patrickvonplaten
commited on
Commit
•
cddd1ca
1
Parent(s):
a6eb496
Upload tokenizer
Browse files- tokenizer_config.json +1 -1
- vocab.json +59 -59
tokenizer_config.json
CHANGED
@@ -2,7 +2,7 @@
|
|
2 |
"bos_token": "<s>",
|
3 |
"do_lower_case": false,
|
4 |
"eos_token": "</s>",
|
5 |
-
"name_or_path": "/tmp/
|
6 |
"pad_token": "<pad>",
|
7 |
"replace_word_delimiter_char": " ",
|
8 |
"special_tokens_map_file": null,
|
|
|
2 |
"bos_token": "<s>",
|
3 |
"do_lower_case": false,
|
4 |
"eos_token": "</s>",
|
5 |
+
"name_or_path": "/tmp/tmpg6eq0vrc",
|
6 |
"pad_token": "<pad>",
|
7 |
"replace_word_delimiter_char": " ",
|
8 |
"special_tokens_map_file": null,
|
vocab.json
CHANGED
@@ -1,65 +1,65 @@
|
|
1 |
{
|
2 |
-
"\"":
|
3 |
-
"%":
|
4 |
-
"'":
|
5 |
-
",":
|
6 |
-
"-":
|
7 |
-
".":
|
8 |
-
":":
|
9 |
-
";":
|
10 |
"</s>": 2,
|
11 |
"<pad>": 0,
|
12 |
"<s>": 1,
|
13 |
"<unk>": 3,
|
14 |
-
"?":
|
15 |
-
"A":
|
16 |
-
"B":
|
17 |
-
"C":
|
18 |
-
"D":
|
19 |
-
"E":
|
20 |
-
"F":
|
21 |
-
"G":
|
22 |
-
"H":
|
23 |
-
"I":
|
24 |
-
"J":
|
25 |
-
"K":
|
26 |
-
"L":
|
27 |
-
"M":
|
28 |
-
"N":
|
29 |
-
"O":
|
30 |
-
"P":
|
31 |
-
"R":
|
32 |
-
"S":
|
33 |
-
"T":
|
34 |
-
"U":
|
35 |
-
"V":
|
36 |
-
"W":
|
37 |
-
"Y":
|
38 |
-
"a":
|
39 |
-
"b":
|
40 |
-
"c":
|
41 |
-
"d":
|
42 |
-
"e":
|
43 |
-
"f":
|
44 |
-
"g":
|
45 |
-
"h":
|
46 |
-
"i":
|
47 |
-
"j":
|
48 |
-
"k":
|
49 |
-
"l":
|
50 |
-
"m":
|
51 |
-
"n":
|
52 |
-
"o":
|
53 |
-
"p":
|
54 |
-
"q":
|
55 |
-
"r":
|
56 |
-
"s":
|
57 |
-
"t":
|
58 |
-
"u":
|
59 |
-
"v":
|
60 |
-
"w":
|
61 |
-
"x":
|
62 |
-
"y":
|
63 |
-
"z":
|
64 |
-
"|":
|
65 |
}
|
|
|
1 |
{
|
2 |
+
"\"": 33,
|
3 |
+
"%": 22,
|
4 |
+
"'": 11,
|
5 |
+
",": 15,
|
6 |
+
"-": 14,
|
7 |
+
".": 62,
|
8 |
+
":": 60,
|
9 |
+
";": 20,
|
10 |
"</s>": 2,
|
11 |
"<pad>": 0,
|
12 |
"<s>": 1,
|
13 |
"<unk>": 3,
|
14 |
+
"?": 37,
|
15 |
+
"A": 18,
|
16 |
+
"B": 36,
|
17 |
+
"C": 23,
|
18 |
+
"D": 27,
|
19 |
+
"E": 41,
|
20 |
+
"F": 40,
|
21 |
+
"G": 51,
|
22 |
+
"H": 10,
|
23 |
+
"I": 16,
|
24 |
+
"J": 53,
|
25 |
+
"K": 58,
|
26 |
+
"L": 4,
|
27 |
+
"M": 7,
|
28 |
+
"N": 44,
|
29 |
+
"O": 5,
|
30 |
+
"P": 59,
|
31 |
+
"R": 54,
|
32 |
+
"S": 34,
|
33 |
+
"T": 57,
|
34 |
+
"U": 47,
|
35 |
+
"V": 24,
|
36 |
+
"W": 17,
|
37 |
+
"Y": 32,
|
38 |
+
"a": 8,
|
39 |
+
"b": 26,
|
40 |
+
"c": 31,
|
41 |
+
"d": 38,
|
42 |
+
"e": 50,
|
43 |
+
"f": 61,
|
44 |
+
"g": 39,
|
45 |
+
"h": 28,
|
46 |
+
"i": 19,
|
47 |
+
"j": 43,
|
48 |
+
"k": 48,
|
49 |
+
"l": 49,
|
50 |
+
"m": 13,
|
51 |
+
"n": 46,
|
52 |
+
"o": 25,
|
53 |
+
"p": 55,
|
54 |
+
"q": 30,
|
55 |
+
"r": 9,
|
56 |
+
"s": 21,
|
57 |
+
"t": 29,
|
58 |
+
"u": 42,
|
59 |
+
"v": 45,
|
60 |
+
"w": 52,
|
61 |
+
"x": 12,
|
62 |
+
"y": 56,
|
63 |
+
"z": 35,
|
64 |
+
"|": 6
|
65 |
}
|