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

Isn't 2048 max lenght would out of context if 6 images put in? #16

Closed
OpenJarvisAI opened this issue Apr 17, 2024 · 1 comment
Closed

Comments

@OpenJarvisAI
Copy link

As I known, clip vit 336 would produce 576 visual tokens per image, the UHD are stack 6 of them, that is 3000+ visual tokens.
How does it able to send to LLM?

@guozonghao96
Copy link
Collaborator

We use a resampler to down-sample the feature tokes, ie, a slice 576 to 144. So that, 6 slices only need 1008 tokens to represent.

Moreover, our repository has been fully improved, and almost all bugs have been eliminated. For details, please refer to the main branch and the LLaVA-UHD v1 branch. This issue is now closed. If there are any new problems, feel free to open a new issue.

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