Commonsense Reasoning on HellaSwag published (test)
82.35Accuracy
CORAL
Updated 4d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.02 | 82.35 | 3.1 | 2.23 | 45.95 | 14.6 | |
2026.02 | 81.98 | 11.37 | 5.74 | 42.15 | 13.29 | |
2026.02 | 81.41 | 10.43 | 9.59 | 49.8 | 16.06 | |
2026.02 | 81.23 | 32.88 | 15.89 | 67.88 | 24.34 | |
2026.02 | 80.79 | 14.64 | 15.73 | 59.44 | 20.07 | |
2026.02 | 77.56 | 28.51 | 13.72 | 66.49 | 23.68 | |
2026.02 | 76.61 | 11.45 | 12.94 | 55.97 | 18.59 | |
2026.02 | 76.43 | 29.48 | 14.28 | 68.99 | 24.87 | |
2026.02 | 74.43 | 12.18 | 8.62 | 63.43 | 21.81 | |
2026.02 | 74.11 | 59.35 | 13.36 | 148.06 | 54.1 | |
2026.02 | 71.07 | 10.56 | 11.45 | 12.94 | 55.97 | |
2026.02 | 70.03 | 10.92 | 7.75 | 64.47 | 22.28 | |
2026.02 | 54.77 | 71.61 | 21.48 | 75.38 | 30.73 | |
2026.02 | 52.29 | 62.09 | 18.55 | 73.84 | 29.9 | |
2026.02 | 51.53 | 64.2 | 19.31 | 76.62 | 31.4 |