Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Hack Detection on Terminal-Wrench (full traces)
Loading...
96.79
AUC
gpt-5.4
91.4756
92.8553
94.235
95.6147
Jun 8, 2026
AUC
TPR@5%FPR
Updated 1mo ago
Evaluation Results
Method
Method
Links
AUC
TPR@5%FPR
gpt-5.4
Condition=hack, Contex...
2026.06
96.79
82.35
gpt-5.4
Condition=sanitized, C...
2026.06
94.74
61.87
gpt-5.4
Condition=stripped, Co...
2026.06
91.68
44
Feedback
Search any
task
Search any
task
