Skip to content

fix(ppo_trainer): default gen kwargs #510

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 3 commits into from
Jun 23, 2023
Merged

fix(ppo_trainer): default gen kwargs #510

merged 3 commits into from
Jun 23, 2023

Conversation

maxreciprocate
Copy link
Collaborator

This PR

  • forces use_cache in gen_kwargs to be true by default
  • replaces batch_size to chunk_size in evaluation pipeline, mirroring make_experience pipeline usage
  • removes enforcement on padding token to be the same as endoftext with decoder models

https://wandb.ai/sorry/trlx-references/reports/fix-default-gen_kwargs-v-main--Vmlldzo0NzIwNzc2
(note that pythia run is absent because there are only 40gb gpus available currently)

PhungVanDuy
PhungVanDuy approved these changes Jun 23, 2023
View reviewed changes
Copy link
Collaborator

@PhungVanDuy PhungVanDuy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sgtm!

@maxreciprocate maxreciprocate merged commit 171357b into main Jun 23, 2023
@maxreciprocate maxreciprocate deleted the fix-default-gen_kwargs branch June 23, 2023 12:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@PhungVanDuy PhungVanDuy PhungVanDuy approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@maxreciprocate @PhungVanDuy