diff --git a/example/20_grouped_conv_bwd_weight/common.hpp b/example/20_grouped_conv_bwd_weight/common.hpp
index 2386e35b48..a7cdacd0a3 100644
--- a/example/20_grouped_conv_bwd_weight/common.hpp
+++ b/example/20_grouped_conv_bwd_weight/common.hpp
@@ -41,15 +41,13 @@ struct CommonLayoutSetting
     using WeightLayout = WeightLay;
     using OutputLayout = OutputLay;
 };
-using ALayout  = ck::tensor_layout::convolution::NHWGC;
-using BLayout = ck::tensor_layout::convolution::GKYXC;
-using ELayout = ck::tensor_layout::convolution::NHWGK;
+
 namespace ctl = ck::tensor_layout::convolution;
 template <ck::index_t NDimSpatial>
 struct CommonLayoutSettingSelector
     : CommonLayoutSetting<ck::tuple_element_t<NDimSpatial - 1,
                                               ck::Tuple<ck::tensor_layout::convolution::GNWC,
-                                                        ck::tensor_layout::convolution::NHWGC,
+                                                        ck::tensor_layout::convolution::GNHWC,
                                                         ck::tensor_layout::convolution::GNDHWC>>,
                           ck::tuple_element_t<NDimSpatial - 1,
                                               ck::Tuple<ck::tensor_layout::convolution::GKXC,
@@ -57,7 +55,7 @@ struct CommonLayoutSettingSelector
                                                         ck::tensor_layout::convolution::GKZYXC>>,
                           ck::tuple_element_t<NDimSpatial - 1,
                                               ck::Tuple<ck::tensor_layout::convolution::GNWK,
-                                                        ck::tensor_layout::convolution::NHWGK,
+                                                        ck::tensor_layout::convolution::GNHWK,
                                                         ck::tensor_layout::convolution::GNDHWK>>>
 {
 };
diff --git a/example/20_grouped_conv_bwd_weight/grouped_conv_bwd_weight_xdl_fp16.cpp b/example/20_grouped_conv_bwd_weight/grouped_conv_bwd_weight_xdl_fp16.cpp
index b24125ed3a..d6f7ad44d2 100644
--- a/example/20_grouped_conv_bwd_weight/grouped_conv_bwd_weight_xdl_fp16.cpp
+++ b/example/20_grouped_conv_bwd_weight/grouped_conv_bwd_weight_xdl_fp16.cpp
@@ -34,7 +34,7 @@ static constexpr index_t SizeOfType = 2;
 static constexpr index_t Tile_Align_W = Tile_W + 1;
 static constexpr index_t ShareMemSize = Tile_H * Tile_Align_W * N_Pack * SizeOfType;
 static constexpr index_t ScratchSize = ShareMemSize / 64 / 4;
-static constexpr index_t Num_Wave = 1;
+static constexpr index_t Num_Wave = 4;
 #define MergeShareMem 1
 template <typename T>
 __device__ T warp_shuffle_up(const T& v_local, uint32_t lane_delta)
@@ -605,9 +605,9 @@ struct DeviceGroupedConvBwdWeightNaive
 }
 }
 
-using ALayout = ck::tensor_layout::convolution::NHWGC;
+using ALayout = ck::tensor_layout::convolution::GNHWC;
 using BLayout = ck::tensor_layout::convolution::GKYXC;
-using ELayout = ck::tensor_layout::convolution::NHWGK;
+using ELayout = ck::tensor_layout::convolution::GNHWK;
 
 template <ck::index_t NDimSpatial>
 using DeviceConvBwdWeightInstance =