mridulpant2010/llm_from_scratch ? reverse-engineered prompt

I want you to help me build a simple learning project for understanding how an LLM works from scratch, using Jupyter notebooks.

Please make it beginner friendly and focused on the core concepts, like tokenization, embeddings, attention, and a basic transformer style model, so I can follow along step by step and see how each part works. I’d like the notebooks to be organized by chapter, with clear explanations in the cells, some small examples, and code that actually runs so I can experiment as I learn.

Keep it educational rather than production ready, and make sure the code is written in a clean, human understandable way. If you need to check current docs or examples online to keep things accurate, go ahead and do that.