Upload tokenizer

special_tokens_map.json

@@ -1,23 +1,23 @@
 {
   "additional_special_tokens": [
-    "<|im_start|>user",
-    "<|im_start|>assistant",
-    "<|im_start|>system",
-    "<|im_start|>function-call",
-    "<|im_start|>function-response",
-    "<|im_end|>",
-    "<s>",
-    "<pad>"
+    "<start_of_turn>",
+    "<end_of_turn>"
   ],
   "bos_token": {
-    "content": "<s>",
+    "content": "<bos>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
     "single_word": false
   },
-  "eos_token": "<|im_end|>",
-  "pad_token": "<|im_end|>",
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1edfa481a0c0013cddd64b0d40d4e91e31073b043c4bd1f6db0ef6df5eeb9f8
-size 17519695
+oid sha256:cd1df914b205b134e52f15364524a919423b48c9056d725b45c5b89187541de0
+size 17518965

tokenizer_config.json

@@ -1,6 +1,6 @@
 {
   "add_bos_token": true,
-  "add_eos_token": false,
+  "add_eos_token": true,
   "added_tokens_decoder": {
     "0": {
       "content": "<pad>",
@@ -1640,7 +1640,7 @@
       "normalized": false,
       "rstrip": false,
       "single_word": false,
-      "special": true
+      "special": false
     },
     "205": {
       "content": "<sub>",
@@ -1737,72 +1737,17 @@
       "rstrip": false,
       "single_word": false,
       "special": false
-    },
-    "256000": {
-      "content": "<|im_end|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "256001": {
-      "content": "<|im_start|>user",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "256002": {
-      "content": "<|im_start|>assistant",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "256003": {
-      "content": "<|im_start|>system",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "256004": {
-      "content": "<|im_start|>function-call",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "256005": {
-      "content": "<|im_start|>function-response",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "additional_special_tokens": [
-    "<|im_start|>user",
-    "<|im_start|>assistant",
-    "<|im_start|>system",
-    "<|im_start|>function-call",
-    "<|im_start|>function-response",
-    "<|im_end|>",
-    "<s>",
-    "<pad>"
+    "<start_of_turn>",
+    "<end_of_turn>"
   ],
-  "bos_token": "<s>",
-  "chat_template": "{% for message in messages %}\n{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% if loop.last and add_generation_prompt %}{{'<|im_start|>assistant\n' }}{% endif %}{% endfor %}",
+  "bos_token": "<bos>",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<|im_end|>",
+  "eos_token": "<eos>",
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "<|im_end|>",
+  "pad_token": "<eos>",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "GemmaTokenizer",