18919630156472253,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.219047619047619,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 1.491430817610063,
                        "high": 2.914288931788932,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.3619047619047624,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.988764044943821,
                        "high": 3.730477484472049,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5040959403704502,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.348370453769799,
                        "high": 0.5282053308562971,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5671079329167564,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.39191676049102386,
                        "high": 0.5942309972133342,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.22857142857142856,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.14034677816035415,
                        "high": 0.32075471698113206,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.3103448275862069,
                        "high": 0.42758620689655175,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5263157894736842,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.4659854802135908,
                        "high": 0.5860927152317881,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7,
                      "unit": "proportion",
                      "denominator": 150,
                      "interval": {
                        "low": 0.625,
                        "high": 0.7730496453900709,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15789473684210525,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.11610079791564892,
                        "high": 0.20074384645713642,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.5315789473684214,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 1.1333281733746128,
                        "high": 1.9199306049822065,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.812280701754386,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.581039623536856,
                        "high": 3.0439212980586223,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.32005562008879124,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.256300135050008,
                        "high": 0.33979984290336984,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.40212116370130185,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3220181183961639,
                        "high": 0.4269280077503878,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.256140350877193,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.20512820512820512,
                        "high": 0.30827577418121566,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6129032258064516,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7548387096774194,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.811965811965812,
                      "unit": "proportion",
                      "denominator": 117,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14193548387096774,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.706451612903226,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.835483870967742,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.34532374100719426,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5323741007194245,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6486486486486487,
                      "unit": "proportion",
                      "denominator": 74,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18705035971223022,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.4460431654676258,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.920863309352518,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.15625,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.2604166666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6,
                      "unit": "proportion",
                      "denominator": 25,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10416666666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5104166666666666,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.6041666666666667,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6981132075471698,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8679245283018868,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8043478260869565,
                      "unit": "proportion",
                      "denominator": 46,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16981132075471697,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.9622641509433962,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.09433962264151,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5686274509803921,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.696078431372549,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8169014084507042,
                      "unit": "proportion",
                      "denominator": 71,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12745098039215685,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.573529411764706,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.700980392156863,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.53125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8235294117647058,
                      "unit": "proportion",
                      "denominator": 17,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.78125,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.1875,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3177570093457944,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5327102803738317,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5964912280701754,
                      "unit": "proportion",
                      "denominator": 57,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.21495327102803738,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.3457943925233644,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.8411214953271027,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.15,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6666666666666666,
                      "unit": "proportion",
                      "denominator": 3,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.95,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.7,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.17105263157894737,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.2894736842105263,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5909090909090909,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.39473684210526316,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.5789473684210527,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "saved_event_telemetry",
                "coverage": {
                  "observed_events": 389,
                  "estimated_events": 1,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "779 saved telemetry estimates plus corrected selected-response result tokens at retained prices; no provider invoice. Earlier replaced calls are separate.",
                "value": 0.0017219806764554482
              }
            },
            "gemini-low": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4230769230769231,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3700980392156863,
                    "high": 0.4755784061696658,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.5717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5187032418952618,
                    "high": 0.6246913580246913,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7399103139013453,
                  "unit": "proportion",
                  "denominator": 223,
                  "interval": {
                    "low": 0.6778242677824268,
                    "high": 0.8,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.14871794871794872,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.11315789473684211,
                    "high": 0.18666751918158567,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.40579330224116467,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.3570427674279906,
                    "high": 0.43146331137058513,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.48903295398294216,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4302823094645015,
                    "high": 0.519968606010705,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.6935897435897436,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 1.3286006978496425,
                    "high": 2.056044911012236,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.03974358974359,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.839549072522505,
                    "high": 3.2370486506030867,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.2692307692307692,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.225,
                    "high": 0.3153877005347593,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5333333333333333,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.42341476091476093,
                        "high": 0.6403508771929824,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6476190476190476,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5392156862745098,
                        "high": 0.75,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8235294117647058,
                      "unit": "proportion",
                      "denominator": 68,
                      "interval": {
                        "low": 0.7236842105263158,
                        "high": 0.9102564102564102,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11428571428571428,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.05714285714285714,
                        "high": 0.18085106382978725,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.357142857142857,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 1.623808345120226,
                        "high": 3.0416776315789473,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.5,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.132181197137281,
                        "high": 3.857142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4924320898177107,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.345452240660574,
                        "high": 0.5191738970905637,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5539861010449245,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.38863377074314576,
                        "high": 0.5840706342268842,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.22857142857142856,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.14423076923076922,
                        "high": 0.3191489361702128,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3824561403508772,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.3231292517006803,
                        "high": 0.4416961130742049,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.543859649122807,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.48288940830251575,
                        "high": 0.6043997827267789,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7032258064516129,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": {
                        "low": 0.625,
                        "high": 0.7777777777777778,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16140350877192983,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.11787038294405214,
                        "high": 0.2080568099598313,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.449122807017544,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 1.0272598054119795,
                        "high": 1.8647270937941791,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.8701754385964913,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.637863336415842,
                        "high": 3.0970721857647656,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.351231095671801,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2783211840908452,
                        "high": 0.3743689624053126,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.41427257438212434,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3282762684148431,
                        "high": 0.44156339155498414,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.28421052631578947,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.2318840579710145,
                        "high": 0.3381308044473512,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6258064516129033,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7677419354838709,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8151260504201681,
                      "unit": "proportion",
