TheTsar1209
commited on
Commit
•
50d4dc0
1
Parent(s):
752edc5
Upload tokenizer
Browse files- special_tokens_map.json +7 -1
- tokenizer_config.json +1 -1
special_tokens_map.json
CHANGED
@@ -13,5 +13,11 @@
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
}
|
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "<|end_of_text|>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
}
|
23 |
}
|
tokenizer_config.json
CHANGED
@@ -2062,7 +2062,7 @@
|
|
2062 |
"clean_up_tokenization_spaces": true,
|
2063 |
"eos_token": "<|im_end|>",
|
2064 |
"model_max_length": 1000000000000000019884624838656,
|
2065 |
-
"pad_token": "<|
|
2066 |
"tokenizer_class": "PreTrainedTokenizerFast",
|
2067 |
"unk_token": null
|
2068 |
}
|
|
|
2062 |
"clean_up_tokenization_spaces": true,
|
2063 |
"eos_token": "<|im_end|>",
|
2064 |
"model_max_length": 1000000000000000019884624838656,
|
2065 |
+
"pad_token": "<|end_of_text|>",
|
2066 |
"tokenizer_class": "PreTrainedTokenizerFast",
|
2067 |
"unk_token": null
|
2068 |
}
|