diff --git a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp
index 6cba2e18d0..9510cedc43 100644
--- a/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp
+++ b/include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp
@@ -38,7 +38,7 @@ namespace {
 /*
  * \brief Wrapper function of GridwiseGemm Wmma Cshuffle V3 to realize grouped forward convolution.
  *
- * \tparam ComputePtrOffsetOfBatch Class that computes the base pointer offsets of A, B, D and E
+ * \tparam ComputePtrOffset Class that computes the base pointer offsets of A, B, D and E
  * matrices for groups or splitN. Currently it works for identical strides, but this can be extended
  * to other layouts. The returned offset can be either \p index_t or \p long_index_t. If it returns
  * \p long_index_t, we are not subject to the 2GB limitations.
@@ -46,14 +46,6 @@ namespace {
  * \tparam Block2ETileMap Block2ETileMap::CalculateBottomIndex() takes in the id of a workgroup and
  * returns the 2D index of the tile that it computes. \see
  * GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v2r3::Run().
- *
- * \note Using \p ComputePtrOffsetOfBatch gives us the flexibility that 2 workgroups can compute 2
- * tiles from different matrices. Keep in mind that these 2 matrices can share the same grid
- * descriptor (like in BatchedGEMM), or use their own grid descriptors (in GroupedGemm). \link
- * impl/device_conv3d_fwd_xdl_ndhwc_kzyxc_ndhwk.hpp kernel_gemm_xdlops_v2r3_for_conv3d \endlink for
- * \link DeviceConv3d \endlink uses the same concept, but currently does NOT encapsulate the
- * computing of pointer offset into \p ComputePtrOffsetOfStridedBatch.
- *
  */
 template <typename GridwiseGemm,
           typename AGridDesc_AK0_M_AK1,
@@ -124,8 +116,6 @@ __launch_bounds__(CK_MAX_THREAD_PER_BLOCK, MinimumOccupancy)
 #endif // End of if (!defined(__HIP_DEVICE_COMPILE__) || defined(__gfx11__) || defined(__gfx12__))
 }
 
-// TODO: Implement 2lds kernel?
-
 } // namespace
 
 template <typename T>
@@ -791,12 +781,6 @@ struct DeviceGroupedConvFwdMultipleABD_Wmma_CShuffle_V3
             }
 
