HOW-LLMS-WORK4 MIN READ
Sort Training, Tuning, and Inference
Distinguish pretraining, fine-tuning, prompting, retrieval, and inference-time controls.
Sort each technique by where it primarily changes the system. Changes learned model behavior Changes runtime context Changes generation behavior Measures the result Pretraining on broad corpora Fine-tuning on task examples Prompt instructions Retrieving current policy passages Lowering temperature Structured decoding or schema constraints Regression eval set Human scoring rubric lifecycle
Read the full lesson
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in