Calibration

Decision Index 0.3 Arts → Decision Index 0.3 Tools

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

Decision Index 0.3 Tools is estimated from Decision Index 0.3 Arts with a offset logistic curve fitted on 112 models measured on both: y = 0.0000 + (0.8633 − 0.0000) / (1 + exp(−7.36·(x − 0.2311))), R² = 0.81, cross-validated error 9.6 pp. It is used for 0 estimates.