Overview
Fine-tuned OpenLLaMA-3B-v2 with an uncensored/unfiltered Wizard-Vicuna conversation dataset ehartford/wizard_vicuna_70k_unfiltered. Used QLoRA for fine-tuning. Trained for one epoch on a 48GB GPU (NVIDIA A6000) instance, took ~9 hours to train.
Training code
Code used to train the model is made available here by georgesung.