LLM Dictionary

A public glossary of LLM concepts, written in simple Markdown.

This repository explains large language model ideas in simple language, with short notes on how models work, how they are trained, and how people use them in real applications.

What's inside

The docs are organized into a few main sections:

• Fundamentals: hardware, tokens, embeddings, and attention
• Training: pretraining, fine-tuning, and RLHF
• Applications: interacting with LLMs, RAG, prompting, and agents
• Concepts: softmax, transformers, APIs, session state, quantization, and more
• Adjacent ideas: topics like PCA, OCR, and Python f-strings

Start Reading

• Docs index
• Interacting with LLMs
• Prompting
• RAG
• Pretraining
• Tokenization
• Embeddings
• Attention

Goal

The goal of this project is to build a clear, beginner-friendly dictionary for modern LLM topics. It should be easy to browse, easy to edit, and useful as a quick learning reference.