chargoddard
commited on
Commit
·
a9eaabe
1
Parent(s):
fde04db
Upload tokenizer
Browse files- special_tokens_map.json +7 -1
- tokenizer_config.json +1 -0
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_config.json
CHANGED
@@ -29,6 +29,7 @@
|
|
29 |
},
|
30 |
"additional_special_tokens": [],
|
31 |
"bos_token": "<s>",
|
|
|
32 |
"clean_up_tokenization_spaces": false,
|
33 |
"eos_token": "</s>",
|
34 |
"legacy": true,
|
|
|
29 |
},
|
30 |
"additional_special_tokens": [],
|
31 |
"bos_token": "<s>",
|
32 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{' ***' + {'user': 'Query', 'system': 'System', 'assistant': 'Response'}[message['role']] + ':' + message['content']}}{% endfor %}{% if add_generation_prompt %}{{ ' ***Response:' }}{% endif %}",
|
33 |
"clean_up_tokenization_spaces": false,
|
34 |
"eos_token": "</s>",
|
35 |
"legacy": true,
|