Training Variable Long Sequences with Data-Centric Parallel
Source
Geng Zhang, Xuanlei Zhao, Kai Wang, Yang You
Author
Geng Zhang, Xuanlei Zhao, Kai Wang, Yang You
Date
Terms in this piece · Glossary
fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.
Why it matters
Teams fine-tuningTaking a trained model and training it a bit more on your own examples so it gets better at one specific job.Full definition → or training on variable-length data get a low-friction way to recover throughput lost to static parallel configurations and workload imbalance.
