Fix tokenizer content

special_tokens_map.json

@@ -1,27 +1,48 @@
 {
   "bos_token": {
-    "content": "<BOS>",
+    "content": "[BOS]",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
   "eos_token": {
-    "content": "<EOS>",
+    "content": "[EOS]",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
   "pad_token": {
-    "content": "<PAD>",
+    "content": "[PAD]",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
   "unk_token": {
-    "content": "<UNK>",
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer.json

@@ -5,7 +5,7 @@
   "added_tokens": [
     {
       "id": 0,
-      "content": "<PAD>",
+      "content": "[PAD]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -14,7 +14,7 @@
     },
     {
       "id": 1,
-      "content": "<UNK>",
+      "content": "[UNK]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -23,7 +23,7 @@
     },
     {
       "id": 2,
-      "content": "<BOS>",
+      "content": "[BOS]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -32,7 +32,7 @@
     },
     {
       "id": 3,
-      "content": "<EOS>",
+      "content": "[EOS]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -41,7 +41,7 @@
     },
     {
       "id": 4,
-      "content": "<INST>",
+      "content": "[CLS]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -50,7 +50,7 @@
     },
     {
       "id": 5,
-      "content": "</INST>",
+      "content": "[SEP]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -59,9 +59,9 @@
     },
     {
       "id": 6,
-      "content": "<MASK>",
+      "content": "[MASK]",
       "single_word": false,
-      "lstrip": false,
+      "lstrip": true,
       "rstrip": false,
       "normalized": false,
       "special": true
@@ -2396,7 +2396,7 @@
         "single": [
           {
             "SpecialToken": {
-              "id": "<BOS>",
+              "id": "[CLS]",
               "type_id": 0
             }
           },
@@ -2405,12 +2405,18 @@
               "id": "A",
               "type_id": 0
             }
+          },
+          {
+            "SpecialToken": {
+              "id": "[SEP]",
+              "type_id": 0
+            }
           }
         ],
         "pair": [
           {
             "SpecialToken": {
-              "id": "<BOS>",
+              "id": "[CLS]",
               "type_id": 0
             }
           },
@@ -2422,25 +2428,85 @@
           },
           {
             "SpecialToken": {
-              "id": "<BOS>",
-              "type_id": 1
+              "id": "[SEP]",
+              "type_id": 0
             }
           },
           {
             "Sequence": {
               "id": "B",
-              "type_id": 1
+              "type_id": 0
+            }
+          },
+          {
+            "SpecialToken": {
+              "id": "[SEP]",
+              "type_id": 0
             }
           }
         ],
         "special_tokens": {
-          "<BOS>": {
-            "id": "<BOS>",
+          "[PAD]": {
+            "id": "[PAD]",
+            "ids": [
+              0
+            ],
+            "tokens": [
+              "[PAD]"
+            ]
+          },
+		      "[UNK]": {
+            "id": "[UNK]",
+            "ids": [
+              1
+            ],
+            "tokens": [
+              "[UNK]"
+            ]
+          },
+		      "[BOS]": {
+            "id": "[BOS]",
             "ids": [
               2
             ],
             "tokens": [
-              "<BOS>"
+              "[BOS]"
+            ]
+          },
+		      "[EOS]": {
+            "id": "[EOS]",
+            "ids": [
+              3
+            ],
+            "tokens": [
+              "[EOS]"
+            ]
+          },
+		      "[CLS]": {
+            "id": "[CLS]",
+            "ids": [
+              4
+            ],
+            "tokens": [
+              "[CLS]"
+            ]
+          },
+		      "[SEP]": {
+            "id": "[SEP]",
+            "ids": [
+              5
+            ],
+            "tokens": [
+              "[SEP]"
+            ]
+          },
+		      "[MASK]": {
+            "id": "[MASK]",
+            "ids": [
+              6
+            ],
+            "tokens": [
+              "[MASK]"
             ]
           }
         }
@@ -2456,20 +2522,20 @@
   "model": {
     "type": "BPE",
     "dropout": null,
-    "unk_token": "<UNK>",
+    "unk_token": "[UNK]",
     "continuing_subword_prefix": null,
     "end_of_word_suffix": null,
     "fuse_unk": false,
     "byte_fallback": false,
     "ignore_merges": true,
     "vocab": {
-      "<PAD>": 0,
-      "<UNK>": 1,
-      "<BOS>": 2,
-      "<EOS>": 3,
-      "<INST>": 4,
-      "</INST>": 5,
-      "<MASK>": 6,
+      "[PAD]": 0,
+      "[UNK]": 1,
+      "[BOS]": 2,
+      "[EOS]": 3,
+      "[CLS]": 4,
+      "[SEP]": 5,
+      "[MASK]": 6,
       "<RESERVED_TOKEN_1>": 7,
       "<RESERVED_TOKEN_2>": 8,
       "<RESERVED_TOKEN_3>": 9,

tokenizer_config.json

@@ -1,7 +1,7 @@
 {
   "added_tokens_decoder": {
     "0": {
-      "content": "<PAD>",
+      "content": "[PAD]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -9,7 +9,7 @@
       "special": true
     },
     "1": {
-      "content": "<UNK>",
+      "content": "[UNK]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -17,7 +17,7 @@
       "special": true
     },
     "2": {
-      "content": "<BOS>",
+      "content": "[BOS]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,7 +25,7 @@
       "special": true
     },
     "3": {
-      "content": "<EOS>",
+      "content": "[EOS]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -33,7 +33,7 @@
       "special": true
     },
     "4": {
-      "content": "<INST>",
+      "content": "[CLS]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -41,7 +41,7 @@
       "special": true
     },
     "5": {
-      "content": "</INST>",
+      "content": "[SEP]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -49,8 +49,8 @@
       "special": true
     },
     "6": {
-      "content": "<MASK>",
-      "lstrip": false,
+      "content": "[MASK]",
+      "lstrip": true,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
@@ -2097,11 +2097,18 @@
       "special": true
     }
   },
-  "bos_token": "<BOS>",
+  "bos_token": "[BOS]",
   "clean_up_tokenization_spaces": true,
-  "eos_token": "<EOS>",
-  "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "<PAD>",
+  "eos_token": "[EOS]",
+  "model_max_length": 8192,
+  "pad_token": "[PAD]",
   "tokenizer_class": "PreTrainedTokenizerFast",
-  "unk_token": "<UNK>"
+  "unk_token": "[UNK]",
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "sep_token": "[SEP]",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ]
 }