update tigerbot-7b-base-v3-tokenizer

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "\n\n### Instruction:\n": 60512,
+  "\n\n### Response:\n": 60513,
+  "<pad>": 60514
+}

special_tokens_map.json

@@ -1,4 +1,8 @@
 {
+  "additional_special_tokens": [
+    "\n\n### Instruction:\n",
+    "\n\n### Response:\n"
+  ],
   "bos_token": {
     "content": "<s>",
     "lstrip": false,
@@ -13,6 +17,7 @@
     "rstrip": false,
     "single_word": false
   },
+  "pad_token": "<pad>",
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer.json

@@ -29,6 +29,33 @@
       "rstrip": false,
       "normalized": true,
       "special": true
+    },
+    {
+      "id": 60512,
+      "content": "\n\n### Instruction:\n",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 60513,
+      "content": "\n\n### Response:\n",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 60514,
+      "content": "<pad>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
     }
   ],
   "normalizer": {
@@ -51,12 +78,6 @@
   "post_processor": {
     "type": "TemplateProcessing",
     "single": [
-      {
-        "SpecialToken": {
-          "id": "<s>",
-          "type_id": 0
-        }
-      },
       {
         "Sequence": {
           "id": "A",
@@ -65,24 +86,12 @@
       }
     ],
     "pair": [
-      {
-        "SpecialToken": {
-          "id": "<s>",
-          "type_id": 0
-        }
-      },
       {
         "Sequence": {
           "id": "A",
           "type_id": 0
         }
       },
-      {
-        "SpecialToken": {
-          "id": "<s>",
-          "type_id": 1
-        }
-      },
       {
         "Sequence": {
           "id": "B",
@@ -90,17 +99,7 @@
         }
       }
     ],
-    "special_tokens": {
-      "<s>": {
-        "id": "<s>",
-        "ids": [
-          1
-        ],
-        "tokens": [
-          "<s>"
-        ]
-      }
-    }
+    "special_tokens": {}
   },
   "decoder": {
     "type": "Sequence",

tokenizer_config.json

@@ -1,6 +1,4 @@
 {
-  "add_bos_token": true,
-  "add_eos_token": false,
   "bos_token": {
     "__type": "AddedToken",
     "content": "<s>",
@@ -30,5 +28,6 @@
     "normalized": true,
     "rstrip": false,
     "single_word": false
-  }
+  },
+  "use_fast": true
 }