{
  "format": "independent-steward-results-audit/1",
  "passed": true,
  "scenes": 1200,
  "methods": 11,
  "prediction_records_hash_verified": true,
  "generated_test_hash_unchanged": true,
  "disjoint_fresh_split_ids": true,
  "scientific_freeze_matches": {
    "136/protocol/model-plan.json": true,
    "136/tools/evaluate_steward.py (preserved pre-serialization-fix evaluator)": true,
    "136/tools/evaluate_steward.py metric functions unchanged": true,
    "136/tools/make_steward_data.py": true,
    "136/tools/steward_model.py": true,
    "136/tools/test_steward_model.py": true,
    "136/tools/train_steward.py": true,
    "136/model/steward-core.js": true,
    "136/model/steward-model.js": true,
    "136/runs/seed-136001/weights.npz": true,
    "136/runs/seed-136001/initial.npz": true,
    "136/runs/seed-136001/manifest.json": true,
    "136/runs/seed-136002/weights.npz": true,
    "136/runs/seed-136002/initial.npz": true,
    "136/runs/seed-136002/manifest.json": true,
    "136/runs/seed-136003/weights.npz": true,
    "136/runs/seed-136003/initial.npz": true,
    "136/runs/seed-136003/manifest.json": true,
    "135/runs/full-v2/weights.npz": true,
    "135/tools/field_model.py": true,
    "135/tools/field_model_v2.py": true,
    "135/tools/tensorize.py": true,
    "135/tools/make_exercises.py": true
  },
  "primary_f1": 0.9912423786523081,
  "primary_above_point91": true,
  "primary_exceeds_fresh_copy": true,
  "primary_improves_required_additions": true,
  "primary_improves_required_removals": true,
  "independent_scores": {
    "copy": {
      "micro": {
        "tp": 28329,
        "fp": 3019,
        "fn": 3150,
        "f1": 0.9018097314848712
      },
      "added": {
        "tp": 0,
        "fp": 0,
        "fn": 3150,
        "f1": 0.0
      },
      "removed": {
        "tp": 0,
        "fp": 0,
        "fn": 3019,
        "f1": 0.0
      },
      "persistent": {
        "correct": 28329,
        "total": 28329
      },
      "exact_set_accuracy": 0.2658333333333333,
      "needed_edits": 6169,
      "missed_edits": 6169,
      "unnecessary_edits": 0,
      "by_relation": {
        "0": {
          "tp": 2415,
          "fp": 2753,
          "fn": 254,
          "f1": 0.6163072604312875
        },
        "1": {
          "tp": 4990,
          "fp": 266,
          "fn": 2666,
          "f1": 0.7729244114002478
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5259,
          "fp": 0,
          "fn": 230,
          "f1": 0.978600669892073
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "prior135": {
      "micro": {
        "tp": 17451,
        "fp": 14192,
        "fn": 14028,
        "f1": 0.5529292481226831
      },
      "added": {
        "tp": 1705,
        "fp": 13890,
        "fn": 1445,
        "f1": 0.18191517738063484
      },
      "removed": {
        "tp": 2717,
        "fp": 12583,
        "fn": 302,
        "f1": 0.29663191222228286
      },
      "persistent": {
        "correct": 15746,
        "total": 28329
      },
      "exact_set_accuracy": 0.1875,
      "needed_edits": 6169,
      "missed_edits": 1747,
      "unnecessary_edits": 26473,
      "by_relation": {
        "0": {
          "tp": 174,
          "fp": 212,
          "fn": 2495,
          "f1": 0.11391162029459902
        },
        "1": {
          "tp": 4228,
          "fp": 1443,
          "fn": 3428,
          "f1": 0.6345013881593757
        },
        "2": {
          "tp": 1724,
          "fp": 219,
          "fn": 3469,
          "f1": 0.48318385650224216
        },
        "3": {
          "tp": 3344,
          "fp": 2343,
          "fn": 1896,
          "f1": 0.6120618651047863
        },
        "4": {
          "tp": 3586,
          "fp": 2390,
          "fn": 1903,
          "f1": 0.6255560401221107
        },
        "5": {
          "tp": 4395,
          "fp": 7585,
          "fn": 837,
          "f1": 0.5106902161282826
        }
      }
    },
    "copy_edges_predicted_count": {
      "micro": {
        "tp": 28329,
        "fp": 3019,
        "fn": 3150,
        "f1": 0.9018097314848712
      },
      "added": {
        "tp": 0,
        "fp": 0,
        "fn": 3150,
        "f1": 0.0
      },
      "removed": {
        "tp": 0,
        "fp": 0,
        "fn": 3019,
        "f1": 0.0
      },
      "persistent": {
        "correct": 28329,
        "total": 28329
      },
      "exact_set_accuracy": 0.2658333333333333,
      "needed_edits": 6169,
      "missed_edits": 6169,
      "unnecessary_edits": 0,
      "by_relation": {
        "0": {
          "tp": 2415,
          "fp": 2753,
          "fn": 254,
          "f1": 0.6163072604312875
        },
        "1": {
          "tp": 4990,
          "fp": 266,
          "fn": 2666,
          "f1": 0.7729244114002478
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5259,
          "fp": 0,
          "fn": 230,
          "f1": 0.978600669892073
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "initial_editor": {
      "micro": {
        "tp": 6834,
        "fp": 25986,
        "fn": 24645,
        "f1": 0.2125694023235198
      },
      "added": {
        "tp": 2595,
        "fp": 25963,
        "fn": 555,
        "f1": 0.163681089945755
      },
      "removed": {
        "tp": 2996,
        "fp": 24090,
        "fn": 23,
        "f1": 0.1990367048663013
      },
      "persistent": {
        "correct": 4239,
        "total": 28329
      },
      "exact_set_accuracy": 0.13833333333333334,
      "needed_edits": 6169,
      "missed_edits": 578,
      "unnecessary_edits": 50053,
      "by_relation": {
        "0": {
          "tp": 0,
          "fp": 0,
          "fn": 2669,
          "f1": 0.0
        },
        "1": {
          "tp": 3406,
          "fp": 10018,
          "fn": 4250,
          "f1": 0.32314990512333963
        },
        "2": {
          "tp": 3028,
          "fp": 13122,
          "fn": 2165,
          "f1": 0.2837464274000843
        },
        "3": {
          "tp": 395,
          "fp": 2782,
          "fn": 4845,
          "f1": 0.0938576690032078
        },
        "4": {
          "tp": 5,
          "fp": 55,
          "fn": 5484,
          "f1": 0.0018021265092809514
        },
        "5": {
          "tp": 0,
          "fp": 9,
          "fn": 5232,
          "f1": 0.0
        }
      }
    },
    "seed136001": {
      "micro": {
        "tp": 31296,
        "fp": 370,
        "fn": 183,
        "f1": 0.9912423786523081
      },
      "added": {
        "tp": 3090,
        "fp": 370,
        "fn": 60,
        "f1": 0.9349470499243571
      },
      "removed": {
        "tp": 3019,
        "fp": 123,
        "fn": 0,
        "f1": 0.9800357084888817
      },
      "persistent": {
        "correct": 28206,
        "total": 28329
      },
      "exact_set_accuracy": 0.7666666666666667,
      "needed_edits": 6169,
      "missed_edits": 60,
      "unnecessary_edits": 493,
      "by_relation": {
        "0": {
          "tp": 2634,
          "fp": 88,
          "fn": 35,
          "f1": 0.9771841958820257
        },
        "1": {
          "tp": 7568,
          "fp": 35,
          "fn": 88,
          "f1": 0.9919391834327282
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5429,
          "fp": 247,
          "fn": 60,
          "f1": 0.9725033587102553
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "seed136002": {
      "micro": {
        "tp": 31321,
        "fp": 377,
        "fn": 158,
        "f1": 0.9915317283188502
      },
      "added": {
        "tp": 3104,
        "fp": 377,
        "fn": 46,
        "f1": 0.9362087166339919
      },
      "removed": {
        "tp": 3019,
        "fp": 112,
        "fn": 0,
        "f1": 0.9817886178861789
      },
      "persistent": {
        "correct": 28217,
        "total": 28329
      },
      "exact_set_accuracy": 0.7733333333333333,
      "needed_edits": 6169,
      "missed_edits": 46,
      "unnecessary_edits": 489,
      "by_relation": {
        "0": {
          "tp": 2625,
          "fp": 68,
          "fn": 44,
          "f1": 0.97911227154047
        },
        "1": {
          "tp": 7588,
          "fp": 43,
          "fn": 68,
          "f1": 0.992738928501341
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5443,
          "fp": 266,
          "fn": 46,
          "f1": 0.9721378817646008
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "seed136003": {
      "micro": {
        "tp": 31331,
        "fp": 396,
        "fn": 148,
        "f1": 0.991393222162453
      },
      "added": {
        "tp": 3105,
        "fp": 396,
        "fn": 45,
        "f1": 0.9336941813261164
      },
      "removed": {
        "tp": 3019,
        "fp": 103,
        "fn": 0,
        "f1": 0.9832274873799055
      },
      "persistent": {
        "correct": 28226,
        "total": 28329
      },
      "exact_set_accuracy": 0.7741666666666667,
      "needed_edits": 6169,
      "missed_edits": 45,
      "unnecessary_edits": 499,
      "by_relation": {
        "0": {
          "tp": 2630,
          "fp": 64,
          "fn": 39,
          "f1": 0.9807943315308596
        },
        "1": {
          "tp": 7592,
          "fp": 38,
          "fn": 64,
          "f1": 0.9933272275284574
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5444,
          "fp": 294,
          "fn": 45,
          "f1": 0.9698049345328227
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "raw_seed136001": {
      "micro": {
        "tp": 31300,
        "fp": 394,
        "fn": 179,
        "f1": 0.9909296693207541
      },
      "added": {
        "tp": 3094,
        "fp": 394,
        "fn": 56,
