Fix xLAM-70b OOM: reduce gpu_memory_utilization#178
Merged
Conversation
v0.1.1: OOM during backward pass due to activation memory. Reduce policy_mini_batch_size to decrease gradient accumulation steps and peak activation memory during backward pass. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
dzorlu
force-pushed
the
fix/xlam-70b-oom
branch
from
ea04dc8 to
a652b7f
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
xLAM-70b training OOM during backward pass:
OOM in MLP layer during gradient checkpointing recomputation - activation memory too high.
Fix
policy_mini_batch_sizefrom 16 → 8Note:
gradient_checkpointingis already enabled by default inppo_base_config.yaml.Test plan
🤖 Generated with Claude Code