diff --git a/bindings/python/py_src/tokenizers/__init__.py b/bindings/python/py_src/tokenizers/__init__.py
index dfbf0333..efd57429 100644
--- a/bindings/python/py_src/tokenizers/__init__.py
+++ b/bindings/python/py_src/tokenizers/__init__.py
@@ -89,6 +89,7 @@ from .tokenizers import (
     pre_tokenizers,
     processors,
     trainers,
+    __version__,
 )
 from .implementations import (
     BertWordPieceTokenizer,
diff --git a/bindings/python/src/tokenizer.rs b/bindings/python/src/tokenizer.rs
index a491a036..7462c359 100644
--- a/bindings/python/src/tokenizer.rs
+++ b/bindings/python/src/tokenizer.rs
@@ -75,7 +75,7 @@ impl PyAddedToken {
             single_word: None,
             lstrip: None,
             rstrip: None,
-            normalized: None,
+            normalized: Some(!special.unwrap_or(true)),
         }
     }
 
diff --git a/bindings/python/src/trainers.rs b/bindings/python/src/trainers.rs
index 54fcc8b0..1c1c9310 100644
--- a/bindings/python/src/trainers.rs
+++ b/bindings/python/src/trainers.rs
@@ -226,7 +226,7 @@ impl PyBpeTrainer {
                     if let Ok(content) = token.extract::<String>() {
                         Ok(tk::tokenizer::AddedToken::from(content, true))
                     } else if let Ok(mut token) = token.extract::<PyRefMut<PyAddedToken>>() {
-                        token.special = false;
+                        token.special = true;
                         Ok(token.get_token())
                     } else {
                         Err(exceptions::PyTypeError::new_err(
diff --git a/bindings/python/tests/bindings/test_tokenizer.py b/bindings/python/tests/bindings/test_tokenizer.py
index abfa8f90..5c002046 100644
--- a/bindings/python/tests/bindings/test_tokenizer.py
+++ b/bindings/python/tests/bindings/test_tokenizer.py
@@ -24,7 +24,7 @@ class TestAddedToken:
         assert added_token.special == False
         added_token.special = True
         assert added_token.special == True
-
+        added_token.special = False
         assert str(added_token) == "<mask>"
         assert (
             repr(added_token)
diff --git a/bindings/python/tests/bindings/test_trainers.py b/bindings/python/tests/bindings/test_trainers.py
index 2bb8f014..48b69c26 100644
--- a/bindings/python/tests/bindings/test_trainers.py
+++ b/bindings/python/tests/bindings/test_trainers.py
@@ -34,8 +34,8 @@ class TestBpeTrainer:
         assert trainer.min_frequency == 12
         assert trainer.show_progress == False
         assert trainer.special_tokens == [
-            AddedToken("1"),
-            AddedToken("2"),
+            AddedToken("1", special = True),
+            AddedToken("2", special = True),
         ]
         assert trainer.limit_alphabet == 13
         assert sorted(trainer.initial_alphabet) == ["a", "b", "c"]
@@ -91,8 +91,8 @@ class TestWordPieceTrainer:
         assert trainer.min_frequency == 12
         assert trainer.show_progress == False
         assert trainer.special_tokens == [
-            AddedToken("1"),
-            AddedToken("2"),
+            AddedToken("1", special = True),
+            AddedToken("2", special = True),
         ]
         assert trainer.limit_alphabet == 13
         assert sorted(trainer.initial_alphabet) == ["a", "b", "c"]
@@ -131,8 +131,8 @@ class TestWordLevelTrainer:
         assert trainer.min_frequency == 12
         assert trainer.show_progress == False
         assert trainer.special_tokens == [
-            AddedToken("1"),
-            AddedToken("2"),
+            AddedToken("1", special = True),
+            AddedToken("2", special = True),
         ]
 
         # Modify these
@@ -272,8 +272,8 @@ class TestUnigram:
         assert trainer.vocab_size == 12345
         assert trainer.show_progress == False
         assert trainer.special_tokens == [
-            AddedToken("1", normalized=False),
-            AddedToken("2", lstrip=True, normalized=False),
+            AddedToken("1", normalized=False, special = True),
+            AddedToken("2", lstrip=True, normalized=False, special = True),
         ]
         assert sorted(trainer.initial_alphabet) == ["a", "b", "c"]
 
diff --git a/tokenizers/src/tokenizer/added_vocabulary.rs b/tokenizers/src/tokenizer/added_vocabulary.rs
index 77a72f62..49ced2db 100644
--- a/tokenizers/src/tokenizer/added_vocabulary.rs
+++ b/tokenizers/src/tokenizer/added_vocabulary.rs
@@ -609,7 +609,7 @@ mod tests {
         );
         assert_eq!(vocab.len(), 3);
 
-        assert_eq!(vocab.get_vocab_r()[&0], added_token);
+        assert_eq!(vocab.get_added_tokens_decoder()[&0], added_token);
     }
 
     #[test]
@@ -650,7 +650,7 @@ mod tests {
         assert_eq!(vocab.len(), 3); // New token was added
         assert!(vocab.is_special_token("test"));
         assert_eq!(
-            *vocab.get_vocab_r(),
+            *vocab.get_added_tokens_decoder(),
             HashMap::from([
                 (0, AddedToken::from("test", true)),
                 (2, AddedToken::from("added_token_1", true)),
diff --git a/tokenizers/src/tokenizer/mod.rs b/tokenizers/src/tokenizer/mod.rs
index 97390d91..915fe7bf 100644
--- a/tokenizers/src/tokenizer/mod.rs
+++ b/tokenizers/src/tokenizer/mod.rs
@@ -669,7 +669,7 @@ where
         // TODO ArthurZ THIS IS WRONG! We need to measure the length of the `set` because
         // now some tokens can be both in the added_tokens_encoder and in the vocab
         if with_added_tokens {
-            self.get_vocab(with_added_tokens).len()
+            self.get_vocab(true).len()
         } else {
             self.model.get_vocab_size()
         }