Steven10429's picture
llama.cpp
61b850a
raw
history blame contribute delete
98 Bytes
#include "common.cuh"
void ggml_cuda_argmax(ggml_backend_cuda_context & ctx, ggml_tensor * dst);