Upload tokenizer

special_tokens_map.json

@@ -1,19 +1,9 @@
 {
   "additional_special_tokens": [
-    {
-      "content": "<|im_start|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "<|im_end|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    }
+    "<|im_start|>system",
+    "<|im_start|>user",
+    "<|im_start|>assistant",
+    "<|im_end|>"
   ],
   "bos_token": {
     "content": "<s>",
@@ -23,13 +13,19 @@
     "single_word": false
   },
   "eos_token": {
-    "content": "</s>",
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "</s>",
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer.json

@@ -2,7 +2,7 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 2200,
+    "max_length": 2048,
     "strategy": "LongestFirst",
     "stride": 0
   },
@@ -37,7 +37,7 @@
     },
     {
       "id": 32000,
-      "content": "<|im_start|>",
+      "content": "[PAD]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -46,6 +46,33 @@
     },
     {
       "id": 32001,
+      "content": "<|im_start|>system",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 32002,
+      "content": "<|im_start|>user",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 32003,
+      "content": "<|im_start|>assistant",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 32004,
       "content": "<|im_end|>",
       "single_word": false,
       "lstrip": false,

tokenizer_config.json

@@ -27,7 +27,7 @@
       "special": true
     },
     "32000": {
-      "content": "<|im_start|>",
+      "content": "[PAD]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -35,6 +35,30 @@
       "special": true
     },
     "32001": {
+      "content": "<|im_start|>system",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|im_start|>user",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|im_start|>assistant",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
       "content": "<|im_end|>",
       "lstrip": false,
       "normalized": false,
@@ -44,19 +68,26 @@
     }
   },
   "additional_special_tokens": [
-    "<|im_start|>",
+    "<|im_start|>system",
+    "<|im_start|>user",
+    "<|im_start|>assistant",
     "<|im_end|>"
   ],
   "bos_token": "<s>",
-  "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 %}",
+  "chat_template": "\n    {% for message in messages %}\n    {% if message['role'] == 'user' %}\n    {{'<|im_start|>user\n' + message['content'].strip() + '<|im_end|>' }}\n    {% elif message['role'] == 'system' %}\n    {{'<|im_start|>system\n' + message['content'].strip() + '<|im_end|>' }}\n    {% elif message['role'] == 'assistant' %}\n    {{'<|im_start|>assistant\n'  + message['content'] + '<|im_end|>' }}\n    {% endif %}\n    {% endfor %}\n    ",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "</s>",
+  "eos_token": "<|im_end|>",
   "legacy": true,
+  "max_length": 2048,
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "</s>",
+  "pad_token": "[PAD]",
+  "padding_side": "right",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
+  "stride": 0,
   "tokenizer_class": "LlamaTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
   "unk_token": "<unk>",
   "use_default_system_prompt": false
 }