Upload tokenizer
Browse files- tokenizer_config.json +1 -2
- vocab.json +40 -40
tokenizer_config.json
CHANGED
|
@@ -42,10 +42,9 @@
|
|
| 42 |
}
|
| 43 |
},
|
| 44 |
"bos_token": "<s>",
|
| 45 |
-
"clean_up_tokenization_spaces":
|
| 46 |
"do_lower_case": false,
|
| 47 |
"eos_token": "</s>",
|
| 48 |
-
"extra_special_tokens": {},
|
| 49 |
"model_max_length": 1000000000000000019884624838656,
|
| 50 |
"pad_token": "[PAD]",
|
| 51 |
"replace_word_delimiter_char": " ",
|
|
|
|
| 42 |
}
|
| 43 |
},
|
| 44 |
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": true,
|
| 46 |
"do_lower_case": false,
|
| 47 |
"eos_token": "</s>",
|
|
|
|
| 48 |
"model_max_length": 1000000000000000019884624838656,
|
| 49 |
"pad_token": "[PAD]",
|
| 50 |
"replace_word_delimiter_char": " ",
|
vocab.json
CHANGED
|
@@ -1,47 +1,47 @@
|
|
| 1 |
{
|
| 2 |
-
"'":
|
| 3 |
-
"A":
|
| 4 |
-
"D":
|
| 5 |
-
"[":
|
| 6 |
"[DA]": 44,
|
| 7 |
"[PAD]": 43,
|
| 8 |
"[UNK]": 42,
|
| 9 |
-
"]":
|
| 10 |
-
"a":
|
| 11 |
-
"b":
|
| 12 |
-
"c":
|
| 13 |
-
"d":
|
| 14 |
-
"e":
|
| 15 |
-
"f":
|
| 16 |
-
"g":
|
| 17 |
-
"h":
|
| 18 |
-
"i":
|
| 19 |
-
"j":
|
| 20 |
"k": 3,
|
| 21 |
-
"l":
|
| 22 |
"m": 25,
|
| 23 |
-
"n":
|
| 24 |
-
"o":
|
| 25 |
-
"p":
|
| 26 |
-
"q":
|
| 27 |
-
"r":
|
| 28 |
-
"s":
|
| 29 |
-
"t":
|
| 30 |
-
"u":
|
| 31 |
-
"v":
|
| 32 |
-
"w":
|
| 33 |
-
"x":
|
| 34 |
-
"y":
|
| 35 |
-
"z":
|
| 36 |
-
"|":
|
| 37 |
-
"«":
|
| 38 |
-
"»":
|
| 39 |
-
"å":
|
| 40 |
-
"æ":
|
| 41 |
-
"é":
|
| 42 |
-
"í":
|
| 43 |
-
"ó":
|
| 44 |
-
"ø":
|
| 45 |
-
"–":
|
| 46 |
-
"—":
|
| 47 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"'": 31,
|
| 3 |
+
"A": 20,
|
| 4 |
+
"D": 16,
|
| 5 |
+
"[": 1,
|
| 6 |
"[DA]": 44,
|
| 7 |
"[PAD]": 43,
|
| 8 |
"[UNK]": 42,
|
| 9 |
+
"]": 30,
|
| 10 |
+
"a": 14,
|
| 11 |
+
"b": 22,
|
| 12 |
+
"c": 5,
|
| 13 |
+
"d": 0,
|
| 14 |
+
"e": 32,
|
| 15 |
+
"f": 15,
|
| 16 |
+
"g": 21,
|
| 17 |
+
"h": 10,
|
| 18 |
+
"i": 18,
|
| 19 |
+
"j": 6,
|
| 20 |
"k": 3,
|
| 21 |
+
"l": 17,
|
| 22 |
"m": 25,
|
| 23 |
+
"n": 35,
|
| 24 |
+
"o": 23,
|
| 25 |
+
"p": 11,
|
| 26 |
+
"q": 40,
|
| 27 |
+
"r": 37,
|
| 28 |
+
"s": 9,
|
| 29 |
+
"t": 26,
|
| 30 |
+
"u": 27,
|
| 31 |
+
"v": 4,
|
| 32 |
+
"w": 34,
|
| 33 |
+
"x": 7,
|
| 34 |
+
"y": 28,
|
| 35 |
+
"z": 24,
|
| 36 |
+
"|": 38,
|
| 37 |
+
"«": 19,
|
| 38 |
+
"»": 13,
|
| 39 |
+
"å": 12,
|
| 40 |
+
"æ": 2,
|
| 41 |
+
"é": 36,
|
| 42 |
+
"í": 41,
|
| 43 |
+
"ó": 33,
|
| 44 |
+
"ø": 39,
|
| 45 |
+
"–": 29,
|
| 46 |
+
"—": 8
|
| 47 |
}
|