Upload tokenizer
Browse files- tokenizer_config.json +7 -0
tokenizer_config.json
CHANGED
|
@@ -104,9 +104,16 @@
|
|
| 104 |
"clean_up_tokenization_spaces": true,
|
| 105 |
"eos_token": "</s>",
|
| 106 |
"extra_ids": 100,
|
|
|
|
| 107 |
"model_max_length": 512,
|
|
|
|
| 108 |
"pad_token": "<pad>",
|
|
|
|
|
|
|
| 109 |
"sp_model_kwargs": {},
|
|
|
|
| 110 |
"tokenizer_class": "T5Tokenizer",
|
|
|
|
|
|
|
| 111 |
"unk_token": "<unk>"
|
| 112 |
}
|
|
|
|
| 104 |
"clean_up_tokenization_spaces": true,
|
| 105 |
"eos_token": "</s>",
|
| 106 |
"extra_ids": 100,
|
| 107 |
+
"max_length": 480,
|
| 108 |
"model_max_length": 512,
|
| 109 |
+
"pad_to_multiple_of": null,
|
| 110 |
"pad_token": "<pad>",
|
| 111 |
+
"pad_token_type_id": 0,
|
| 112 |
+
"padding_side": "right",
|
| 113 |
"sp_model_kwargs": {},
|
| 114 |
+
"stride": 0,
|
| 115 |
"tokenizer_class": "T5Tokenizer",
|
| 116 |
+
"truncation_side": "right",
|
| 117 |
+
"truncation_strategy": "longest_first",
|
| 118 |
"unk_token": "<unk>"
|
| 119 |
}
|