Calibration

BioMysteryBench (human-difficult) → AA-HLE

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

AA-HLE is estimated from BioMysteryBench (human-difficult) with a offset logistic curve fitted on 6 models measured on both: y = 0.3408 + (0.5496 − 0.3408) / (1 + exp(−200.00·(x − 0.4315))), R² = 0.79, cross-validated error 10.1 pp. It is used for 0 estimates.