[WIP]: Autotune Chunk Size #395
Draft
Conversation
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
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Outdated suggestions cannot be applied.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.
Summary
Our chunked loss functions currently statically set the chunk size to 1. However, this might lead to underutilized gpu memory resources. In this PR we show how the chunk size can be dynamically set, based on the additional peak memory consumed by the forward pass of the chunked loss being used.
Testing Done
On A100 80 GB SMX
Benchmarking against chunked orpo shows very marginal improvement in speed compared to a static chunk size.
make testto ensure correctness (wip)make checkstyleto ensure code stylemake test-convergenceto ensure convergence