Upload tokenizer

added_tokens.json

@@ -1,3 +1,6 @@
 {
-  "<|startoftext|>": 50257
+  "<|assistant|>": 50259,
+  "<|pad|>": 50257,
+  "<|system|>": 50260,
+  "<|user|>": 50258
 }

special_tokens_map.json

@@ -1,14 +1,21 @@
 {
   "additional_special_tokens": [
     {
-      "content": "<|startoftext|>",
+      "content": "<|user|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false
     },
     {
-      "content": "<|endoftext|>",
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|system|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,11 +32,17 @@
   "eos_token": {
     "content": "<|endoftext|>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<|endoftext|>",
   "unk_token": {
     "content": "<|endoftext|>",
     "lstrip": false,

tokenizer.json

@@ -7,14 +7,12 @@
     "stride": 0
   },
   "padding": {
-    "strategy": {
-      "Fixed": 256
-    },
+    "strategy": "BatchLongest",
     "direction": "Right",
     "pad_to_multiple_of": null,
-    "pad_id": 50256,
+    "pad_id": 50257,
     "pad_type_id": 0,
-    "pad_token": "<|endoftext|>"
+    "pad_token": "<|pad|>"
   },
   "added_tokens": [
     {
@@ -28,7 +26,34 @@
     },
     {
       "id": 50257,
-      "content": "<|startoftext|>",
+      "content": "<|pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 50258,
+      "content": "<|user|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 50259,
+      "content": "<|assistant|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 50260,
+      "content": "<|system|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,

tokenizer_config.json

@@ -11,7 +11,31 @@
       "special": true
     },
     "50257": {
-      "content": "<|startoftext|>",
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50258": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50259": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50260": {
+      "content": "<|system|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -20,8 +44,9 @@
     }
   },
   "additional_special_tokens": [
-    "<|startoftext|>",
-    "<|endoftext|>"
+    "<|user|>",
+    "<|assistant|>",
+    "<|system|>"
   ],
   "bos_token": "<|endoftext|>",
   "clean_up_tokenization_spaces": true,
@@ -29,7 +54,7 @@
   "errors": "replace",
   "extra_special_tokens": {},
   "model_max_length": 1024,
-  "pad_token": "<|endoftext|>",
+  "pad_token": "<|pad|>",
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "<|endoftext|>"
 }