-
Notifications
You must be signed in to change notification settings - Fork 50
Memory budget strategy for activation checkpointing #297
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
Merged
+37
−1
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
Loading status checks…
Loading status checks…
Loading status checks…
Loading status checks…
Olmo2 on 4 B100s w/ ac budget = 0.5
|
Now with ac budget = 0.2
|
Loading status checks…
Loading status checks…
Loading status checks…
epwalsh
approved these changes
Jul 8, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice!
Loading status checks…
Loading status checks…
TianhuaTao
pushed a commit
that referenced
this pull request
Jul 10, 2025
See https://pytorch.org/blog/activation-checkpointing-techniques/ for more details, but essentially this is an easy way to try to enable selective activation checkpointing without fiddling with a bunch of different options to try to make it fast but stay within your GPU memory allowance.  > We observe a 50% memory reduction by recomputing only pointwise ops, with a steady drop-off as you recompute more and more of your matmuls. Attention is the most expensive, so you tend to want to recompute those last.
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Outdated suggestions cannot be applied.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.
See https://pytorch.org/blog/activation-checkpointing-techniques/ for more details, but essentially this is an easy way to try to enable selective activation checkpointing without fiddling with a bunch of different options to try to make it fast but stay within your GPU memory allowance.