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

Commonsense Benchmarks

Benchmarks

Task NameDataset NameSOTA ResultTrend
Commonsense ReasoningCommonsense Benchmarks (ARC-e, ARC-c, OBQA, PIQA, Hella) zero-shot
Average Accuracy60.34
7
Showing 1 of 1 rows