Commonsense Reasoning on Hellaswag (HS and Accuracy)
50HS Score
Pre-unlearn
Updated 16d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2025.10 | 50 | - | |
2025.10 | 44.7 | - | |
2026.03 | 44.5 | 94.1 | |
2026.03 | 44.22 | 92.55 | |
2026.03 | 42.91 | 93.83 | |
2025.10 | 42.7 | - | |
2025.10 | 42.2 | - | |
2025.10 | 42.1 | - | |
2025.10 | 41.7 | - | |
2025.10 | 39.4 | - | |
2025.10 | 36.7 | - | |
2025.10 | 35.8 | - | |
2025.10 | 33.2 | - | |
2025.10 | 31.1 | - | |
2025.10 | 31 | - | |
2025.10 | 30.8 | - | |
2026.03 | 30.55 | - | |
2026.03 | 30.4 | - | |
2026.03 | 30.35 | 91.98 | |
2025.10 | 29.7 | - | |
2026.03 | 29.27 | - | |
2026.03 | 29.13 | - | |
2026.03 | 28.99 | - | |
2026.03 | 28.65 | - | |
2025.10 | 28.4 | - | |
2026.03 | 26.88 | - | |
2026.03 | 26.03 | - | |
2026.03 | 26.03 | - | |
2026.03 | 25.93 | - | |
2026.03 | 25.45 | - | |
2026.03 | 16.43 | 88.47 | |
2026.03 | 15.15 | 92.81 | |
2026.03 | 14.11 | 93.78 | |
2026.03 | 13.82 | 93.7 | |
2026.03 | 10.69 | 92.65 | |
2026.03 | 8.72 | 92.56 | |
2026.03 | 6.86 | 92.33 | |
2026.03 | 2.71 | 92.22 | |
2026.03 | 2.22 | 88.57 | |
2026.03 | 0.81 | 88.76 | |
2026.03 | 0.76 | 91.62 | |
2026.03 | 0.48 | 88.45 | |
2026.03 | 0.47 | 88.42 | |
2025.01 | - | 50.8 | |
2025.01 | - | 50.82 | |
2025.01 | - | 52.03 | |
2025.01 | - | 52.97 | |
2025.01 | - | 53.56 | |
2025.01 | - | 54.12 |