OpenGVLab / InternVL

[CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型
https://internvl.readthedocs.io/en/latest/
MIT License
6.09k stars 474 forks source link

[Feature] Attention Visualization #700

Open paulgavrikov opened 1 week ago

paulgavrikov commented 1 week ago

Motivation

Could you kindly share code to visualize the attention to, both, the prompt (tokens) and the input image in InternVL?

Related resources

No response

Additional context

No response

sms-s commented 13 hours ago

+1