Upload tokenizer
Browse files- vocab.json +62 -72
vocab.json
CHANGED
@@ -1,74 +1,64 @@
|
|
1 |
{
|
2 |
-
"!":
|
3 |
-
"\"":
|
4 |
-
"'":
|
5 |
-
"(":
|
6 |
-
")":
|
7 |
-
",":
|
8 |
-
"-":
|
9 |
-
".":
|
10 |
-
":":
|
11 |
-
";":
|
12 |
-
"?":
|
13 |
-
"A":
|
14 |
-
"B":
|
15 |
-
"C":
|
16 |
-
"D":
|
17 |
-
"E":
|
18 |
-
"F":
|
19 |
-
"G":
|
20 |
-
"H":
|
21 |
-
"I":
|
22 |
-
"J":
|
23 |
-
"K":
|
24 |
-
"L":
|
25 |
-
"M":
|
26 |
-
"N":
|
27 |
-
"O":
|
28 |
-
"P":
|
29 |
-
"R":
|
30 |
-
"S":
|
31 |
-
"T":
|
32 |
-
"U":
|
33 |
-
"V":
|
34 |
-
"
|
35 |
-
"[
|
36 |
-
"
|
37 |
-
"
|
38 |
-
"
|
39 |
-
"
|
40 |
-
"
|
41 |
-
"
|
42 |
-
"
|
43 |
-
"
|
44 |
-
"
|
45 |
-
"
|
46 |
-
"
|
47 |
-
"
|
48 |
-
"
|
49 |
-
"
|
50 |
-
"
|
51 |
-
"
|
52 |
-
"
|
53 |
-
"
|
54 |
-
"
|
55 |
-
"
|
56 |
-
"
|
57 |
-
"
|
58 |
-
"
|
59 |
-
"
|
60 |
-
"
|
61 |
-
"
|
62 |
-
"
|
63 |
-
"
|
64 |
-
"Ĵ": 44,
|
65 |
-
"–": 17,
|
66 |
-
"—": 40,
|
67 |
-
"‘": 22,
|
68 |
-
"’": 64,
|
69 |
-
"“": 12,
|
70 |
-
"”": 39,
|
71 |
-
"„": 49,
|
72 |
-
"fi": 9,
|
73 |
-
"fl": 38
|
74 |
}
|
|
|
1 |
{
|
2 |
+
"!": 6,
|
3 |
+
"\"": 3,
|
4 |
+
"'": 54,
|
5 |
+
"(": 26,
|
6 |
+
")": 1,
|
7 |
+
",": 39,
|
8 |
+
"-": 4,
|
9 |
+
".": 51,
|
10 |
+
":": 21,
|
11 |
+
";": 12,
|
12 |
+
"?": 49,
|
13 |
+
"A": 56,
|
14 |
+
"B": 47,
|
15 |
+
"C": 45,
|
16 |
+
"D": 38,
|
17 |
+
"E": 32,
|
18 |
+
"F": 46,
|
19 |
+
"G": 59,
|
20 |
+
"H": 48,
|
21 |
+
"I": 17,
|
22 |
+
"J": 30,
|
23 |
+
"K": 13,
|
24 |
+
"L": 50,
|
25 |
+
"M": 33,
|
26 |
+
"N": 19,
|
27 |
+
"O": 31,
|
28 |
+
"P": 18,
|
29 |
+
"R": 40,
|
30 |
+
"S": 24,
|
31 |
+
"T": 34,
|
32 |
+
"U": 52,
|
33 |
+
"V": 23,
|
34 |
+
"[PAD]": 61,
|
35 |
+
"[UNK]": 60,
|
36 |
+
"a": 53,
|
37 |
+
"b": 29,
|
38 |
+
"c": 37,
|
39 |
+
"d": 27,
|
40 |
+
"e": 42,
|
41 |
+
"f": 8,
|
42 |
+
"g": 57,
|
43 |
+
"h": 25,
|
44 |
+
"i": 35,
|
45 |
+
"j": 20,
|
46 |
+
"k": 36,
|
47 |
+
"l": 44,
|
48 |
+
"m": 22,
|
49 |
+
"n": 0,
|
50 |
+
"o": 14,
|
51 |
+
"p": 9,
|
52 |
+
"q": 5,
|
53 |
+
"r": 16,
|
54 |
+
"s": 28,
|
55 |
+
"t": 43,
|
56 |
+
"u": 2,
|
57 |
+
"v": 7,
|
58 |
+
"w": 15,
|
59 |
+
"y": 11,
|
60 |
+
"z": 10,
|
61 |
+
"|": 41,
|
62 |
+
"Ĵ": 58,
|
63 |
+
"—": 55
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
64 |
}
|