diff --git a/example/20_grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_dl_v4.hpp b/example/20_grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_dl_v4.hpp
index fb7eff4610..0f39c3b751 100644
--- a/example/20_grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_dl_v4.hpp
+++ b/example/20_grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_dl_v4.hpp
@@ -1444,12 +1444,13 @@ struct DeviceGroupedConvBwdWeightDlV4 : public DeviceGroupedConvBwdWeight<NDimSp
             << "Dilation<" << Dilation_Y << ", " << Dilation_X << ">, "
             << "Stride<" << Stride_H << ", " << Stride_W<< ">, "
             << "Pad<" << Pad_H << ", " << Pad_W<< ">, "
-            << "NBatch: " << NBatch<< ", "
-            << "NumWavePerTile: " << NumWavePerTile<< ", "
-            << "InScalarPerVector: " << InScalarPerVector<< ", "
-            << "OutScalarPerVector: " << OutScalarPerVector<< ", "
-            << "DstScalarPerVector: " << DstScalarPerVector<< ", "
-            << "RequirePadding: " << RequirePadding << ">"
+            << "NBatch: " << NBatch << ", "
+            << "NumWavePerTile: " << NumWavePerTile << ", "
+            << "InScalarPerVector: " << InScalarPerVector << ", "
+            << "OutScalarPerVector: " << OutScalarPerVector << ", "
+            << "DstScalarPerVector: " << DstScalarPerVector << ", "
+            << "RequirePadding: " << RequirePadding << ", "
+            << "WSplit: " << WSplit << ">"
             << std::endl;
         // clang-format on
 
diff --git a/example/20_grouped_conv_bwd_weight/run_grouped_conv_bwd_weight_example.inc b/example/20_grouped_conv_bwd_weight/run_grouped_conv_bwd_weight_example.inc
index 1a20e6ded4..605d0acdbf 100644
--- a/example/20_grouped_conv_bwd_weight/run_grouped_conv_bwd_weight_example.inc
+++ b/example/20_grouped_conv_bwd_weight/run_grouped_conv_bwd_weight_example.inc
@@ -98,9 +98,9 @@ bool run_grouped_conv_bwd_weight(const ExecutionConfig& config,
     }
 
 #if ENABLE_CONV_FACTORY
-    std::size_t best_tflops = 0;
-    std::size_t best_gb_per_sec = 0;
-    std::size_t best_avg_time = 0;
+    float best_tflops = 0;
+    float best_gb_per_sec = 0;
+    float best_avg_time = 0;
     std::string best_kernel = "";
     ck::index_t split_k_array[] = {1, 2, 4, 8, 16, 32};
     ck::index_t split_k_count = 6;