Update tokenizer_config.json

tokenizer_config.json

@@ -3,15 +3,14 @@
   "add_prefix_space": true,
   "added_tokens_decoder": {
     "50256": {
-      "content": "<|user|>",
-      "lstrip": false,
+      "content": "<|endoftext|>",
+      "lstrip": true,
       "normalized": false,
-      "rstrip": false,
-      "single_word": false,
+      "rstrip": true,
       "special": true
     },
     "50257": {
-      "content": "<|assistant|>",
+      "content": "<|im_end|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -19,21 +18,21 @@
       "special": true
     },
     "50258": {
-      "content": "<|system|>",
-      "lstrip": false,
+      "content": "<|im_start|>",
+      "lstrip": true,
       "normalized": false,
-      "rstrip": false,
+      "rstrip": true,
       "single_word": false,
       "special": true
     }
   },
-  "bos_token": "<|startoftext|>",
-  "chat_template": "{% for message in messages %}{{ message['role'] + '\\n' + message['content'] + '\\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\\n' }}{% endif %}",
+  "bos_token": "<|endoftext|>",
+  "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
   "clean_up_tokenization_spaces": true,
-  "eos_token": "<|endoftext|>",
+  "eos_token": "<|im_end|>",
   "errors": "replace",
-  "model_max_length": 2048,
-  "pad_token": "<|pad|>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|endoftext|>",
   "tokenizer_class": "GPT2Tokenizer",
-  "unk_token": "<|unknown|>"
+  "unk_token": "<|endoftext|>"
 }