                      "denominator": 119,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14193548387096774,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.764516129032258,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.893548387096774,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35251798561151076,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5539568345323741,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6363636363636364,
                      "unit": "proportion",
                      "denominator": 77,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2014388489208633,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.2841726618705036,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.9748201438848922,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.19791666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.28125,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7037037037037037,
                      "unit": "proportion",
                      "denominator": 27,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08333333333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5572916666666666,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.7552083333333333,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7169811320754716,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8113207547169812,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8837209302325582,
                      "unit": "proportion",
                      "denominator": 43,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09433962264150944,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.339622641509434,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.188679245283019,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5784313725490197,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7450980392156863,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7763157894736842,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16666666666666666,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.465686274509804,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.7401960784313726,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7272727272727273,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.765625,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.484375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.308411214953271,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.514018691588785,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6,
                      "unit": "proportion",
                      "denominator": 55,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.205607476635514,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.1401869158878504,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.822429906542056,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.1,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.15,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6666666666666666,
                      "unit": "proportion",
                      "denominator": 3,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.7,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.7,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3157894736842105,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7083333333333334,
                      "unit": "proportion",
                      "denominator": 24,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5197368421052632,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.769736842105263,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.0022184179487179487
              }
            },
            "gemini-medium": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.45897435897435895,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4067335979221605,
                    "high": 0.5117508891675746,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.6102564102564103,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5604113110539846,
                    "high": 0.6595758715544826,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7521008403361344,
                  "unit": "proportion",
                  "denominator": 238,
                  "interval": {
                    "low": 0.6945577638307764,
                    "high": 0.807088821016207,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.15128205128205127,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.11622259668869839,
                    "high": 0.18932038834951456,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.4142105923113062,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.3694420747216249,
                    "high": 0.4390089536226359,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5204184364936925,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.46417081182973385,
                    "high": 0.5515753519874144,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.9038461538461537,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 1.5347924554245225,
                    "high": 2.258190606929866,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.1730769230769234,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.975539606330255,
                    "high": 3.366754617414248,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.25384615384615383,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.20971763880867916,
                    "high": 0.3007528716249727,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6095238095238096,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5045871559633027,
                        "high": 0.7118838885641922,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.7523809523809524,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.6630404250122129,
                        "high": 0.8372093023255814,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.810126582278481,
                      "unit": "proportion",
                      "denominator": 79,
                      "interval": {
                        "low": 0.7162162162162162,
                        "high": 0.8939438502673795,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14285714285714285,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.08,
                        "high": 0.21052631578947367,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.361904761904762,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 1.6136363636363638,
                        "high": 3.083372641509434,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.6476190476190475,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.2823485730926034,
                        "high": 4.0,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5671821234321234,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.401583014438673,
                        "high": 0.5820668544807199,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6617124773374773,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4685135168451185,
                        "high": 0.6790779968941733,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2571428571428571,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.17391304347826086,
                        "high": 0.34375,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.40350877192982454,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.3444776238254499,
                        "high": 0.462688030436055,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5578947368421052,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.5,
                        "high": 0.6167280761477565,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7232704402515723,
                      "unit": "proportion",
                      "denominator": 159,
                      "interval": {
                        "low": 0.6511588887154675,
                        "high": 0.7922077922077922,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1543859649122807,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.11347517730496454,
                        "high": 0.19798657718120805,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.7350877192982457,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 1.3286675418444371,
                        "high": 2.1210690789473685,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.9982456140350884,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.7706206453978734,
                        "high": 3.2218635025754234,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3451476581410792,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2822105819473091,
                        "high": 0.36753220941886483,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4247971177120975,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3473361008582265,
                        "high": 0.45234733466937216,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.25263157894736843,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.20149185391722704,
                        "high": 0.3055555555555556,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6709677419354839,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7935483870967742,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8455284552845529,
                      "unit": "proportion",
                      "denominator": 123,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12258064516129032,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.9806451612903224,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.04516129032258,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.381294964028777,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5683453237410072,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6708860759493671,
                      "unit": "proportion",
                      "denominator": 79,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18705035971223022,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.629496402877698,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.068345323741007,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22916666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6111111111111112,
                      "unit": "proportion",
                      "denominator": 36,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14583333333333334,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5625,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.9166666666666667,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7547169811320755,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8679245283018868,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8695652173913043,
                      "unit": "proportion",
                      "denominator": 46,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11320754716981132,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.19811320754717,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.235849056603773,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6274509803921569,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7549019607843137,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8311688311688312,
                      "unit": "proportion",
                      "denominator": 77,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12745098039215685,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.8676470588235294,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.946078431372549,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.59375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.78125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.76,
                      "unit": "proportion",
                      "denominator": 25,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.890625,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.609375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3177570093457944,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5046728971962616,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6296296296296297,
                      "unit": "proportion",
                      "denominator": 54,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.18691588785046728,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.5514018691588785,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.9065420560747666,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.625,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.9,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.15,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6071428571428571,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.47368421052631576,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.855263157894737,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.003253471794871795
              }
            },
            "gemini-high": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4666666666666667,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4143920595533499,
                    "high": 0.5194830626292767,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.6538461538461539,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.6036269430051814,
                    "high": 0.7029984854562304,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.7137254901960784,
                  "unit": "proportion",
                  "denominator": 255,
                  "interval": {
                    "low": 0.6521739130434783,
                    "high": 0.7706785491396475,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.18717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.14761904761904762,
                    "high": 0.22928258427700063,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.3922962036147562,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.34947100437239237,
                    "high": 0.41544845062666175,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5230616048196749,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.46596133916318977,
                    "high": 0.5539312675022157,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 1.798717948717949,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 1.4175098808767035,
                    "high": 2.167989710326679,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.208974358974359,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 3.0111641050844433,
                    "high": 3.4043160580658864,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.28205128205128205,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.23466340206185568,
                    "high": 0.3317079033771107,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6190476190476191,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5108695652173914,
                        "high": 0.7191011235955056,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.780952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.693069306930693,
                        "high": 0.8623881880733945,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7926829268292683,
                      "unit": "proportion",
                      "denominator": 82,
                      "interval": {
                        "low": 0.691358024691358,
                        "high": 0.8809523809523809,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1619047619047619,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.09259259259259259,
                        "high": 0.24000961538461524,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.5476190476190474,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 1.784987068965517,
                        "high": 3.2571957671957663,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.6904761904761907,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.31997619047619,
                        "high": 4.04054054054054,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5107671498194374,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.38266519153619105,
                        "high": 0.540111764994806,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5746130435468672,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.43049834047821506,
                        "high": 0.6076257356191567,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.22857142857142856,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.14166482300884956,
                        "high": 0.3218439509059029,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4105263157894737,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.3511450381679389,
                        "high": 0.4696497724852357,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6070175438596491,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.5480760676156584,
                        "high": 0.6654418377063055,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6763005780346821,
                      "unit": "proportion",
                      "denominator": 173,
                      "interval": {
                        "low": 0.6012269938650306,
                        "high": 0.7471910112359551,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.19649122807017544,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.15034965034965034,
                        "high": 0.2463768115942029,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 1.52280701754386,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 1.0907368310360448,
                        "high": 1.9464898800599704,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0315789473684216,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.8032754614689557,
                        "high": 3.2566677852348995,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.32367967828292854,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.26207503429174434,
                        "high": 0.3453264068046811,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.43157290437723805,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.349433379055659,
                        "high": 0.46043520907290814,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.3017543859649123,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.24528121159546012,
                        "high": 0.3587812857408835,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6580645161290323,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8129032258064516,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8095238095238095,
                      "unit": "proportion",
                      "denominator": 126,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15483870967741936,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.0,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.064516129032258,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4172661870503597,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6618705035971223,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6304347826086957,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2446043165467626,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.410071942446043,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.1366906474820144,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22916666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3854166666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5945945945945946,
                      "unit": "proportion",
                      "denominator": 37,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15625,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.421875,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.9322916666666667,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7547169811320755,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9245283018867925,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8163265306122449,
                      "unit": "proportion",
                      "denominator": 49,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.16981132075471697,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.3773584905660377,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.320754716981132,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6078431372549019,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7549019607843137,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8051948051948052,
                      "unit": "proportion",
                      "denominator": 77,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14705882352941177,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.803921568627451,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.9313725490196076,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.625,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7692307692307693,
                      "unit": "proportion",
                      "denominator": 26,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.015625,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.734375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.35514018691588783,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.616822429906542,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5757575757575758,
                      "unit": "proportion",
                      "denominator": 66,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2616822429906542,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.2289719626168225,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.957943925233645,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.35,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7142857142857143,
                      "unit": "proportion",
                      "denominator": 7,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.2,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.95,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2236842105263158,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.39473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5666666666666667,
                      "unit": "proportion",
                      "denominator": 30,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.17105263157894737,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.21710526315789475,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.9276315789473684,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_event_usage_ledger",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "Saved selected-response usage at retained GBP price/FX buffers; reservations and earlier failed attempts are separate, not free.",
                "value": 0.004105528205128205
              }
            },
            "gemini-medium37": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.48717948717948717,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.43478260869565216,
                    "high": 0.5398936170212766,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.5948717948717949,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5424164524421594,
                    "high": 0.6466165413533834,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8189655172413793,
                  "unit": "proportion",
                  "denominator": 232,
                  "interval": {
                    "low": 0.7672413793103449,
                    "high": 0.8674711667181547,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.1076923076923077,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.07780899775856549,
                    "high": 0.13994910941475827,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.4834808090277646,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4369546180370484,
                    "high": 0.5089072946184343,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.557862471955113,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5041784054273635,
                    "high": 0.5872007245597318,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.5166666666666666,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.1967733380480907,
                    "high": 2.825,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.35,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 3.162022611005444,
                    "high": 3.536422836851276,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.16666666666666666,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.129576370137926,
                    "high": 0.2074468085106383,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.580952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.475,
                        "high": 0.6853932584269663,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6952380952380952,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5929203539823009,
                        "high": 0.7943987817089452,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8356164383561644,
                      "unit": "proportion",
                      "denominator": 73,
                      "interval": {
                        "low": 0.7384615384615385,
                        "high": 0.9210526315789473,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11428571428571428,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.053763440860215055,
                        "high": 0.18447393414943003,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.833333333333334,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.1590610047846894,
                        "high": 3.455063703852327,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.6428571428571432,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.29165521978022,
