I think this is what is used in the Python code
const size_t nb2 = dst->nb[2];
const size_t nb3 = dst->nb[3];
- const ggml_float eps = 1e-5f; // TODO: make this a parameter
+ const ggml_float eps = 1e-6f; // TODO: make this a parameter
// TODO: optimize
for (int i03 = 0; i03 < ne03; i03++) {
mean /= ne00;
float * y = (float *) ((char *) dst->data + i01*nb1 + i02*nb2 + i03*nb3);
-
+
memcpy(y, x, ne00 * sizeof(float));
// for (int i00 = 0; i00 < ne00; i00++) {
// y[i00] = x[i00];