ninjasaid13B to LocalLLaMA@poweruser.forumEnglish · 11 months agoLoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70Barxiv.orgexternal-linkmessage-square13fedilinkarrow-up11arrow-down10
arrow-up11arrow-down1external-linkLoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70Barxiv.orgninjasaid13B to LocalLLaMA@poweruser.forumEnglish · 11 months agomessage-square13fedilink
minus-squarea_beautiful_rhindBlinkfedilinkEnglisharrow-up1·11 months agoYea, no shit. I did it to vicuna using proxy logs. The LLM attacks are waaaay more effective once you find the proper string. I’d run the now working 4-bit version on more models, it’s just that I tend to boycott censored weights instead.
Yea, no shit. I did it to vicuna using proxy logs. The LLM attacks are waaaay more effective once you find the proper string.
I’d run the now working 4-bit version on more models, it’s just that I tend to boycott censored weights instead.