macarious commited on
Commit
5f18170
1 Parent(s): a512eb6

Upload tokenizer

Browse files
Files changed (1) hide show
  1. vocab.json +32 -30
vocab.json CHANGED
@@ -1,32 +1,34 @@
1
  {
2
- "'": 0,
3
- "[PAD]": 29,
4
- "[UNK]": 28,
5
- "a": 1,
6
- "b": 2,
7
- "c": 3,
8
- "d": 4,
9
- "e": 5,
10
- "f": 6,
11
- "g": 7,
12
- "h": 8,
13
- "i": 9,
14
- "j": 10,
15
- "k": 11,
16
- "l": 12,
17
- "m": 13,
18
- "n": 14,
19
- "o": 15,
20
- "p": 16,
21
- "q": 17,
22
- "r": 18,
23
- "s": 19,
24
- "t": 20,
25
- "u": 21,
26
- "v": 22,
27
- "w": 23,
28
- "x": 24,
29
- "y": 25,
30
- "z": 26,
31
- "|": 27
 
 
32
  }
 
1
  {
2
+ "'": 4,
3
+ "</s>": 2,
4
+ "<s>": 1,
5
+ "[PAD]": 0,
6
+ "[UNK]": 3,
7
+ "a": 5,
8
+ "b": 6,
9
+ "c": 7,
10
+ "d": 8,
11
+ "e": 9,
12
+ "f": 10,
13
+ "g": 11,
14
+ "h": 12,
15
+ "i": 13,
16
+ "j": 14,
17
+ "k": 15,
18
+ "l": 16,
19
+ "m": 17,
20
+ "n": 18,
21
+ "o": 19,
22
+ "p": 20,
23
+ "q": 21,
24
+ "r": 22,
25
+ "s": 23,
26
+ "t": 24,
27
+ "u": 25,
28
+ "v": 26,
29
+ "w": 27,
30
+ "x": 28,
31
+ "y": 29,
32
+ "z": 30,
33
+ "|": 31
34
  }