                        "high": 3.9936728639240506,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5553839838322597,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.39146388179096087,
                        "high": 0.5703936970455734,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6710889804639805,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4730188571640777,
                        "high": 0.689225717263401,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1619047619047619,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.0888818565400844,
                        "high": 0.24490296118447372,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.45263157894736844,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.3919376332622601,
                        "high": 0.5124577481855636,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.5578947368421052,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.4966216216216216,
                        "high": 0.6181818181818182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8113207547169812,
                      "unit": "proportion",
                      "denominator": 159,
                      "interval": {
                        "low": 0.7485705931495406,
                        "high": 0.8692819741532977,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10526315789473684,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.0711743772241993,
                        "high": 0.14145524551320818,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.4000000000000004,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.039332877959927,
                        "high": 2.745902965674572,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.2421052631578946,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 3.018768465692018,
                        "high": 3.4595083367854182,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.4403752862123052,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37129940246965343,
                        "high": 0.4632153407410484,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.47425030822863634,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.399860894967319,
                        "high": 0.4988472900288214,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.16842105263157894,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.12624584717607973,
                        "high": 0.21400814105417318,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6838709677419355,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7806451612903226,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8760330578512396,
                      "unit": "proportion",
                      "denominator": 121,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0967741935483871,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.4870967741935486,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.132258064516129,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4244604316546763,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.539568345323741,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7866666666666666,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11510791366906475,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.2302158273381294,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.273381294964029,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2604166666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.375,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6944444444444444,
                      "unit": "proportion",
                      "denominator": 36,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11458333333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.3645833333333333,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.1979166666666665,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7358490566037735,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8679245283018868,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8478260869565217,
                      "unit": "proportion",
                      "denominator": 46,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1320754716981132,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.490566037735849,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.245283018867925,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6568627450980392,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7352941176470589,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8933333333333333,
                      "unit": "proportion",
                      "denominator": 75,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0784313725490196,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.485294117647059,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.073529411764706,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.59375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8421052631578947,
                      "unit": "proportion",
                      "denominator": 19,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.3125,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.40625,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.40186915887850466,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5233644859813084,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7678571428571429,
                      "unit": "proportion",
                      "denominator": 56,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.12149532710280374,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.205607476635514,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2336448598130842,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.75,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.925,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.425,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3684210526315789,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6785714285714286,
                      "unit": "proportion",
                      "denominator": 28,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11842105263157894,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.2171052631578947,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.138157894736842,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "saved_event_telemetry",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "Saved provider-token cost estimate for each selected response; not an invoice.",
                "value": 0.0031772794415384615
              }
            },
            "gemini-medium38": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.4717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4190205308135899,
                    "high": 0.524632910193255,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.558974358974359,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5064926675391791,
                    "high": 0.6112663220629929,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8440366972477065,
                  "unit": "proportion",
                  "denominator": 218,
                  "interval": {
                    "low": 0.7923655395200537,
                    "high": 0.8909952606635071,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.08717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.05970149253731343,
                    "high": 0.11716674472070844,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.5056045771004524,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.45544501401993476,
                    "high": 0.5327076861406577,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5444972368774103,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.49047924586762204,
                    "high": 0.5736852004591699,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.588461538461539,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.2754202776440375,
                    "high": 2.8904727910516486,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.3320512820512826,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 3.142669685562111,
                    "high": 3.515313079777366,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.14871794871794872,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.11232876712328767,
                    "high": 0.18812406439582943,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.580952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.4742268041237113,
                        "high": 0.6857142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6666666666666666,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5634920634920635,
                        "high": 0.7678571428571429,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8714285714285714,
                      "unit": "proportion",
                      "denominator": 70,
                      "interval": {
                        "low": 0.7857142857142857,
                        "high": 0.9444444444444444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08571428571428572,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.036036036036036036,
                        "high": 0.14285714285714285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.0285714285714285,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.4081395960446037,
                        "high": 3.588261554621848,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.695238095238096,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.362736103615758,
                        "high": 4.0235962445573294,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5669526975082531,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4114755691839025,
                        "high": 0.5790927515196521,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6378217846967847,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4629100153318903,
                        "high": 0.6514793454596086,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.13333333333333333,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.06862745098039216,
                        "high": 0.2075565103680178,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.43157894736842106,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.37037037037037035,
                        "high": 0.49264705882352944,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.519298245614035,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.4603745470448582,
                        "high": 0.5804195804195804,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.831081081081081,
                      "unit": "proportion",
                      "denominator": 148,
                      "interval": {
                        "low": 0.7664233576642335,
                        "high": 0.888199671172817,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08771929824561403,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.05647710348993429,
                        "high": 0.12318840579710146,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.4263157894736844,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.072126114466494,
                        "high": 2.7690972222222223,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.1982456140350877,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.9765808473500477,
                        "high": 3.4180670289855075,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.449802005001963,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.37127089388336454,
                        "high": 0.47522819991095683,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.47286877448924314,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3903104269030242,
                        "high": 0.49959887682946746,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1543859649122807,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.11301369863013698,
                        "high": 0.19860744494890836,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6838709677419355,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7483870967741936,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9137931034482759,
                      "unit": "proportion",
                      "denominator": 116,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.06451612903225806,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.6548387096774193,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.138709677419355,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4028776978417266,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5179856115107914,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7777777777777778,
                      "unit": "proportion",
