Cyr-Ch/lora-qlora-101

Parameter-efficient fine-tuning toolkit for LLMs using LoRA and QLoRA. Colab-ready notebooks with detailed theory, code explanations, and production deployment guides. Supports Llama-2, Mistral, and other open-source models.

语言：Jupyter Notebook · GitHub 星标：4 · Fork：1

话题：finetuning · huggingface · llm · lora · peft · peft-fine-tuning-llm · pytorch · qlora

在 GitHub 上查看原仓库