One-click model liberation toolkit for removing LLM refusal behaviors
Fully automatic censorship removal for transformer-based language models via abliteration
