Supervised Fine-Tuning

AI and Machine Learning · Language Models · 2021 · also: SFT, Instruction tuning · supervised-fine-tuning.yaml

Continues next-token training on curated demonstrations of the behaviour wanted. Cheap, stable, and bounded by how well the desired behaviour can be written down.

• supersedescorrects · extends
• classifiesspecializes · part-of
• substitutes forapproximates · alternative-to
• depends onrequires · validates

Colour is the family; a dashed line is the second member of it.

This node

• extendsadds capability to Autoregressive Language Modeltrains on demonstrations of the behaviour wanted

Referenced by

• requiresRLHF does not work without thisthe policy is initialised from a supervised pass

References

• Finetuned Language Models Are Zero-Shot Learners — Wei et al. — 2021 · link
• Background — Fine-tuning (deep learning) on Wikipedia