未登录FlashAttention-T: Towards Fully Tensorized Attention by Exploiting Tensor-Vector ParallelismFlashAttention-T:通过利用张量-向量并行实现完全张量化注意力
Xu, Jianxing; Wen, Yuanbo; Bi, Jun; Xu, Ruibai; Xu, Guanglin; Zhang, Rui; Li, Wei; Li, Ling; Chen, Tianshi; Guo, Qi; Chen, Yunji
分享
收藏