diff --git a/include/micm/cuda/util/cuda_dense_matrix.hpp b/include/micm/cuda/util/cuda_dense_matrix.hpp
index 192807ff0..efd3f07f2 100644
--- a/include/micm/cuda/util/cuda_dense_matrix.hpp
+++ b/include/micm/cuda/util/cuda_dense_matrix.hpp
@@ -68,7 +68,6 @@ namespace micm
     CudaMatrixParam param_;
 
    public:
-
     CudaDenseMatrix()
         : VectorMatrix<T, L>()
     {
@@ -213,8 +212,7 @@ namespace micm
       {
         // the cudaMemset function only works for integer types and is an asynchronous function:
         // https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__MEMORY.html#group__CUDART__MEMORY_1gf7338650f7683c51ee26aadc6973c63a
-        CHECK_CUDA_ERROR(
-            cudaMemset(this->param_.d_data_, val, sizeof(T) * this->param_.number_of_elements_), "cudaMemset");
+        CHECK_CUDA_ERROR(cudaMemset(this->param_.d_data_, val, sizeof(T) * this->param_.number_of_elements_), "cudaMemset");
       }
       else
       {
diff --git a/include/micm/cuda/util/cuda_sparse_matrix.hpp b/include/micm/cuda/util/cuda_sparse_matrix.hpp
index 25eb07cfd..64cdb7316 100644
--- a/include/micm/cuda/util/cuda_sparse_matrix.hpp
+++ b/include/micm/cuda/util/cuda_sparse_matrix.hpp
@@ -107,8 +107,7 @@ namespace micm
       {
         // the cudaMemset function only works for integer types and is an asynchronous function:
         // https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__MEMORY.html#group__CUDART__MEMORY_1gf7338650f7683c51ee26aadc6973c63a
-        CHECK_CUDA_ERROR(
-            cudaMemset(this->param_.d_data_, val, sizeof(T) * this->param_.number_of_elements_), "cudaMemset");
+        CHECK_CUDA_ERROR(cudaMemset(this->param_.d_data_, val, sizeof(T) * this->param_.number_of_elements_), "cudaMemset");
       }
       else
       {
diff --git a/src/util/cuda_matrix.cu b/src/util/cuda_matrix.cu
index 0f79bd133..97e92a77e 100644
--- a/src/util/cuda_matrix.cu
+++ b/src/util/cuda_matrix.cu
@@ -85,8 +85,12 @@ namespace micm
     template cudaError_t MallocVector<int>(CudaMatrixParam& param, std::size_t number_of_elements);
     template cudaError_t CopyToDevice<double>(CudaMatrixParam& param, std::vector<double>& h_data);
     template cudaError_t CopyToHost<double>(CudaMatrixParam& param, std::vector<double>& h_data);
-    template cudaError_t CopyToDeviceFromDevice<double>(CudaMatrixParam& vectorMatrixDest, const CudaMatrixParam& vectorMatrixSrc);
-    template cudaError_t CopyToDeviceFromDevice<int>(CudaMatrixParam& vectorMatrixDest, const CudaMatrixParam& vectorMatrixSrc);
+    template cudaError_t CopyToDeviceFromDevice<double>(
+        CudaMatrixParam& vectorMatrixDest,
+        const CudaMatrixParam& vectorMatrixSrc);
+    template cudaError_t CopyToDeviceFromDevice<int>(
+        CudaMatrixParam& vectorMatrixDest,
+        const CudaMatrixParam& vectorMatrixSrc);
     template cudaError_t FillCudaMatrix<double>(CudaMatrixParam& param, double val);
   }  // namespace cuda
 }  // namespace micm