Alignment-Lab-AI
commited on
Commit
•
39782eb
1
Parent(s):
3f8e529
Upload tokenizer
Browse files- special_tokens_map.json +1 -7
- tokenizer.json +27 -0
- tokenizer_config.json +1 -1
special_tokens_map.json
CHANGED
@@ -13,13 +13,7 @@
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
-
"pad_token":
|
17 |
-
"content": "</s>",
|
18 |
-
"lstrip": false,
|
19 |
-
"normalized": false,
|
20 |
-
"rstrip": false,
|
21 |
-
"single_word": false
|
22 |
-
},
|
23 |
"unk_token": {
|
24 |
"content": "<unk>",
|
25 |
"lstrip": false,
|
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
+
"pad_token": "</s>",
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
"unk_token": {
|
18 |
"content": "<unk>",
|
19 |
"lstrip": false,
|
tokenizer.json
CHANGED
@@ -62,6 +62,12 @@
|
|
62 |
"id": "A",
|
63 |
"type_id": 0
|
64 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
}
|
66 |
],
|
67 |
"pair": [
|
@@ -77,6 +83,12 @@
|
|
77 |
"type_id": 0
|
78 |
}
|
79 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
{
|
81 |
"SpecialToken": {
|
82 |
"id": "<s>",
|
@@ -88,9 +100,24 @@
|
|
88 |
"id": "B",
|
89 |
"type_id": 1
|
90 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
}
|
92 |
],
|
93 |
"special_tokens": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
"<s>": {
|
95 |
"id": "<s>",
|
96 |
"ids": [
|
|
|
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 |
"type_id": 0
|
84 |
}
|
85 |
},
|
86 |
+
{
|
87 |
+
"SpecialToken": {
|
88 |
+
"id": "</s>",
|
89 |
+
"type_id": 0
|
90 |
+
}
|
91 |
+
},
|
92 |
{
|
93 |
"SpecialToken": {
|
94 |
"id": "<s>",
|
|
|
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": [
|
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>",
|
|
|
1 |
{
|
2 |
"add_bos_token": true,
|
3 |
+
"add_eos_token": true,
|
4 |
"added_tokens_decoder": {
|
5 |
"0": {
|
6 |
"content": "<unk>",
|