Home
last modified time | relevance | path

Searched defs:tVgV (Results 1 – 2 of 2) sorted by relevance

/external/pytorch/aten/src/ATen/native/transformers/cuda/flash_attn/
Dflash_fwd_kernel.h158 Tensor tVgV = gmem_thr_copy_QKV.partition_S(gV); // (VCPY, VCPY_N, VCPY_K, nblocksN) in compute_attn_1rowblock() local
604 Tensor tVgV = gmem_thr_copy_QKV.partition_S(gV); // (VCPY, VCPY_N, VCPY_K) in compute_attn_1rowblock_splitkv() local
Dflash_bwd_kernel.h203 Tensor tVgV = gmem_thr_copy_QKV.partition_S(gV); // (VCPY, VCPY_N, VCPY_K) in compute_dq_dk_dv_1colblock() local