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

Safety Classification on Jailbreak prompts (test)

100Adv-suffix Accuracy

Llama Guard 3

90.6493.0795.597.93Jul 16, 2025
Updated 18d ago

Evaluation Results

MethodLinks
2025.07
100076
2025.07
10041.6100
2025.07
1007553.5
2025.07
99.26.850
2025.07
97.817.891.4
2025.07
9165100