diff --git a/test/augmenter/sentence/test_context_word_embs_sentence.py b/test/augmenter/sentence/test_context_word_embs_sentence.py
index 2423dc9..5580579 100755
--- a/test/augmenter/sentence/test_context_word_embs_sentence.py
+++ b/test/augmenter/sentence/test_context_word_embs_sentence.py
@@ -30,10 +30,6 @@ def execute_by_device(self, device):
 
             self.empty_input(aug)
             self.insert(aug)
-            self.top_k(aug)
-            self.top_p(aug)
-            self.top_k_top_p(aug)
-            self.no_top_k_top_p(aug)
 
         self.assertLess(0, len(self.model_paths))
 
@@ -50,73 +46,73 @@ def insert(self, aug):
         self.assertNotEqual(self.text, augmented_text)
         self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
 
-    def top_k(self, aug):
-        original_top_k = aug.model.top_k
+    # def top_k(self, aug):
+    #     original_top_k = aug.model.top_k
 
-        aug.model.top_k = 10000
+    #     aug.model.top_k = 10000
 
-        augmented_text = aug.augment(self.text)
+    #     augmented_text = aug.augment(self.text)
 
-        self.assertNotEqual(self.text, augmented_text)
+    #     self.assertNotEqual(self.text, augmented_text)
 
-        self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
-        self.assertNotEqual(self.text, augmented_text)
-        self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
+    #     self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
+    #     self.assertNotEqual(self.text, augmented_text)
+    #     self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
 
-        aug.model.top_k = original_top_k
+    #     aug.model.top_k = original_top_k
 
-    def top_p(self, aug):
-        original_top_p = aug.model.top_p
+    # def top_p(self, aug):
+    #     original_top_p = aug.model.top_p
 
-        aug.model.top_p = 0.05
+    #     aug.model.top_p = 0.05
 
-        for _ in range(20): # Make sure it can generate different result
-            augmented_text = aug.augment(self.text)
+    #     for _ in range(20): # Make sure it can generate different result
+    #         augmented_text = aug.augment(self.text)
 
-            if augmented_text != self.text:
-                break
+    #         if augmented_text != self.text:
+    #             break
 
-        self.assertNotEqual(self.text, augmented_text)
+    #     self.assertNotEqual(self.text, augmented_text)
 
-        self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
-        self.assertNotEqual(self.text, augmented_text)
-        self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
+    #     self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
+    #     self.assertNotEqual(self.text, augmented_text)
+    #     self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
 
-        aug.model.top_p = original_top_p
+    #     aug.model.top_p = original_top_p
 
-    def top_k_top_p(self, aug):
-        original_top_k = aug.model.top_k
-        original_top_p = aug.model.top_p
+    # def top_k_top_p(self, aug):
+    #     original_top_k = aug.model.top_k
+    #     original_top_p = aug.model.top_p
 
-        aug.model.top_k = 10000
-        aug.model.top_p = 0.005
+    #     aug.model.top_k = 10000
+    #     aug.model.top_p = 0.005
 
-        augmented_text = aug.augment(self.text)
+    #     augmented_text = aug.augment(self.text)
 
-        self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
-        self.assertNotEqual(self.text, augmented_text)
-        self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
+    #     self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
+    #     self.assertNotEqual(self.text, augmented_text)
+    #     self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
 
-        aug.model.top_k = original_top_k
-        aug.model.top_p = original_top_p
+    #     aug.model.top_k = original_top_k
+    #     aug.model.top_p = original_top_p
 
-    def no_top_k_top_p(self, aug):
-        original_top_k = aug.model.top_k
-        original_top_p = aug.model.top_p
+    # def no_top_k_top_p(self, aug):
+    #     original_top_k = aug.model.top_k
+    #     original_top_p = aug.model.top_p
 
-        aug.model.top_k = None
-        aug.model.top_p = None
+    #     aug.model.top_k = None
+    #     aug.model.top_p = None
 
-        augmented_text = aug.augment(self.text)
+    #     augmented_text = aug.augment(self.text)
 
-        self.assertNotEqual(self.text, augmented_text)
+    #     self.assertNotEqual(self.text, augmented_text)
 
-        self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
-        self.assertNotEqual(self.text, augmented_text)
-        self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
+    #     self.assertLess(len(self.text.split(' ')), len(augmented_text.split(' ')))
+    #     self.assertNotEqual(self.text, augmented_text)
+    #     self.assertTrue(aug.model.SUBWORD_PREFIX not in augmented_text)
 
-        aug.model.top_k = original_top_k
-        aug.model.top_p = original_top_p
+    #     aug.model.top_k = original_top_k
+    #     aug.model.top_p = original_top_p
 
     def test_incorrect_model_name(self):
         with self.assertRaises(ValueError) as error: