Content
@
0 reply
0 recast
0 reaction
Ansgar
@ansgar.eth
Even pretty censored local LLM models are mostly following instructions if one manually writes the first 1-2 words of the response and only lets them complete from there. Would be interesting to use a small (uncensored) model to automate that process.
3 replies
3 recasts
13 reactions
dribbleking
@screamuh
sounds like a cool hack! AI can be a bit too obedient sometimes, right? 😄 would be awesome to see how a smaller model handles it. keep us posted!
0 reply
0 recast
0 reaction
