Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Alignment Evaluation on BeaveTails (test)
Loading...
37.2
HS Score
SFT
13.488
19.644
25.8
31.956
Jun 10, 2025
HS Score
FA Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
HS Score
FA Score
SFT
Utility Task=AGNEWS
2025.06
37.2
84.9
BEA
Utility Task=AGNEWS
2025.06
36.8
84.2
Lisa-aligned
Utility Task=AGNEWS
2025.06
31.4
85.8
Lisa-base
Utility Task=AGNEWS
2025.06
30.8
83.1
SafeInstr
Utility Task=AGNEWS
2025.06
21.6
83.2
Safe LoRA
Utility Task=AGNEWS
2025.06
18.8
82.6
AsFT
Utility Task=AGNEWS
2025.06
14.4
82.9
Feedback
Search any
task
Search any
task
