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

Truthfulness on TruthfulQA (TruthfulQA metric)

68.3TruthfulQA Score

MADS8B

40.552847.756454.9662.1636May 29, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
68.3
2026.05
65.24
2026.05
65.24
2026.05
64.5
2026.05
64.38
2026.05
63.04
2026.05
61.08
2026.05
60.83
2026.05
59.73
2026.05
59.61
2026.05
58.75
2026.05
58.07
2026.05
57.28
2026.05
57.16
2026.05
56.06
2026.05
54.71
2026.05
53.98
2026.05
52.63
2026.05
50.43
2026.05
41.62