Calibration
HLE-Verified → Vals GPQA Diamond
Vals GPQA Diamond is estimated from HLE-Verified with a linear curve fitted on 6 models measured on both: y = 0.2216·x + 0.8172, R² = 0.75, cross-validated error 7.2 pp. It is used for 0 estimates.
Vals GPQA Diamond is estimated from HLE-Verified with a linear curve fitted on 6 models measured on both: y = 0.2216·x + 0.8172, R² = 0.75, cross-validated error 7.2 pp. It is used for 0 estimates.