KEMBAR78
Add Nemotron-H prompt format, fix cut-to-conversation custom attr propagation by pzelasko · Pull Request #13963 · NVIDIA-NeMo/NeMo · GitHub
Skip to content

Conversation

@pzelasko
Copy link
Collaborator

Important

The Update branch button must only be pressed in very rare occassions.
An outdated branch is never blocking the merge of a PR.
Please reach out to the automation team before pressing that button.

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Collection: [Note which collection this PR will affect]

Changelog

  • Add specific line by line info of high level changes in this PR.

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this 

GitHub 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The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.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:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you add or update any necessary documentation?
  • Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
    • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

  • New Feature
  • Bugfix
  • Documentation

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

  • Related to # (issue)

Signed-off-by: Piotr Żelasko <petezor@gmail.com>
Signed-off-by: Piotr Żelasko <petezor@gmail.com>
@pzelasko pzelasko changed the title Add Nemotron-H prompt format Add Nemotron-H prompt format, fix cut-to-conversation custom attr propagation Jun 24, 2025
Signed-off-by: Piotr Żelasko <petezor@gmail.com>
@pzelasko pzelasko marked this pull request as ready for review June 24, 2025 14:56
Signed-off-by: Piotr Żelasko <petezor@gmail.com>
@github-actions
Copy link
Contributor

[🤖]: Hi @pzelasko 👋,

We wanted to let you know that a CICD pipeline for this PR just finished successfully.

So it might be time to merge this PR or get some approvals.

//cc @chtruong814 @ko3n1g @pablo-garay @thomasdhc

Copy link
Collaborator

@zhehuaichen zhehuaichen left a comment

Choose a reason for hiding this comment

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

Can you also document what else needs to be done to enbale nemotron-h

assert canary_tokenizer.ids_to_text(ans["context_ids"].tolist()) == '<|startoftranscript|><|en|><|transcribe|><|en|><|pnc|>'
assert canary_tokenizer.ids_to_text(ans["answer_ids"].tolist()) == ' TEST<|endoftext|>'
assert ans["mask"].tolist() == [False] * 5 + [True] * 5
assert ans["mask"].shape[0] == ans["input_ids"].shape[0]
Copy link
Collaborator

Choose a reason for hiding this comment

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

why do we need to update this? same question for other tests

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

just to simplify editing the unit tests; technically I think this specific assertion may even be removed as there is a separate test for the masking logic in prompt formatter

@pzelasko
Copy link
Collaborator Author

Can you also document what else needs to be done to enbale nemotron-h

Some changes in how SALM accesses various modules and properties of self.llm, since Nemotron-H (as released currently) doesn't seem to have the same structure as other models in HF. Probably needs to be reflected in FSDP2 setup too. Otherwise I think it will work.

@pzelasko pzelasko merged commit e1a4efc into main Jun 27, 2025
207 of 213 checks passed
@pzelasko pzelasko deleted the nemotron-h-prompt-format branch June 27, 2025 17:33
AmirHussein96 pushed a commit to AmirHussein96/NeMo that referenced this pull request Jul 23, 2025
…pagation (NVIDIA-NeMo#13963)

* Add Nemotron-H prompt format

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix propagation of custom attr in cut_to_conversation

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix CI

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Unit test for the conversion fix

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

---------

Signed-off-by: Piotr Żelasko <petezor@gmail.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Amir Hussein <amhussein@nvidia.com>
AmirHussein96 pushed a commit to AmirHussein96/NeMo that referenced this pull request Aug 5, 2025
…pagation (NVIDIA-NeMo#13963)

* Add Nemotron-H prompt format

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix propagation of custom attr in cut_to_conversation

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix CI

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Unit test for the conversion fix

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

---------

Signed-off-by: Piotr Żelasko <petezor@gmail.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Amir Hussein <amhussein@nvidia.com>
AmirHussein96 pushed a commit to AmirHussein96/NeMo that referenced this pull request Aug 5, 2025
…pagation (NVIDIA-NeMo#13963)

* Add Nemotron-H prompt format

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix propagation of custom attr in cut_to_conversation

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix CI

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Unit test for the conversion fix

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

---------

Signed-off-by: Piotr Żelasko <petezor@gmail.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Amir Hussein <amhussein@nvidia.com>
nasretdinovr pushed a commit to nasretdinovr/NeMo that referenced this pull request Aug 8, 2025
…pagation (NVIDIA-NeMo#13963)

* Add Nemotron-H prompt format

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix propagation of custom attr in cut_to_conversation

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Fix CI

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

* Unit test for the conversion fix

Signed-off-by: Piotr Żelasko <petezor@gmail.com>

---------

Signed-off-by: Piotr Żelasko <petezor@gmail.com>
Co-authored-by: Charlie Truong <chtruong@nvidia.com>
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.

4 participants