[hybrid_dev] Nemotron-H mamba2

[oleksost]

Mamba2 implementation as in nemotron h. This also uses correct mamba2 kernels.

Motivation:

• we want to use the same mamba2 implementation as in vllm

TODOs:

• added per head local RMSNorm

🔍 Type of change

Select all that apply:

• 🐛 Bug fix (non-breaking change that addresses a specific issue)
• 🚀 New feature (non-breaking change that adds functionality)
• ⚠️ Breaking change (a change that could affect existing functionality)
• 📈 Performance improvement/optimization (improves speed, memory usage, or efficiency)
• 🛠️ Code refactor (non-functional changes that improve code readability, structure, etc.)
• 📦 Dependency bump (updates dependencies, including Dockerfile or package changes)
• 📝 Documentation change (updates documentation, including new content or typo fixes)
• 🔧 Infrastructure/Build change (affects build process, CI/CD, or dependencies)

📝 Changes

List the key changes introduced in this PR:

• adds nemotron h modelling file
• adds mamba2 from nemotron into modelling modelling_ssm_hybrid_apriel15b
• adds fast-llm implementaiton of Mamba2 from nemotron H

✅ Checklist

Make sure the following tasks are completed before submitting the PR:

General

• 📜 I have read and followed the contributing guidelines.
• 🏷️ I am using a clear and descriptive PR title that summarizes the key change or feature introduced.
• 🎉 The functionality is complete, and I have tested the changes.
• 📝 I have updated the documentation if needed.
• ⚠️ The change does not introduce any new issues (e.g., runtime warnings, type checker errors, linting problems, unhandled edge cases).
• 🧩 I have commented my code, especially in hard-to-understand areas.

Dependencies and Configuration

• 🐋 I have updated the Docker configuration or dependencies, if applicable.
• 🔄 I have ensured compatibility with the existing setup after dependency changes.

Testing

• 🧪 I have added or updated tests to cover my changes.
• ✔️ New and existing tests pass locally with my changes.
• 🚦 I have tested these changes on GPUs and verified training stability.
• 🏋️ I have tested the changes on realistic training workloads, if applicable.

Performance Impact

• 📊 I have run benchmarks where applicable to evaluate the performance impact.
• ✅ The benchmarks show no performance regression.
• 🚀 The benchmarks indicate a potential performance improvement.
• ⚠️ The benchmarks indicate a potential performance degradation.
• 📈 I have provided benchmark results and detailed any performance impact below, if applicable.

📊 Performance Impact Details

If there is any impact on performance, describe it and provide benchmark results, if applicable:

---

🗒️ Additional Notes

Include any additional context, information, or considerations here, such as known issues, follow-up tasks, or backward compatibility concerns.

[RaymondLi0]

Very nice!
Didn't have a detailed look at this yet. But running the tests/models/test_checkpoint tests is a good way to test the conversion and the modeling file. For this you can add a new model config in tests/utils/model_configs.py that would use your new nm2 block. I can help for this if needed

[RaymondLi0]

Missing ... = ?

[oleksost]

Addressed, added model for test, the tests seem to pass.

pytest tests/models/test_checkpoint.py --models hybrid_nm2
....
1 failed, 23 passed, 219 skipped, 34 warnings in 45.57s

Not sure if the failed one (tests/models/test_checkpoint.py::test_save_and_load_in_parallel[hybrid_nm2]@dependency_group_3) is expected to fail (it fails on previous mamba models as well)