failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5

failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5 is a 70 billion parameter instruction-tuned causal language model derived from Meta-Llama-3-70B-Instruct. This model utilizes an orthogonalization methodology to specifically inhibit refusal behaviors, making it "uncensored" without altering other core functionalities. It maintains the original model's knowledge and training while surgically removing the tendency to refuse user requests, offering a distinct approach compared to traditional fine-tuning.