isbondarev commited on
Commit
104e0f2
·
verified ·
1 Parent(s): a2b5383

Upload tokenizer

Browse files
Files changed (2) hide show
  1. chat_template.jinja +1 -0
  2. tokenizer_config.json +0 -1
chat_template.jinja ADDED
@@ -0,0 +1 @@
 
 
1
+ {% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ '<unk>' + system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'reserved_0' + content + 'reserved_1' }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
tokenizer_config.json CHANGED
@@ -35,7 +35,6 @@
35
  ]
36
  },
37
  "bos_token": "<s>",
38
- "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ '<unk>' + system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'reserved_0' + content + 'reserved_1' }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}",
39
  "clean_up_tokenization_spaces": false,
40
  "decode_with_prefix_space": false,
41
  "eos_token": "</s>",
 
35
  ]
36
  },
37
  "bos_token": "<s>",
 
38
  "clean_up_tokenization_spaces": false,
39
  "decode_with_prefix_space": false,
40
  "eos_token": "</s>",