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.

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

Issue since most recent transformers update #580

Open
siddharthverma314 opened this issue Nov 11, 2023 · 1 comment
Open

Issue since most recent transformers update #580

siddharthverma314 opened this issue Nov 11, 2023 · 1 comment
Labels
bug Something isn't working

Comments

@siddharthverma314
Copy link

🐛 Describe the bug

The new version of transformers saves model checkpoints in safetensors format, which is incompatible with the way that trlx loads checkpoints using PreTrainedModelWrapper.from_pretrained. Specifically, the code fails pytorch_model.bin is not present anymore.

Which trlX version are you using?

trlx==0.7

Additional system and package information

transformers==4.34.0
@siddharthverma314 siddharthverma314 added the bug Something isn't working label Nov 11, 2023
@LouisCastricato LouisCastricato mentioned this issue Nov 13, 2023
Open
@RylanSchaeffer RylanSchaeffer mentioned this issue Jul 17, 2024
Open
@RylanSchaeffer
Copy link

RylanSchaeffer commented Jul 17, 2024 •
edited
Loading

I just hit this problem (my models are saved on HuggingFace as safetensors and thus loading pytorch_model.bin fails). Any updates on this issue?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@RylanSchaeffer @siddharthverma314