Commonsense Inference on HellaSwag
52.5Accuracy
Base
Updated 4d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.02 | 52.5 | |
2026.02 | 52.25 | |
2026.02 | 52.1 | |
2026.02 | 52.05 | |
2026.02 | 51.8 | |
2026.02 | 50.25 | |
2026.02 | 49.9 | |
2026.02 | 49.72 | |
2026.02 | 49.45 | |
2026.02 | 43.48 | |
2026.02 | 43.15 | |
2026.02 | 42.9 | |
2026.02 | 42.55 |