gacky1601 commited on
Commit
c3f2c44
·
verified ·
1 Parent(s): 43fea32

Upload tokenizer

Browse files
chat_template.jinja ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {% if messages[0]['role'] == 'system' %}{{ messages[0]['content'] + eos_token }}{% set loop_messages = messages[1:] %}{% else %}{{ 'You are given a problem.
2
+ Think about the problem and provide your working out.
3
+ Place it between <reasoning> and </reasoning>.
4
+ Then, provide your solution between <output></output>' + eos_token }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{{ message['content'] }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<reasoning>' }}{% endif %}
special_tokens_map.json CHANGED
@@ -15,7 +15,7 @@
15
  "<|video_pad|>"
16
  ],
17
  "eos_token": {
18
- "content": "<|endoftext|>",
19
  "lstrip": false,
20
  "normalized": false,
21
  "rstrip": false,
 
15
  "<|video_pad|>"
16
  ],
17
  "eos_token": {
18
+ "content": "<|im_end|>",
19
  "lstrip": false,
20
  "normalized": false,
21
  "rstrip": false,
tokenizer_config.json CHANGED
@@ -227,12 +227,11 @@
227
  "<|video_pad|>"
228
  ],
229
  "bos_token": null,
230
- "chat_template": "{% if messages[0]['role'] == 'system' %}{{ messages[0]['content'] + eos_token }}{% set loop_messages = messages[1:] %}{% else %}{{ 'You are given a problem.\nThink about the problem and provide your working out.\nPlace it between <reasoning> and </reasoning>.\nThen, provide your solution between <output></output>' + eos_token }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{{ message['content'] }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<reasoning>' }}{% endif %}",
231
  "clean_up_tokenization_spaces": false,
232
- "eos_token": "<|endoftext|>",
233
  "errors": "replace",
234
  "extra_special_tokens": {},
235
- "model_max_length": 32768,
236
  "pad_token": "<|vision_pad|>",
237
  "padding_side": "left",
238
  "split_special_tokens": false,
 
227
  "<|video_pad|>"
228
  ],
229
  "bos_token": null,
 
230
  "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
  "errors": "replace",
233
  "extra_special_tokens": {},
234
+ "model_max_length": 40960,
235
  "pad_token": "<|vision_pad|>",
236
  "padding_side": "left",
237
  "split_special_tokens": false,