mirror of
https://github.com/ikawrakow/ik_llama.cpp.git
synced 2026-03-05 19:40:19 +00:00
154e0d75fc
* Merging mainline - WIP * Merging mainline - WIP AVX2 and CUDA appear to work. CUDA performance seems slightly (~1-2%) lower as it is so often the case with llama.cpp/ggml after some "improvements" have been made. * Merging mainline - fix Metal * Remove check --------- Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
23 lines
498 B
Plaintext
23 lines
498 B
Plaintext
#version 450
|
|
|
|
#include "generic_head.comp"
|
|
#include "types.comp"
|
|
|
|
#extension GL_EXT_control_flow_attributes : enable
|
|
|
|
layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
|
|
|
|
layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
|
|
layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
|
|
|
|
void main() {
|
|
const uint i = gl_GlobalInvocationID.x;
|
|
|
|
if (i >= p.KX) {
|
|
return;
|
|
}
|
|
|
|
const float xi = float(data_a[i]);
|
|
data_d[i] = D_TYPE(xi / (1.0f + exp(-xi)));
|
|
}
|