CPU/CUDA: Gemma 2 FlashAttention support (#8542)

* CPU/CUDA: Gemma 2 FlashAttention support

* apply logit_softcap to scale in kernel

* disable logit softcapping tests on Metal

* remove metal check
This commit is contained in:
Johannes Gäßler 2024-08-24 21:34:59 +02:00 committed by GitHub
parent 8f824ffe8e
commit e11bd856d5
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
12 changed files with 319 additions and 79 deletions

View file

@ -22,6 +22,7 @@ typedef void (* fattn_kernel_t)(
const float m0,
const float m1,
const uint32_t n_head_log2,
const float logit_softcap,
const int ne00,
const int ne01,
const int ne02,
@ -657,11 +658,17 @@ void launch_fattn(
const dim3 blocks_num(parallel_blocks*((Q->ne[1] + cols_per_block - 1) / cols_per_block), Q->ne[2], Q->ne[3]);
const int shmem = 0;
float scale = 1.0f;
float max_bias = 0.0f;
float scale = 1.0f;
float max_bias = 0.0f;
float logit_softcap = 0.0f;
memcpy(&scale, (float *) KQV->op_params + 0, sizeof(float));
memcpy(&max_bias, (float *) KQV->op_params + 1, sizeof(float));
memcpy(&scale, (float *) KQV->op_params + 0, sizeof(float));
memcpy(&max_bias, (float *) KQV->op_params + 1, sizeof(float));
memcpy(&logit_softcap, (float *) KQV->op_params + 2, sizeof(float));
if (logit_softcap != 0.0f) {
scale /= logit_softcap;
}
const uint32_t n_head = Q->ne[2];
const uint32_t n_head_log2 = 1u << (uint32_t) floorf(log2f((float) n_head));
@ -675,7 +682,7 @@ void launch_fattn(
V_data,
mask ? ((const char *) mask->data) : nullptr,
(parallel_blocks) == 1 ? (float *) KQV->data : dst_tmp.ptr, dst_tmp_meta.ptr,
scale, max_bias, m0, m1, n_head_log2,
scale, max_bias, m0, m1, n_head_log2, logit_softcap,
Q->ne[0], Q->ne[1], Q->ne[2], Q->ne[3],
K->ne[0], K->ne[1], K->ne[2], K->ne[3],
mask ? mask->ne[1] : 0, mask ? mask->nb[1] : 0,