Update tokenizer_config.json

- formatted the JSON
- added character_id key/value pairs for 3 special tokens (unk/sep/pad)

tokenizer_config.json

@@ -1 +1,21 @@
-{"name_or_path": "helboukkouri/character-bert", "tokenizer_class": "CharacterBertTokenizer", "max_word_length": 50, "do_lower_case": true, "do_basic_tokenize": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "auto_map": {"AutoTokenizer": ["tokenization_character_bert.CharacterBertTokenizer", null]}}
+{
+    "name_or_path": "helboukkouri/character-bert",
+    "tokenizer_class": "CharacterBertTokenizer",
+    "max_word_length": 50,
+    "do_lower_case": true,
+    "do_basic_tokenize": true,
+    "bow_character_id": 258,
+    "eow_character_id": 259,
+    "pad_character_id": 260,
+    "unk_token": "[UNK]",
+    "sep_token": "[SEP]",
+    "pad_token": "[PAD]",
+    "cls_token": "[CLS]",
+    "mask_token": "[MASK]",
+    "auto_map": {
+        "AutoTokenizer": [
+            "tokenization_character_bert.CharacterBertTokenizer",
+            null
+        ]
+    }
+}