llmfinetuning

This repository contains 29 complete modules covering everything from fundamental concepts to cutting-edge fine-tuning techniques for Large Language Models. Each module includes detailed Jupyter notebooks with theoretical explanations, practical code examples, and best practices.

End-to-end text summarization project that fine-tunes PEGASUS (`google/pegasus-cnn_dailymail`) on the SAMSum dataset and serves predictions through a FastAPI API.

LLM Post-Training, RLHF, PPO, DPO, etc

This paper studies prompt robustness and ambiguity handling for small instruction-tuned LLMs (Qwen2.5-1.5B/3B) in educational tutoring. It evaluates corruption-augmented supervised fine-tuning on GSM8K and DPO in two roles: i augmenting robustness for math reasoning under noisy prompts, ii inducing clarification-seeking behavior on ambiguous prompt

LLM fine-tuning pipeline using Unsloth, LoRA and QLoRA for efficient domain-specific training

End-to-end LLM financial assistant: real-time news ingestion (Bytewax), QLoRA fine-tuning, RAG inference (LangChain + Qdrant), Streamlit & Gradio UI

Fine-tune Large Language Models using LoRA, QLoRA, and Unsloth on Windows with this verified, all-in-one setup guide.