ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-01-18 10:46:28 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Fix FlashAttention debug test, FP32 assert (llama/7684)

Browse Source
This commit is contained in:
Johannes Gäßler 2024-06-01 23:26:10 +02:00 committed by Georgi Gerganov
parent a16137d13d
commit 9b3d784020
1 changed files with 0 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
ggml-cuda/fattn-vec-f32.cuh
View File

@ -278,14 +278,10 @@ void ggml_cuda_flash_attn_ext_vec_f32_case_impl(ggml_backend_cuda_context & ctx,
template <int D, ggml_type type_K, ggml_type type_V>
void ggml_cuda_flash_attn_ext_vec_f32_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
ggml_tensor * KQV = dst;
ggml_tensor * Q = dst->src[0];
ggml_tensor * K = dst->src[1];
ggml_tensor * V = dst->src[2];
const int32_t precision = KQV->op_params[2];
GGML_ASSERT(precision == GGML_PREC_DEFAULT);
GGML_ASSERT(K->type == type_K);
GGML_ASSERT(V->type == type_V);