Skip to content
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.

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

Add rng_seed parameter to trainers #2254

Merged
merged 2 commits into from
Mar 31, 2023
Merged

Conversation

andreaskoepf
Copy link
Collaborator

@andreaskoepf andreaskoepf commented Mar 28, 2023

Add rng_seed command line/config parameter. Command line arg will overwrite yaml-config value.

@andreaskoepf
Copy link
Collaborator Author

I tested sft with two independent runs on galactica-125m-base-finetuned .. both loss-curves were nearly identical:

image

Only problem I see is the random state manipulation in PerDatasetSampler utils.py#L92.

@andreaskoepf andreaskoepf enabled auto-merge (squash) March 31, 2023 16:23
@andreaskoepf andreaskoepf merged commit df50632 into main Mar 31, 2023
@andreaskoepf andreaskoepf deleted the reproductible_training_seed branch March 31, 2023 16:25
yk pushed a commit that referenced this pull request Apr 2, 2023
Add `rng_seed` command line/config parameter. Command line arg will
overwrite yaml-config value.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants