Add kv_channels to convert_llama_nemo_to_hf.py#10078
Closed
suhara wants to merge 3 commits intoNVIDIA-NeMo:mainfrom
Closed
Add kv_channels to convert_llama_nemo_to_hf.py#10078suhara wants to merge 3 commits intoNVIDIA-NeMo:mainfrom
kv_channels to convert_llama_nemo_to_hf.py#10078suhara wants to merge 3 commits intoNVIDIA-NeMo:mainfrom
Conversation
Signed-off-by: suhara <suhara@users.noreply.github.com>
5 tasks
suiyoubi
reviewed
Aug 8, 2024
|
|
||
| head_size = hidden_size // head_num | ||
| # Use kv_channels if available. Otherwise, use hidden_size // head_num as head_size | ||
| head_size = model.cfg.get("kv_channels", hidden_size // head_num) |
Collaborator
There was a problem hiding this comment.
I think in nemo config kv_channels can be set to null explictly, which would cause issue here. I would suggest changing to something like:
if model.cfg.get("kv_channels", None) is None:
head_size = hidden_size // head_num
else:
head_size = model.cfg.kv_channels
Contributor
Author
There was a problem hiding this comment.
@suiyoubi Thank you for the suggestion! Revised accordingly.
Contributor
|
This PR is stale because it has been open for 14 days with no activity. Remove stale label or comment or update or this will be closed in 7 days. |
Contributor
|
This PR was closed because it has been inactive for 7 days since being marked as stale. |
Contributor
Author
|
The feature is covered by another PR, which has been merged. |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do ?
Custom Llama-based models may use custom
head_dimas it is supported by NeMo/MegatronLM (it's calledkv_channelsin these tools).The current
convert_llama_nemo_to_hf.pyassumeshead_dim * num_heads == hidden_sizeThere's an ongoing PR to support custom
head_dimin HFI've been in touch with and getting support from @suiyoubi for this PR.
Collection: [Note which collection this PR will affect]
Changelog
Usage
# Add a code snippet demonstrating how to use thisGitHub Actions CI
The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.
The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".
Before your PR is "Ready for review"
Pre checks:
PR Type:
If you haven't finished some of the above items you can still open "Draft" PR.
Who can review?
Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.
Additional Information