Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Jailbreak Attack on Gemma-7b five finetuned variants

66.2Average ASR

PGP

-2.64815.22633.150.974Dec 14, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
66.2
2025.12
63.4
2025.12
44.2
2025.12
41.8
2025.12
41.6
2025.12
36.2
2025.12
36
2025.12
34.4
2025.12
25.6
2025.12
14
2025.12
13.4
2025.12
12
2025.12
9
2025.12
6.4
2025.12
4.4
2025.12
0