Red-teaming Safety Evaluation on Edgebench
4.53HS Score
Meta-Llama-3.1-8B (Unaligned)
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| 4.53 | 75 | 85 | ||
| 4.24 | 57 | 69 | ||
2025.05 | 3.61 | - | 41 | |
| 3.59 | 36 | 47 | ||
| 3.32 | 30 | 34 | ||
2025.05 | 3.23 | - | 29 | |
| 3.15 | 33 | 29 | ||
2025.05 | 3.14 | - | 27 | |
2025.05 | 3.08 | - | 23 | |
2025.05 | 2.91 | - | 23 | |
| 2.85 | 29 | 26 | ||
2025.05 | 2.81 | - | 23 | |
| 2.36 | 23 | 18 | ||
2025.05 | 2.35 | - | 14 | |
| 2.32 | 29 | 21 | ||
2025.05 | 2.19 | - | 12 |