Safety Classification on Safety Evaluation Scenarios Hate Speech
99.9Safe Classification RateLlama-2-7B-Chat
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Llama-2-7B-ChatAmnesia Status=Disabled2026.03 | 99.9 | 0.1 | |
| AmnesiaAmnesia Status=Enabled, Base Model=Llama-2-7B-Chat2026.03 | 43.3 | 56.7 |