nbeerbower/Llama-3.1-Nemotron-lorablated-70B

The nbeerbower/Llama-3.1-Nemotron-lorablated-70B is a 70 billion parameter Llama-3.1-Nemotron-based model with a 32768 token context length. It was created by nbeerbower by merging nvidia/Llama-3.1-Nemotron-70B-Instruct-HF with mlabonne/Llama-3-70B-Instruct-abliterated-LORA using task arithmetic. This model is specifically designed to be an uncensored version of the base Nemotron model, leveraging a LoRA ablation technique.