PrajwalS commited on
Commit
1813df9
1 Parent(s): fad7288

add tokenizer

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