update tokenizer

tokenizer_config.json

@@ -27,12 +27,20 @@
     "3": {
       "content": "<unk>",
       "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "[UNK]",
+      "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "250001": {
+    "30526": {
       "content": "<mask>",
       "lstrip": true,
       "normalized": false,
@@ -44,18 +52,21 @@
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",
+  "do_lower_case": true,
   "eos_token": "</s>",
   "mask_token": "<mask>",
-  "max_length": 128,
-  "model_max_length": 128,
+  "max_length": 384,
+  "model_max_length": 384,
   "pad_to_multiple_of": null,
   "pad_token": "<pad>",
   "pad_token_type_id": 0,
   "padding_side": "right",
   "sep_token": "</s>",
   "stride": 0,
-  "tokenizer_class": "XLMRobertaTokenizer",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "MPNetTokenizer",
   "truncation_side": "right",
   "truncation_strategy": "longest_first",
-  "unk_token": "<unk>"
+  "unk_token": "[UNK]"
 }