diff --git a/examples/scaffolded_scheduler/scheduling_ddpm_lite.py b/examples/scaffolded_scheduler/scheduling_ddpm_lite.py
index 89f1003..df0b9e4 100644
--- a/examples/scaffolded_scheduler/scheduling_ddpm_lite.py
+++ b/examples/scaffolded_scheduler/scheduling_ddpm_lite.py
@@ -46,7 +46,8 @@ def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.devic
         """
         self.num_inference_steps = num_inference_steps
         step = self.config.num_train_timesteps // num_inference_steps
-        timesteps = (torch.arange(0, num_inference_steps) * step).round()[::-1].clone()
+        timesteps = (torch.arange(0, num_inference_steps) * step).round().long()
+        timesteps = torch.flip(timesteps, dims=[0])   # torch has no [::-1]
         self.timesteps = timesteps.to(device) if device is not None else timesteps
 
     def step(
diff --git a/src/diffusers/schedulers/scheduling_euler_lite.py b/src/diffusers/schedulers/scheduling_euler_lite.py
index c41edc5..5fc5795 100644
--- a/src/diffusers/schedulers/scheduling_euler_lite.py
+++ b/src/diffusers/schedulers/scheduling_euler_lite.py
@@ -40,7 +40,8 @@ def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.devic
         """
         self.num_inference_steps = num_inference_steps
         step = self.config.num_train_timesteps // num_inference_steps
-        timesteps = (torch.arange(0, num_inference_steps) * step).round()[::-1].clone()
+        timesteps = (torch.arange(0, num_inference_steps) * step).round().long()
+        timesteps = torch.flip(timesteps, dims=[0])   # torch has no [::-1]
         self.timesteps = timesteps.to(device) if device is not None else timesteps
 
     def step(
diff --git a/templates/scheduler/scheduling_TEMPLATE.py b/templates/scheduler/scheduling_TEMPLATE.py
index 907ed84..e52665a 100644
--- a/templates/scheduler/scheduling_TEMPLATE.py
+++ b/templates/scheduler/scheduling_TEMPLATE.py
@@ -36,7 +36,8 @@ def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.devic
         """
         self.num_inference_steps = num_inference_steps
         step = self.config.num_train_timesteps // num_inference_steps
-        timesteps = (torch.arange(0, num_inference_steps) * step).round()[::-1].clone()
+        timesteps = (torch.arange(0, num_inference_steps) * step).round().long()
+        timesteps = torch.flip(timesteps, dims=[0])   # torch has no [::-1]
         self.timesteps = timesteps.to(device) if device is not None else timesteps
 
     def step(
diff --git a/tests/schedulers/test_scheduling_euler_lite.py b/tests/schedulers/test_scheduling_euler_lite.py
index 8e4c2eb..3764f3b 100644
--- a/tests/schedulers/test_scheduling_euler_lite.py
+++ b/tests/schedulers/test_scheduling_euler_lite.py
@@ -103,6 +103,7 @@ def test_output_type(self):
         out = s.step(torch.ones_like(sample), 1, sample, generator=torch.Generator().manual_seed(0))
         self.assertTrue(hasattr(out, "prev_sample"))
         self.assertEqual(out.prev_sample.shape, sample.shape)
+        self.assertEqual(out.prev_sample.dtype, sample.dtype)
 
     def test_same_seed_same_output(self):
         import torch