diff --git a/curvlinops/_torch_base.py b/curvlinops/_torch_base.py
index 2266b5ad..00e71975 100644
--- a/curvlinops/_torch_base.py
+++ b/curvlinops/_torch_base.py
@@ -191,7 +191,7 @@ def __check_tensor_and_preprocess(
         Raises:
             ValueError: If the input tensor has an invalid shape.
         """
-        if X.ndim > 2 or X.shape[0] != self.shape[1]:  # noqa: PLR2004
+        if X.ndim > 2 or X.shape[0] != self.shape[1]:
             raise ValueError(
                 f"Input tensor must have shape ({self.shape[1]},) or "
                 + f"({self.shape[1]}, K), with K arbitrary. Got {X.shape}."
diff --git a/curvlinops/diagonal/hutchinson.py b/curvlinops/diagonal/hutchinson.py
index 4f3e99c8..b7647a5a 100644
--- a/curvlinops/diagonal/hutchinson.py
+++ b/curvlinops/diagonal/hutchinson.py
@@ -62,7 +62,7 @@ def __init__(self, A: LinearOperator):
         Raises:
             ValueError: If the operator is not square.
         """
-        if len(A.shape) != 2 or A.shape[0] != A.shape[1]:  # noqa: PLR2004
+        if len(A.shape) != 2 or A.shape[0] != A.shape[1]:
             raise ValueError(f"A must be square. Got shape {A.shape}.")
         self._A = A
 
diff --git a/curvlinops/fisher.py b/curvlinops/fisher.py
index d5bc9d9d..d4a1e5e5 100644
--- a/curvlinops/fisher.py
+++ b/curvlinops/fisher.py
@@ -292,7 +292,7 @@ def sample_grad_output(self, output: Tensor, num_samples: int, y: Tensor) -> Ten
             NotImplementedError: If the prediction does not have two dimensions.
             NotImplementedError: If binary classification labels are not binary.
         """
-        if output.ndim != 2:  # noqa: PLR2004
+        if output.ndim != 2:
             raise NotImplementedError(f"Only 2d outputs supported. Got {output.shape}")
 
         C = output.shape[1]
diff --git a/curvlinops/kfac.py b/curvlinops/kfac.py
index b0435e33..1f6e488e 100644
--- a/curvlinops/kfac.py
+++ b/curvlinops/kfac.py
@@ -548,7 +548,7 @@ def draw_label(self, output: Tensor) -> Tensor:
             ValueError: If the output is not 2d.
             NotImplementedError: If the loss function is not supported.
         """
-        if output.ndim != 2:  # noqa: PLR2004
+        if output.ndim != 2:
             raise ValueError("Only a 2d output is supported.")
 
         if isinstance(self._loss_func, MSELoss):
diff --git a/curvlinops/kfac_utils.py b/curvlinops/kfac_utils.py
index 679cbf27..d7061739 100644
--- a/curvlinops/kfac_utils.py
+++ b/curvlinops/kfac_utils.py
@@ -100,7 +100,7 @@ def loss_hessian_matrix_sqrt(
         NotImplementedError: If the loss function is ``BCEWithLogitsLoss`` but the
             target is not binary.
     """
-    if output_one_datum.ndim != 2 or output_one_datum.shape[0] != 1:  # noqa: PLR2004
+    if output_one_datum.ndim != 2 or output_one_datum.shape[0] != 1:
         raise ValueError(
             f"Expected 'output_one_datum' to be 2d with shape [1, C], got "
             f"{output_one_datum.shape}"
diff --git a/curvlinops/trace/hutchinson.py b/curvlinops/trace/hutchinson.py
index 26f5e8a8..a2d1506c 100644
--- a/curvlinops/trace/hutchinson.py
+++ b/curvlinops/trace/hutchinson.py
@@ -58,7 +58,7 @@ def __init__(self, A: LinearOperator):
         Raises:
             ValueError: If the operator is not square.
         """
-        if len(A.shape) != 2 or A.shape[0] != A.shape[1]:  # noqa: PLR2004
+        if len(A.shape) != 2 or A.shape[0] != A.shape[1]:
             raise ValueError(f"A must be square. Got shape {A.shape}.")
         self._A = A
 
diff --git a/docs/examples/basic_usage/example_benchmark.py b/docs/examples/basic_usage/example_benchmark.py
index 3e3ca718..925febfb 100644
--- a/docs/examples/basic_usage/example_benchmark.py
+++ b/docs/examples/basic_usage/example_benchmark.py
@@ -185,7 +185,7 @@ def setup_problem(
         for m in supported_layers:
             # ignore the last layer of GPT because it has 50k outputs, which
             # will yield an extremely large Kronecker factor
-            if all(d <= 50_000 for d in m.weight.shape):  # noqa: PLR2004
+            if all(d <= 50_000 for d in m.weight.shape):
                 params.extend([p for p in m.parameters() if p.requires_grad])
     else:
         params = [p for p in model.parameters() if p.requires_grad]
diff --git a/docs/examples/basic_usage/example_inverses.py b/docs/examples/basic_usage/example_inverses.py
index b68840c0..849babd3 100644
--- a/docs/examples/basic_usage/example_inverses.py
+++ b/docs/examples/basic_usage/example_inverses.py
@@ -249,7 +249,7 @@
 # of the matrix to be inverted:
 max_eigval = eigsh(damped_GGN, k=1, which="LM", return_eigenvectors=False)[0]
 # eigenvalues (scale * damped_GGN_mat) are in [0; 2)
-scale = 1.0 if max_eigval < 2.0 else 1.99 / max_eigval  # noqa: PLR2004
+scale = 1.0 if max_eigval < 2.0 else 1.99 / max_eigval
 
 # %%
 #
diff --git a/test/test_inverse.py b/test/test_inverse.py
index 312c6e5e..eae8528b 100644
--- a/test/test_inverse.py
+++ b/test/test_inverse.py
@@ -137,7 +137,7 @@ def test_Neumann_inverse_damped_GGN_matvec(inv_case, delta: float = 1e-2):
 
     # set scale such that Neumann series converges
     eval_max = eigh(damped_GGN_functorch)[0][-1]
-    scale = 1.0 if eval_max < 2 else 1.9 / eval_max  # noqa: PLR2004
+    scale = 1.0 if eval_max < 2 else 1.9 / eval_max
 
     # NOTE This may break when other cases are added because slow convergence
     inv_GGN = NeumannInverseLinearOperator(GGN + damping, num_terms=7_000, scale=scale)
diff --git a/test/utils.py b/test/utils.py
index aa913a00..f36210db 100644
--- a/test/utils.py
+++ b/test/utils.py
@@ -250,7 +250,7 @@ def forward(self, x: Tensor) -> Tensor:
         # Example: Transformer for translation: (batch, sequence_length, c)
         # (although second and third dimension would have to be transposed for
         # classification)
-        if x.ndim > 2 and self.loss == "CE":  # noqa: PLR2004
+        if x.ndim > 2 and self.loss == "CE":
             x = rearrange(x, "batch ... c -> batch c ...")
         return x