Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BUG] <title>多图单轮对话全量SFT,loss 变为零 #679

Open
2 tasks done
Evan9978 opened this issue Nov 25, 2024 · 5 comments
Open
2 tasks done

[BUG] <title>多图单轮对话全量SFT,loss 变为零 #679

Evan9978 opened this issue Nov 25, 2024 · 5 comments

Comments

@Evan9978
Copy link

是否已有关于该错误的issue或讨论? | Is there an existing issue / discussion for this?

  • 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答? | Is there an existing answer for this in FAQ?

  • 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

训练过程中 loss 突然降为零
截屏2024-11-25 12 16 11

期望行为 | Expected Behavior

No response

复现方法 | Steps To Reproduce

No response

运行环境 | Environment

- OS:
- Python:
- Transformers:
- PyTorch:
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`):

备注 | Anything else?

No response

@wuzhanglin123
Copy link

我也遇到了单轮对话全量SFT,loss 变为零的情况,8卡训练会一直出现,4卡训练基本不出现

@MickeyFei
Copy link

遇到了单论对话lora sft,loss 从step2 开始就为0的情况 8卡训练出现,单卡训练不出现

@Evan9978
Copy link
Author

Evan9978 commented Dec 1, 2024

把 tune vision 设置为 false 之后这个问题不再出现了

@LDLINGLINGLING
Copy link
Collaborator

有没有可能是设置的有效长度太短,导致loss都不计算了

@Evan9978
Copy link
Author

Evan9978 commented Dec 5, 2024

有没有可能是设置的有效长度太短,导致loss都不计算了

这种情况应该不会后面loss全都变零

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants