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

BoolQ, Winogrande, PIQA, OpenBookQA, HellaSwag, ARC

Benchmarks

Task NameDataset NameSOTA ResultTrend
Zero-shot Question AnsweringBoolQ, Winogrande, PIQA, OpenBookQA, HellaSwag, ARC-Easy, ARC-Challenge Zero-shot
BoolQ Accuracy (Zero-shot)83.76
21
Showing 1 of 1 rows