List your company

Evaluation Suite (ARC-e, ARC-c, PIQA, HellaS., WinoG., BoolQ)

Benchmarks

TaskDatasetSOTA resultTrendResults
Evaluation Suite (ARC-e, ARC-c, PIQA, HellaS., WinoG., BoolQ) Zero-shot (test)
92.59ARC-e Accuracy
30