Orenguteng/Llama-3.1-8B-Lexi-Uncensored

Orenguteng/Llama-3.1-8B-Lexi-Uncensored is an 8 billion parameter instruction-tuned causal language model based on Meta's Llama-3.1 architecture. This model is specifically fine-tuned to be uncensored, aiming to improve evaluation scores by removing biases and refusals without adding new knowledge. It retains original instruct capabilities while demonstrating enhanced performance, making it suitable for applications requiring highly compliant and unfiltered responses.