Delete tokenizer.json
Browse files- tokenizer.json +0 -144
tokenizer.json
DELETED
|
@@ -1,144 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"version": "1.0",
|
| 3 |
-
"truncation": null,
|
| 4 |
-
"padding": null,
|
| 5 |
-
"added_tokens": [
|
| 6 |
-
{
|
| 7 |
-
"id": 0,
|
| 8 |
-
"content": "<unk>",
|
| 9 |
-
"single_word": false,
|
| 10 |
-
"lstrip": false,
|
| 11 |
-
"rstrip": false,
|
| 12 |
-
"normalized": false,
|
| 13 |
-
"special": true
|
| 14 |
-
},
|
| 15 |
-
{
|
| 16 |
-
"id": 1,
|
| 17 |
-
"content": "<s>",
|
| 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 |
-
"normalizer": {
|
| 35 |
-
"type": "Sequence",
|
| 36 |
-
"normalizers": [
|
| 37 |
-
{
|
| 38 |
-
"type": "Prepend",
|
| 39 |
-
"prepend": "▁"
|
| 40 |
-
},
|
| 41 |
-
{
|
| 42 |
-
"type": "Replace",
|
| 43 |
-
"pattern": {
|
| 44 |
-
"String": " "
|
| 45 |
-
},
|
| 46 |
-
"content": "▁"
|
| 47 |
-
}
|
| 48 |
-
]
|
| 49 |
-
},
|
| 50 |
-
"pre_tokenizer": null,
|
| 51 |
-
"post_processor": {
|
| 52 |
-
"type": "TemplateProcessing",
|
| 53 |
-
"single": [
|
| 54 |
-
{
|
| 55 |
-
"SpecialToken": {
|
| 56 |
-
"id": "<s>",
|
| 57 |
-
"type_id": 0
|
| 58 |
-
}
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"Sequence": {
|
| 62 |
-
"id": "A",
|
| 63 |
-
"type_id": 0
|
| 64 |
-
}
|
| 65 |
-
}
|
| 66 |
-
],
|
| 67 |
-
"pair": [
|
| 68 |
-
{
|
| 69 |
-
"SpecialToken": {
|
| 70 |
-
"id": "<s>",
|
| 71 |
-
"type_id": 0
|
| 72 |
-
}
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"Sequence": {
|
| 76 |
-
"id": "A",
|
| 77 |
-
"type_id": 0
|
| 78 |
-
}
|
| 79 |
-
},
|
| 80 |
-
{
|
| 81 |
-
"SpecialToken": {
|
| 82 |
-
"id": "<s>",
|
| 83 |
-
"type_id": 1
|
| 84 |
-
}
|
| 85 |
-
},
|
| 86 |
-
{
|
| 87 |
-
"Sequence": {
|
| 88 |
-
"id": "B",
|
| 89 |
-
"type_id": 1
|
| 90 |
-
}
|
| 91 |
-
}
|
| 92 |
-
],
|
| 93 |
-
"special_tokens": {
|
| 94 |
-
"<s>": {
|
| 95 |
-
"id": "<s>",
|
| 96 |
-
"ids": [
|
| 97 |
-
1
|
| 98 |
-
],
|
| 99 |
-
"tokens": [
|
| 100 |
-
"<s>"
|
| 101 |
-
]
|
| 102 |
-
}
|
| 103 |
-
}
|
| 104 |
-
},
|
| 105 |
-
"decoder": {
|
| 106 |
-
"type": "Sequence",
|
| 107 |
-
"decoders": [
|
| 108 |
-
{
|
| 109 |
-
"type": "Replace",
|
| 110 |
-
"pattern": {
|
| 111 |
-
"String": "▁"
|
| 112 |
-
},
|
| 113 |
-
"content": " "
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"type": "ByteFallback"
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"type": "Fuse"
|
| 120 |
-
},
|
| 121 |
-
{
|
| 122 |
-
"type": "Strip",
|
| 123 |
-
"content": " ",
|
| 124 |
-
"start": 1,
|
| 125 |
-
"stop": 0
|
| 126 |
-
}
|
| 127 |
-
]
|
| 128 |
-
},
|
| 129 |
-
"model": {
|
| 130 |
-
"type": "BPE",
|
| 131 |
-
"dropout": null,
|
| 132 |
-
"unk_token": "<unk>",
|
| 133 |
-
"continuing_subword_prefix": null,
|
| 134 |
-
"end_of_word_suffix": null,
|
| 135 |
-
"fuse_unk": true,
|
| 136 |
-
"byte_fallback": true,
|
| 137 |
-
"vocab": {
|
| 138 |
-
"<unk>": 0,
|
| 139 |
-
"<s>": 1,
|
| 140 |
-
"</s>": 2
|
| 141 |
-
},
|
| 142 |
-
"merges": []
|
| 143 |
-
}
|
| 144 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|