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