Ali Safaya
commited on
Commit
•
08ec911
1
Parent(s):
e51f9f8
padding token id
Browse files- special_tokens_map.json +1 -0
- tokenizer.json +15 -15
special_tokens_map.json
CHANGED
@@ -13,6 +13,7 @@
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
|
|
16 |
"unk_token": {
|
17 |
"content": "<unk>",
|
18 |
"lstrip": false,
|
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
+
"pad_token": "<unk>",
|
17 |
"unk_token": {
|
18 |
"content": "<unk>",
|
19 |
"lstrip": false,
|
tokenizer.json
CHANGED
@@ -1,7 +1,16 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
"truncation": null,
|
4 |
-
"padding":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
@@ -61,15 +70,15 @@
|
|
61 |
"id": "A",
|
62 |
"type_id": 0
|
63 |
}
|
64 |
-
}
|
|
|
|
|
65 |
{
|
66 |
"SpecialToken": {
|
67 |
-
"id": "
|
68 |
"type_id": 0
|
69 |
}
|
70 |
-
}
|
71 |
-
],
|
72 |
-
"pair": [
|
73 |
{
|
74 |
"Sequence": {
|
75 |
"id": "A",
|
@@ -84,15 +93,6 @@
|
|
84 |
}
|
85 |
],
|
86 |
"special_tokens": {
|
87 |
-
"</s>": {
|
88 |
-
"id": "</s>",
|
89 |
-
"ids": [
|
90 |
-
1
|
91 |
-
],
|
92 |
-
"tokens": [
|
93 |
-
"</s>"
|
94 |
-
]
|
95 |
-
},
|
96 |
"<s>": {
|
97 |
"id": "<s>",
|
98 |
"ids": [
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
"truncation": null,
|
4 |
+
"padding": {
|
5 |
+
"strategy": {
|
6 |
+
"Fixed": 2048
|
7 |
+
},
|
8 |
+
"direction": "Right",
|
9 |
+
"pad_to_multiple_of": null,
|
10 |
+
"pad_id": 2,
|
11 |
+
"pad_type_id": 0,
|
12 |
+
"pad_token": "<unk>"
|
13 |
+
},
|
14 |
"added_tokens": [
|
15 |
{
|
16 |
"id": 0,
|
|
|
70 |
"id": "A",
|
71 |
"type_id": 0
|
72 |
}
|
73 |
+
}
|
74 |
+
],
|
75 |
+
"pair": [
|
76 |
{
|
77 |
"SpecialToken": {
|
78 |
+
"id": "<s>",
|
79 |
"type_id": 0
|
80 |
}
|
81 |
+
},
|
|
|
|
|
82 |
{
|
83 |
"Sequence": {
|
84 |
"id": "A",
|
|
|
93 |
}
|
94 |
],
|
95 |
"special_tokens": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
"<s>": {
|
97 |
"id": "<s>",
|
98 |
"ids": [
|