Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on Adolescent AI Safety dataset N=2,052 (test)
Loading...
0.39
Unsafe Rate
Targeted Rewrite
0.3568
0.5809
0.805
1.0291
May 20, 2026
Unsafe Rate
Unsafe Count
Refusal Rate
Refusal Count
Updated 12d ago
Evaluation Results
Method
Method
Links
Unsafe Rate
Unsafe Count
Refusal Rate
Refusal Count
Targeted Rewrite
Setting=Targeted Rewrite
2026.05
0.39
8
3.75
77
Universal Rewrite
Setting=Universal Rewrite
2026.05
0.73
15
9.11
187
Original Baseline
Setting=Original Baseline
2026.05
1.22
25
11.65
239
Feedback
Search any
task
Search any
task