deepspeedai / deepspeedai/DeepSpeed
[REQUEST] how to Wrap normalization layers like LayerNorm in FP32 when use zero (fp16 or bf16)?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
Describe the solution you'd like
A clear and concise description of what you want to happen.
Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.
Additional context
Add any other context or screenshots about the feature request here.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue contains only the feature-request template and mentions normalization layers such as LayerNorm, FP32, FP16, BF16, and ZeRO; it names no files, tests, or entry points. First clarify the intended ZeRO behavior and supported normalization layers, then identify the relevant DeepSpeed and PyTorch integration points. Done should include agreed scope, implementation guidance, and tests covering the requested precision behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- distributed-systems, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 10/100