vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Use F32 sqrtf instead of F64 sqrt

Browse source 
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
This commit is contained in:
Calvin Laurenson 2024-06-16 08:00:05 -07:00 • committed by GitHub
parent fea1dc98c0
commit 3591f1cc8c
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda/unary.cu
View file

@ -98,7 +98,7 @@ static __global__ void sqrt_f32(const float * x, float * dst, const int k) {
if (i >= k) {
return;
}
dst[i] = sqrt(x[i]);
dst[i] = sqrtf(x[i]);
}
static void gelu_f32_cuda(const float * x, float * dst, const int k, cudaStream_t stream) {