A small, educational codebase and notebook that implements and documents core concepts behind large language models (LLMs) from first principles. This repository collects notes, math derivations, and simple implementations so you can learn the building blocks of modern LLMs and experiment with training & inference.
This repo is intended for learning and research — not a production LLM.
tutorial.ipynb— the core learning notebook with notes, math, and runnable code.explanability/— supporting code that documents explainability concepts used in the repo.