Upload tokenizer.json
Browse files- tokenizer.json +231 -0
tokenizer.json
ADDED
@@ -0,0 +1,231 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"version": "1.0",
|
3 |
+
"truncation": null,
|
4 |
+
"padding": null,
|
5 |
+
"added_tokens": [
|
6 |
+
{
|
7 |
+
"id": 0,
|
8 |
+
"content": "<s>",
|
9 |
+
"single_word": false,
|
10 |
+
"lstrip": false,
|
11 |
+
"rstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"special": true
|
14 |
+
},
|
15 |
+
{
|
16 |
+
"id": 1,
|
17 |
+
"content": "<pad>",
|
18 |
+
"single_word": false,
|
19 |
+
"lstrip": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"normalized": false,
|
22 |
+
"special": true
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"id": 2,
|
26 |
+
"content": "</s>",
|
27 |
+
"single_word": false,
|
28 |
+
"lstrip": false,
|
29 |
+
"rstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"special": true
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"id": 3,
|
35 |
+
"content": "<unk>",
|
36 |
+
"single_word": false,
|
37 |
+
"lstrip": false,
|
38 |
+
"rstrip": false,
|
39 |
+
"normalized": false,
|
40 |
+
"special": true
|
41 |
+
},
|
42 |
+
{
|
43 |
+
"id": 79,
|
44 |
+
"content": "<mask>",
|
45 |
+
"single_word": false,
|
46 |
+
"lstrip": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"id": 80,
|
53 |
+
"content": "<ctc_blank>",
|
54 |
+
"single_word": false,
|
55 |
+
"lstrip": false,
|
56 |
+
"rstrip": false,
|
57 |
+
"normalized": false,
|
58 |
+
"special": true
|
59 |
+
}
|
60 |
+
],
|
61 |
+
"normalizer": {
|
62 |
+
"type": "Precompiled",
|
63 |
+
"precompiled_charsmap": null
|
64 |
+
},
|
65 |
+
"pre_tokenizer": {
|
66 |
+
"type": "Sequence",
|
67 |
+
"pretokenizers": [
|
68 |
+
{
|
69 |
+
"type": "WhitespaceSplit"
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"type": "Metaspace",
|
73 |
+
"replacement": "\u2581",
|
74 |
+
"add_prefix_space": true
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"type": "Split",
|
78 |
+
"pattern": {
|
79 |
+
"Regex": ""
|
80 |
+
},
|
81 |
+
"behavior": "Isolated",
|
82 |
+
"invert": false
|
83 |
+
}
|
84 |
+
]
|
85 |
+
},
|
86 |
+
"post_processor": {
|
87 |
+
"type": "TemplateProcessing",
|
88 |
+
"single": [
|
89 |
+
{
|
90 |
+
"Sequence": {
|
91 |
+
"id": "A",
|
92 |
+
"type_id": 0
|
93 |
+
}
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"SpecialToken": {
|
97 |
+
"id": "</s>",
|
98 |
+
"type_id": 0
|
99 |
+
}
|
100 |
+
}
|
101 |
+
],
|
102 |
+
"pair": [
|
103 |
+
{
|
104 |
+
"Sequence": {
|
105 |
+
"id": "A",
|
106 |
+
"type_id": 0
|
107 |
+
}
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"SpecialToken": {
|
111 |
+
"id": "</s>",
|
112 |
+
"type_id": 0
|
113 |
+
}
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"Sequence": {
|
117 |
+
"id": "B",
|
118 |
+
"type_id": 0
|
119 |
+
}
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"SpecialToken": {
|
123 |
+
"id": "</s>",
|
124 |
+
"type_id": 0
|
125 |
+
}
|
126 |
+
}
|
127 |
+
],
|
128 |
+
"special_tokens": {
|
129 |
+
"</s>": {
|
130 |
+
"id": "</s>",
|
131 |
+
"ids": [
|
132 |
+
2
|
133 |
+
],
|
134 |
+
"tokens": [
|
135 |
+
"</s>"
|
136 |
+
]
|
137 |
+
}
|
138 |
+
}
|
139 |
+
},
|
140 |
+
"decoder": {
|
141 |
+
"type": "Metaspace",
|
142 |
+
"replacement": "\u2581",
|
143 |
+
"add_prefix_space": true
|
144 |
+
},
|
145 |
+
"model": {
|
146 |
+
"unk_id": 2,
|
147 |
+
"vocab": {
|
148 |
+
"<s>": 0,
|
149 |
+
"<pad>": 1,
|
150 |
+
"</s>": 2,
|
151 |
+
"<unk>": 3,
|
152 |
+
"\u2581": 4,
|
153 |
+
"e": 5,
|
154 |
+
"t": 6,
|
155 |
+
"a": 7,
|
156 |
+
"o": 8,
|
157 |
+
"n": 9,
|
158 |
+
"i": 10,
|
159 |
+
"h": 11,
|
160 |
+
"s": 12,
|
161 |
+
"r": 13,
|
162 |
+
"d": 14,
|
163 |
+
"l": 15,
|
164 |
+
"u": 16,
|
165 |
+
"c": 17,
|
166 |
+
"m": 18,
|
167 |
+
"f": 19,
|
168 |
+
"w": 20,
|
169 |
+
"g": 21,
|
170 |
+
"y": 22,
|
171 |
+
",": 23,
|
172 |
+
"p": 24,
|
173 |
+
"b": 25,
|
174 |
+
".": 26,
|
175 |
+
"v": 27,
|
176 |
+
"k": 28,
|
177 |
+
"\"": 29,
|
178 |
+
"I": 30,
|
179 |
+
"'": 31,
|
180 |
+
"T": 32,
|
181 |
+
"A": 33,
|
182 |
+
"S": 34,
|
183 |
+
"H": 35,
|
184 |
+
";": 36,
|
185 |
+
"x": 37,
|
186 |
+
"W": 38,
|
187 |
+
"-": 39,
|
188 |
+
"B": 40,
|
189 |
+
"?": 41,
|
190 |
+
"C": 42,
|
191 |
+
"M": 43,
|
192 |
+
"!": 44,
|
193 |
+
"q": 45,
|
194 |
+
"j": 46,
|
195 |
+
"E": 47,
|
196 |
+
"N": 48,
|
197 |
+
"P": 49,
|
198 |
+
"O": 50,
|
199 |
+
"D": 51,
|
200 |
+
"L": 52,
|
201 |
+
"G": 53,
|
202 |
+
"R": 54,
|
203 |
+
"F": 55,
|
204 |
+
"Y": 56,
|
205 |
+
"z": 57,
|
206 |
+
"J": 58,
|
207 |
+
":": 59,
|
208 |
+
"K": 60,
|
209 |
+
"U": 61,
|
210 |
+
"V": 62,
|
211 |
+
")": 63,
|
212 |
+
"(": 64,
|
213 |
+
"Q": 65,
|
214 |
+
"Z": 66,
|
215 |
+
"]": 67,
|
216 |
+
"[": 68,
|
217 |
+
"X": 69,
|
218 |
+
"\u2014": 70,
|
219 |
+
"/": 71,
|
220 |
+
"\u00e6": 72,
|
221 |
+
"\u00e9": 73,
|
222 |
+
"{": 74,
|
223 |
+
"}": 75,
|
224 |
+
"\u00ea": 76,
|
225 |
+
"\u0153": 77,
|
226 |
+
"\u0304": 78,
|
227 |
+
"<mask>": 79,
|
228 |
+
"<ctc_blank>": 80
|
229 |
+
}
|
230 |
+
}
|
231 |
+
}
|