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

Common Sense Reasoning on HellaSwag (0-shot)

84.4Accuracy

fp16

42.155253.122664.0975.0574Oct 30, 2025Nov 27, 2025Dec 26, 2025Jan 24, 2026Feb 22, 2026Mar 23, 2026Apr 21, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
84.4
2026.02
83.1
2026.02
81.7
2026.02
81.1
2026.02
79.7
2025.12
79.2
2026.02
76.9
2026.02
76.2
2026.02
76.2
2025.12
74.33
2026.02
74.2
2025.12
73.85
2026.02
73.4
2025.12
73.02
2026.04
72.96
2025.12
72.43
2026.02
72.2
2025.12
70.21
2026.02
70.2
2025.12
63.07
2025.12
62.84
2025.12
61.4
2026.04
58.54
2025.10
58.25
2026.04
57.7
2025.10
57.43
2025.10
56.88
2025.10
56.53
2025.10
55.76
2025.10
55.67
2025.10
53.42
2025.10
52.91
2026.04
51.54
2025.10
50.93
2025.10
50.23
2025.10
49.53
2025.10
49.16
2025.12
43.78