MMLU
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MMLU
65.83Accuracy
18
MMLU f_con o=5
99.5Accuracy (Exact)
18
MMLU-Pro
69.3Pass@1
18
MMLU-Sci
54.96Score
18
MMLU Biology
0.882Accuracy
18
MMLU Physics
60.78Accuracy
18
MMLU Math
51.81Accuracy
18
MMLU
84.76Accuracy
17
MMLU Math
61.09Accuracy
17
MMLU-Pro (test)
95.6MAR
17
MMLU-Pro
81.39Score
17
MMLU-Redux
90Knowledge (MMLU-Redux) Score
17
MMLU pass@1
88pass@1 Accuracy
17
MMLU
0.2ΔMMLU
17
MMLU Redux
87.47Accuracy
17
MMLU-Pro
61.1Accuracy (MMLU-Pro)
17
MMLU HELM
59.48Balanced Accuracy
17
MMLU ProX Health
76.56Accuracy (ZH)
17
Global MMLU-Lite
64.5Accuracy
17
MMLU Security
93.2Accuracy
17
MMLU
61.2Acc
17
MMLU (test)
77.4Accuracy (5-shot)
17
MMLU
71.4Accuracy
16
MMLU
71.4Accuracy
16
MMLU
70.5Accuracy
16