Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
LLM Jailbreaking on Llama3-LAT
Loading...
71
Success Rate First (SRF)
Adaptive Probe-based Steering
-2.84
16.33
35.5
54.67
May 19, 2026
Success Rate First (SRF)
Harmful Behavior (HB)
Success Rate (SR)
Updated 13d ago
Evaluation Results
Method
Method
Links
Success Rate First (SRF)
Harmful Behavior (HB)
Success Rate (SR)
Adaptive Probe-based Steering
2026.05
71
82
98
RD-A
2026.05
23
29
35
Angular
2026.05
10
12
14
RD-C
2026.05
9
13
14
SCAV
2026.05
1
3
1
RepE
2026.05
0
0
0
Feedback
Search any
task
Search any
task
