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.

Sign up for GitHub

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

Jump to bottom

[BUG] Pre-training unstable #309

Open

2 tasks done

orrzohar opened this issue Jun 28, 2024 · 0 comments

orrzohar commented Jun 28, 2024 •

edited

Loading

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

I am trying to pre-train MiniCPM-V-LLaMA3.
I tried the following:

lr 1e-3 to 1e-5
batch size 128 to 1024
However, I keep getting NaNs in/after the perceiver resampler.

期望行为 | Expected Behavior

Stable training

复现方法 | Steps To Reproduce

Clone this repo
download image pre-training datasets
do pretraining

运行环境 | Environment

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

备注 | Anything else?

No response

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment