Glossary: NLP with Transformers
Learning rate
Also written: learning_rate
The size of each weight-update step. Fine-tuning uses small values like 5e-5 so the pretrained knowledge is adjusted rather than overwritten.
Glossary: NLP with Transformers
Also written: learning_rate
The size of each weight-update step. Fine-tuning uses small values like 5e-5 so the pretrained knowledge is adjusted rather than overwritten.