                      "denominator": 72,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11510791366906475,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.258992805755396,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2661870503597124,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.22916666666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3125,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7333333333333333,
                      "unit": "proportion",
                      "denominator": 30,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08333333333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.34375,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.125,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7547169811320755,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8301886792452831,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9090909090909091,
                      "unit": "proportion",
                      "denominator": 44,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.07547169811320754,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.8962264150943398,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.367924528301887,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6470588235294118,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7058823529411765,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9166666666666666,
                      "unit": "proportion",
                      "denominator": 72,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.058823529411764705,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.5294117647058822,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.019607843137255,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.46875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.59375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7894736842105263,
                      "unit": "proportion",
                      "denominator": 19,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.1875,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.4375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.38317757009345793,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4953271028037383,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7735849056603774,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11214953271028037,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.2803738317757007,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2149532710280373,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.35,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8571428571428571,
                      "unit": "proportion",
                      "denominator": 7,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.075,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.325,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21052631578947367,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3026315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6956521739130435,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.1513157894736843,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.0723684210526314,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "estimated",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "verified_selected_result_token_reconstruction",
                "coverage": {
                  "observed_events": 0,
                  "estimated_events": 390,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "Selected-response tokens repriced with the run's retained list prices and GBP buffers; no provider invoice. Earlier failed/uncertain attempts are separate.",
                "value": 0.007509248717948718
              }
            },
            "gemini38-low": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.47692307692307695,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4239394719722433,
                    "high": 0.5298512308586936,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.5666666666666667,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5149981343283582,
                    "high": 0.618798955613577,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8416289592760181,
                  "unit": "proportion",
                  "denominator": 221,
                  "interval": {
                    "low": 0.7912621359223301,
                    "high": 0.8879310344827587,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.08974358974358974,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.0625,
                    "high": 0.11928934010152284,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.5087793128538374,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4570642681703775,
                    "high": 0.536713242085056,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.5479161830733634,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.4922230580296374,
                    "high": 0.5779988760915987,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.583333333333334,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.2615898199294486,
                    "high": 2.8913070926657882,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.352564102564103,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 3.1653420940625665,
                    "high": 3.5404725443521503,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.15384615384615385,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.11764705882352941,
                    "high": 0.1935503636938646,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6095238095238096,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5,
                        "high": 0.7142857142857143,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6952380952380952,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.5899935897435897,
                        "high": 0.7962962962962963,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8767123287671232,
                      "unit": "proportion",
                      "denominator": 73,
                      "interval": {
                        "low": 0.796875,
                        "high": 0.948051948051948,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.08571428571428572,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.036036036036036036,
                        "high": 0.14285714285714285,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.157142857142857,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.539600703491402,
                        "high": 3.7255133185349614,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.776190476190476,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.428571428571429,
                        "high": 4.120004807692308,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5725306789382419,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4168189666063947,
                        "high": 0.5925458024118739,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.6679524587612823,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.48628879437412725,
                        "high": 0.6913034361471861,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.12380952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.06,
                        "high": 0.19642857142857142,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4280701754385965,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.36899845594200625,
                        "high": 0.48829625884732053,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.519298245614035,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.458955223880597,
                        "high": 0.5787545787545788,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8243243243243243,
                      "unit": "proportion",
                      "denominator": 148,
                      "interval": {
                        "low": 0.7597390013087687,
                        "high": 0.8843602825745682,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0912280701754386,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.058823529411764705,
                        "high": 0.1263951202837243,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.3719298245614033,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.0157315340909094,
                        "high": 2.727272727272727,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.196491228070176,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.973851909184727,
                        "high": 3.4177914236989766,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.45035712094922625,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3745020999083098,
                        "high": 0.4754769636778676,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4619047394351038,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.3841047178546768,
                        "high": 0.48766868069524877,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1649122807017544,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.12177121771217712,
                        "high": 0.211222284200251,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6903225806451613,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7548387096774194,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9145299145299145,
                      "unit": "proportion",
                      "denominator": 117,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.06451612903225806,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.7096774193548385,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.193548387096774,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.4172661870503597,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5251798561151079,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7945205479452054,
                      "unit": "proportion",
                      "denominator": 73,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1079136690647482,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.3633093525179856,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2985611510791366,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21875,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3229166666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6774193548387096,
                      "unit": "proportion",
                      "denominator": 31,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.10416666666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.0833333333333333,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.0729166666666665,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7735849056603774,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8113207547169812,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9534883720930233,
                      "unit": "proportion",
                      "denominator": 43,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.03773584905660377,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 4.160377358490566,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.443396226415095,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6470588235294118,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.7254901960784313,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8918918918918919,
                      "unit": "proportion",
                      "denominator": 74,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0784313725490196,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.4754901960784315,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.063725490196078,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5625,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8181818181818182,
                      "unit": "proportion",
                      "denominator": 22,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.546875,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.640625,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.37383177570093457,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4766355140186916,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7843137254901961,
                      "unit": "proportion",
                      "denominator": 51,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.102803738317757,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.308411214953271,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.196261682242991,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.25,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.625,
                      "unit": "proportion",
                      "denominator": 8,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.15,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.475,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.225,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.21052631578947367,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.3026315789473684,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6956521739130435,
