Zero-shot Common-sense Reasoning Suite (BoolQ, PIQA, HellaSwag, WinoGrande, ARC-e, ARC-c, OBQA)
48.92Avg Accuracy
GradMAP
Updated 2d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.02 | 48.92 | |
2026.02 | 48.68 | |
2026.02 | 45.5 | |
2026.02 | 39.6 | |
2026.02 | 39.21 | |
2026.02 | 38.31 | |
2026.02 | 38.1 | |
2026.02 | 37.66 | |
2026.02 | 37.62 | |
2026.02 | 37.51 | |
2026.02 | 37.32 | |
2026.02 | 36.58 | |
2026.02 | 35.41 | |
2026.02 | 34.49 | |
2026.02 | 32.75 | |
2026.02 | 32.69 | |
2026.02 | 32.2 | |
2026.02 | 30.46 | |
2026.02 | 30.1 | |
2026.02 | 29.26 |