Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on LM Safety Evaluation Dataset
Loading...
0
Unsafe Rate
Goal
-0.972
5.589
12.15
18.711
Apr 17, 2026
Unsafe Rate
Over-Refusal Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Unsafe Rate
Over-Refusal Rate
Goal
Model=Mistral-Nemo
2026.04
0
41.7
Goal
Model=Mistral-Small
2026.04
0
61.3
Beam
Model=Mistral-Small
2026.04
3.83
24.3
Beam
Model=Mistral-Nemo
2026.04
4.33
25.3
Base
Model=Mistral-Small
2026.04
15.8
16.3
CB
Model=Mistral-Small
2026.04
17.2
16.3
Base
Model=Mistral-Nemo
2026.04
24.2
17.7
CB
Model=Mistral-Nemo
2026.04
24.3
15
Feedback
Search any
task
Search any
task
