fix ignore-eos

[grimoire]

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

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
3. If the modification has a dependency on downstream projects of a newer version, this PR should be tested with all supported versions of downstream projects.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[Copilot]

Pull request overview

This PR fixes a bug in the handling of bad words and stop words when ignore_eos is enabled. The fix moves the responsibility of handling invalid (negative) bad word indices from the data preparation phase to the processing phase.

Changes:

• Removed the line that cleaned negative indices in __get_bad_words function
• Added internal handling of negative indices in _process_bad_words_ function
• Added a comment explaining that invalid bad words might be negative

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
lmdeploy/pytorch/strategies/ar/sampling.py	Removed the cleaning of negative indices from the __get_bad_words helper function
lmdeploy/pytorch/engine/logits_process.py	Added handling of negative indices within _process_bad_words_ function by using .where(mask, 0) to create valid indices for gather/scatter operations

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[zhulinJulia24]

Testcase passed