deepspeedai / deepspeedai/DeepSpeed
How can we set zero2 not to save the overall weight
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
How to set deepspeed's zero2 to save checkpoint, how to set ds_config.json to not save the overall parameter weight of the adapter fusion model?
Does the parameter stage3_gather_16bit_weights_on_model_save work for zero2？
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
Start by checking how ds_config.json handles ZeRO-2 checkpoint saving and whether stage3_gather_16bit_weights_on_model_save applies to that stage. Confirm how adapter-fusion checkpoints handle the overall parameter weights; done means documenting the supported configuration or identifying the required change.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- distributed-systems, machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100