| { |
| "prompt": 2, |
| "orders": 2, |
| "method": "temperature per question type and per option-count bucket (choice), by NLL, fitted on the out-of-distribution calibration split only; thresholds per type: the most permissive threshold whose accepted calibration decisions have a one-sided 95% Clopper-Pearson upper bound <= max_error; a type without one defers", |
| "temperature_per_prim": { |
| "choice": 1.1721576888192515, |
| "noul": 1.8077142767822016, |
| "score": 1.7817974362806785 |
| }, |
| "temperature_per_bucket": { |
| "choice": { |
| "2-10": 1.5874010519681994, |
| "11-26": 0.8655365610061431, |
| "27-255": 0.805245165974627 |
| } |
| }, |
| "thresholds": { |
| "0.01": { |
| "confidence": null, |
| "method": "per type: the most permissive threshold whose accepted calibration decisions have a one-sided 95% Clopper-Pearson upper bound <= max_error; a type without one defers", |
| "per_type": { |
| "choice": { |
| "confidence": null, |
| "heldout_coverage": 0.0, |
| "heldout_error": null |
| }, |
| "noul": { |
| "confidence": null, |
| "heldout_coverage": 0.0, |
| "heldout_error": null |
| }, |
| "score": { |
| "confidence": null, |
| "heldout_coverage": 0.0, |
| "heldout_error": null |
| } |
| }, |
| "heldout_coverage": 0.0, |
| "heldout_error": null |
| }, |
| "0.05": { |
| "confidence": null, |
| "method": "per type: the most permissive threshold whose accepted calibration decisions have a one-sided 95% Clopper-Pearson upper bound <= max_error; a type without one defers", |
| "per_type": { |
| "choice": { |
| "confidence": 0.9886388083772373, |
| "heldout_coverage": 0.04135085059794509, |
| "heldout_error": 0.010183299389002087 |
| }, |
| "noul": { |
| "confidence": 0.9369879604949581, |
| "heldout_coverage": 0.1076631977294229, |
| "heldout_error": 0.0035149384885764245 |
| }, |
| "score": { |
| "confidence": null, |
| "heldout_coverage": 0.0, |
| "heldout_error": null |
| } |
| }, |
| "heldout_coverage": 0.05263941997318369, |
| "heldout_error": 0.006603773584905714 |
| } |
| }, |
| "gates": { |
| "ece_lofo": { |
| "choice": { |
| "pooled": 0.03552901963898983, |
| "n": 11874, |
| "gate": 0.05, |
| "pass": true |
| }, |
| "noul": { |
| "pooled": 0.020483490631501188, |
| "n": 5285, |
| "gate": 0.05, |
| "pass": true |
| }, |
| "score": { |
| "pooled": 0.009868532056588154, |
| "n": 2978, |
| "gate": 0.05, |
| "pass": true |
| } |
| }, |
| "ece_worst_fold": 0.10641607620521439, |
| "worst_fold": "agent", |
| "automated_at": { |
| "0.01": 0.0, |
| "0.05": 0.05263941997318369 |
| }, |
| "automated_error": { |
| "0.01": null, |
| "0.05": 0.006603773584905714 |
| }, |
| "note": "held-out numbers: every evaluation row predicted by the fit without its family" |
| }, |
| "test": { |
| "T": { |
| "choice": { |
| "acc": 0.7091965639211724, |
| "ece": 0.03517645160708048, |
| "n": 11874 |
| }, |
| "noul": { |
| "acc": 0.7496688741721854, |
| "ece": 0.021456977493178177, |
| "n": 5285 |
| }, |
| "score": { |
| "acc": 0.5204835460040296, |
| "ece": 0.008651888864610293, |
| "n": 2978 |
| }, |
| "all": { |
| "acc": 0.6919104136663853, |
| "ece": 0.02142799871209621, |
| "n": 20137 |
| } |
| }, |
| "note": "evaluation pool (development and test splits), full fit" |
| }, |
| "n": { |
| "fit": 3560, |
| "many": 1500, |
| "eval": 20137 |
| }, |
| "engine": "fast", |
| "fitted_on": { |
| "split": "out-of-distribution calibration split (sealed-seed BH-, JDI- and agent-shaped families, many-option items)", |
| "orders": 2, |
| "readouts": "basal-serve --mode fast, bf16" |
| }, |
| "certified_on": "the same split", |
| "evaluated_on": "development and test splits, leave-one-family-out and leave-one-language-out" |
| } |
|
|
