diff --git a/include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp b/include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp
index c2c7652085..e34c19bdf0 100644
--- a/include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp
+++ b/include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp
@@ -700,7 +700,7 @@ struct DeviceBatchedGemmMultiD_Xdl : public DeviceBatchedGemmMultiD<ALayout,
             << BlockSize << ", "
             << MPerBlock << ", "
             << NPerBlock << ", "
-            << KPerBlock
+            << KPerBlock << ", "
             << AK1 << ", "
             << BK1 << ", "
             << getGemmSpecializationString(GemmSpec)