Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Calibration on TriviaQA

0.0845Brier Score

Probe (train on TriviaQA)

0.0751160.1384580.20180.265142Jan 6, 2026Jan 12, 2026Jan 19, 2026Jan 25, 2026Feb 1, 2026Feb 7, 2026Feb 14, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2026.02
0.0845--
2026.02
0.1079--
2026.02
0.1113--
2026.02
0.1266--
2026.02
0.1577--
2026.01
0.17020.0226-
2026.01
0.17240.0309-
2026.02
0.1756--
2026.01
0.17890.112-
2026.01
0.1850.0387-
2026.01
0.1850.0895-
2026.02
0.1885--
2026.01
0.19230.1393-
2026.02
0.1933--
2026.01
0.19390.0286-
2026.01
0.19470.0269-
2026.01
0.19660.0314-
2026.01
0.20460.1769-
2026.01
0.2090.1725-
2026.02
0.2101--
2026.01
0.21340.2113-
2026.01
0.21970.191-
2026.01
0.23020.2443-
2026.01
0.23930.2055-
2026.02
0.2431--
2026.01
0.24650.2121-
2026.01
0.25060.2476-
2026.02
0.255--
2026.02
0.2624--
2026.02
0.2639--
2026.02
0.2648--
2026.01
0.27210.2054-
2026.02
0.2745--
2026.02
0.2865--
2026.02
0.297--
2026.02
0.2977--
2026.01
0.30310.2889-
2026.01
0.31180.3406-
2026.01
0.31910.3583-
2026.05
-11.430.8189
2026.05
-7.960.8189
2026.05
-8.110.819
2026.05
-9.580.813
2026.05
-7.990.8574
2026.05
-8.710.8573
2026.05
-8.840.863
2026.05
-9.130.7924
2026.05
-11.050.8006
2026.05
-6.880.8219
2026.05
-4.250.8612