Picture for David Atanasov

David Atanasov

Representation noising effectively prevents harmful fine-tuning on LLMs

Add code
May 23, 2024
Figure 1 for Representation noising effectively prevents harmful fine-tuning on LLMs
Figure 2 for Representation noising effectively prevents harmful fine-tuning on LLMs
Figure 3 for Representation noising effectively prevents harmful fine-tuning on LLMs
Figure 4 for Representation noising effectively prevents harmful fine-tuning on LLMs
Viaarxiv icon