Calibration

AA-HLE → MMLU-Redux

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

MMLU-Redux is estimated from AA-HLE with a offset logistic curve fitted on 7 models measured on both: y = 0.0000 + (0.9483 − 0.0000) / (1 + exp(−200.00·(x − 0.0784))), R² = 0.94, cross-validated error 4.6 pp. It is used for 0 estimates.