【GLM】subbatch performance and weight bug fix #2661
Merged
+212
−45
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
Codecov Report❌ Patch coverage is
❌ Your patch status has failed because the patch coverage (5.88%) is below the target coverage (80.00%). You can increase the patch coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## develop #2661 +/- ##
==========================================
Coverage ? 29.63%
==========================================
Files ? 311
Lines ? 54682
Branches ? 0
==========================================
Hits ? 16204
Misses ? 38478
Partials ? 0 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
danleifeng
changed the title
lugimzzz
reviewed
Sep 24, 2025
paddleformers/trainer/trainer.py
Outdated
| saved_signal_path = os.path.join(output_dir, f"saved_signal_{dist.get_rank()}") | ||
|
|
||
| if self.args.unified_checkpoint and self.args.offload_optim: | ||
| if self.args.unified_checkpoint and (self.args.offload_optim or self.args.tensorwise_offload_optimizer): |
There was a problem hiding this comment.
这个会造成这保存optimizer参数的时候显存异常上涨，不建议这么写
examples/run_finetune.py
Outdated
| os.environ["USE_CASUAL_MASK"] = "False" | ||
|
|
||
|
|
||
| def mock_offload_optimizer(): |
There was a problem hiding this comment.
unified_checkpoint_config: ignore_merge_optimizer
optim: adamw_custom
tensorwise_offload_optimizer: True
训练添加上述也可以做到offload optimizer降低显存，暂时先把optimizer相关修改删除，这个PR可以先合一版
| moe_group=moe_group, | ||
| ) | ||
| if hasattr(dist, "fleet") and dist.is_initialized() and expert_parallel_degree > 1: | ||
| # for p in self.experts.parameters(): |
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.
Before submitting
testsfolder. If there are codecov issues, please add tests cases first.PR types
PR changes
Description
配置新增：
本地glm sft训练改动依赖pr：
sp：#2621
fused_loss修复：#2648