now emit the json format

Author: mike dupont

examples/main/main.cpp

@@ -659,7 +659,7 @@ int main(int argc, char ** argv) {
         if (input_echo) {
             for (auto id : embd) {
                 const std::string token_str = llama_token_to_piece(ctx, id);
-                printf("%s", token_str.c_str());
+                printf("TOKEN:%s\n", token_str.c_str());
 
                 if (embd.size() > 1) {
                     input_tokens.push_back(id);

ggml-cuda.cu

@@ -7623,12 +7623,10 @@ static void ggml_cuda_mul_mat(const ggml_tensor * src0, const ggml_tensor * src1
 #endif
 
     // debug helpers
-    printf("src0: %8d %8d %8d %8d\n", src0->ne[0], src0->ne[1], src0->ne[2], src0->ne[3]);
-    printf("      %8d %8d %8d %8d\n", src0->nb[0], src0->nb[1], src0->nb[2], src0->nb[3]);
-    printf("src1: %8d %8d %8d %8d\n", src1->ne[0], src1->ne[1], src1->ne[2], src1->ne[3]);
-    printf("      %8d %8d %8d %8d\n", src1->nb[0], src1->nb[1], src1->nb[2], src1->nb[3]);
-    printf("src0 is contiguous %d, transposed %d, type = %s, name = %s\n", ggml_is_contiguous(src0), ggml_is_transposed(src0), ggml_type_name(src0->type), src0->name);
-    printf("src1 is contiguous %d, transposed %d, type = %s, name = %s\n", ggml_is_contiguous(src1), ggml_is_transposed(src1), ggml_type_name(src1->type), src1->name);
+    printf("JSON { \"data\":{ \"src0\": { \"ne\" : [ %8d, %8d, %8d, %8d ],	   \"nb\" : [ %8d, %8d, %8d, %8d ], \"contiguous\":\"%d\", \"transposed\":\"%d\", \"type\": \"%s\", \"name\" : \"%s\"}, \"src1\": { \"ne\" : [ %8d, %8d, %8d, %8d ], \"nb\" : [ %8d, %8d, %8d, %8d ], \"contiguous\":\"%d\", \"transposed\":\"%d\", \"type\": \"%s\", \"name\" : \"%s\"}}}\n",
+	   src0->ne[0], src0->ne[1], src0->ne[2], src0->ne[3], src0->nb[0], src0->nb[1], src0->nb[2], src0->nb[3],
+	   ggml_is_contiguous(src0), ggml_is_transposed(src0), ggml_type_name(src0->type), src0->name,
+	   src1->ne[0], src1->ne[1], src1->ne[2], src1->ne[3], src1->nb[0], src1->nb[1], src1->nb[2], src1->nb[3], ggml_is_contiguous(src1), ggml_is_transposed(src1), ggml_type_name(src1->type), src1->name);
 
     if (!split && all_on_device && !use_tensor_cores && src0->type == GGML_TYPE_F16 && ggml_is_permuted(src0) && ggml_is_permuted(src1) && src1->ne[1] == 1) {
         // KQ single-batch