Fine-tune Decision
12%
Lesson 1 of 8

When should you fine-tune?

Before using training compute, decide whether the problem is instructions, knowledge or repeated behavior.

Active reading
Prompting

Change instructions at run time.

RAG

Add current external knowledge.

Fine-tuning

Change learned behavior patterns.

LoRA

Train a smaller adapter instead of every parameter.

Fine-tuning is not the first fix

If the problem is missing facts, use retrieval. If the problem is an unclear instruction, improve prompting. Fine-tuning becomes more relevant when you need a repeated behavior, style or task pattern that examples can demonstrate consistently.

Try it
Describe the problem before choosing training.
Common mistake

Do not fine-tune to “upload knowledge.”

Frequently changing facts are usually a better fit for retrieval or another external knowledge mechanism.

Key takeaway

Choose the smallest intervention that solves the problem: prompting first, retrieval for knowledge, fine-tuning for repeatable learned behavior.

Next lesson

Training Data

Continue to the next lesson.

Continue →