Thanks to visit codestin.com
Credit goes to github.com

Skip to content

anhhuyalex/openrl

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Setup UV:

uv sync

To run:

cd src
accelerate launch --config_file accelerate_config.yaml train.py

Configs are in src/config.yaml and src/accelerate_config.yaml.

TODO:

  • We can't add more wandb metrics for now. Fix this
  • Add more evals, look into evalchemy
  • Add sweep code
  • Add more metrics that we're interested in

About

Replicating Deepseek R1 and OpenAI O series training

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%