{
  "questions": 1024,
  "split": "validation",
  "selection_use": "Quantization recipe comparison; not an independent test",
  "loops": 2,
  "probability_calibration": false,
  "calibration_examples": 256,
  "calibration_source": "Training examples disjoint from validation by ID and source group",
  "weights_sha256": "62bc20a6fbafaf157154650dfb4e9ec32a19d67e8e1f60ddcc3318fd37d6ccb4",
  "metrics": {
    "questions": 1024,
    "correct": 0.78515625,
    "brier": 0.256938595885194,
    "cross_entropy": 0.9667092188545374
  },
  "by_kind": {
    "choice": {
      "questions": 665,
      "correct": 0.7819548872180451,
      "brier": 0.29931089665593197,
      "cross_entropy": 1.021091694949712
    },
    "noul": {
      "questions": 144,
      "correct": 0.9097222222222222,
      "brier": 0.146691427234091,
      "cross_entropy": 0.4264544071293922
    },
    "score": {
      "questions": 215,
      "correct": 0.7116279069767442,
      "brier": 0.19972004831876652,
      "cross_entropy": 1.160348038785374
    }
  }
}