        "f1": 0.932208496535101
      },
      "removed": {
        "tp": 3019,
        "fp": 123,
        "fn": 0,
        "f1": 0.9800357084888817
      },
      "persistent": {
        "correct": 28206,
        "total": 28329
      },
      "exact_set_accuracy": 0.7558333333333334,
      "needed_edits": 6169,
      "missed_edits": 56,
      "unnecessary_edits": 517,
      "by_relation": {
        "0": {
          "tp": 2634,
          "fp": 88,
          "fn": 35,
          "f1": 0.9771841958820257
        },
        "1": {
          "tp": 7568,
          "fp": 35,
          "fn": 88,
          "f1": 0.9919391834327282
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5433,
          "fp": 271,
          "fn": 56,
          "f1": 0.9707853122487269
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "raw_seed136002": {
      "micro": {
        "tp": 31325,
        "fp": 415,
        "fn": 154,
        "f1": 0.9909995412771477
      },
      "added": {
        "tp": 3108,
        "fp": 415,
        "fn": 42,
        "f1": 0.9315150606923422
      },
      "removed": {
        "tp": 3019,
        "fp": 112,
        "fn": 0,
        "f1": 0.9817886178861789
      },
      "persistent": {
        "correct": 28217,
        "total": 28329
      },
      "exact_set_accuracy": 0.7541666666666667,
      "needed_edits": 6169,
      "missed_edits": 42,
      "unnecessary_edits": 527,
      "by_relation": {
        "0": {
          "tp": 2625,
          "fp": 68,
          "fn": 44,
          "f1": 0.97911227154047
        },
        "1": {
          "tp": 7588,
          "fp": 43,
          "fn": 68,
          "f1": 0.992738928501341
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5447,
          "fp": 304,
          "fn": 42,
          "f1": 0.9692170818505338
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "raw_seed136003": {
      "micro": {
        "tp": 31335,
        "fp": 436,
        "fn": 144,
        "f1": 0.9908300395256917
      },
      "added": {
        "tp": 3109,
        "fp": 436,
        "fn": 41,
        "f1": 0.9287528005974608
      },
      "removed": {
        "tp": 3019,
        "fp": 103,
        "fn": 0,
        "f1": 0.9832274873799055
      },
      "persistent": {
        "correct": 28226,
        "total": 28329
      },
      "exact_set_accuracy": 0.7533333333333333,
      "needed_edits": 6169,
      "missed_edits": 41,
      "unnecessary_edits": 539,
      "by_relation": {
        "0": {
          "tp": 2630,
          "fp": 64,
          "fn": 39,
          "f1": 0.9807943315308596
        },
        "1": {
          "tp": 7592,
          "fp": 38,
          "fn": 64,
          "f1": 0.9933272275284574
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5448,
          "fp": 334,
          "fn": 41,
          "f1": 0.9667287729571467
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    },
    "seed136001_without_additions": {
      "micro": {
        "tp": 31126,
        "fp": 123,
        "fn": 353,
        "f1": 0.9924116821833949
      },
      "added": {
        "tp": 2920,
        "fp": 123,
        "fn": 230,
        "f1": 0.9430001614726304
      },
      "removed": {
        "tp": 3019,
        "fp": 123,
        "fn": 0,
        "f1": 0.9800357084888817
      },
      "persistent": {
        "correct": 28206,
        "total": 28329
      },
      "exact_set_accuracy": 0.7341666666666666,
      "needed_edits": 6169,
      "missed_edits": 230,
      "unnecessary_edits": 246,
      "by_relation": {
        "0": {
          "tp": 2634,
          "fp": 88,
          "fn": 35,
          "f1": 0.9771841958820257
        },
        "1": {
          "tp": 7568,
          "fp": 35,
          "fn": 88,
          "f1": 0.9919391834327282
        },
        "2": {
          "tp": 5193,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "3": {
          "tp": 5240,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        },
        "4": {
          "tp": 5259,
          "fp": 0,
          "fn": 230,
          "f1": 0.978600669892073
        },
        "5": {
          "tp": 5232,
          "fp": 0,
          "fn": 0,
          "f1": 1.0
        }
      }
    }
  },
  "limits": [
    "Scores independently recomputed from recorded triples; model inference and teacher truth not independently recreated here.",
    "Generator manifest pins the retained test hash; pretest freeze itself did not separately include the dataset manifest.",
    "A post-freeze JSON serialization failure was repaired and evaluation rerun; the original evaluator is preserved, and metric function ASTs are unchanged.",
    "This is another local agent audit, not external replication or authenticated preregistration.",
    "None of these numerical checks establishes preservation of philosophical meaning."
  ]
}
