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

commonsense inference on HellaSwag (test)

84.89Accuracy

FP16

72.46275.688578.91582.1415Mar 30, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.03
84.89
2024.03
79.06
2024.03
77.33
2024.03
72.94