Drop RoPE when filling KV cache#3346
Closed
GD06 wants to merge 1 commit into
Closed
Conversation
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
jianyuh
approved these changes
Nov 21, 2024
Contributor
|
@GD06 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 21, 2024
Summary: This PR provides CUDA kernels to fill KV cache without applying ROPE. Reviewed By: jianyuh Differential Revision: D66307820 Pulled By: GD06
facebook-github-bot
force-pushed
the
genai_nope_qkv
branch
from
77e3b5c to
3aab114
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66307820 |
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 21, 2024
Summary: This PR provides CUDA kernels to fill KV cache without applying ROPE. Reviewed By: jianyuh Differential Revision: D66307820 Pulled By: GD06
facebook-github-bot
force-pushed
the
genai_nope_qkv
branch
from
3aab114 to
ae41dfa
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66307820 |
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 21, 2024
Summary: This PR provides CUDA kernels to fill KV cache without applying ROPE. Reviewed By: jianyuh Differential Revision: D66307820 Pulled By: GD06
facebook-github-bot
force-pushed
the
genai_nope_qkv
branch
from
ae41dfa to
25e2d83
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66307820 |
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 21, 2024
Summary: X-link: facebookresearch/FBGEMM#488 This PR provides CUDA kernels to fill KV cache without applying ROPE. Reviewed By: jianyuh Differential Revision: D66307820 Pulled By: GD06
facebook-github-bot
force-pushed
the
genai_nope_qkv
branch
from
25e2d83 to
5cbab5a
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66307820 |
Summary: X-link: facebookresearch/FBGEMM#488 This PR provides CUDA kernels to fill KV cache without applying ROPE. Reviewed By: jianyuh Differential Revision: D66307820 Pulled By: GD06
facebook-github-bot
force-pushed
the
genai_nope_qkv
branch
from
5cbab5a to
b5f25d5
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66307820 |
1 similar comment
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D66307820 |
Contributor
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.
This PR provides CUDA kernels to fill KV cache without applying ROPE.