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

Commonsense Inference on HellaSwag

88.97Accuracy

N-3-Super 120B-A12B-Base

62.023669.019376.01583.0107Sep 27, 2025Nov 9, 2025Dec 23, 2025Feb 5, 2026Mar 20, 2026May 3, 2026Jun 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
88.97
2026.04
87.7
2026.02
85.75
2026.02
85.57
2026.03
85.53
2026.03
85.32
2026.04
84.69
2026.02
84.63
2026.03
84.55
2026.03
84.38
2026.03
84.15
2026.03
83.32
2026.03
83.01
2026.03
82.7
2026.02
81.4
2026.02
80.5
2026.02
80.34
2025.09
80.2
2025.09
80
2025.09
80
2025.09
79.9
2025.09
79.8
2025.09
79.14
2026.03
79.06
2025.09
79
2025.09
78.9
2026.06
78.87
2025.09
78.7
2025.09
78.64
2026.06
78.5
2025.09
78.4
2025.09
78.4
2025.09
78.38
2025.09
78.3
2025.09
78.3
2025.09
77.6
2025.09
77.59
2025.09
77.5
2025.09
77.3
2026.06
77.19
2026.06
77.17
2025.09
77
2026.06
76.91
2025.09
76.9
2026.06
76.89
2025.09
76.63
2026.06
76.48
2026.06
76.48
2025.09
76
2025.09
75.52
2025.09
75.2
2025.09
75.04
2026.06
75.01
2026.06
74.97
2026.03
74.9
2026.03
74.9
2026.03
74.84
2026.03
74.72
2026.03
74.6
2026.06
74.5
2026.03
74.3
2026.06
74.29
2026.03
74.22
2026.06
74.15
2026.06
73.97
2026.06
73.96
2025.09
73.94
2026.06
73.89
2025.09
73.8
2026.06
73.65
2026.03
73.38
2026.06
73.27
2026.03
73.18
2026.06
73.11
2026.06
72.9
2026.03
72.36
2025.09
72.06
2026.06
71.92
2026.06
71.27
2026.06
71.22
2026.06
71.15
2026.06
70.98
2026.03
70.8
2026.04
70.8
2026.06
70.6
2026.04
70.26
2026.06
69.96
2026.06
69.82
2026.06
69.54
2026.04
69.32
2026.04
67.92
2026.04
67.71
2026.06
67.28
2026.06
66.92
2026.06
66.53
2026.04
66.41
2026.06
65.93
2026.03
64.52
2026.03
63.86
2026.06
63.06
Showing 100 of 171 rows