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

Common Sense Reasoning on HellaSwag (Accuracy)

91.84Accuracy

Mistral-Small

22.856840.765958.67576.5841May 20, 2025Jul 28, 2025Oct 5, 2025Dec 13, 2025Feb 20, 2026Apr 30, 2026Jul 8, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2025.05
91.84
2025.05
88.28
2025.05
88.25
2025.05
88.22
2025.05
87.91
2025.05
87.81
2025.05
87.76
2025.05
87.62
2025.05
87.42
2025.05
79.83
2026.07
79.8
2026.07
79.34
2026.07
79.34
2026.07
79.3
2026.07
79.07
2026.07
79
2026.07
78.84
2026.07
78.79
2026.07
77.6
2026.07
76.38
2026.07
75.4
2026.07
75.23
2026.07
69.56
2025.11
64.9
2025.11
64.2
2026.07
63.74
2025.11
63.3
2025.11
62
2025.11
61.9
2025.11
61.3
2025.11
61
2025.11
60.8
2025.11
59.5
2025.11
58.7
2025.11
58.3
2025.11
58.1
2025.11
58
2025.11
57.8
2025.11
57.3
2025.11
57.2
2025.11
57.1
2025.11
57
2025.11
57
2025.11
56.5
2025.11
56.3
2025.11
55.5
2025.11
55
2025.11
54
2025.11
52.6
2026.07
51.65
2025.11
51.4
2025.11
51.1
2025.11
51
2025.11
50.4
2025.11
48.1
2025.11
47
2026.06
46.5
2026.06
46.3
2025.11
45.1
2025.11
45
2026.05
44.81
2026.05
44.75
2025.11
44.6
2025.11
44.3
2026.05
44.18
2025.11
44
2026.05
43.94
2025.11
43.1
2025.11
42.7
2025.11
41
2026.06
39
2026.06
38.7
2025.11
31.2
2025.11
31.2
2026.06
30
2026.06
29.6
2026.05
27.13
2026.05
27.07
2026.05
26.85
2026.05
26.22
2026.05
26.17
2026.05
26.07
2026.05
25.75
2026.05
25.7
2026.05
25.51