Upload tokenizer
Browse files- added_tokens.json +4 -2
- tokenizer_config.json +10 -10
- vocab.json +64 -66
added_tokens.json
CHANGED
@@ -1,4 +1,6 @@
|
|
1 |
{
|
2 |
-
"</s>":
|
3 |
-
"<s>":
|
|
|
|
|
4 |
}
|
|
|
1 |
{
|
2 |
+
"</s>": 67,
|
3 |
+
"<s>": 66,
|
4 |
+
"[PAD]": 69,
|
5 |
+
"[UNK]": 68
|
6 |
}
|
tokenizer_config.json
CHANGED
@@ -1,23 +1,23 @@
|
|
1 |
{
|
2 |
"added_tokens_decoder": {
|
3 |
"66": {
|
4 |
-
"content": "
|
5 |
-
"lstrip":
|
6 |
"normalized": false,
|
7 |
-
"rstrip":
|
8 |
"single_word": false,
|
9 |
-
"special":
|
10 |
},
|
11 |
"67": {
|
12 |
-
"content": "
|
13 |
-
"lstrip":
|
14 |
"normalized": false,
|
15 |
-
"rstrip":
|
16 |
"single_word": false,
|
17 |
-
"special":
|
18 |
},
|
19 |
"68": {
|
20 |
-
"content": "
|
21 |
"lstrip": false,
|
22 |
"normalized": false,
|
23 |
"rstrip": false,
|
@@ -25,7 +25,7 @@
|
|
25 |
"special": true
|
26 |
},
|
27 |
"69": {
|
28 |
-
"content": "
|
29 |
"lstrip": false,
|
30 |
"normalized": false,
|
31 |
"rstrip": false,
|
|
|
1 |
{
|
2 |
"added_tokens_decoder": {
|
3 |
"66": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
"single_word": false,
|
9 |
+
"special": true
|
10 |
},
|
11 |
"67": {
|
12 |
+
"content": "</s>",
|
13 |
+
"lstrip": false,
|
14 |
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
"single_word": false,
|
17 |
+
"special": true
|
18 |
},
|
19 |
"68": {
|
20 |
+
"content": "[UNK]",
|
21 |
"lstrip": false,
|
22 |
"normalized": false,
|
23 |
"rstrip": false,
|
|
|
25 |
"special": true
|
26 |
},
|
27 |
"69": {
|
28 |
+
"content": "[PAD]",
|
29 |
"lstrip": false,
|
30 |
"normalized": false,
|
31 |
"rstrip": false,
|
vocab.json
CHANGED
@@ -1,70 +1,68 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"
|
7 |
-
"
|
8 |
-
"
|
9 |
-
"a": 46,
|
10 |
"b": 6,
|
11 |
-
"c":
|
12 |
-
"d":
|
13 |
-
"e":
|
14 |
-
"f":
|
15 |
-
"g":
|
16 |
-
"h":
|
17 |
-
"i":
|
18 |
-
"j":
|
19 |
-
"k":
|
20 |
-
"l":
|
21 |
-
"m":
|
22 |
-
"n":
|
23 |
-
"o":
|
24 |
-
"p":
|
25 |
-
"q":
|
26 |
-
"r":
|
27 |
-
"s":
|
28 |
"t": 22,
|
29 |
-
"u":
|
30 |
-
"v":
|
31 |
-
"w":
|
32 |
-
"x":
|
33 |
-
"y":
|
34 |
-
"z":
|
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 |
-
"
|
65 |
-
"
|
66 |
-
"
|
67 |
-
"
|
68 |
-
"
|
69 |
-
"…": 17
|
70 |
}
|
|
|
1 |
{
|
2 |
+
" ": 17,
|
3 |
+
"&": 12,
|
4 |
+
"'": 39,
|
5 |
+
"(": 19,
|
6 |
+
")": 8,
|
7 |
+
"=": 46,
|
8 |
+
"a": 36,
|
|
|
9 |
"b": 6,
|
10 |
+
"c": 16,
|
11 |
+
"d": 50,
|
12 |
+
"e": 41,
|
13 |
+
"f": 44,
|
14 |
+
"g": 10,
|
15 |
+
"h": 4,
|
16 |
+
"i": 15,
|
17 |
+
"j": 2,
|
18 |
+
"k": 11,
|
19 |
+
"l": 28,
|
20 |
+
"m": 25,
|
21 |
+
"n": 55,
|
22 |
+
"o": 48,
|
23 |
+
"p": 21,
|
24 |
+
"q": 5,
|
25 |
+
"r": 49,
|
26 |
+
"s": 29,
|
27 |
"t": 22,
|
28 |
+
"u": 52,
|
29 |
+
"v": 57,
|
30 |
+
"w": 32,
|
31 |
+
"x": 31,
|
32 |
+
"y": 37,
|
33 |
+
"z": 33,
|
34 |
+
"«": 20,
|
35 |
+
"»": 56,
|
36 |
+
"ß": 42,
|
37 |
+
"à": 26,
|
38 |
+
"á": 40,
|
39 |
+
"â": 47,
|
40 |
+
"ç": 27,
|
41 |
+
"è": 3,
|
42 |
+
"é": 23,
|
43 |
+
"ê": 34,
|
44 |
+
"ë": 64,
|
45 |
+
"î": 30,
|
46 |
+
"ï": 60,
|
47 |
+
"ó": 51,
|
48 |
+
"ô": 7,
|
49 |
+
"ö": 54,
|
50 |
+
"ø": 14,
|
51 |
+
"ù": 65,
|
52 |
+
"ú": 62,
|
53 |
+
"û": 61,
|
54 |
+
"ü": 53,
|
55 |
+
"ć": 43,
|
56 |
+
"ē": 63,
|
57 |
+
"ō": 38,
|
58 |
+
"œ": 18,
|
59 |
+
"š": 59,
|
60 |
+
"ū": 13,
|
61 |
+
"̀": 1,
|
62 |
+
"́": 45,
|
63 |
+
"̂": 9,
|
64 |
+
"̧": 58,
|
65 |
+
"—": 24,
|
66 |
+
"’": 35,
|
67 |
+
"…": 0
|
|
|
68 |
}
|