diff --git a/tensorflow_addons/layers/__init__.py b/tensorflow_addons/layers/__init__.py
index d527e16362..e6c9d0055c 100644
--- a/tensorflow_addons/layers/__init__.py
+++ b/tensorflow_addons/layers/__init__.py
@@ -18,11 +18,11 @@
 from __future__ import division
 from __future__ import print_function
 
-from tensorflow_addons.layers.gelu import GeLU
+from tensorflow_addons.layers.gelu import GELU
 from tensorflow_addons.layers.maxout import Maxout
 from tensorflow_addons.layers.normalizations import GroupNormalization
 from tensorflow_addons.layers.normalizations import InstanceNormalization
 from tensorflow_addons.layers.optical_flow import CorrelationCost
 from tensorflow_addons.layers.poincare import PoincareNormalize
 from tensorflow_addons.layers.sparsemax import Sparsemax
-from tensorflow_addons.layers.wrappers import WeightNormalization
\ No newline at end of file
+from tensorflow_addons.layers.wrappers import WeightNormalization
diff --git a/tensorflow_addons/layers/gelu.py b/tensorflow_addons/layers/gelu.py
index 31420f76de..d2dcfa87e5 100644
--- a/tensorflow_addons/layers/gelu.py
+++ b/tensorflow_addons/layers/gelu.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 # ==============================================================================
-"""Implements GeLU activation."""
+"""Implements GELU activation."""
 
 from __future__ import absolute_import
 from __future__ import division
@@ -23,7 +23,7 @@
 
 
 @tf.keras.utils.register_keras_serializable(package='Addons')
-class GeLU(tf.keras.layers.Layer):
+class GELU(tf.keras.layers.Layer):
     """Gaussian Error Linear Unit.
 
     A smoother version of ReLU generally used
@@ -40,7 +40,7 @@ class GeLU(tf.keras.layers.Layer):
     """
 
     def __init__(self, approximate=True, **kwargs):
-        super(GeLU, self).__init__(**kwargs)
+        super(GELU, self).__init__(**kwargs)
         self.approximate = approximate
         self.supports_masking = True
 
@@ -49,7 +49,7 @@ def call(self, inputs):
 
     def get_config(self):
         config = {'approximate': self.approximate}
-        base_config = super(GeLU, self).get_config()
+        base_config = super(GELU, self).get_config()
         return dict(list(base_config.items()) + list(config.items()))
 
     def compute_output_shape(self, input_shape):
diff --git a/tensorflow_addons/layers/gelu_test.py b/tensorflow_addons/layers/gelu_test.py
index 99331fb44e..d31badcab1 100644
--- a/tensorflow_addons/layers/gelu_test.py
+++ b/tensorflow_addons/layers/gelu_test.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 # ==============================================================================
-"""Tests for GeLU activation."""
+"""Tests for GELU activation."""
 
 from __future__ import absolute_import
 from __future__ import division
@@ -21,18 +21,18 @@
 import numpy as np
 import tensorflow as tf
 from absl.testing import parameterized
-from tensorflow_addons.layers.gelu import GeLU
+from tensorflow_addons.layers.gelu import GELU
 from tensorflow_addons.utils import test_utils
 
 
 @parameterized.parameters([np.float16, np.float32, np.float64])
 @test_utils.run_all_in_graph_and_eager_modes
-class TestGeLU(tf.test.TestCase):
+class TestGELU(tf.test.TestCase):
     def test_random(self, dtype):
         x = np.array([[0.5, 1.2, -0.3]]).astype(dtype)
         val = np.array([[0.345714, 1.0617027, -0.11462909]]).astype(dtype)
         test_utils.layer_test(
-            GeLU, kwargs={'dtype': dtype}, input_data=x, expected_output=val)
+            GELU, kwargs={'dtype': dtype}, input_data=x, expected_output=val)
 
 
 if __name__ == '__main__':