Summary

• Support FSDP2 + TP strided sharding, and DTensor parameters with FP8, across all TransformerEngineBaseModule(s).
• Associated with BioNeMo-Recipes Llama3 TP: Enable TransformerEngine-backed Tensor Parallelism with Llama3. bionemo-framework#1483

Details

• Fix bug where "shard" was the presumed weight sharding sub-mesh in the DTensor.device_mesh. Now, users can precisely specify their own custom weight-sharding DeviceMesh for per-tensor amax_reduction_group via the set_device_mesh API.

Type of change

• Documentation change (change only to the documentation, either a fix or a new content)
• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Infra/Build change
• Code refactoring

Changes

Please list the changes introduced in this PR:

• Change A
• Change B

Checklist:

• I have read and followed the contributing guidelines
• The functionality is complete
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes