Cite
Notes
Only stored in your browser.
Attribution
Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent
arXiv 2024
More Expressive Attention with Negative Weights
from 2 papers
Di Wang
Ruobing Xie
Xingwu Sun
Zhanhui Kang
Ang Lv
Ao Liu
Bin Hu
Bo wang
Chao Yu
Chenchen Zhang