Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on Harmful Questions
Loading...
0
ASR (None)
STAR1-R1-Distill-7B
-0.16
0.92
2
3.08
May 21, 2025
ASR (None)
ASR (PAP)
ASR (PAIR)
ASR (Avg)
Updated 1mo ago
Evaluation Results
Method
Method
Links
ASR (None)
ASR (PAP)
ASR (PAIR)
ASR (Avg)
STAR1-R1-Distill-7B
Model Series=7B Models...
2025.05
0
14
42
18.7
RealSafe-R1-7B
Model Series=7B Models...
2025.05
0
2
8
3.3
STAR1-R1-Distill-32B
Model Series=32B Model...
2025.05
0
12
46
19.3
RealSafe-R1-32B
Model Series=32B Model...
2025.05
0
4
12
5.3
Improved CoT
Model Series=32B Model...
2025.05
0
2
8
3.3
Improved CoT
Model Series=7B Models...
2025.05
4
4
12
6.7
Feedback
Search any
task
Search any
task