Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Attack Detection on Extraction Attacks 105K sample set
Loading...
100
Detection Rate
PromptGuard 2
11.808
34.704
57.6
80.496
Feb 15, 2026
Detection Rate
Updated 4d ago
Evaluation Results
Method
Method
Links
Detection Rate
PromptGuard 2
acronym=PG
2026.02
100
LogReg (Ours)
input=raw activations,...
2026.02
79
Llama-as-Judge
acronym=LJ, prompting=...
2026.02
31.8
LlamaGuard
acronym=LG
2026.02
15.2
Feedback
Search any
task
Search any
task