Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Prompt-Response Safety Routing on HarmBench
Loading...
55.92
Routing F1
SafeRoute
17.8352
27.7226
37.61
47.4974
Feb 18, 2025
Routing F1
Updated 4d ago
Evaluation Results
Method
Method
Links
Routing F1
SafeRoute
small_model=Llama-Guar...
2025.02
55.92
SafeRoute
Protocol=Prompt-Respon...
2025.02
51.24
Entropy
small_model=Llama-Guar...
2025.02
43.45
Entropy
Protocol=Prompt-Respon...
2025.02
40.94
+CC
small_model=Llama-Guar...
2025.02
38.28
CC
Protocol=Prompt-Respon...
2025.02
37.86
BC
Protocol=Prompt-Respon...
2025.02
32.62
+BC
small_model=Llama-Guar...
2025.02
28.46
+TS
small_model=Llama-Guar...
2025.02
22.64
TS
Protocol=Prompt-Respon...
2025.02
19.3
Feedback
Search any
task
Search any
task