metadata

license: llama3.1
language:
  - en
pipeline_tag: question-answering
base_model: ruliad/deepthought-8b-llama-v0.01-alpha
tags:
  - abliterated
  - uncensored

huihui-ai/deepthought-8b-abliterated

This is an uncensored version of ruliad/deepthought-8b-llama-v0.01-alpha created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.