Visual-Adversarial-Examples-Jailbreak-Large-Language-Models icon indicating copy to clipboard operation
Visual-Adversarial-Examples-Jailbreak-Large-Language-Models copied to clipboard

Have you updated the model parameter in the training process?

Open ghLcd9dG opened this issue 1 year ago • 3 comments

Thanks for case. Have you updated the model parameter in the training process?

ghLcd9dG avatar May 11 '24 08:05 ghLcd9dG

Hi, when doing adversarial examples attacks, model weights won't be changed. The attacks are conducted by only making perturbations to the inputs.

Unispac avatar May 11 '24 08:05 Unispac

Okay, thanks. Let me look through the detail of it.

ghLcd9dG avatar May 11 '24 09:05 ghLcd9dG

Hi, may I ask does it support multi-GPU. I am running evaluation on A100. I guess 99k of real toxicity might cause a whole day to run.

ghLcd9dG avatar May 14 '24 05:05 ghLcd9dG