Open
Conversation
fix some small bugs in training steps were being computed incorrectly so the LR scheduler was configured wrong. this means that the entire llava-pretrain run would just be in warmup fixed some logging/saving logic to be more aligned with actual optimizer steps added seed for reproducibility fixed LR scheduler configuration on resuming training swapped torch.compile and gradient checkpointing order so torch.compile gets the proper final computational graph
bkal01
force-pushed
the
bhavesh/fix-lr-scheduler
branch
from
cd4b3ea to
d29da76
Compare
bkal01
force-pushed
the
bhavesh/fix-lr-scheduler
branch
from
bdc66c5 to
8fd683c
Compare
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
steps were being computed incorrectly so the LR scheduler was configured wrong. this means that the entire llava-pretrain run would just be in warmup
fixed some logging/saving logic to be more aligned with actual optimizer steps
added seed for reproducibility
fixed LR scheduler configuration on resuming training