| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| ARC Easy | QTIP | Zero-shot Acc (ARC Easy)78.2 | 67 | 2mo ago | |
| 6-task zero-shot (MMLU, PIQA, ARC-E, ARC-C, Winogrande, OBQA) | Avg. Accuracy (Zero-Shot)63.89 | 59 | 2mo ago | ||
| ARC Challenge | Accuracy51.1 | 46 | 4mo ago | ||
| Zero-shot Evaluation Suite (PIQA, HellaSwag, MMLU, HumanEval, BoolQ, WinoGrande, ARC-E, ARC-C) (test) | PIQA Accuracy82.54 | 37 | 2mo ago | ||
| WinoGrande | Zero-shot Accuracy77.3 | 30 | 4mo ago | ||
| PIQA | Zero-shot PIQA Accuracy81.5 | 30 | 4mo ago | ||
| Reasoning Suite Zero-shot (PIQA, Hella Swag, LAMBADA, ARC-e, ARC-c, SciQ, Race, MMLU) | PIQA80.7 | 21 | 4mo ago |