Upload tokenizer

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "additional_special_tokens": [
+    ">>MIDDLE<<",
+    "<|prompter|>",
+    ">>ABSTRACT<<",
+    ">>DOMAIN<<",
+    "<|prefix_begin|>",
+    "<|assistant|>",
+    ">>SUMMARY<<",
+    "<|prefix_end|>",
+    ">>INTRODUCTION<<",
+    ">>TITLE<<",
+    ">>PREFIX<<",
+    ">>SUFFIX<<",
+    ">>ANSWER<<",
+    "<|system|>",
+    ">>QUESTION<<",
+    ">>COMMENT<<"
+  ],
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>",
+  "sep_token": "<|endoftext|>"
+}

tokenizer_config.json

@@ -0,0 +1,8 @@
+{
+  "add_prefix_space": false,
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 2048,
+  "special_tokens_map_file": null,
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}