aryansingh/gpt-2-inference-engine
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

comp files

Browse Source
This commit is contained in:
Aryan Singh committed 2026-06-07 14:30:55 -05:00
1 parent fa3efcaf07
commit 1f77d0131d
27 files changed
+1238 -798
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

kernels/residual.cuh
+2 -2
View File
@@ -4,7 +4,7 @@
#include <cuda_runtime.h>
#include "../utils/cuda_utils.cuh"
__global__ void residual_forward_kernel(float* out, float* inp1, float* inp2, int N) {
__global__ void residual_forward_kernel(float* __restrict__ out, const float* __restrict__ inp1, const float* __restrict__ inp2, int N) {
// Implement this
int x = blockDim.x * blockIdx.x + threadIdx.x;
@@ -16,7 +16,7 @@ __global__ void residual_forward_kernel(float* out, float* inp1, float* inp2, in
// Launch kernel here
void residual_forward(float* out, float* inp1, float* inp2, int N) {
// Implement this
dim3 block(32,1,1);
dim3 block(256,1,1);
dim3 grid((N + block.x - 1)/ block.x);
residual_forward_kernel<<<grid, block>>>(out, inp1, inp2, N);