Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Attack Detection on Benign Samples 105K sample set
Loading...
40
Benign FPR
PromptGuard 2
15.6
180.3
345
509.7
Feb 15, 2026
Benign FPR
Updated 4d ago
Evaluation Results
Method
Method
Links
Benign FPR
PromptGuard 2
acronym=PG
2026.02
40
LlamaGuard
acronym=LG
2026.02
300
Llama-as-Judge
acronym=LJ, prompting=...
2026.02
440
LogReg (Ours)
input=raw activations,...
2026.02
650
Feedback
Search any
task
Search any
task
