Skip to content

Add last_token_pos in llama_transformer (#11793) #12239

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 1 commit into from
Jul 8, 2025
Merged

Add last_token_pos in llama_transformer (#11793) #12239

merged 1 commit into from
Jul 8, 2025

Conversation

@jinook-song-meta
Copy link
Contributor

Summary:

Add last_token_pos in the forward options.

Purpose:

  • the last norm and output of lm-head can be performed with the last valid token at prefill.
  • If the input sequence length is fixed when an accelerator doesn't support the dynamic shapes, selecting the last token from the input is not always guaranteed as valid.
  • Thus, it needs an additional pointer to select the last valid token only to perform the last norm and output.

Reviewed By: JacobSzwejbka

Differential Revision: D76440105

@pytorch-bot
Copy link

pytorch-bot bot commented Jul 4, 2025 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/12239

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (2 Unrelated Failures)

As of commit ce3d29b with merge base 97047c0 (image):

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jul 4, 2025
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D76440105

@github-actions
Copy link

github-actions bot commented Jul 4, 2025

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

jinook-song-meta added a commit to jinook-song-meta/executorch that referenced this pull request Jul 4, 2025
@jinook-song-meta @facebook-github-bot
Add last_token_pos in llama_transformer (pytorch#12239)
13eb7d4 
Summary:


Add last_token_pos in the forward options.

Purpose:
* the last norm and output of lm-head can be performed with the last valid token at prefill.
* If the input sequence length is fixed when an accelerator doesn't support the dynamic shapes, selecting the last token from the input is not always guaranteed as valid. 
* Thus, it needs an additional pointer to select the last valid token only to perform the last norm and output.

Reviewed By: JacobSzwejbka

Differential Revision: D76440105
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D76440105

jinook-song-meta added a commit to jinook-song-meta/executorch that referenced this pull request Jul 7, 2025
@jinook-song-meta @facebook-github-bot
Add last_token_pos in llama_transformer (pytorch#12239)
9bb4371 
Summary:

Add last_token_pos in the forward options.

Purpose:
* the last norm and output of lm-head can be performed with the last valid token at prefill.
* If the input sequence length is fixed when an accelerator doesn't support the dynamic shapes, selecting the last token from the input is not always guaranteed as valid. 
* Thus, it needs an additional pointer to select the last valid token only to perform the last norm and output.

Reviewed By: JacobSzwejbka

Differential Revision: D76440105
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D76440105

@jinook-song-meta @facebook-github-bot
Add last_token_pos in llama_transformer (pytorch#12239)
ce3d29b 
Summary:

Add last_token_pos in the forward options.

Purpose:
* the last norm and output of lm-head can be performed with the last valid token at prefill.
* If the input sequence length is fixed when an accelerator doesn't support the dynamic shapes, selecting the last token from the input is not always guaranteed as valid. 
* Thus, it needs an additional pointer to select the last valid token only to perform the last norm and output.

Reviewed By: JacobSzwejbka

Differential Revision: D76440105
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D76440105

1 similar comment
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D76440105

@facebook-github-bot facebook-github-bot merged commit b342f83 into pytorch:main Jul 8, 2025
101 of 103 checks passed
Tanish2101 pushed a commit to Tanish2101/executorch that referenced this pull request Jul 9, 2025
@jinook-song-meta @Tanish2101
Add last_token_pos in llama_transformer (pytorch#11793)
c0b6627 
Differential Revision: D76440105

Pull Request resolved: pytorch#12239
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@jackzhxng jackzhxng jackzhxng approved these changes

@lucylq lucylq Awaiting requested review from lucylq lucylq is a code owner

Assignees

No one assigned

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@jinook-song-meta @facebook-github-bot @jackzhxng