Fix Llama-70B-FSDP model loading issue#752
Merged
vivekgoe merged 1 commit intohuggingface:mainfrom Mar 7, 2024
Merged
Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
vivekgoe
requested changes
Mar 4, 2024
Collaborator
vivekgoe
left a comment
vivekgoe
left a comment
There was a problem hiding this comment.
Please add a brief description of changes done in this PR in 1st section of PR.
vivekgoe
approved these changes
Mar 7, 2024
vivekgoe
added
run-test
puneeshkhanna
pushed a commit
to puneeshkhanna/optimum-habana-fork
that referenced
this pull request
Mar 11, 2024
Fix Llama-70B-FSDP model loading issue (HabanaAI#63)
HolyFalafel
pushed a commit
to HabanaAI/optimum-habana-fork
that referenced
this pull request
Mar 11, 2024
Fix Llama-70B-FSDP model loading issue (#63)
3 tasks
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When running large models with FSDP, if we try to load full model params to each card we will run out of CPU RAM (refer to this these PRs huggingface/transformers#25107 huggingface/accelerate#1777. This is avoided by loading model params only to card 0 and broadcasting params to other cards from card 0. Enabled CPU_RAM_EFFICIENT_LOADING configuration in accelerate to support above described functionality.