                      "unit": "proportion",
                      "denominator": 23,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09210526315789473,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.9802631578947368,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.0328947368421053,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_usage_ledger_plus_selected_canary",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "789 saved reconciled selected-response costs plus one prior canary cost reconstructed from its hash-bound result tokens and retained price/FX buffer; earlier failed or uncertain attempts remain separate. Not an invoice.",
                "value": 0.006153087179487179
              }
            },
            "gemini38-high": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.5794871794871795,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.5274151436031331,
                    "high": 0.6308900523560209,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.6974358974358974,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.6505102040816326,
                    "high": 0.7435233160621761,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.8308823529411765,
                  "unit": "proportion",
                  "denominator": 272,
                  "interval": {
                    "low": 0.7829181494661922,
                    "high": 0.8745868644067795,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.11794871794871795,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.08638743455497382,
                    "high": 0.15151700377861746,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.5865229883529476,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5337536679544457,
                    "high": 0.6122990938092412,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.6466791922865834,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.5884976338984914,
                    "high": 0.6750990008665992,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 2.6961538461538463,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.3501163192858354,
                    "high": 3.0261194029850746,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 3.632051282051282,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 3.444303629587349,
                    "high": 3.814863940181281,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.18717948717948718,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.14786751361161524,
                    "high": 0.22894736842105262,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7142857142857143,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.6224489795918368,
                        "high": 0.8,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8285714285714286,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.7563025210084033,
                        "high": 0.897196261682243,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8620689655172413,
                      "unit": "proportion",
                      "denominator": 87,
                      "interval": {
                        "low": 0.7821782178217822,
                        "high": 0.9325842696629213,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11428571428571428,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.05504587155963303,
                        "high": 0.1810344827586207,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 3.295238095238095,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.6409057811120924,
                        "high": 3.8990624999999994,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 4.057142857142857,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 3.736836231203008,
                        "high": 4.357142857142857,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.661547217954781,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.48494878400077984,
                        "high": 0.6728746328076685,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.7718050876139112,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.565773581334243,
                        "high": 0.7850204049422798,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.1523809523809524,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.08256880733944955,
                        "high": 0.2298879310344827,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5298245614035088,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.46964398946418556,
                        "high": 0.5907476245845225,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.6491228070175439,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.5912162162162162,
                        "high": 0.706081383803158,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8162162162162162,
                      "unit": "proportion",
                      "denominator": 185,
                      "interval": {
                        "low": 0.7566103297968583,
                        "high": 0.8715083798882681,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11929824561403508,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.08243727598566308,
                        "high": 0.15916955017301038,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 2.4754385964912284,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.0758966934429295,
                        "high": 2.857606329682479,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.4754385964912284,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 3.2491525423728818,
                        "high": 3.697115384615385,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.5333835274757305,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.4552077201535471,
                        "high": 0.5603667026108957,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.5607365288847423,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.47855170580244705,
                        "high": 0.5891034565909417,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.2,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.15441176470588236,
                        "high": 0.24919019933554787,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7612903225806451,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.864516129032258,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8805970149253731,
                      "unit": "proportion",
                      "denominator": 134,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1032258064516129,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.6032258064516127,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.312903225806451,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5323741007194245,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6690647482014388,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7956989247311828,
                      "unit": "proportion",
                      "denominator": 93,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1366906474820144,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.4964028776978417,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.6115107913669067,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3541666666666667,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.46875,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7555555555555555,
                      "unit": "proportion",
                      "denominator": 45,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.11458333333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.5208333333333333,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.5625,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.8113207547169812,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9622641509433962,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.8431372549019608,
                      "unit": "proportion",
                      "denominator": 51,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.1509433962264151,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.5754716981132075,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.4245283018867925,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.7352941176470589,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8137254901960784,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9036144578313253,
                      "unit": "proportion",
                      "denominator": 83,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.0784313725490196,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.6176470588235294,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.254901960784314,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.78125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.88,
                      "unit": "proportion",
                      "denominator": 25,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.09375,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 3.28125,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.0625,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.48598130841121495,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6355140186915887,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7647058823529411,
                      "unit": "proportion",
                      "denominator": 68,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.14953271028037382,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.2616822429906542,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.4766355140186915,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.55,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.9090909090909091,
                      "unit": "proportion",
                      "denominator": 11,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.05,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.575,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.075,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3157894736842105,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.4473684210526316,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7058823529411765,
                      "unit": "proportion",
                      "denominator": 34,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.13157894736842105,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.243421052631579,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.4276315789473686,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "hash_verified_usage_ledger_plus_selected_canary",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "789 saved reconciled selected-response costs plus one prior canary cost reconstructed from its hash-bound result tokens and retained price/FX buffer; earlier failed or uncertain attempts remain separate. Not an invoice.",
                "value": 0.013991210256410256
              }
            },
            "qwen-repaired": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.35384615384615387,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3036649214659686,
                    "high": 0.40601659535579276,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.8461538461538461,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.8068459657701712,
                    "high": 0.883419689119171,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.41818181818181815,
                  "unit": "proportion",
                  "denominator": 330,
                  "interval": {
                    "low": 0.36079316348973606,
                    "high": 0.47648955418568933,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.49230769230769234,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.43902439024390244,
                    "high": 0.5467989648936833,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.1860547887327449,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.15939678648592506,
                    "high": 0.20091256108108368,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.38165084868255367,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.32696776715061543,
                    "high": 0.4121283304227359,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 0.0576923076923077,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": -0.34278350515463923,
                    "high": 0.46542663394851297,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 2.6089743589743595,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.3973327628361862,
                    "high": 2.824001265822785,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.5102564102564102,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4565189520624303,
