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

SFT Tokenizer Fix #1142

Merged

Conversation

ChrisCates
Copy link
Contributor

After experiencing issues with GPTQ Conversion, discussed in this issue #1138.

I was able to resolve it by passing the Tokenizer into the SFTTrainer.

I also noticed that lowering gradient_accumulation_steps helps resolve the linalg.cholesky error experienced with pytorch.

I've left a note under Best Practices in the sft_trainer.mdx file.

We can decide where to best put this note in regards to GPTQ Conversion.

Copy link
Contributor

@younesbelkada younesbelkada left a comment

Choose a reason for hiding this comment

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

Thanks a lot for adding this!
Can you just run the styling checks? make precommit then we can merge

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@ChrisCates ChrisCates force-pushed the @chriscates/sft-tokenizer-fix branch from fed4e1b to af52453 Compare December 26, 2023 16:38
@ChrisCates
Copy link
Contributor Author

@younesbelkada, updated the commit with the pre-commit fixes. Let me know if any other issues.

Copy link
Contributor

@younesbelkada younesbelkada left a comment

Choose a reason for hiding this comment

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

Thanks a lot for your contribution !

@younesbelkada younesbelkada merged commit 18a33ff into huggingface:main Dec 27, 2023
9 checks passed
lapp0 pushed a commit to lapp0/trl that referenced this pull request May 10, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants