LLM basics¶
A model consumes a token context and produces more tokens. The harness determines what those tokens can do in the world.
Working notes¶
- [TODO: illustrate tokens and the finite context window.]
- [TODO: explain KV caching as reuse of prior computation, not durable project memory.]
- [TODO: distinguish reasoning tokens, model size and mixture-of-experts routing.]
- [TODO: connect context growth to tool output and compaction.]
Before the course¶
- [TODO: select a compact worked example and its verification artifact.]
- [TODO: rehearse the live walkthrough and prepare a fallback.]