Upload tokenizer
Browse files- tokenizer.json +9 -27
- tokenizer_config.json +9 -1
tokenizer.json
CHANGED
|
@@ -29,6 +29,15 @@
|
|
| 29 |
"rstrip": false,
|
| 30 |
"normalized": false,
|
| 31 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
}
|
| 33 |
],
|
| 34 |
"normalizer": {
|
|
@@ -62,12 +71,6 @@
|
|
| 62 |
"id": "A",
|
| 63 |
"type_id": 0
|
| 64 |
}
|
| 65 |
-
},
|
| 66 |
-
{
|
| 67 |
-
"SpecialToken": {
|
| 68 |
-
"id": "</s>",
|
| 69 |
-
"type_id": 0
|
| 70 |
-
}
|
| 71 |
}
|
| 72 |
],
|
| 73 |
"pair": [
|
|
@@ -83,12 +86,6 @@
|
|
| 83 |
"type_id": 0
|
| 84 |
}
|
| 85 |
},
|
| 86 |
-
{
|
| 87 |
-
"SpecialToken": {
|
| 88 |
-
"id": "</s>",
|
| 89 |
-
"type_id": 0
|
| 90 |
-
}
|
| 91 |
-
},
|
| 92 |
{
|
| 93 |
"SpecialToken": {
|
| 94 |
"id": "<s>",
|
|
@@ -100,24 +97,9 @@
|
|
| 100 |
"id": "B",
|
| 101 |
"type_id": 1
|
| 102 |
}
|
| 103 |
-
},
|
| 104 |
-
{
|
| 105 |
-
"SpecialToken": {
|
| 106 |
-
"id": "</s>",
|
| 107 |
-
"type_id": 1
|
| 108 |
-
}
|
| 109 |
}
|
| 110 |
],
|
| 111 |
"special_tokens": {
|
| 112 |
-
"</s>": {
|
| 113 |
-
"id": "</s>",
|
| 114 |
-
"ids": [
|
| 115 |
-
2
|
| 116 |
-
],
|
| 117 |
-
"tokens": [
|
| 118 |
-
"</s>"
|
| 119 |
-
]
|
| 120 |
-
},
|
| 121 |
"<s>": {
|
| 122 |
"id": "<s>",
|
| 123 |
"ids": [
|
|
|
|
| 29 |
"rstrip": false,
|
| 30 |
"normalized": false,
|
| 31 |
"special": true
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"id": 32000,
|
| 35 |
+
"content": "[PAD]",
|
| 36 |
+
"single_word": false,
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"rstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"special": true
|
| 41 |
}
|
| 42 |
],
|
| 43 |
"normalizer": {
|
|
|
|
| 71 |
"id": "A",
|
| 72 |
"type_id": 0
|
| 73 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
}
|
| 75 |
],
|
| 76 |
"pair": [
|
|
|
|
| 86 |
"type_id": 0
|
| 87 |
}
|
| 88 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
{
|
| 90 |
"SpecialToken": {
|
| 91 |
"id": "<s>",
|
|
|
|
| 97 |
"id": "B",
|
| 98 |
"type_id": 1
|
| 99 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
}
|
| 101 |
],
|
| 102 |
"special_tokens": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 103 |
"<s>": {
|
| 104 |
"id": "<s>",
|
| 105 |
"ids": [
|
tokenizer_config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"add_bos_token": true,
|
| 3 |
-
"add_eos_token":
|
| 4 |
"added_tokens_decoder": {
|
| 5 |
"0": {
|
| 6 |
"content": "<unk>",
|
|
@@ -25,6 +25,14 @@
|
|
| 25 |
"rstrip": false,
|
| 26 |
"single_word": false,
|
| 27 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
}
|
| 29 |
},
|
| 30 |
"additional_special_tokens": [],
|
|
|
|
| 1 |
{
|
| 2 |
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
"added_tokens_decoder": {
|
| 5 |
"0": {
|
| 6 |
"content": "<unk>",
|
|
|
|
| 25 |
"rstrip": false,
|
| 26 |
"single_word": false,
|
| 27 |
"special": true
|
| 28 |
+
},
|
| 29 |
+
"32000": {
|
| 30 |
+
"content": "[PAD]",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
}
|
| 37 |
},
|
| 38 |
"additional_special_tokens": [],
|