paxml

Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry leading model flop utilization rates.

GitHub
448 stars
16 watching
68 forks
Language: Python
last commit: 5 days ago
Linked from 1 awesome list

c4gptjaxlarge-language-modelsllmmodel-flopsparallelism

Backlinks from these awesome lists: