hugpadhugpad

failspy / llama-3-70B-Instruct-abliterated-GGUF

Other56534Updated May 8, 2024
Hugging Face

Downloads

565

Likes

34

Last updated

May 8, 2024

License

Not available

This is meta-llama/Llama-3-70B-Instruct with orthogonalized bfloat16 safetensor weights, generated with the methodology that was described in the preview paper/blog post: 'Refusal in LLMs is mediated by a single direction' which I encourage you to read to understand more.

Tasks & library

  • transformers

Tags

  • transformers
  • gguf
  • conversational

Technical specifications

ParametersNot available
Librarytransformers
LanguagesNot available
LicenseNot available