Calibration

MathVision w/ Python → MathVision

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

MathVision is estimated from MathVision w/ Python with a offset logistic curve fitted on 5 models measured on both: y = 0.8980 + (0.9513 − 0.8980) / (1 + exp(−200.00·(x − 0.9648))), R² = 0.98, cross-validated error 0.7 pp. It is used for 0 estimates.