mit-han-lab / streaming-llm

[ICLR 2024] Efficient Streaming Language Models with Attention Sinks
https://arxiv.org/abs/2309.17453
MIT License
6.59k stars 361 forks source link

Can you provide the code related to the visualization in the paper? #86

Open micelvrice opened 1 month ago

micelvrice commented 1 month ago

Thank you for your excellent work, regarding Figure2 in the paper, the phenomenon is really very interesting, but I can't get similar results, could you please provide the code and data related to that visualization. 11