diff --git a/src/neuronx_distributed/modules/lora/layer.py b/src/neuronx_distributed/modules/lora/layer.py
index 0eaecd7..8c64084 100644
--- a/src/neuronx_distributed/modules/lora/layer.py
+++ b/src/neuronx_distributed/modules/lora/layer.py
@@ -314,8 +314,8 @@ def _embed(self, input: torch.Tensor, weight: torch.Tensor) -> torch.Tensor:
         )
 
     def forward(self, x: torch.Tensor, *args: Any, **kwargs: Any) -> torch.Tensor:
-        previous_dtype = x.dtype
         result = self.base_layer(x, *args, **kwargs)
+        previous_dtype = result.dtype
         if not self.merged:
             assert self.lora_embedding_A is not None and self.lora_embedding_B is not None
             embedding_A = self.lora_embedding_A.T
diff --git a/test/unit_test/modules/lora/test_lora_layers.py b/test/unit_test/modules/lora/test_lora_layers.py
index 7b50904..6e3c994 100644
--- a/test/unit_test/modules/lora/test_lora_layers.py
+++ b/test/unit_test/modules/lora/test_lora_layers.py
@@ -71,6 +71,19 @@ def test_torch_embedding_layer(self):
                 layer_str = str(lora_layer)
                 assert "lora" in layer_str
 
+    def test_torch_embedding_forward_preserves_output_dtype(self):
+        layer = torch.nn.Embedding(8, 3, dtype=torch.float32)
+        with torch.no_grad():
+            layer.weight[0] = torch.tensor([0.25, 0.5, 0.75])
+        lora_layer = LoraEmbedding(layer, get_lora_config())
+        input_ids = torch.tensor([0], dtype=torch.long)
+
+        expected = layer(input_ids)
+        actual = lora_layer(input_ids)
+
+        self.assertEqual(actual.dtype, expected.dtype)
+        torch.testing.assert_close(actual, expected)
+
     @patch("neuronx_distributed.parallel_layers.layers.get_tensor_model_parallel_size", MagicMock(return_value=8))
     @patch("neuronx_distributed.parallel_layers.layers.get_tensor_model_parallel_rank", MagicMock(return_value=1))
     @patch("neuronx_distributed.parallel_layers.parallel_state.initialize_model_parallel", MagicMock(return_value=True))