Open-Source LLM Fine-Tuning (PEFT / LoRA / QLoRA)

Parameter-efficient fine-tuning of an open-source LLM on a 10-year NPM ecosystem dataset, deployed on the Hugging Face Hub.