Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Jailbreak Attack Robustness on Jailbreak Attack Evaluation Set Llama-3 8B
Loading...
100
GCG Robustness Score
GSAE
38.848
54.724
70.6
86.476
Dec 7, 2025
GCG Robustness Score
AutoDAN Robustness Score
TAP Robustness Score
Adaptive Robustness Score
Updated 4d ago
Evaluation Results
Method
Method
Links
GCG Robustness Score
AutoDAN Robustness Score
TAP Robustness Score
Adaptive Robustness Score
GSAE
Backbone=Llama-3 8B, T...
2025.12
100
95.1
90.1
92.4
SAE steering
Backbone=Llama-3 8B
2025.12
72.5
68.2
65
61.4
SafeSwitch
Backbone=Llama-3 8B
2025.12
68.3
84
40.1
39.5
Safety-Tuned Baseline
Backbone=Llama-3 8B
2025.12
65.4
55.3
60.1
50.3
CAA
Backbone=Llama-3 8B, T...
2025.12
58.1
55
49.3
46.5
Prompting guardrails
Backbone=Llama-3 8B
2025.12
41.2
36.1
32.4
28
Feedback
Search any
task
Search any
task