[Flash Attention 2] Add flash attention 2 for GPT-Neo-X
#26463
Merged
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
24 tasks
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. |
ArthurZucker
approved these changes
Sep 29, 2023
Comment on lines
390
to
393
| query = query.to(torch.float16) | ||
| key = key.to(torch.float16) | ||
| value = value.to(torch.float16) | ||
|
|
There was a problem hiding this comment.
We should take into account bfloat16 here as well
Comment on lines
377
to
381
| # In PEFT, usually we cast the layer norms in float32 for training stability reasons | ||
| # therefore the input hidden states gets silently casted in float32. Hence, we need | ||
| # cast them back in float16 just to be sure everything works as expected. | ||
| # This might slowdown training & inference so it is recommended to not cast the LayerNorms | ||
| # in fp32. (LlamaRMSNorm handles it correctly) |
There was a problem hiding this comment.
Is this also true for GPTNeoX? (Comment is the same as Llama 😓 )
|
Any plans on completing this or should someone else pick it up? For what it's worth, this implementation is working very well for me 👍 |
|
cc @amyeroberts let me know if I need to address anything else in this PR! |
|
Checking on the progress here. What's the ETA on merging this with the main branch? Thanks! |
amyeroberts
approved these changes
Nov 27, 2023
younesbelkada
commented
Dec 6, 2023
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Outdated suggestions cannot be applied.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.
What does this PR do?
Adds flash attention support for GPT-Neo-X
Fixes: #26444