emrgnt-cmplxty
commited on
Commit
•
5fedfd8
1
Parent(s):
0ab3c12
Upload tokenizer
Browse files- special_tokens_map.json +1 -2
- tokenizer.json +12 -12
- tokenizer_config.json +13 -16
special_tokens_map.json
CHANGED
@@ -17,8 +17,7 @@
|
|
17 |
"[Irrelevant]",
|
18 |
"[Utility:4]",
|
19 |
"</paragraph>",
|
20 |
-
"[Fully supported]"
|
21 |
-
"<pad>"
|
22 |
],
|
23 |
"bos_token": "<s>",
|
24 |
"eos_token": "</s>",
|
|
|
17 |
"[Irrelevant]",
|
18 |
"[Utility:4]",
|
19 |
"</paragraph>",
|
20 |
+
"[Fully supported]"
|
|
|
21 |
],
|
22 |
"bos_token": "<s>",
|
23 |
"eos_token": "</s>",
|
tokenizer.json
CHANGED
@@ -7,27 +7,27 @@
|
|
7 |
"id": 0,
|
8 |
"content": "<unk>",
|
9 |
"single_word": false,
|
10 |
-
"lstrip":
|
11 |
-
"rstrip":
|
12 |
-
"normalized":
|
13 |
"special": true
|
14 |
},
|
15 |
{
|
16 |
"id": 1,
|
17 |
"content": "<s>",
|
18 |
"single_word": false,
|
19 |
-
"lstrip":
|
20 |
-
"rstrip":
|
21 |
-
"normalized":
|
22 |
"special": true
|
23 |
},
|
24 |
{
|
25 |
"id": 2,
|
26 |
"content": "</s>",
|
27 |
"single_word": false,
|
28 |
-
"lstrip":
|
29 |
-
"rstrip":
|
30 |
-
"normalized":
|
31 |
"special": true
|
32 |
},
|
33 |
{
|
@@ -169,9 +169,9 @@
|
|
169 |
"id": 32015,
|
170 |
"content": "<pad>",
|
171 |
"single_word": false,
|
172 |
-
"lstrip":
|
173 |
-
"rstrip":
|
174 |
-
"normalized":
|
175 |
"special": true
|
176 |
}
|
177 |
],
|
|
|
7 |
"id": 0,
|
8 |
"content": "<unk>",
|
9 |
"single_word": false,
|
10 |
+
"lstrip": true,
|
11 |
+
"rstrip": true,
|
12 |
+
"normalized": false,
|
13 |
"special": true
|
14 |
},
|
15 |
{
|
16 |
"id": 1,
|
17 |
"content": "<s>",
|
18 |
"single_word": false,
|
19 |
+
"lstrip": true,
|
20 |
+
"rstrip": true,
|
21 |
+
"normalized": false,
|
22 |
"special": true
|
23 |
},
|
24 |
{
|
25 |
"id": 2,
|
26 |
"content": "</s>",
|
27 |
"single_word": false,
|
28 |
+
"lstrip": true,
|
29 |
+
"rstrip": true,
|
30 |
+
"normalized": false,
|
31 |
"special": true
|
32 |
},
|
33 |
{
|
|
|
169 |
"id": 32015,
|
170 |
"content": "<pad>",
|
171 |
"single_word": false,
|
172 |
+
"lstrip": true,
|
173 |
+
"rstrip": true,
|
174 |
+
"normalized": false,
|
175 |
"special": true
|
176 |
}
|
177 |
],
|
tokenizer_config.json
CHANGED
@@ -1,28 +1,26 @@
|
|
1 |
{
|
2 |
-
"add_bos_token": true,
|
3 |
-
"add_eos_token": false,
|
4 |
"added_tokens_decoder": {
|
5 |
"0": {
|
6 |
"content": "<unk>",
|
7 |
-
"lstrip":
|
8 |
-
"normalized":
|
9 |
-
"rstrip":
|
10 |
"single_word": false,
|
11 |
"special": true
|
12 |
},
|
13 |
"1": {
|
14 |
"content": "<s>",
|
15 |
-
"lstrip":
|
16 |
-
"normalized":
|
17 |
-
"rstrip":
|
18 |
"single_word": false,
|
19 |
"special": true
|
20 |
},
|
21 |
"2": {
|
22 |
"content": "</s>",
|
23 |
-
"lstrip":
|
24 |
-
"normalized":
|
25 |
-
"rstrip":
|
26 |
"single_word": false,
|
27 |
"special": true
|
28 |
},
|
@@ -148,9 +146,9 @@
|
|
148 |
},
|
149 |
"32015": {
|
150 |
"content": "<pad>",
|
151 |
-
"lstrip":
|
152 |
-
"normalized":
|
153 |
-
"rstrip":
|
154 |
"single_word": false,
|
155 |
"special": true
|
156 |
}
|
@@ -173,8 +171,7 @@
|
|
173 |
"[Irrelevant]",
|
174 |
"[Utility:4]",
|
175 |
"</paragraph>",
|
176 |
-
"[Fully supported]"
|
177 |
-
"<pad>"
|
178 |
],
|
179 |
"bos_token": "<s>",
|
180 |
"clean_up_tokenization_spaces": false,
|
|
|
1 |
{
|
|
|
|
|
2 |
"added_tokens_decoder": {
|
3 |
"0": {
|
4 |
"content": "<unk>",
|
5 |
+
"lstrip": true,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": true,
|
8 |
"single_word": false,
|
9 |
"special": true
|
10 |
},
|
11 |
"1": {
|
12 |
"content": "<s>",
|
13 |
+
"lstrip": true,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": true,
|
16 |
"single_word": false,
|
17 |
"special": true
|
18 |
},
|
19 |
"2": {
|
20 |
"content": "</s>",
|
21 |
+
"lstrip": true,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": true,
|
24 |
"single_word": false,
|
25 |
"special": true
|
26 |
},
|
|
|
146 |
},
|
147 |
"32015": {
|
148 |
"content": "<pad>",
|
149 |
+
"lstrip": true,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": true,
|
152 |
"single_word": false,
|
153 |
"special": true
|
154 |
}
|
|
|
171 |
"[Irrelevant]",
|
172 |
"[Utility:4]",
|
173 |
"</paragraph>",
|
174 |
+
"[Fully supported]"
|
|
|
175 |
],
|
176 |
"bos_token": "<s>",
|
177 |
"clean_up_tokenization_spaces": false,
|