             {
-                // Original effective calculation of MBlock and NBlock
-                // const auto M = e_grid_desc_m_n.GetLength(I0);
-                // const auto N = e_grid_desc_m_n.GetLength(I1);
-                // const auto MBlock = M / MPerBlock;
-                // const auto NBlock = N / NPerBlock;
-
                 const index_t GemmM = a_grid_desc_ak0_m_ak1_.GetLength(I1);
                 const index_t GemmN = b_grid_desc_bk0_n_bk1_.GetLength(I1);
                 const auto MBlock   = CTranspose ? GridwiseGemmCTranspose::CalculateMBlock(GemmN)
@@ -1149,9 +1133,14 @@ struct DeviceGroupedConvFwdMultipleABD_Wmma_CShuffle_V3
                             ck::utility::flush_icache();
                             // rotating mem
                             rotating_mem.Next();
-                            // clear c mem
-                            // TODO: this E clearing does not look correct. Fix when implementing
-                            // splitK. if(arg_.KBatch > 1)
+                            // clear E mem
+
+                            // TODO: The calculation of the E buffer size may not be correct in all
+                            // cases, for example if the memory is not contiguous due to padding or
+                            // unusual strides. Investigate when implementing splitK. It may be
+                            // safer to use GetElementSpaceSize().
+
+                            // if(arg_.KBatch > 1)
                             //     HIP_CHECK_ERROR(hipMemsetAsync(arg_.p_e_grid,
                             //                                    0,
                             //                                    arg_.M * arg_.N *
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_comp_instance.cpp
index fbbc19895c..712aa7b18a 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
index e4e9e37c6a..b56c9f196d 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp
index d3f2ddecf3..5a2d9c3aec 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp
index 9d071cf30b..41ee820f97 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/comp/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
index 585652e81b..e5922cd27f 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 GNHWC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_f16_instance.cpp
index a31f3d33c9..7847c1d24a 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_f16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[g, n, hi, wi, c] * wei[g, k, y, x, c] = out[g, n, ho, wo, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_gnhwc_gkyxc_gnhwk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 GNHWC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_16x16_instance.cpp
index e173a2c9c7..68b28e4c05 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instance.in b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instance.in
index 739cf5a9c1..8b1f50452a 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instance.in
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instance.in
@@ -21,7 +21,6 @@ using device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instances
                                                                 PassThrough,
                                                                 PassThrough>>>;
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 template <int Shards, int ShardIndex>
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instances_shard(
     device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_instances& instances)
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_16x16_instance.cpp
index 80f4d11044..c1bc35123a 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instance.in b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instance.in
index 307ceab1b0..b59c957259 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instance.in
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instance.in
@@ -21,7 +21,6 @@ using device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instances
                                                                 PassThrough,
                                                                 PassThrough>>>;
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 template <int Shards, int ShardIndex>
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instances_shard(
     device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_instances& instances)
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_bf16_instance.cpp
index f4ecc86067..5ea8c0c8ab 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_bf16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_f16_instance.cpp
index 8d16b21329..bc36d60d7d 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_f16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_int8_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_int8_instance.cpp
index e38652d2a8..9395115252 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_int8_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_int8_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkyxc_ngkhw_int8_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
index 0ef4e5f99f..5cae4c5548 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
index d7604a72ce..8a3b51222c 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instance.cpp
index 89bcf9b9e5..27314cdadc 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instance.cpp
index 0fe43a92c1..d0fed048ef 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_instance.cpp
index 51042a1cd6..0977ef9442 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_inter_instance.cpp
index 45da66418b..4af95245b3 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_intra_instance.cpp
index c2376c9e46..0e73a7ca43 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_bf16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_inter_instance.cpp
index 7d744b380e..32a9c52dd5 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_intra_instance.cpp
index 2c4e5b5d99..eecc2cb938 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_ngchw_gkcyx_ngkhw_f16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
index a0997fff61..edee843cca 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
index 3d9277d1da..89d1058ad0 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp
index 09cbdd32e1..466fa567f5 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp
index ba66416e21..7d91942bc9 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.in b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.in
index 1997bd55b7..19ea7f8162 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.in
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter_instance.in
@@ -21,7 +21,6 @@ using device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter
                                                                 PassThrough,
                                                                 PassThrough>>>;
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 template <int Shards, int ShardIndex>
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter_instances_shard(
     device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_inter_instances& instances)
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.in b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.in
index c770f2442b..4438a2830f 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.in
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/mem/device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra_instance.in
@@ -21,7 +21,6 @@ using device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra
                                                                 PassThrough,
                                                                 PassThrough>>>;
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 template <int Shards, int ShardIndex>
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra_instances_shard(
     device_grouped_conv2d_fwd_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_int8_mem_intra_instances& instances)
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_bf16_instance.cpp
index 00a513fbb4..f603720f0c 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_bf16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_f16_instance.cpp
index 08030dd596..edd109c4da 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_f16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_ngchw_gkcyx_ngkhw_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NGCHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
index a79fdc3eb7..5953c399d2 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instance.cpp
index b9915b655a..d6da0f4915 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_int8_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_int8_instance.cpp
index 2f602d8bd8..6f796967f7 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_int8_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/wmma/merged_groups/device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_int8_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_int8_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
index 03bf021b5f..10b942d4d9 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp
index e964de4c0a..09ef8ced40 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/comp/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
index c78004e94a..98354619cf 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
index 986c525ac4..35b9e58747 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp
index 1e0664cdc8..4549779019 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
index e3b6f678fb..16731b71d6 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
index 60b209da1e..6646cd0c83 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
index b161a48107..096ada2922 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp
index ff0c1372d4..42bbf69597 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp
index 17bdb2ae70..7aaa6a0c83 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/mem/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
index 0c07022b88..fc4064e117 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
index db689b231b..0d98791036 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_bias_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
index 4054c12ba8..93a87c0d6c 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp
index ce72c810cf..c34825c2f3 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/comp/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_comp_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_comp_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
index 06b5b57a55..cd0a747383 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
index 0389af2356..020fcc50f6 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp
index de8c6c817c..1df82bdc70 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
index bf597ef659..459a707d9e 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
index 5d8490e5e0..d67cf409cf 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
index 1d8fcc958b..03900bfd88 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp
index dcde3cbc4b..ffb0e911ad 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_inter_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_inter_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp
index 0280e82252..941fe8890b 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/mem/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_fp16_mem_intra_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_mem_intra_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
index 8f07e44661..0c97e52f07 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_bf16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
index fa19121e2a..86bae39d88 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_clamp/wmma/merged_groups/device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_fp16_instance.cpp
@@ -9,7 +9,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv2d_fwd_clamp_wmma_cshufflev3_merged_groups_nhwgc_gkyxc_nhwgk_f16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
                                                                 NHWGC,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_16x16_instance.cpp
index 547223a21e..480f1cb728 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 NGCDHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_instance.in b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_instance.in
index 8f46e4da99..40a4e67927 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_instance.in
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_instance.in
@@ -7,7 +7,6 @@
 
 namespace ck::tensor_operation::device::instance {
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 using device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_bf16_instances =
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 NGCDHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_16x16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_16x16_instance.cpp
index 08a64b7ee2..36a2b743fa 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_16x16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_16x16_instance.cpp
@@ -8,7 +8,6 @@ namespace ck {
 namespace tensor_operation {
 namespace device {
 namespace instance {
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 void add_device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_16x16_instances(
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 NGCDHW,
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_instance.in b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_instance.in
index a36b3317a4..80865caf5a 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_instance.in
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd/wmma/device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_instance.in
@@ -7,7 +7,6 @@
 
 namespace ck::tensor_operation::device::instance {
 
-// Compilation parameters for in[n, hi, wi, g, c] * wei[g, k, y, x, c] = out[n, ho, wo, g, k]
 using device_grouped_conv3d_fwd_wmma_cshufflev3_ngcdhw_gkczyx_ngkdhw_f16_instances =
     std::vector<std::unique_ptr<DeviceGroupedConvFwdMultipleABD<3,
                                                                 NGCDHW,