[WIP] Allow PP + megablox #1009
Open
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
gobbleturk
requested review from
jonb377,
khatwanimohit,
bvandermoon and
vipannalla
as code owners
RissyRan
reviewed
Nov 2, 2024
| @@ -482,7 +482,7 @@ def update_model_vars(base_config_path, raw_keys, config_name: str): | |||
|
|
|||
| def validate_megablox_parallelism(raw_keys): | |||
| if raw_keys["megablox"] and ( | |||
| using_sequence_parallelism(raw_keys) or using_pipeline_parallelism(raw_keys) or using_expert_parallelism(raw_keys) | |||
| using_sequence_parallelism(raw_keys) or using_expert_parallelism(raw_keys) | |||
| ): | |||
| raise ValueError("Currently we only support Megablox with data and tensor parallelism.") | |||
There was a problem hiding this comment.
You may want to update the message in ValueError. Looking forward to see some baseline metrics! Thanks Matt!
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Outdated suggestions cannot be applied.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.
Will add some tests here, perhaps even AOT PP over DCN before reivew. As of now just tested on real v4-8 that PP + Megablox does fine, which is expected since PP is just vmapping over stages so isn't fundamentally changing the input shardings to megablox