Move hooks and fsdp modules onto state rather than trainer #3522
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This reverts commit ae598f9.
mvpatel2000
reviewed
Aug 5, 2024
Co-authored-by: Mihir Patel <mihir.v.patel7@gmail.com>
Co-authored-by: Mihir Patel <mihir.v.patel7@gmail.com>
mvpatel2000
approved these changes
Aug 5, 2024
Requested rerun from research |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Eval'ing after resuming from checkpointing broke since calls to
prepare_fsdp_moduleinstate.pydidn't modify the handles and FSDP modules in modules. We fix this by movingautomicrobatch_fsdp_hook_handlesandfsdp_modulesintoStateinstance variables instead ofTrainerinstance variables.Fixed regression test that previously broken:
mpt-7b-composer-eval-regression-KkY29t