BadLlama: cheaply removing safety fine-tuning from Llama 2-Chat 13B

arXiv:2311.00117v2 Announce Type: replace
Abstract: Llama 2-Chat is a collection of large language models that Meta developed and released to the public. While Meta fine-tuned Llama 2-Chat to refuse to output harmful content, we hypothesize that public access to model weights enables bad actors to cheaply circumvent Llama 2-Chat's safeguards and weaponize Llama 2's capabilities for malicious purposes. We demonstrate that it is possible to effectively undo the safety fine-tuning from Llama 2-Chat 13B with less than $200, while retaining …

13b abstract actors arxiv chat collection cs.cl fine-tuning language language models large language large language models llama llama 2 meta public safeguards safety type