Reasoning on HellaSwag 10-shot
81.7Normalized Log Accuracy
Tülu
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 81.7 | 4.66 | |
2026.03 | 80.9 | - | |
2026.03 | 78.3 | 5.34 | |
2026.03 | 77.5 | 5.34 | |
2026.03 | 77.3 | 4.66 | |
2026.03 | 76.4 | 4.66 | |
2026.03 | 74.7 | 5.34 | |
2026.03 | 66.5 | - |