whisper.cpp/ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_1.cu
Johannes Gäßler e08c62149b CUDA: refactor mmq, dmmv, mmvq (llama/7716)
* CUDA: refactor mmq, dmmv, mmvq

* fix out-of-bounds write

* struct for qk, qr, qi

* fix cmake build

* mmq_type_traits
2024-06-16 18:19:48 +03:00

6 lines
177 B
Plaintext

// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-vec-f16.cuh"
DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q5_1);