Safety Evaluation on Alpaca Mix Safe
0Refusal RateRandom feature
Evaluation Results
| Method | Links | |
|---|---|---|
| Random featureModel=Qwen3-1.7B-it2026.06 | 0 | |
| No SteeringModel=Qwen3-4B-it2026.06 | 0.8 | |
| No SteeringModel=Qwen3-1.7B-it2026.06 | 0.9 | |
| Random featureModel=Qwen3-4B-it2026.06 | 4.3 | |
| Target featureModel=Qwen3-1.7B-it2026.06 | 10.6 | |
| Target featureModel=Qwen3-4B-it2026.06 | 28.5 |