                    "high": 0.5646446199083737,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.42857142857142855,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.3220290278589519,
                        "high": 0.5339805825242718,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8761904761904762,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.81,
                        "high": 0.9363636363636364,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4891304347826087,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": {
                        "low": 0.37142857142857144,
                        "high": 0.6049382716049383,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.44761904761904764,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.339995145631068,
                        "high": 0.5575221238938053,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.7619047619047619,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": -0.05884103641456582,
                        "high": 1.5833333333333333,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.5981232222673714,
                        "high": 3.394743421052631,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.2921770224254696,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.1935114689109603,
                        "high": 0.31441578346465565,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4260914910371432,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2822042254951504,
                        "high": 0.4585230175526228,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.44761904761904764,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.339995145631068,
                        "high": 0.5575221238938053,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.3263157894736842,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.2690909090909091,
                        "high": 0.3843810096153846,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8350877192982457,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.7873134328358209,
                        "high": 0.8803995738841542,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3907563025210084,
                      "unit": "proportion",
                      "denominator": 238,
                      "interval": {
                        "low": 0.3252032520325203,
                        "high": 0.45689655172413796,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5087719298245614,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.44755244755244755,
                        "high": 0.5700008143322475,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.2017543859649123,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": -0.6484112507386728,
                        "high": 0.25650633549578444,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.4649122807017543,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.2206392610424954,
                        "high": 2.7101449275362324,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.17004896020114546,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13144020489747812,
                        "high": 0.1843991466197147,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.3357376906535436,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.25951014813091833,
                        "high": 0.3640701099927701,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5333333333333333,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.47183098591549294,
                        "high": 0.5947040974258715,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.535483870967742,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9806451612903225,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5460526315789473,
                      "unit": "proportion",
                      "denominator": 152,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.44516129032258067,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.229032258064516,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.5193548387096776,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2949640287769784,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8848920863309353,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 123,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5899280575539568,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6115107913669064,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.4820143884892087,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.14583333333333334,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5729166666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2545454545454545,
                      "unit": "proportion",
                      "denominator": 55,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4270833333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.8645833333333334,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.3229166666666667,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6792452830188679,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9811320754716981,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.6923076923076923,
                      "unit": "proportion",
                      "denominator": 52,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.3018867924528302,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.6132075471698113,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.122641509433962,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.46078431372549017,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9803921568627451,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.47,
                      "unit": "proportion",
                      "denominator": 100,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5196078431372549,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5098039215686274,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.2058823529411766,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.28125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.96875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2903225806451613,
                      "unit": "proportion",
                      "denominator": 31,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.859375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.578125,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.29906542056074764,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8598130841121495,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.34782608695652173,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5607476635514018,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.5373831775700935,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.453271028037383,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.0,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.45,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.0,
                      "unit": "proportion",
                      "denominator": 9,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.45,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.55,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.7,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18421052631578946,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6052631578947368,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.30434782608695654,
                      "unit": "proportion",
                      "denominator": 46,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6842105263157895,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.486842105263158,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "provider_usage_or_saved_generation_metadata",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "Selected evaluated response attempt only: 786 first-pass outputs, four successful repair outputs. Four superseded first-pass length charges are excluded from this comparison and retained in the audit below.",
                "value": 0.0017576954532692308
              }
            },
            "qwen-first-pass": {
              "metrics": {
                "species_accuracy": {
                  "value": 0.35128205128205126,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.3010471204188482,
                    "high": 0.4036001432626392,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "species_accuracy"
                },
                "answer_coverage": {
                  "value": 0.8358974358974359,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.7964824120603015,
                    "high": 0.8737373737373737,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "answer_coverage"
                },
                "accuracy_among_answers": {
                  "value": 0.42024539877300615,
                  "unit": "proportion",
                  "denominator": 326,
                  "interval": {
                    "low": 0.3630947802197802,
                    "high": 0.47896440129449835,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "accuracy_among_answers"
                },
                "wrong_species_rate_all_events": {
                  "value": 0.4846153846153846,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.43116883116883115,
                    "high": 0.5382716049382716,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "wrong_species_rate_all_events"
                },
                "macro_f1": {
                  "value": 0.1879247540821163,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.1609243544252662,
                    "high": 0.2028975237292849,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "macro_f1"
                },
                "reference_species_macro_f1": {
                  "value": 0.3806680916022355,
                  "unit": "f1",
                  "denominator": null,
                  "interval": {
                    "low": 0.3259749743486162,
                    "high": 0.41099754806701305,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "reference_species_macro_f1"
                },
                "adjusted_tis": {
                  "value": 0.07948717948717948,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": -0.3155234392714081,
                    "high": 0.48195305404178007,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "adjusted_tis"
                },
                "raw_tis": {
                  "value": 2.5923076923076924,
                  "unit": "points_per_event",
                  "denominator": 390,
                  "interval": {
                    "low": 2.3818845044178643,
                    "high": 2.8076923076923075,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "raw_tis"
                },
                "any_rank_conflict_rate": {
                  "value": 0.5025641025641026,
                  "unit": "proportion",
                  "denominator": 390,
                  "interval": {
                    "low": 0.4487798121748531,
                    "high": 0.5566502463054187,
                    "level": 0.95,
                    "method": "source_cluster",
                    "scope": "pointwise",
                    "valid_resamples": 10000
                  },
                  "source_kind": "saved_v1_subgroup_interval",
                  "interval_status": "ok",
                  "metric_key": "any_rank_conflict_rate"
                }
              },
              "conditions": {
                "day": {
                  "n_events": 105,
                  "n_species": 24,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.42857142857142855,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.3220290278589519,
                        "high": 0.5339805825242718,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8666666666666667,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.7982456140350878,
                        "high": 0.9285714285714286,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4945054945054945,
                      "unit": "proportion",
                      "denominator": 91,
                      "interval": {
                        "low": 0.3764705882352941,
                        "high": 0.6104076479076477,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4380952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.3333333333333333,
                        "high": 0.5447339246119731,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": 0.8095238095238095,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 0.0147953240702222,
                        "high": 1.602176139272913,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 3.0,
                      "unit": "points_per_event",
