-
Notifications
You must be signed in to change notification settings - Fork 1.9k
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
fixing the full state path in checkpoint handler+loss report calculation #51
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@HamidShojanazeri Thanks for the fixes. Please see comments inline.
…bookresearch/llama-recipes into checkpoint_handler_path_fix
Seems like this is supposed to fix #65? I am going to try this out and will keep you folks updated! |
@DhruvaBansal00 thanks for trying out the PR, I am not able to repro on my end can you pls share the command your running? and you env using |
@HamidShojanazeri I hadn't applied your changes correctly before. Just finished running three epochs and seems like everything is running as expected. Thanks for the changes and appreciate the fast turnaround :) |
Thanks very much @DhruvaBansal00 for confirmation. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @HamidShojanazeri for all the fixes and updates.
Thanks @jeonsworld your changes for PR #24 have also been rolled into this PR
This PR fixes