Skip to content

Conversation

@YashasviChaurasia
Copy link
Contributor

Description of the change

HF Models now allow users to load chat template via separate chat_template.jinja file which is unlike the chat template inside tokenizer config file. This means users who launch experiments via fms-hf-tuning on models where they want to update chat template need to copy paste it to the data config.

The PR would add support for chat_template_path in data_config.yaml so users can specify path to chat_template.jinja.
Fixes a few typos along the way.

Related issue number

How to verify the PR

Was the PR tested

  • I have added >=1 unit test(s) for every new method I have added.
  • I have ensured all unit tests pass

@github-actions
Copy link

Thanks for making a pull request! 😃
One of the maintainers will review and advise on the next steps.

Copy link
Collaborator

@dushyantbehl dushyantbehl left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM apart from minor changes.

Copy link
Collaborator

@dushyantbehl dushyantbehl left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

@dushyantbehl dushyantbehl merged commit f337875 into foundation-model-stack:main Oct 6, 2025
9 checks passed
@dushyantbehl dushyantbehl changed the title feat: add chat_template_path in data_config feat: Allow chat template to be specified via a path in data config. Oct 6, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants