Glossary · Phase 6 · Modern LLM engineering
Headline metric
The one number you agree in advance to judge a model by, so you can't pick whichever number looks best afterwards.
Learn it properly
Capstone: a tuned, evaluated assistant · Fine-tune, evaluate and deploy a specialist assistant. About 45 minutes.
Other terms from the same lesson
- release gate: A test a model must pass before it ships, with the pass mark written down in advance.