Upload tokenizer
Browse files- added_tokens.json +1 -4
- special_tokens_map.json +28 -4
- tokenizer.json +1 -29
- tokenizer_config.json +2 -3
added_tokens.json
CHANGED
@@ -1,8 +1,5 @@
|
|
1 |
{
|
2 |
"\n\n### Instruction:\n": 65107,
|
3 |
"\n\n### Response:\n": 65108,
|
4 |
-
"
|
5 |
-
"<pad>": 65109,
|
6 |
-
"<s>": 1,
|
7 |
-
"<unk>": 0
|
8 |
}
|
|
|
1 |
{
|
2 |
"\n\n### Instruction:\n": 65107,
|
3 |
"\n\n### Response:\n": 65108,
|
4 |
+
"<pad>": 65109
|
|
|
|
|
|
|
5 |
}
|
special_tokens_map.json
CHANGED
@@ -3,8 +3,32 @@
|
|
3 |
"\n\n### Instruction:\n",
|
4 |
"\n\n### Response:\n"
|
5 |
],
|
6 |
-
"bos_token":
|
7 |
-
|
8 |
-
|
9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
}
|
|
|
3 |
"\n\n### Instruction:\n",
|
4 |
"\n\n### Response:\n"
|
5 |
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<s>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": true,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "</s>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": true,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "<pad>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"unk_token": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": true,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
}
|
34 |
}
|
tokenizer.json
CHANGED
@@ -78,12 +78,6 @@
|
|
78 |
"post_processor": {
|
79 |
"type": "TemplateProcessing",
|
80 |
"single": [
|
81 |
-
{
|
82 |
-
"SpecialToken": {
|
83 |
-
"id": "<s>",
|
84 |
-
"type_id": 0
|
85 |
-
}
|
86 |
-
},
|
87 |
{
|
88 |
"Sequence": {
|
89 |
"id": "A",
|
@@ -92,24 +86,12 @@
|
|
92 |
}
|
93 |
],
|
94 |
"pair": [
|
95 |
-
{
|
96 |
-
"SpecialToken": {
|
97 |
-
"id": "<s>",
|
98 |
-
"type_id": 0
|
99 |
-
}
|
100 |
-
},
|
101 |
{
|
102 |
"Sequence": {
|
103 |
"id": "A",
|
104 |
"type_id": 0
|
105 |
}
|
106 |
},
|
107 |
-
{
|
108 |
-
"SpecialToken": {
|
109 |
-
"id": "<s>",
|
110 |
-
"type_id": 1
|
111 |
-
}
|
112 |
-
},
|
113 |
{
|
114 |
"Sequence": {
|
115 |
"id": "B",
|
@@ -117,17 +99,7 @@
|
|
117 |
}
|
118 |
}
|
119 |
],
|
120 |
-
"special_tokens": {
|
121 |
-
"<s>": {
|
122 |
-
"id": "<s>",
|
123 |
-
"ids": [
|
124 |
-
1
|
125 |
-
],
|
126 |
-
"tokens": [
|
127 |
-
"<s>"
|
128 |
-
]
|
129 |
-
}
|
130 |
-
}
|
131 |
},
|
132 |
"decoder": {
|
133 |
"type": "Sequence",
|
|
|
78 |
"post_processor": {
|
79 |
"type": "TemplateProcessing",
|
80 |
"single": [
|
|
|
|
|
|
|
|
|
|
|
|
|
81 |
{
|
82 |
"Sequence": {
|
83 |
"id": "A",
|
|
|
86 |
}
|
87 |
],
|
88 |
"pair": [
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
{
|
90 |
"Sequence": {
|
91 |
"id": "A",
|
92 |
"type_id": 0
|
93 |
}
|
94 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
95 |
{
|
96 |
"Sequence": {
|
97 |
"id": "B",
|
|
|
99 |
}
|
100 |
}
|
101 |
],
|
102 |
+
"special_tokens": {}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
103 |
},
|
104 |
"decoder": {
|
105 |
"type": "Sequence",
|
tokenizer_config.json
CHANGED
@@ -1,6 +1,4 @@
|
|
1 |
{
|
2 |
-
"add_bos_token": false,
|
3 |
-
"add_eos_token": false,
|
4 |
"added_tokens_decoder": {
|
5 |
"0": {
|
6 |
"content": "<unk>",
|
@@ -56,6 +54,7 @@
|
|
56 |
"\n\n### Response:\n"
|
57 |
],
|
58 |
"bos_token": "<s>",
|
|
|
59 |
"clean_up_tokenization_spaces": false,
|
60 |
"eos_token": "</s>",
|
61 |
"legacy": true,
|
@@ -64,5 +63,5 @@
|
|
64 |
"sp_model_kwargs": {},
|
65 |
"tokenizer_class": "LlamaTokenizer",
|
66 |
"unk_token": "<unk>",
|
67 |
-
"use_default_system_prompt":
|
68 |
}
|
|
|
1 |
{
|
|
|
|
|
2 |
"added_tokens_decoder": {
|
3 |
"0": {
|
4 |
"content": "<unk>",
|
|
|
54 |
"\n\n### Response:\n"
|
55 |
],
|
56 |
"bos_token": "<s>",
|
57 |
+
"chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ '\n\n### Instruction:\n' }}{% elif message['role'] == 'assistant' %}{{ '\n\n### Response:\n' }}{% endif %}{{ message['content'] }}{{ eos_token }}{% endfor %}",
|
58 |
"clean_up_tokenization_spaces": false,
|
59 |
"eos_token": "</s>",
|
60 |
"legacy": true,
|
|
|
63 |
"sp_model_kwargs": {},
|
64 |
"tokenizer_class": "LlamaTokenizer",
|
65 |
"unk_token": "<unk>",
|
66 |
+
"use_default_system_prompt": false
|
67 |
}
|