@hn50
@AmpBenzScientist perhaps this could be of interest for you.

Tool that seems to be reversing the model, finding and then avoiding the censor branch. Seems to me like a big cracking tool for llms.

Follow

@PawelK The Xbox One was finally cracked. The security on it is insane.

Sign in to participate in the conversation
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.