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

ENH: Warn when disabling adapters and bias != 'none' #741

Conversation

BenjaminBossan
Copy link
Member

As discussed internally.

For LoRA, given that bias='all' or bias='none', when doing inference with a model in the disable_adapter() context, the output will not be identical to the output of the base model. This may be surprising to users. Therefore, a warning is given. Furthermore, the docstring has been extended to reflect this fact.

As discussed internally.

For LoRA, given that bias='all' or bias='none', when doing inference
with a model in the disable_adapter context, the output will not be
identical to the output of the base model. This may be surprising to
users. Therefore, a warning is given. Furthermore, the docstring has
been extended to reflect this fact.
Copy link
Contributor

@pacman100 pacman100 left a comment

Choose a reason for hiding this comment

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

Thank you @BenjaminBossan, LGTM! 🤗

@HuggingFaceDocBuilderDev
Copy link

HuggingFaceDocBuilderDev commented Jul 21, 2023

The documentation is not available anymore as the PR was closed or merged.

@BenjaminBossan BenjaminBossan merged commit a955ef1 into huggingface:main Jul 24, 2023
@BenjaminBossan BenjaminBossan deleted the warning-when-disabling-adapters-with-modified-biases branch July 24, 2023 08:34
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