Commonsense Reasoning Suite
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Commonsense Reasoning Suite (Arc, Hellaswag, Obqa, Piqa, Race, Siqa, Winogrande) (test)
26.54Arc-c
4
Commonsense Reasoning Suite (OBQA, HellaSwag, ARC-E, WSC, Winogrande, BoolQ, PIQA)
49.4Average Accuracy
2