Update tokenizer_config.json

tokenizer_config.json

@@ -25,38 +25,23 @@
       "rstrip": false,
       "single_word": false,
       "special": true
-    },
-    "32000": {
-      "content": "<|end_of_turn|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "32001": {
-      "content": "<|pad_0|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "additional_special_tokens": [
-    "<|end_of_turn|>",
-    "<|pad_0|>"
+    "<unk>",
+    "<s>",
+    "</s>"
   ],
   "bos_token": "<s>",
-  "chat_template": "{{ bos_token }}{% for message in messages %}{{ 'GPT4 Correct ' + message['role'].title() + ': ' + message['content'] + '<|end_of_turn|>'}}{% endfor %}{% if add_generation_prompt %}{{ 'GPT4 Correct Assistant:' }}{% endif %}",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<|end_of_turn|>",
+  "eos_token": "</s>",
   "legacy": true,
-  "model_max_length": 1000000000000000019884624838656,
-  "pad_token": null,
+  "model_max_length": 8192,
+  "pad_token": "<s>",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": "<unk>",
+  "chat_template": "{{ bos_token }}{% for message in messages %}{{ 'GPT4 Correct ' + message['role'].title() + ': ' + message['content'] + eos_token}}{% endfor %}{% if add_generation_prompt %}{{ 'GPT4 Correct Assistant:' }}{% endif %}",
   "use_default_system_prompt": true
 }