AI-Hypercomputer / maxtext

A simple, performant and scalable Jax LLM!
Apache License 2.0
1.47k stars 275 forks source link

Add eval to convergence test and log metrics #876

Closed aireenmei closed 1 week ago

aireenmei commented 2 weeks ago

Tested (fewer steps): hf: https://cloudlogging.app.goo.gl/UrxmBToWXDAefotc6 grain: https://cloudlogging.app.goo.gl/HbUVQdDSXK5Z6wyQA tfds: https://cloudlogging.app.goo.gl/SEzeAxoEJWmXwrdT8