sequence parallel default dtype #7364
Merged
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
Signed-off-by: Stas Bekman <[email protected]>
tjruwase
approved these changes
Jun 17, 2025
Collaborator
Author
|
so what's special about a6000 gpu that it fails? I'm looking at the breakage basically it's this call that fails: for older pytorch versions, like pt-2.4, but it succeeds with pt-2.7.1. Aha! This workflow uses: so it reports 2.6 but that The caller is here: Is there a reason why this workflow is locked onto this pytorch version? If so we have to run it on an older edit: The half-baked pt-2.6 comes from |
Signed-off-by: Stas Bekman <[email protected]>
Signed-off-by: Stas Bekman <[email protected]>
Antlera
pushed a commit
to Antlera/DeepSpeed
that referenced
this pull request
Jun 27, 2025
the newly released nccl finally started to use fp32 accumulation for reduction ops! * Floating point summation is always done in fp32 accumulators (with the exception of fp8 on NVLS, where it uses fp16 inside the switch). Thus, the accuracy with fp8 and fp16 data types should be much improved. NVIDIA/nccl@72d2432 So we should change the fp32 comms default for SP to the same dtype as inputs if `nccl>=2.27.3` - the user can still override the default. --------- Signed-off-by: Stas Bekman <[email protected]> Co-authored-by: Olatunji Ruwase <[email protected]>
mauryaavinash95
pushed a commit
to DataStates/DeepSpeed
that referenced
this pull request
Oct 4, 2025
the newly released nccl finally started to use fp32 accumulation for reduction ops! * Floating point summation is always done in fp32 accumulators (with the exception of fp8 on NVLS, where it uses fp16 inside the switch). Thus, the accuracy with fp8 and fp16 data types should be much improved. NVIDIA/nccl@72d2432 So we should change the fp32 comms default for SP to the same dtype as inputs if `nccl>=2.27.3` - the user can still override the default. --------- Signed-off-by: Stas Bekman <[email protected]> Co-authored-by: Olatunji Ruwase <[email protected]>
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.
the newly released nccl finally started to use fp32 accumulation for reduction ops!
exception of fp8 on NVLS, where it uses fp16 inside the switch). Thus,
the accuracy with fp8 and fp16 data types should be much improved.
NVIDIA/nccl@72d2432
So we should change the fp32 comms default for SP to the same dtype as inputs if
nccl>=2.27.3- the user can still override the default.