diff --git a/pandas/core/arrays/sparse/accessor.py b/pandas/core/arrays/sparse/accessor.py
index fc7debb1f31e4..67bb417865475 100644
--- a/pandas/core/arrays/sparse/accessor.py
+++ b/pandas/core/arrays/sparse/accessor.py
@@ -92,8 +92,8 @@ def from_coo(cls, A, dense_index: bool = False) -> Series:
         ...     ([3.0, 1.0, 2.0], ([1, 0, 0], [0, 2, 3])), shape=(3, 4)
         ... )
         >>> A
-        <3x4 sparse matrix of type '<class 'numpy.float64'>'
-        with 3 stored elements in COOrdinate format>
+        <COOrdinate sparse matrix of dtype 'float64'
+            with 3 stored elements and shape (3, 4)>
 
         >>> A.todense()
         matrix([[0., 0., 1., 2.],
@@ -178,8 +178,8 @@ def to_coo(self, row_levels=(0,), column_levels=(1,), sort_labels: bool = False)
         ...     row_levels=["A", "B"], column_levels=["C", "D"], sort_labels=True
         ... )
         >>> A
-        <3x4 sparse matrix of type '<class 'numpy.float64'>'
-        with 3 stored elements in COOrdinate format>
+        <COOrdinate sparse matrix of dtype 'float64'
+            with 3 stored elements and shape (3, 4)>
         >>> A.todense()
         matrix([[0., 0., 1., 3.],
         [3., 0., 0., 0.],
@@ -350,8 +350,8 @@ def to_coo(self):
         --------
         >>> df = pd.DataFrame({"A": pd.arrays.SparseArray([0, 1, 0, 1])})
         >>> df.sparse.to_coo()
-        <4x1 sparse matrix of type '<class 'numpy.int64'>'
-                with 2 stored elements in COOrdinate format>
+        <COOrdinate sparse matrix of dtype 'int64'
+            with 2 stored elements and shape (4, 1)>
         """
         import_optional_dependency("scipy")
         from scipy.sparse import coo_matrix