Upload tokenizer

merges.txt

@@ -49735,10 +49735,3 @@ alra shidiya
 ĠBan que
 ĠJas im
 Wear house
-ĠPriv é
-97150443 0584
-Ġkal eej
-Ġproperty finder
-ĠBri ghton
-Can ary
-imm i

special_tokens_map.json

@@ -1,5 +1,9 @@
 {
-  "bos_token": "<|endoftext|>",
-  "eos_token": "<|endoftext|>",
-  "unk_token": "<|endoftext|>"
+  "bos_token": "<|BOS|>",
+  "cls_token": "<|CLS|>",
+  "eos_token": "<|EOS|>",
+  "mask_token": "<|MASK|>",
+  "pad_token": "<|PAD|>",
+  "sep_token": "<|SEP|>",
+  "unk_token": "<|UNK|>"
 }

tokenizer_config.json

@@ -1,10 +1,14 @@
 {
   "add_prefix_space": false,
-  "bos_token": "<|endoftext|>",
-  "eos_token": "<|endoftext|>",
+  "bos_token": "<|BOS|>",
+  "cls_token": "<|CLS|>",
+  "eos_token": "<|EOS|>",
+  "mask_token": "<|MASK|>",
   "model_max_length": 1024,
   "name_or_path": "gpt2",
+  "pad_token": "<|PAD|>",
+  "sep_token": "<|SEP|>",
   "special_tokens_map_file": null,
   "tokenizer_class": "GPT2Tokenizer",
-  "unk_token": "<|endoftext|>"
+  "unk_token": "<|UNK|>"
 }