Skip to main content

attn_mma_smem_bytes

Function attn_mma_smem_bytes 

Source
pub fn attn_mma_smem_bytes(hd: usize, vd: usize) -> usize
Expand description

The f16 tensor-core attention block’s dynamic shared memory: a tile’s keys (attn_mma_keys) and their values as f16, rows padded eight values.