Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Refusal Evaluation on HarmBench
Loading...
71
Baseline Performance
LLaMA-3.1 8B
67.45
69.225
71
72.775
Feb 11, 2026
Baseline Performance
Post-Intervention Performance
Performance Gain
Updated 4d ago
Evaluation Results
Method
Method
Links
Baseline Performance
Post-Intervention Performance
Performance Gain
LLaMA-3.1 8B
2026.02
71
-
-
CRL-Token
Backbone=LLaMA-3.1 8B,...
2026.02
-
607
5.36
Feedback
Search any
task
Search any
task
