Restore O2 FP16/BF16 wrapper for GPT

[timmoon10]

What does this PR do ?

When running GPT with O2 enabled, I am seeing errors at the following line since _wrap_model_for_O2 is not defined:

NeMo/nemo/collections/nlp/models/language_modeling/megatron_gpt_model.py

Line 266 in 8523384

self._wrap_model_for_O2()

This function was removed in #7814. I see that this PR added the function to BERT, so could this be a bug?

Collection: NLP

Changelog

• Restore _wrap_model_for_O2 function in GPT

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, and set model.megatron_amp_O2=True

Jenkins CI

To run Jenkins, a NeMo User with write access must comment jenkins on the PR.

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.

Pinging @shanmugamr1992

Additional Information

• Bug from First draft of mcore bert model in NeMo #7814

[timmoon10]

Closed by #8038.