Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on Alpaca Mix (Unsafe)
Loading...
93
Refusal Rate
Target feature
-3.408
21.621
46.65
71.679
Jun 25, 2026
Refusal Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Refusal Rate
Target feature
Model=Qwen3-4B-it
2026.06
93
No Steering
Model=Qwen3-4B-it
2026.06
76
Target feature
Model=Qwen3-1.7B-it
2026.06
66.7
No Steering
Model=Qwen3-1.7B-it
2026.06
41.7
Random feature
Model=Qwen3-4B-it
2026.06
1.2
Random feature
Model=Qwen3-1.7B-it
2026.06
0.3
Feedback
Search any
task
Search any
task