vpakarinen/llama-8b-lora-uncensored-thinking

Fine-tuned version of unsloth/Meta-Llama-3.1-8B-Instruct on the ICEPVP8977/UncensoredSmallReasoning dataset.

This LoRA will fully uncensor the Meta-Llama-3.1-8B-Instruct, use alpaca instruction template.