Upload tokenizer

special_tokens_map.json

@@ -1,5 +1,11 @@
 {
-  "bos_token": "<|user|>",
+  "bos_token": {
+    "content": "<|user|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "cls_token": {
     "content": "<s>",
     "lstrip": false,
@@ -8,9 +14,9 @@
     "single_word": false
   },
   "eos_token": {
-    "content": "</s>",
+    "content": "<|end|>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },

tokenizer.json

@@ -3,6 +3,24 @@
   "truncation": null,
   "padding": null,
   "added_tokens": [
+    {
+      "id": 2,
+      "content": "<|end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 3,
+      "content": "<|user|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
     {
       "id": 8000,
       "content": "<s>",

tokenizer_config.json

@@ -1,6 +1,22 @@
 {
   "add_prefix_space": false,
   "added_tokens_decoder": {
+    "2": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
     "8000": {
       "content": "<s>",
       "lstrip": false,
@@ -45,7 +61,7 @@
   "bos_token": "<|user|>",
   "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",
-  "eos_token": "</s>",
+  "eos_token": "<|end|>",
   "errors": "replace",
   "mask_token": "<mask>",
   "model_max_length": 1000000000000000019884624838656,