Guilherme34
commited on
Commit
•
51959b7
1
Parent(s):
b6b5ee7
Upload tokenizer
Browse files- special_tokens_map.json +7 -1
- tokenizer.json +0 -27
- tokenizer_config.json +3 -3
special_tokens_map.json
CHANGED
@@ -13,7 +13,13 @@
|
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
"unk_token": {
|
18 |
"content": "<unk>",
|
19 |
"lstrip": false,
|
|
|
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,
|
tokenizer.json
CHANGED
@@ -80,12 +80,6 @@
|
|
80 |
"id": "A",
|
81 |
"type_id": 0
|
82 |
}
|
83 |
-
},
|
84 |
-
{
|
85 |
-
"SpecialToken": {
|
86 |
-
"id": "<|im_end|>",
|
87 |
-
"type_id": 0
|
88 |
-
}
|
89 |
}
|
90 |
],
|
91 |
"pair": [
|
@@ -101,12 +95,6 @@
|
|
101 |
"type_id": 0
|
102 |
}
|
103 |
},
|
104 |
-
{
|
105 |
-
"SpecialToken": {
|
106 |
-
"id": "<|im_end|>",
|
107 |
-
"type_id": 0
|
108 |
-
}
|
109 |
-
},
|
110 |
{
|
111 |
"SpecialToken": {
|
112 |
"id": "<s>",
|
@@ -118,12 +106,6 @@
|
|
118 |
"id": "B",
|
119 |
"type_id": 1
|
120 |
}
|
121 |
-
},
|
122 |
-
{
|
123 |
-
"SpecialToken": {
|
124 |
-
"id": "<|im_end|>",
|
125 |
-
"type_id": 1
|
126 |
-
}
|
127 |
}
|
128 |
],
|
129 |
"special_tokens": {
|
@@ -135,15 +117,6 @@
|
|
135 |
"tokens": [
|
136 |
"<s>"
|
137 |
]
|
138 |
-
},
|
139 |
-
"<|im_end|>": {
|
140 |
-
"id": "<|im_end|>",
|
141 |
-
"ids": [
|
142 |
-
32000
|
143 |
-
],
|
144 |
-
"tokens": [
|
145 |
-
"<|im_end|>"
|
146 |
-
]
|
147 |
}
|
148 |
}
|
149 |
},
|
|
|
80 |
"id": "A",
|
81 |
"type_id": 0
|
82 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
83 |
}
|
84 |
],
|
85 |
"pair": [
|
|
|
95 |
"type_id": 0
|
96 |
}
|
97 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
98 |
{
|
99 |
"SpecialToken": {
|
100 |
"id": "<s>",
|
|
|
106 |
"id": "B",
|
107 |
"type_id": 1
|
108 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
109 |
}
|
110 |
],
|
111 |
"special_tokens": {
|
|
|
117 |
"tokens": [
|
118 |
"<s>"
|
119 |
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
120 |
}
|
121 |
}
|
122 |
},
|
tokenizer_config.json
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
{
|
2 |
"add_bos_token": true,
|
3 |
-
"add_eos_token":
|
|
|
4 |
"added_tokens_decoder": {
|
5 |
"0": {
|
6 |
"content": "<unk>",
|
@@ -50,8 +51,7 @@
|
|
50 |
"eos_token": "<|im_end|>",
|
51 |
"legacy": true,
|
52 |
"model_max_length": 1000000000000000019884624838656,
|
53 |
-
"pad_token": "
|
54 |
-
"padding_side": "left",
|
55 |
"sp_model_kwargs": {},
|
56 |
"spaces_between_special_tokens": false,
|
57 |
"tokenizer_class": "LlamaTokenizer",
|
|
|
1 |
{
|
2 |
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"add_prefix_space": true,
|
5 |
"added_tokens_decoder": {
|
6 |
"0": {
|
7 |
"content": "<unk>",
|
|
|
51 |
"eos_token": "<|im_end|>",
|
52 |
"legacy": true,
|
53 |
"model_max_length": 1000000000000000019884624838656,
|
54 |
+
"pad_token": "</s>",
|
|
|
55 |
"sp_model_kwargs": {},
|
56 |
"spaces_between_special_tokens": false,
|
57 |
"tokenizer_class": "LlamaTokenizer",
|