mit-han-lab/spatten-llm
[HPCA'21] SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning
ScalaMIT
Issues
- 0
missing thirdparty/ramulator2 folder
#1 opened by JJJayyyy
[HPCA'21] SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning
ScalaMIT