Ошибка памяти при запуске на больших данных

#include "cuda_runtime.h"
#include "device_launch_parameters.h"
 
#include <stdio.h>
#include <malloc.h>
#include <string.h>
#include <windows.h>
#include <math.h>
 
#define SIZE_BASE_ARR (int)pow((double)2,(double)14)
#define SIZE_LOOK_ARR (int)pow((double)2,(double)20)
 
#define THREADS_IN_BLOCK (int)pow((double)2,(double)9)
#define BLOCKS_IN_GRID (int)pow((double)2,(double)11)
 
#define gpuErrchk(ans) {gpuAssert((ans), __FILE__, __LINE__);}
void gpuAssert(cudaError_t code, const char *file, int line, bool abort = true)
{
    if (code != cudaSuccess) {
        fprintf(stderr, "GPUassert: %s %s %d\n", cudaGetErrorString(code), file, line);
        //if (abort) exit(code);
    }
}
__global__ void searchElements(float *lookForArr, float *baseArr, float *resultsArr)
{
    int lookForArrId = threadIdx.x + blockDim.x * blockIdx.x;
    int i = 0;
 
    if (lookForArrId < SIZE_LOOK_ARR) {
        for (i = 0; i < SIZE_BASE_ARR; i++) 
            if (baseArr[i] == lookForArr[lookForArrId])
                resultsArr[lookForArrId]++;
    }
}
void initArrs (float *arr, int size)
{
    int i = 0;
    for (i = 0; i < size; i++)
        arr[i] = 1;
}
int main()
{
    float *baseArr = NULL, *lookForArr = NULL, *resultArr = NULL;
    float *baseArrDev = NULL, *lookForArrDev = NULL, *resultArrDev = NULL;
    int i = 0;
 
    baseArr = (float*)malloc(SIZE_BASE_ARR * sizeof(float));
    lookForArr = (float*)malloc(SIZE_LOOK_ARR * sizeof(float));
    resultArr = (float*)malloc(SIZE_LOOK_ARR * sizeof(float));
 
    initArrs(baseArr, SIZE_BASE_ARR); initArrs(lookForArr, SIZE_LOOK_ARR);
    memset(resultArr, 0, SIZE_LOOK_ARR * sizeof(float));
 
 
    cudaMalloc( (void**) &baseArrDev, SIZE_BASE_ARR * sizeof(float));
    cudaMalloc( (void**) &lookForArrDev, SIZE_LOOK_ARR * sizeof(float));
    cudaMalloc( (void**) &resultArrDev, SIZE_LOOK_ARR * sizeof(float));
 
    gpuErrchk( cudaMemcpy(baseArrDev, baseArr, SIZE_BASE_ARR * sizeof(float), cudaMemcpyHostToDevice));
    gpuErrchk( cudaMemcpy(lookForArrDev, lookForArr, SIZE_LOOK_ARR * sizeof(float), cudaMemcpyHostToDevice));
    gpuErrchk( cudaMemcpy(resultArrDev, resultArr, SIZE_LOOK_ARR * sizeof(float), cudaMemcpyHostToDevice));
 
    dim3 threads = dim3(THREADS_IN_BLOCK);
    dim3 blocks = dim3(BLOCKS_IN_GRID);
 
    searchElements<<<blocks, threads>>> (lookForArrDev, baseArrDev, resultArrDev);
    gpuErrchk( cudaPeekAtLastError() );
    gpuErrchk( cudaDeviceSynchronize() );
    gpuErrchk( cudaMemcpy(resultArr, resultArrDev, SIZE_LOOK_ARR * sizeof(float), cudaMemcpyDeviceToHost) );
 
    cudaFree(baseArrDev); cudaFree(lookForArrDev); cudaFree(resultArrDev);
 
    for (i = 0; i < SIZE_LOOK_ARR; i++)
        if (resultArr[i] != SIZE_BASE_ARR) {
            printf("Error\n");
            break;
        }
    printf("Press any key...");
    getchar();
    return 0;
}