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

Commonsense Reasoning on HellaSwag (leaderboard)

95.6Accuracy

Human

81.14484.89788.6592.403Mar 24, 2021
Updated 1mo ago

Evaluation Results

MethodLinks
2021.03
95.6
2021.03
93.9
2021.03
85.5
2021.03
85
81.7