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

General Language Understanding on CMMLU

77.3Overall Accuracy

NBDiff-7B-BASE

54.066460.098266.1372.1618Dec 7, 2025Dec 12, 2025Dec 17, 2025Dec 22, 2025Dec 27, 2025Jan 1, 2026Jan 6, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
77.3
2025.12
69.9
2025.12
65.7
2025.12
60.9
2026.01
55.9
2026.01
55.53
2026.01
55.51
2026.01
55.48
2026.01
55.44
2026.01
55.38
2026.01
55.23
2026.01
55.16
2026.01
55.16
2026.01
54.96