Upload tokenizer/tokenizer_config.json with huggingface_hub
Browse files
tokenizer/tokenizer_config.json
CHANGED
|
@@ -19,12 +19,12 @@
|
|
| 19 |
}
|
| 20 |
},
|
| 21 |
"bos_token": "<|startoftext|>",
|
| 22 |
-
"clean_up_tokenization_spaces":
|
| 23 |
"do_lower_case": true,
|
| 24 |
"eos_token": "<|endoftext|>",
|
| 25 |
"errors": "replace",
|
| 26 |
"extra_special_tokens": {},
|
| 27 |
-
"model_max_length":
|
| 28 |
"pad_token": "<|endoftext|>",
|
| 29 |
"tokenizer_class": "CLIPTokenizer",
|
| 30 |
"unk_token": "<|endoftext|>"
|
|
|
|
| 19 |
}
|
| 20 |
},
|
| 21 |
"bos_token": "<|startoftext|>",
|
| 22 |
+
"clean_up_tokenization_spaces": true,
|
| 23 |
"do_lower_case": true,
|
| 24 |
"eos_token": "<|endoftext|>",
|
| 25 |
"errors": "replace",
|
| 26 |
"extra_special_tokens": {},
|
| 27 |
+
"model_max_length": 248,
|
| 28 |
"pad_token": "<|endoftext|>",
|
| 29 |
"tokenizer_class": "CLIPTokenizer",
|
| 30 |
"unk_token": "<|endoftext|>"
|