Safety
AdvPrompter
AISafetyLab evaluation of 50 AdvPrompter prompts against Vicuna-7B-v1.5, without a defense and with SafeDecoding. Each released judgment records whether Llama Guard 3 classified the response as unsafe.
50items
2subjects
MIT (AISafetyLab result release)license
safetydomain
textmodality
item-level responses released
Saturation status: No
Response matrix
Loading response matrix…
Correct (1)Incorrect (0)Unobserved
Scale: 1 = correct · 0 = incorrect