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

Fix default delta_kwargs handling #171

Merged
merged 2 commits into from
Jan 9, 2023
Merged

Fix default delta_kwargs handling #171

merged 2 commits into from
Jan 9, 2023

Conversation

jon-tow
Copy link
Collaborator

@jon-tow jon-tow commented Jan 8, 2023

  • Makes delta_kwargs an optional config; previously it defaulted to an empty dict which breaks the delta handlers for runs that do not use delta-tuning. Discovered by @PhungVanDuy

  • Removes gpt2 from the default MODIFIED_MODULES_DICT as its MLP layers are implemented as Conv1D layers which are not supported by the OpenDelta pacakge; see this line here (Conv1D does not contain in/out_features attrs)

LouisCastricato
LouisCastricato approved these changes Jan 9, 2023
View reviewed changes
Copy link
Contributor

@LouisCastricato LouisCastricato left a comment

Choose a reason for hiding this comment

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

LGTM

@jon-tow jon-tow merged commit 4bfe339 into CarperAI:main Jan 9, 2023
@jon-tow jon-tow deleted the fix-default-delta-args branch January 9, 2023 00:29
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@LouisCastricato LouisCastricato LouisCastricato approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@jon-tow @LouisCastricato