Adding a new config parameter to combine layers during FSDP #360
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Outdated suggestions cannot be applied.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.
Created a new class called
MultiBlock
, which wraps around multipleBlock
to reduce the number of NCCL communication. Number of blocks to combine can be controlled withnum_blocks_to_combine
parameter GPT/Llama.Ideally the message size should be around 1GB to get the best performance. But when models are running on multiple nodes every layer is split into really small chunks causing the message size to be extremely small resulting in suboptimal usage of network bandwidth. This parameter can be controlled to make sure we send larger message sizes.