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

Commonsense Reasoning on HellaSwag 5-shot

69.4Accuracy

MoE 128e8a1s

55.98459.46762.9566.433May 22, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.05
69.4
2026.05
69.3
2026.05
56.5