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

Would you plan to adapt it to qwen2-7B? #13

Open
King-king424 opened this issue Jul 10, 2024 · 3 comments
Open

Would you plan to adapt it to qwen2-7B? #13

King-king424 opened this issue Jul 10, 2024 · 3 comments

Comments

@King-king424
Copy link

No description provided.

@xiaoachen98
Copy link
Owner

No description provided.

Maybe you can contribute to this part. All you need to do is add a llava_qwen2.py, the corresponding conv_mode, and a preprocess_qwen2 function in the train.py file to handle the corresponding mask.

@King-king424
Copy link
Author

No description provided.

Maybe you can contribute to this part. All you need to do is add a llava_qwen2.py, the corresponding conv_mode, and a preprocess_qwen2 function in the train.py file to handle the corresponding mask.

I have tried using qwen2-7B-instruct as llm, but found that the fine-tuned results were not as expected. It may be that the preprocess was not done well. Now trying to find a public solution

@xiaoachen98
Copy link
Owner

No description provided.

Maybe you can contribute to this part. All you need to do is add a llava_qwen2.py, the corresponding conv_mode, and a preprocess_qwen2 function in the train.py file to handle the corresponding mask.

I have tried using qwen2-7B-instruct as llm, but found that the fine-tuned results were not as expected. It may be that the preprocess was not done well. Now trying to find a public solution

I have conducted the experiments on Qwen2-0.5B and everything is fine.

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

2 participants