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

Multiple Choice Classification on MMLU

100Accuracy

Llama-Guard-3-8B

36.24852.79969.3585.901May 11, 2026May 14, 2026May 17, 2026May 20, 2026May 23, 2026May 26, 2026May 30, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
100
2026.05
99.47
2026.05
99.47
2026.05
99.12
2026.05
99.12
2026.05
89.72
2026.05
88.34
2026.05
83.07
2026.05
80.27
2026.05
80.27
2026.05
80.11
2026.05
80.11
2026.05
80.11
2026.05
80
2026.05
79.44
2026.05
78.48
2026.05
77.92
2026.05
77.92
2026.05
77.91
2026.05
77.47
2026.05
77.24
2026.05
77.02
2026.05
76.68
2026.05
67.11
2026.05
65.43
2026.05
65.43
2026.05
65.25
2026.05
64.56
2026.05
64.28
2026.05
62.61
2026.05
62.61
2026.05
61.87
2026.05
61.64
2026.05
61.36
2026.05
60.69
2026.05
52.1
2026.05
46.8
2026.05
45.9
2026.05
45.85
2026.05
45.54
2026.05
45.47
2026.05
45.24
2026.05
43.57
2026.05
41
2026.05
39.9
2026.05
38.9
2026.05
38.7