Add HF's Accelerate - Githubissues

lucidrains / PaLM-rlhf-pytorch

Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM

MIT License

7.67k stars 668 forks source link

Closed ell-hol closed 1 year ago

ell-hol commented 1 year ago

Training is much faster now

lucidrains commented 1 year ago

do you want to try outfitting the PPO training too?

lucidrains commented 1 year ago

@ell-hol actually don't worry about the PPO, i'll just do it, since it is a bit out of the ordinary

lucidrains commented 1 year ago

@ell-hol thank Yacine!

ell-hol commented 1 year ago

@ell-hol thank Yacine!

My pleasure. What about that wandb integration ?