-
Notifications
You must be signed in to change notification settings - Fork 442
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
replicate kv for some models when tp is divisble by kv_head_num #2874
Open
irexyc
wants to merge
3
commits into
InternLM:main
Choose a base branch
from
irexyc:replicate-kv
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
+36
−1
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
lvhan028
approved these changes
Dec 12, 2024
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
lvhan028
approved these changes
Dec 13, 2024
lzhangzz
reviewed
Dec 17, 2024
lmdeploy/turbomind/deploy/module.py
Outdated
Comment on lines
205
to
224
def _repeat_kv(self, qkvo, kind: str): | ||
"""replicate kv.""" | ||
q, k, v, o = qkvo | ||
head_dim = self.model.model_config.size_per_head | ||
hidden_dim = self.model.model_config.hidden_units | ||
|
||
def _repeat(x): | ||
dim = hidden_dim if kind != 'bias' else 1 | ||
x = x.view(-1, head_dim, dim).repeat(1, self.model.repeat_kv, 1) | ||
x = x.reshape(-1, dim) | ||
return x | ||
|
||
k, v = map(_repeat, (k, v)) | ||
if kind == 'bias': | ||
if o is None: | ||
o = torch.zeros(hidden_dim, dtype=q.dtype, device=q.device) | ||
q, k, v, o = map(torch.squeeze, (q, k, v, o)) | ||
|
||
return (q, k, v, o) | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In this file, all layout adjustment is performed on the transposed weight ([input_dim, output_dim]
), please keep it consistent.
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.
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily receiving feedbacks. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.
Motivation
Please describe the motivation of this PR and the goal you want to achieve through this PR.
Modification
Please briefly describe what modification is made in this PR.
BC-breaking (Optional)
Does the modification introduce changes that break the backward-compatibility of the downstream repositories?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.
Use cases (Optional)
If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.
Checklist