Fails 32 way sharded - bump atol - #27
Open
aidando73 wants to merge 1 commit into
Open
Conversation
meta-cla
Bot
added
the
CLA Signed
codingwithsurya
approved these changes
Nov 2, 2025
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
[rank2]: Mismatched elements: 1 / 262144 (0.0%)
[rank2]: Greatest absolute difference: 0.125 at index (251031,) (up to 0.1 allowed)
[rank2]: Greatest relative difference: 1.0703125 at index (251031,) (up to 0.1 allowed)
[rank1]: Traceback (most recent call last):
[rank1]: File "/workspace/kraken/benchmark/benchmark_all_reduce.py", line 281, in
[rank1]: main(args)
[rank1]: File "/workspace/kraken/benchmark/benchmark_all_reduce.py", line 208, in main
[rank1]: run_experiment(config),
[rank1]: File "/workspace/kraken/benchmark/benchmark_all_reduce.py", line 157, in run_experiment
[rank1]: torch.testing.assert_close(test_o, gloden_o, atol=1e-1, rtol=1e-1)
[rank1]: File "/workspace/kraken/.venv/lib/python3.10/site-packages/torch/testing/_comparison.py", line 1589, in assert_close
[rank1]: raise error_metas[0].to_error(msg)
[rank1]: AssertionError: Tensor-likes are not close!
[rank1]: Mismatched elements: 1 / 262144 (0.0%)
[rank1]: Greatest absolute difference: 0.125 at index (251031,) (up to 0.1 allowed)
[rank1]: Greatest relative difference: 1.0703125 at index (251031,) (up to 0.1 allowed)
[rank0]: Traceback (most recent call last):
[rank0]: File "/workspace/kraken/benchmark/benchmark_all_reduce.py", line 281, in
[rank0]: main(args)
[rank0]: File "/workspace/kraken/benchmark/benchmark_all_reduce.py", line 208, in main
[rank0]: run_experiment(config),
[rank0]: File "/workspace/kraken/benchmark/benchmark_all_reduce.py", line 157, in run_experiment
[rank0]: torch.testing.assert_close(test_o, gloden_o, atol=1e-1, rtol=1e-1)
[rank0]: File "/workspace/kraken/.venv/lib/python3.10/site-packages/torch/testing/_comparison.py", line 1589, in assert_close
[rank0]: raise error_metas[0].to_error(msg)
[rank0]: AssertionError: Tensor-likes are not close!
[rank0]: Mismatched elements: 1 / 262144 (0.0%)
[rank0]: Greatest absolute difference: 0.125 at index (251031,) (up to 0.1 allowed)
[rank0]: Greatest relative difference: 1.0703125 at index (251031,) (up to 0.1 allowed)
[rank0]:[W1029 15:51:21.380868586 ProcessGroupNCCL.cpp:1524] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())