Is supervised fine-tuning the same as pre-training?
No. Pre-training commonly builds broad capabilities from a large data mixture. SFT usually adapts an existing model using examples that demonstrate particular outputs or behaviors.
Explore This Question
Search for additional explanations using these tools:
More FAQs in What is supervised fine-tuning?
Often a retrieval system or database is easier to update and verify. Fine-tuning is more naturally suited to recurring response patterns and task behavior.
Yes. Narrow or inconsistent training can reduce other capabilities or introduce unwanted patterns. Evaluation should check both the target task and important existing behaviors.
Still have questions?
Browse all our FAQs or reach out to our support team