Upload tokenizer
Browse files- tokenizer_config.json +4 -0
tokenizer_config.json
CHANGED
|
@@ -34,9 +34,13 @@
|
|
| 34 |
"eos_token": "</s>",
|
| 35 |
"eos_token_id": 2,
|
| 36 |
"legacy": true,
|
|
|
|
| 37 |
"max_model_input_sizes": 2048,
|
| 38 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 39 |
"pad_token": "</s>",
|
|
|
|
|
|
|
| 40 |
"tokenizer_class": "LlamaTokenizer",
|
| 41 |
"unk_token": "<unk>",
|
| 42 |
"unk_token_id": 0,
|
|
|
|
| 34 |
"eos_token": "</s>",
|
| 35 |
"eos_token_id": 2,
|
| 36 |
"legacy": true,
|
| 37 |
+
"max_length": 512,
|
| 38 |
"max_model_input_sizes": 2048,
|
| 39 |
"model_max_length": 1000000000000000019884624838656,
|
| 40 |
+
"pad_to_multiple_of": null,
|
| 41 |
"pad_token": "</s>",
|
| 42 |
+
"pad_token_type_id": 0,
|
| 43 |
+
"padding_side": "right",
|
| 44 |
"tokenizer_class": "LlamaTokenizer",
|
| 45 |
"unk_token": "<unk>",
|
| 46 |
"unk_token_id": 0,
|