Skip to content

feat: surface truncating & padding options #719

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 18 commits into from
Oct 29, 2025
Merged

feat: surface truncating & padding options #719

merged 18 commits into from
Oct 29, 2025

Conversation

@akoumpa
Copy link
Contributor

@akoumpa akoumpa commented Oct 24, 2025 •
edited
Loading

Changes:

  • Surfaces the padding and truncation options in the formatting utils.
  • For the ColumnMapped dataset, includes above options and also adds a fallback if all labels are empty in the labels.

@copy-pr-bot
Copy link

copy-pr-bot bot commented Oct 24, 2025

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@akoumpa akoumpa linked an issue Oct 24, 2025 that may be closed by this pull request
Add sequence truncation support for long sequences #718
Closed
@akoumpa akoumpa force-pushed the akoumparouli/feat_surface_padding_truncation_options branch from a11d569 to 420a23c Compare October 27, 2025 05:44
@akoumpa
Copy link
Contributor Author

akoumpa commented Oct 27, 2025

/ok to test 7d99857

@akoumpa
Copy link
Contributor Author

akoumpa commented Oct 27, 2025

/ok to test 5e30ab6

@akoumpa
Copy link
Contributor Author

akoumpa commented Oct 27, 2025

/ok to test 2e62dc3

@akoumpa
add truncation & padding options
13e625b 
Signed-off-by: Alexandros Koumparoulis <[email protected]>
@akoumpa
Copy link
Contributor Author

akoumpa commented Oct 29, 2025

/ok to test 13e625b

@akoumpa akoumpa marked this pull request as ready for review October 29, 2025 05:20
HuiyingLi
HuiyingLi reviewed Oct 29, 2025
View reviewed changes
nemo_automodel/components/datasets/llm/formatting_utils.py
return_dict=True,
return_assistant_tokens_mask=template_has_generation_kwd,
padding=padding,
truncation=truncation,
Copy link
Contributor

Choose a reason for hiding this comment

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

If I understand correctly we truncate regardless of the template structure here.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

yes

HuiyingLi
HuiyingLi approved these changes Oct 29, 2025
View reviewed changes
Copy link
Contributor

@HuiyingLi HuiyingLi left a comment

Choose a reason for hiding this comment

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

LGTM thank you!

@HuiyingLi HuiyingLi merged commit 38b330c into main Oct 29, 2025
51 checks passed
@HuiyingLi HuiyingLi deleted the akoumparouli/feat_surface_padding_truncation_options branch October 29, 2025 07:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@HuiyingLi HuiyingLi HuiyingLi approved these changes

@adil-a adil-a Awaiting requested review from adil-a adil-a is a code owner

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

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Add sequence truncation support for long sequences

3 participants

@akoumpa @HuiyingLi