feat add pad_token

special_tokens_map.json

@@ -30,5 +30,12 @@
     "normalized": false,
     "rstrip": false,
     "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
   }
 }

tokenizer.json

@@ -18212,7 +18212,7 @@
     },
     {
       "id": 2023,
-      "content": ">>UNUSED_1897<<",
+      "content": "<|pad|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -20280,7 +20280,7 @@
       ">>UNUSED_1894<<": 2020,
       ">>UNUSED_1895<<": 2021,
       ">>UNUSED_1896<<": 2022,
-      ">>UNUSED_1897<<": 2023,
+      "<|pad|>": 2023,
       "!": 2024,
       "\"": 2025,
       "#": 2026,

tokenizer_config.json

@@ -16186,7 +16186,7 @@
       "special": true
     },
     "2023": {
-      "content": ">>UNUSED_1897<<",
+      "content": "<|pad|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -16227,5 +16227,6 @@
     "attention_mask"
   ],
   "model_max_length": 32768,
+  "pad_token": "<|pad|>",
   "tokenizer_class": "PreTrainedTokenizerFast"
 }