mit-han-lab / streaming-llm

[ICLR 2024] Efficient Streaming Language Models with Attention Sinks
https://arxiv.org/abs/2309.17453
MIT License
6.38k stars 355 forks source link

Request for Code and Details on Figures 2 and 7 #67

Open ZhouZineng opened 8 months ago

ZhouZineng commented 8 months ago

Hello,

Firstly, I would like to express my admiration for the excellent work you've done on this project. The findings are insightful and the presentation of the data is clear and engaging.

I'm particularly interested in the visualizations presented in Figures 2 and 7 of the associated paper. They provide a unique perspective on the data that I find very intriguing.

Would it be possible for you to provide the specific code used to generate these figures, along with the corresponding dataset sentences used? I believe this would greatly help in my understanding of the phenomenon you are presenting and would be immensely beneficial for my ongoing research.

Thank you for your time and consideration. I look forward to your response. :)