Upload tokenizer
Browse files- special_tokens_map.json +0 -1
- tokenizer.json +1 -33
- tokenizer.model +3 -0
- tokenizer_config.json +2 -2
special_tokens_map.json
CHANGED
@@ -19,7 +19,6 @@
|
|
19 |
"rstrip": false,
|
20 |
"single_word": false
|
21 |
},
|
22 |
-
"pad_token": "<unk>",
|
23 |
"unk_token": {
|
24 |
"content": "<unk>",
|
25 |
"lstrip": false,
|
|
|
19 |
"rstrip": false,
|
20 |
"single_word": false
|
21 |
},
|
|
|
22 |
"unk_token": {
|
23 |
"content": "<unk>",
|
24 |
"lstrip": false,
|
tokenizer.json
CHANGED
@@ -1,11 +1,6 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
"direction": "Right",
|
5 |
-
"max_length": 512,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
"padding": null,
|
10 |
"added_tokens": [
|
11 |
{
|
@@ -103,12 +98,6 @@
|
|
103 |
"id": "A",
|
104 |
"type_id": 0
|
105 |
}
|
106 |
-
},
|
107 |
-
{
|
108 |
-
"SpecialToken": {
|
109 |
-
"id": "</s>",
|
110 |
-
"type_id": 0
|
111 |
-
}
|
112 |
}
|
113 |
],
|
114 |
"pair": [
|
@@ -124,12 +113,6 @@
|
|
124 |
"type_id": 0
|
125 |
}
|
126 |
},
|
127 |
-
{
|
128 |
-
"SpecialToken": {
|
129 |
-
"id": "</s>",
|
130 |
-
"type_id": 0
|
131 |
-
}
|
132 |
-
},
|
133 |
{
|
134 |
"SpecialToken": {
|
135 |
"id": "<s>",
|
@@ -141,24 +124,9 @@
|
|
141 |
"id": "B",
|
142 |
"type_id": 1
|
143 |
}
|
144 |
-
},
|
145 |
-
{
|
146 |
-
"SpecialToken": {
|
147 |
-
"id": "</s>",
|
148 |
-
"type_id": 1
|
149 |
-
}
|
150 |
}
|
151 |
],
|
152 |
"special_tokens": {
|
153 |
-
"</s>": {
|
154 |
-
"id": "</s>",
|
155 |
-
"ids": [
|
156 |
-
2
|
157 |
-
],
|
158 |
-
"tokens": [
|
159 |
-
"</s>"
|
160 |
-
]
|
161 |
-
},
|
162 |
"<s>": {
|
163 |
"id": "<s>",
|
164 |
"ids": [
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
|
|
|
|
|
|
|
|
|
|
4 |
"padding": null,
|
5 |
"added_tokens": [
|
6 |
{
|
|
|
98 |
"id": "A",
|
99 |
"type_id": 0
|
100 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
101 |
}
|
102 |
],
|
103 |
"pair": [
|
|
|
113 |
"type_id": 0
|
114 |
}
|
115 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
116 |
{
|
117 |
"SpecialToken": {
|
118 |
"id": "<s>",
|
|
|
124 |
"id": "B",
|
125 |
"type_id": 1
|
126 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
127 |
}
|
128 |
],
|
129 |
"special_tokens": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
130 |
"<s>": {
|
131 |
"id": "<s>",
|
132 |
"ids": [
|
tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:45ccb9c8b6b561889acea59191d66986d314e7cbd6a78abc6e49b139ca91c1e6
|
3 |
+
size 500058
|
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>",
|
@@ -73,7 +73,7 @@
|
|
73 |
"legacy": null,
|
74 |
"middle_token": "β<MID>",
|
75 |
"model_max_length": 1000000000000000019884624838656,
|
76 |
-
"pad_token":
|
77 |
"prefix_token": "β<PRE>",
|
78 |
"sp_model_kwargs": {},
|
79 |
"suffix_token": "β<SUF>",
|
|
|
1 |
{
|
2 |
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
"added_tokens_decoder": {
|
5 |
"0": {
|
6 |
"content": "<unk>",
|
|
|
73 |
"legacy": null,
|
74 |
"middle_token": "β<MID>",
|
75 |
"model_max_length": 1000000000000000019884624838656,
|
76 |
+
"pad_token": null,
|
77 |
"prefix_token": "β<PRE>",
|
78 |
"sp_model_kwargs": {},
|
79 |
"suffix_token": "β<SUF>",
|