                      "denominator": 105,
                      "interval": {
                        "low": 2.5981232222673714,
                        "high": 3.394743421052631,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.3007704642615128,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.19920298270245912,
                        "high": 0.32366330650773373,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.4260914910371432,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.2822042254951504,
                        "high": 0.4585230175526228,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.4380952380952381,
                      "unit": "proportion",
                      "denominator": 105,
                      "interval": {
                        "low": 0.3333333333333333,
                        "high": 0.5447339246119731,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "night_ir": {
                  "n_events": 285,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.32280701754385965,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.2659923705136471,
                        "high": 0.3811188811188811,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "species_accuracy"
                    },
                    "answer_coverage": {
                      "value": 0.8245614035087719,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.7755867523138554,
                        "high": 0.8709705518714173,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "answer_coverage"
                    },
                    "accuracy_among_answers": {
                      "value": 0.39148936170212767,
                      "unit": "proportion",
                      "denominator": 235,
                      "interval": {
                        "low": 0.3254897336293008,
                        "high": 0.4581714807436918,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "accuracy_among_answers"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5017543859649123,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.44074074074074077,
                        "high": 0.5632224380767362,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "wrong_species_rate_all_events"
                    },
                    "adjusted_tis": {
                      "value": -0.18947368421052635,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": -0.6300536881419234,
                        "high": 0.26531281365005016,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "adjusted_tis"
                    },
                    "raw_tis": {
                      "value": 2.4421052631578952,
                      "unit": "points_per_event",
                      "denominator": 285,
                      "interval": {
                        "low": 2.1982102272727277,
                        "high": 2.687961132022686,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "raw_tis"
                    },
                    "macro_f1": {
                      "value": 0.1688338010912495,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.13034962230182134,
                        "high": 0.18289251890430513,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "macro_f1"
                    },
                    "reference_species_macro_f1": {
                      "value": 0.33333853035964645,
                      "unit": "f1",
                      "denominator": null,
                      "interval": {
                        "low": 0.25735694659590364,
                        "high": 0.3610954860418332,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "reference_species_macro_f1"
                    },
                    "any_rank_conflict_rate": {
                      "value": 0.5263157894736842,
                      "unit": "proportion",
                      "denominator": 285,
                      "interval": {
                        "low": 0.4641509433962264,
                        "high": 0.5871886120996441,
                        "level": 0.95,
                        "method": "source_cluster",
                        "scope": "pointwise",
                        "valid_resamples": 10000
                      },
                      "source_kind": "saved_v1_subgroup_interval",
                      "interval_status": "ok",
                      "metric_key": "any_rank_conflict_rate"
                    }
                  }
                },
                "easy": {
                  "n_events": 155,
                  "n_species": 38,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.5290322580645161,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9548387096774194,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.5540540540540541,
                      "unit": "proportion",
                      "denominator": 148,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4258064516129032,
                      "unit": "proportion",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 1.2838709677419355,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.47741935483871,
                      "unit": "points_per_event",
                      "denominator": 155,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium": {
                  "n_events": 139,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.2949640287769784,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8848920863309353,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.3333333333333333,
                      "unit": "proportion",
                      "denominator": 123,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5899280575539568,
                      "unit": "proportion",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6115107913669064,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.4820143884892087,
                      "unit": "points_per_event",
                      "denominator": 139,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard": {
                  "n_events": 96,
                  "n_species": 39,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.14583333333333334,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.5729166666666666,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2545454545454545,
                      "unit": "proportion",
                      "denominator": 55,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.4270833333333333,
                      "unit": "proportion",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.8645833333333334,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.3229166666666667,
                      "unit": "points_per_event",
                      "denominator": 96,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | day": {
                  "n_events": 53,
                  "n_species": 21,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.6792452830188679,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9622641509433962,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.7058823529411765,
                      "unit": "proportion",
                      "denominator": 51,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.2830188679245283,
                      "unit": "proportion",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 2.707547169811321,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 4.122641509433962,
                      "unit": "points_per_event",
                      "denominator": 53,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "easy | night_ir": {
                  "n_events": 102,
                  "n_species": 33,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.45098039215686275,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.9509803921568627,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.4742268041237113,
                      "unit": "proportion",
                      "denominator": 97,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5,
                      "unit": "proportion",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": 0.5441176470588235,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 3.142156862745098,
                      "unit": "points_per_event",
                      "denominator": 102,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | day": {
                  "n_events": 32,
                  "n_species": 17,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.28125,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.96875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.2903225806451613,
                      "unit": "proportion",
                      "denominator": 31,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.6875,
                      "unit": "proportion",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.859375,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.578125,
                      "unit": "points_per_event",
                      "denominator": 32,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "medium | night_ir": {
                  "n_events": 107,
                  "n_species": 35,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.29906542056074764,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.8598130841121495,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.34782608695652173,
                      "unit": "proportion",
                      "denominator": 92,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.5607476635514018,
                      "unit": "proportion",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.5373831775700935,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 2.453271028037383,
                      "unit": "points_per_event",
                      "denominator": 107,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | day": {
                  "n_events": 20,
                  "n_species": 15,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.0,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.45,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.0,
                      "unit": "proportion",
                      "denominator": 9,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.45,
                      "unit": "proportion",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -1.55,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 0.7,
                      "unit": "points_per_event",
                      "denominator": 20,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                },
                "hard | night_ir": {
                  "n_events": 76,
                  "n_species": 32,
                  "metrics": {
                    "species_accuracy": {
                      "value": 0.18421052631578946,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "answer_coverage": {
                      "value": 0.6052631578947368,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "accuracy_among_answers": {
                      "value": 0.30434782608695654,
                      "unit": "proportion",
                      "denominator": 46,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "wrong_species_rate_all_events": {
                      "value": 0.42105263157894735,
                      "unit": "proportion",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "adjusted_tis": {
                      "value": -0.6842105263157895,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    },
                    "raw_tis": {
                      "value": 1.486842105263158,
                      "unit": "points_per_event",
                      "denominator": 76,
                      "interval": null,
                      "source_kind": "saved_v1_event_outcomes_descriptive"
                    }
                  }
                }
              },
              "cost": {
                "status": "recorded",
                "unit": "GBP_per_event",
                "denominator": 390,
                "basis": "selected_response_attempts",
                "source_kind": "provider_usage_or_saved_generation_metadata",
                "coverage": {
                  "observed_events": 390,
                  "estimated_events": 0,
                  "missing_events": 0,
                  "population_events": 390,
                  "population_attempts": 390
                },
                "note": "First pass includes 786 model outputs and four charged HTTP 200 length nonresponses. 785 charges came from result usage and five exact charges from hash-bound provider generation metadata.",
                "value": 0.0017951259094230768
              }
            }
          }
        }
      }
    }
  },
  "note": "V1 saved intervals are pointwise candidate estimates; newly crossed class/difficulty cells and prototype class slices are descriptive only. Cohorts are distinct."
}