Upload tokenizer_config.json with huggingface_hub
Browse files- tokenizer_config.json +18 -0
tokenizer_config.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": null,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"cls_token": "<cls>",
|
| 6 |
+
"do_lower_case": false,
|
| 7 |
+
"eos_token": "</s>",
|
| 8 |
+
"extra_ids": 0,
|
| 9 |
+
"keep_accents": true,
|
| 10 |
+
"legacy": true,
|
| 11 |
+
"mask_token": "<mask>",
|
| 12 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 13 |
+
"pad_token": "<pad>",
|
| 14 |
+
"sep_token": "<sep>",
|
| 15 |
+
"sp_model_kwargs": {},
|
| 16 |
+
"tokenizer_class": "T5Tokenizer",
|
| 17 |
+
"unk_token": "<unk>"
|
| 18 |
+
}
|