Upload tokenizer
Browse files- tokenizer_config.json +1 -1
tokenizer_config.json
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"bos_token": "<s>",
|
|
|
|
| 3 |
"cls_token": "<s>",
|
| 4 |
"do_lower_case": false,
|
| 5 |
"eos_token": "</s>",
|
|
@@ -16,7 +17,6 @@
|
|
| 16 |
"pad_token": "<pad>",
|
| 17 |
"sep_token": "</s>",
|
| 18 |
"sp_model_kwargs": {},
|
| 19 |
-
"special_tokens_map_file": "/root/.cache/huggingface/hub/models--deepset--xlm-roberta-base-squad2/snapshots/54343fc5b757acc423de447f7b494f0e63e4c2ee/special_tokens_map.json",
|
| 20 |
"tokenizer_class": "XLMRobertaTokenizer",
|
| 21 |
"unk_token": "<unk>"
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"bos_token": "<s>",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
"cls_token": "<s>",
|
| 5 |
"do_lower_case": false,
|
| 6 |
"eos_token": "</s>",
|
|
|
|
| 17 |
"pad_token": "<pad>",
|
| 18 |
"sep_token": "</s>",
|
| 19 |
"sp_model_kwargs": {},
|
|
|
|
| 20 |
"tokenizer_class": "XLMRobertaTokenizer",
|
| 21 |
"unk_token": "<unk>"
|
| 22 |
}
|