ming030890
commited on
Commit
•
e0299ea
1
Parent(s):
3141ec1
Upload tokenizer
Browse files- special_tokens_map.json +0 -3
- spiece.model +2 -2
- tokenizer_config.json +0 -3
special_tokens_map.json
CHANGED
@@ -1,5 +1,4 @@
|
|
1 |
{
|
2 |
-
"cls_token": "<s>",
|
3 |
"eos_token": {
|
4 |
"content": "</s>",
|
5 |
"lstrip": false,
|
@@ -7,7 +6,6 @@
|
|
7 |
"rstrip": false,
|
8 |
"single_word": false
|
9 |
},
|
10 |
-
"mask_token": "<mask>",
|
11 |
"pad_token": {
|
12 |
"content": "<pad>",
|
13 |
"lstrip": false,
|
@@ -15,7 +13,6 @@
|
|
15 |
"rstrip": false,
|
16 |
"single_word": false
|
17 |
},
|
18 |
-
"sep_token": "</s>",
|
19 |
"unk_token": {
|
20 |
"content": "<unk>",
|
21 |
"lstrip": false,
|
|
|
1 |
{
|
|
|
2 |
"eos_token": {
|
3 |
"content": "</s>",
|
4 |
"lstrip": false,
|
|
|
6 |
"rstrip": false,
|
7 |
"single_word": false
|
8 |
},
|
|
|
9 |
"pad_token": {
|
10 |
"content": "<pad>",
|
11 |
"lstrip": false,
|
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
|
|
16 |
"unk_token": {
|
17 |
"content": "<unk>",
|
18 |
"lstrip": false,
|
spiece.model
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:828052fd5047daceaebe35af8fe5ddb31abeee929e380863e7525daeee825c18
|
3 |
+
size 513688
|
tokenizer_config.json
CHANGED
@@ -28,14 +28,11 @@
|
|
28 |
},
|
29 |
"additional_special_tokens": [],
|
30 |
"clean_up_tokenization_spaces": true,
|
31 |
-
"cls_token": "<s>",
|
32 |
"eos_token": "</s>",
|
33 |
"extra_ids": 0,
|
34 |
"legacy": false,
|
35 |
-
"mask_token": "<mask>",
|
36 |
"model_max_length": 1000000000000000019884624838656,
|
37 |
"pad_token": "<pad>",
|
38 |
-
"sep_token": "</s>",
|
39 |
"sp_model_kwargs": {},
|
40 |
"tokenizer_class": "T5Tokenizer",
|
41 |
"unk_token": "<unk>"
|
|
|
28 |
},
|
29 |
"additional_special_tokens": [],
|
30 |
"clean_up_tokenization_spaces": true,
|
|
|
31 |
"eos_token": "</s>",
|
32 |
"extra_ids": 0,
|
33 |
"legacy": false,
|
|
|
34 |
"model_max_length": 1000000000000000019884624838656,
|
35 |
"pad_token": "<pad>",
|
|
|
36 |
"sp_model_kwargs": {},
|
37 |
"tokenizer_class": "T5Tokenizer",
|
38 |
"unk_token": "<unk>"
|