fix(afmoe): reduce tokens in test_compile_static_cache to avoid flaky bfloat16 drift#46521
Merged
Conversation
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: afmoe |
Contributor
|
CI Dashboard: View test results in Grafana |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
tarekziade
approved these changes
Jun 9, 2026
louzongzhi
pushed a commit
to louzongzhi/transformers
that referenced
this pull request
Jun 10, 2026
… bfloat16 drift (huggingface#46521) fix Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
louzongzhi
pushed a commit
to louzongzhi/transformers
that referenced
this pull request
Jun 10, 2026
… bfloat16 drift (huggingface#46521) fix Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
torch.compilewithreduce-overhead+ static cache in bfloat16 produces outputs that vary across runs. Switching to float16 makes the compiled outputs consistent across runs, but they differ from the dynamic and static (non-compiled) baselines.This PR reduce
num_tokens_to_generatefrom 24 to 4 keeps the test within the stable bfloat16 range and eliminates the flakiness observed in CI.A run to see it was failing
https://github.com/huggingface/transformers/actions/runs/27200489039
A run to see it works now
https://github.com/huggingface/transformers/actions/runs/27207933386