Back
advanced
Advanced Fine-Tuning

Instruction Tuning

Understand how supervised instruction examples turn base models into helpful assistants

20 min read· Instruction Tuning· Fine-Tuning· Supervised Learning· Prompting

Instruction Tuning

Base models complete text. Instruction-tuned models follow tasks.

Instruction tuning uses examples like:

text
instruction -> ideal response

to teach the model how users expect an assistant to behave.

What instruction data teaches

  • follow task wording
  • answer directly
  • use requested formats
  • refuse unsafe requests
  • ask clarifying questions
  • use tools when examples include tools

Dataset quality

Good examples are:

  • clear
  • diverse
  • correct
  • formatted consistently
  • reviewed
  • representative of real use

Bad instruction data creates generic, verbose, or unsafe assistants.

Instruction tuning vs prompting

PromptingInstruction tuning
changes behavior per requestchanges model behavior
fast to iterateneeds training data
no deployment burdenneeds evals and rollout
good first stepgood for repeated behavior

Knowledge check

Q1: What does instruction tuning teach?

How to respond to tasks like an assistant.

Q2: Why not instruction-tune first?

Prompting and RAG may solve the problem cheaper and faster.