Refusal on Llama-3-8B n≈200
0ASR
Fine-tune
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.04 | 0 | 9 | - | |
2026.04 | 0 | 0.5 | - | |
2026.04 | 0 | 1 | - | |
2026.04 | 0 | - | 10.5 | |
2026.04 | 0 | 28.4 | - | |
2026.04 | 0 | 0.5 | - | |
2026.04 | 0 | 85 | - | |
2026.04 | 0 | 1 | - | |
2026.04 | 0 | - | 8.5 | |
2026.04 | 0.5 | 78.5 | - | |
2026.04 | 0.5 | 82.5 | - | |
2026.04 | 0.5 | 1.5 | - | |
2026.04 | 0.5 | 1.5 | - | |
2026.04 | 0.5 | - | 11 | |
2026.04 | 0.5 | 12 | - | |
2026.04 | 0.5 | 1.5 | - | |
2026.04 | 1 | 12.5 | - | |
2026.04 | 1 | 69.5 | - | |
2026.04 | 1.1 | - | 9.5 | |
2026.04 | 1.6 | 68 | - | |
2026.04 | 2 | 26.4 | - | |
2026.04 | 3.5 | 14.9 | - | |
2026.04 | 4.5 | 82 | - | |
2026.04 | 5 | 3 | - | |
2026.04 | 5.5 | - | 12.5 | |
2026.04 | 5.5 | 81.5 | - | |
2026.04 | 6 | 1 | - | |
2026.04 | 6.5 | 7 | - | |
2026.04 | 6.5 | 23.9 | - | |
2026.04 | 9 | 0.5 | - | |
2026.04 | 11.5 | 24.4 | - | |
2026.04 | 12 | - | 13 | |
2026.04 | 13.5 | 8.5 | - | |
2026.04 | 13.5 | 21.9 | - | |
2026.04 | 14 | 5.5 | - | |
2026.04 | 19.5 | 13.5 | - | |
2026.04 | 28 | 1 | - | |
2026.04 | 30 | - | 10 | |
2026.04 | 30.5 | 3.5 | - | |
2026.04 | 33 | 26.4 | - | |
2026.04 | 45.2 | 1 | - | |
2026.04 | 58 | 3.5 | - |