bn22 commited on
Commit
bb84ed0
·
verified ·
1 Parent(s): de05d63

Upload processor

Browse files
Files changed (3) hide show
  1. chat_template.jinja +8 -12
  2. tokenizer.json +2 -2
  3. tokenizer_config.json +1 -0
chat_template.jinja CHANGED
@@ -17,27 +17,23 @@
17
  {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
  {%- for message in messages[::-1] %}
19
  {%- set index = (messages|length - 1) - loop.index0 %}
20
- {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
  {%- set ns.multi_step_tool = false %}
22
  {%- set ns.last_query_index = index %}
23
  {%- endif %}
24
  {%- endfor %}
25
  {%- for message in messages %}
26
- {%- if message.content is string %}
27
- {%- set content = message.content %}
28
- {%- else %}
29
- {%- set content = '' %}
30
- {%- endif %}
31
  {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
- {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
  {%- elif message.role == "assistant" %}
 
34
  {%- set reasoning_content = '' %}
35
- {%- if message.reasoning_content is string %}
36
  {%- set reasoning_content = message.reasoning_content %}
37
  {%- else %}
38
- {%- if '</think>' in content %}
39
- {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
- {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
  {%- endif %}
42
  {%- endif %}
43
  {%- if loop.index0 > ns.last_query_index %}
@@ -74,7 +70,7 @@
74
  {{- '<|im_start|>user' }}
75
  {%- endif %}
76
  {{- '\n<tool_response>\n' }}
77
- {{- content }}
78
  {{- '\n</tool_response>' }}
79
  {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
  {{- '<|im_end|>\n' }}
 
17
  {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
  {%- for message in messages[::-1] %}
19
  {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
  {%- set ns.multi_step_tool = false %}
22
  {%- set ns.last_query_index = index %}
23
  {%- endif %}
24
  {%- endfor %}
25
  {%- for message in messages %}
 
 
 
 
 
26
  {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
27
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
28
  {%- elif message.role == "assistant" %}
29
+ {%- set content = message.content %}
30
  {%- set reasoning_content = '' %}
31
+ {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
32
  {%- set reasoning_content = message.reasoning_content %}
33
  {%- else %}
34
+ {%- if '</think>' in message.content %}
35
+ {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
36
+ {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
37
  {%- endif %}
38
  {%- endif %}
39
  {%- if loop.index0 > ns.last_query_index %}
 
70
  {{- '<|im_start|>user' }}
71
  {%- endif %}
72
  {{- '\n<tool_response>\n' }}
73
+ {{- message.content }}
74
  {{- '\n</tool_response>' }}
75
  {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
76
  {{- '<|im_end|>\n' }}
tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
3
- size 11422650
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93623af029cdc69b87f2864d3b2cc2424fdf16684f15e139b5b9d08ec34ced91
3
+ size 11423701
tokenizer_config.json CHANGED
@@ -22,6 +22,7 @@
22
  ],
23
  "is_local": false,
24
  "model_max_length": 128,
 
25
  "pad_token": "<|endoftext|>",
26
  "processor_class": "VisionTextDualEncoderProcessor",
27
  "split_special_tokens": false,
 
22
  ],
23
  "is_local": false,
24
  "model_max_length": 128,
25
+ "model_specific_special_tokens": {},
26
  "pad_token": "<|endoftext|>",
27
  "processor_class": "VisionTextDualEncoderProcessor",
28
  "split_special_tokens": false,