Upload tokenizer

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|assistant|>": 32001,
+  "<|prompter|>": 32002,
+  "[PAD]": 32000
+}

special_tokens_map.json

@@ -1,4 +1,11 @@
 {
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>",
+    "<|assistant|>",
+    "<|prompter|>"
+  ],
   "bos_token": {
     "content": "<s>",
     "lstrip": false,
@@ -13,7 +20,13 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "</s>",
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": true,
+    "single_word": false
+  },
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer_config.json

@@ -25,20 +25,50 @@
       "rstrip": false,
       "single_word": false,
       "special": true
+    },
+    "32000": {
+      "content": "[PAD]",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|prompter|>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
     }
   },
-  "additional_special_tokens": [],
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>",
+    "<|assistant|>",
+    "<|prompter|>"
+  ],
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "legacy": true,
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "</s>",
+  "pad_token": "[PAD]",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "LlamaTokenizer",
   "trust_remote_code": true,
   "unk_token": "<unk>",
-  "use_default_system_prompt": false,
+  "use_default_system_prompt": true,
   "use_fast": true
 }