deepspeedai / deepspeedai/DeepSpeed

The training parameters have not changed

Open
#4,758 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug training
Dominant language
Python
Stars
43.1k
Forks
5k
Avg merge
4d 15h
Merged PRs (30d)
112

Description

Hello,I found that the output_layer of chatglm hasn't changed,
image
I trained LoRa and added an OutputLayer to the modules_to_save, but I see an updated value in the updataed_params, but it looks like data replication doesn't pass the value.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the ChatGLM LoRA training case with OutputLayer included in modules_to_save, then trace how the updated_params value is handled during data replication. Compare the output_layer value before and after replication; done means the trained value is preserved across replicas.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
distributed-systems, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.