Elstuhn/llama-3.2-1B-Instruct-abliterated

Elstuhn/llama-3.2-1B-Instruct-abliterated is a 1 billion parameter instruction-tuned Llama-3.2 model developed by Elston. This model is specifically engineered to be uncensored, significantly reducing safety filter refusals by 90% compared to its base model. It is primarily designed for applications requiring a less restrictive language model, offering greater flexibility in content generation.