Upload tokenizer
Browse files- special_tokens_map.json +7 -1
- tokenizer_config.json +6 -1
special_tokens_map.json
CHANGED
|
@@ -1 +1,7 @@
|
|
| 1 |
-
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|bos|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"mask_token": "<|mask|>",
|
| 5 |
+
"pad_token": "<|pad|>",
|
| 6 |
+
"unk_token": "<|unk|>"
|
| 7 |
+
}
|
tokenizer_config.json
CHANGED
|
@@ -41,7 +41,12 @@
|
|
| 41 |
"special": true
|
| 42 |
}
|
| 43 |
},
|
|
|
|
| 44 |
"clean_up_tokenization_spaces": false,
|
|
|
|
|
|
|
| 45 |
"model_max_length": 1000000000000000019884624838656,
|
| 46 |
-
"
|
|
|
|
|
|
|
| 47 |
}
|
|
|
|
| 41 |
"special": true
|
| 42 |
}
|
| 43 |
},
|
| 44 |
+
"bos_token": "<|bos|>",
|
| 45 |
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"eos_token": "<|endoftext|>",
|
| 47 |
+
"mask_token": "<|mask|>",
|
| 48 |
"model_max_length": 1000000000000000019884624838656,
|
| 49 |
+
"pad_token": "<|pad|>",
|
| 50 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 51 |
+
"unk_token": "<|unk|>"
|
| 52 |
}
|