Upload tokenizer
Browse files- merges.txt +0 -0
- special_tokens_map.json +20 -0
- tokenizer_config.json +27 -0
- vocab.json +0 -0
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<special0>",
|
| 4 |
+
"<special1>",
|
| 5 |
+
"<special2>",
|
| 6 |
+
"<special3>",
|
| 7 |
+
"<special4>",
|
| 8 |
+
"<special5>",
|
| 9 |
+
"<special6>",
|
| 10 |
+
"<special7>",
|
| 11 |
+
"<special8>",
|
| 12 |
+
"<special9>"
|
| 13 |
+
],
|
| 14 |
+
"bos_token": "<s>",
|
| 15 |
+
"cls_token": "</s>",
|
| 16 |
+
"mask_token": "<special1>",
|
| 17 |
+
"pad_token": "<s>",
|
| 18 |
+
"sep_token": "</s>",
|
| 19 |
+
"unk_token": "<unk>"
|
| 20 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<special0>",
|
| 4 |
+
"<special1>",
|
| 5 |
+
"<special2>",
|
| 6 |
+
"<special3>",
|
| 7 |
+
"<special4>",
|
| 8 |
+
"<special5>",
|
| 9 |
+
"<special6>",
|
| 10 |
+
"<special7>",
|
| 11 |
+
"<special8>",
|
| 12 |
+
"<special9>"
|
| 13 |
+
],
|
| 14 |
+
"bos_token": "<s>",
|
| 15 |
+
"cls_token": "</s>",
|
| 16 |
+
"do_lowercase_and_remove_accent": true,
|
| 17 |
+
"id2lang": null,
|
| 18 |
+
"lang2id": null,
|
| 19 |
+
"mask_token": "<special1>",
|
| 20 |
+
"model_max_length": 512,
|
| 21 |
+
"name_or_path": "xlm-mlm-en-2048",
|
| 22 |
+
"pad_token": "<pad>",
|
| 23 |
+
"sep_token": "</s>",
|
| 24 |
+
"special_tokens_map_file": null,
|
| 25 |
+
"tokenizer_class": "XLMTokenizer",
|
| 26 |
+
"unk_token": "<unk>"
|
| 27 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|