Change sharding annotation for activation_embed_and_logits_batch #1015
Conversation
This file contains 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
khatwanimohit
requested review from
gobbleturk,
jonb377,
bvandermoon and
vipannalla
as code owners
jonb377
previously approved these changes
Nov 6, 2024
jonb377
reviewed
Nov 6, 2024
| @@ -214,7 +214,7 @@ logical_axis_rules: [ | |||
| # For pipeline parallelism the pre and post decoder layer tensors' batch dimension is sharded by stages. | |||
| # Microbatches are sharded by stage, so moving out of and into this sharding should be a local reshape. | |||
| # The "stage" needs to be listed first since the microbatch dimension is first before the reshape. | |||
There was a problem hiding this comment.
Oh I didn't see this comment, I'll defer to @gobbleturk.
There was a problem hiding this comment.
This sound be fine, perhaps the order here needs to match the order of mesh axes
gobbleturk
approved these changes
Nov 6, 2024
| @@ -214,7 +214,7 @@ logical_axis_rules: [ | |||
| # For pipeline parallelism the pre and post decoder layer tensors' batch dimension is sharded by stages. | |||
| # Microbatches are sharded by stage, so moving out of and into this sharding should be a local reshape. | |||
| # The "stage" needs to be listed first since the microbatch dimension is first before the reshape. | |||
There was a problem hiding this comment.
This sound be fine, perhaps the order here needs to match the order of mesh axes
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Outdated suggestions cannot be applied.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.
No description provided.