Debug FP8 support in Apex distributed optimizer with old Transformer Engine versions

[timmoon10]

What does this PR do ?

#12004 updated the Apex distributed optimizer to handle API changes in Transformer Engine 2.0. However, it had some basic Python bugs in the backward-compatible path.

Collection: Core

Changelog

• Debug FP8 support in Apex distributed optimizer with old Transformer Engine versions

Usage

Run GPT, e.g. with the config at https://github.com/NVIDIA/NeMo/blob/main/examples/nlp/language_modeling/conf/megatron_gpt_config.yaml.

Some relevant options:

Distributed optimizer: model.optim.name=distributed_fused_adam
Transformer Engine: model.transformer_engine=True
FP8 support: model.fp8=True
FP8 params: model.fp8_params=True

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The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.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

• Bugfix for Debug Apex distributed optimizer to handle Transformer Engine 2.0 #12004