LLM-FineTuning-Large-Language-Models

Why we included this project

This is a working collection of Jupyter notebooks for fine-tuning open-weight models, and it earns its place because it shows the actual recipes teams use, not just the theory. The notebooks walk through Llama-3, Mistral 7B, Falcon, Phi, and CodeLlama using PEFT, QLoRA, GPTQ, DPO, and ORPO, so you can follow a concrete path instead of piecing together documentation. Each notebook has a matching YouTube walkthrough, which helps if you are new to parameter-efficient fine-tuning. There are also explainer notebooks on LoRA rank, rotary embeddings, quantization, and chat templates that get into the mechanics behind the code. For a solo developer or small team working with a single GPU, this is a handy reference to keep next to the official docs.