# LLM Post-Training Glossary

The canonical language for this workspace. Covers how large language models are trained
and, especially, *post-trained* (fine-tuning, RLHF/RL, alignment).

> **How this file grows:** a term is added here only once you've *demonstrated* you
> understand it (e.g. answered correctly in a lesson), not when it's first introduced.
> That keeps the glossary a record of compressed knowledge rather than a dictionary.
> Terms will start landing here after Lesson 0001.

## Terms

_(none yet — you're about to earn the first ones in Lesson 0001)_