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

Reasoning on HellaSwag (Accuracy)

76Accuracy

Base

32.600843.867955.13566.4021Oct 7, 2025Nov 18, 2025Dec 30, 2025Feb 10, 2026Mar 24, 2026May 5, 2026Jun 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
76
2025.10
67.83
2025.10
60.96
2025.10
44.28
2026.06
37.93
2026.06
36.98
2026.06
36.35
2026.06
34.61
2026.06
34.31
2026.06
34.27