dmax123 commited on
Commit
5306bd3
·
verified ·
1 Parent(s): 46710e7

Add files using upload-large-folder tool

Browse files
Files changed (50) hide show
  1. .gitattributes +3 -0
  2. chat_template.jinja +195 -0
  3. hf_quant_config.json +0 -0
  4. model-00003-of-00113.json +20 -0
  5. model-00004-of-00113.json +0 -0
  6. model-00007-of-00113.json +20 -0
  7. model-00008-of-00113.json +14 -0
  8. model-00009-of-00113.json +0 -0
  9. model-00013-of-00113.json +0 -0
  10. model-00016-of-00113.json +0 -0
  11. model-00021-of-00113.json +20 -0
  12. model-00023-of-00113.json +20 -0
  13. model-00026-of-00113.json +20 -0
  14. model-00027-of-00113.json +0 -0
  15. model-00029-of-00113.json +0 -0
  16. model-00032-of-00113.json +20 -0
  17. model-00034-of-00113.json +0 -0
  18. model-00035-of-00113.json +20 -0
  19. model-00037-of-00113.json +20 -0
  20. model-00038-of-00113.json +0 -0
  21. model-00041-of-00113.json +0 -0
  22. model-00042-of-00113.json +20 -0
  23. model-00044-of-00113.json +20 -0
  24. model-00045-of-00113.json +0 -0
  25. model-00046-of-00113.json +20 -0
  26. model-00047-of-00113.json +0 -0
  27. model-00052-of-00113.json +0 -0
  28. model-00054-of-00113.json +0 -0
  29. model-00055-of-00113.json +20 -0
  30. model-00056-of-00113.json +0 -0
  31. model-00057-of-00113.json +20 -0
  32. model-00063-of-00113.json +0 -0
  33. model-00065-of-00113.json +14 -0
  34. model-00067-of-00113.json +20 -0
  35. model-00072-of-00113.json +0 -0
  36. model-00073-of-00113.json +20 -0
  37. model-00075-of-00113.json +0 -0
  38. model-00079-of-00113.json +0 -0
  39. model-00080-of-00113.json +20 -0
  40. model-00083-of-00113.json +14 -0
  41. model-00084-of-00113.json +0 -0
  42. model-00086-of-00113.json +0 -0
  43. model-00095-of-00113.json +0 -0
  44. model-00100-of-00113.json +0 -0
  45. model-00102-of-00113.json +0 -0
  46. model-00108-of-00113.json +0 -0
  47. model-00109-of-00113.json +228 -0
  48. model-00112-of-00113.json +245 -0
  49. special_tokens_map.json +30 -0
  50. ultra_v3_reasoning_parser.py +28 -0
.gitattributes CHANGED
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ config.json filter=lfs diff=lfs merge=lfs -text
37
+ model.safetensors.index.json filter=lfs diff=lfs merge=lfs -text
38
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
chat_template.jinja ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {% macro render_extra_keys(json_dict, handled_keys) %}
2
+ {%- if json_dict is mapping %}
3
+ {%- for json_key in json_dict if json_key not in handled_keys %}
4
+ {%- if json_dict[json_key] is mapping or (json_dict[json_key] is sequence and json_dict[json_key] is not string) %}
5
+ {{- '\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | tojson | safe) ~ '</' ~ json_key ~ '>' }}
6
+ {%- else %}
7
+ {{-'\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | string) ~ '</' ~ json_key ~ '>' }}
8
+ {%- endif %}
9
+ {%- endfor %}
10
+ {%- endif %}
11
+ {% endmacro %}
12
+ {%- set enable_thinking = enable_thinking if enable_thinking is defined else True %}
13
+ {%- set medium_effort = medium_effort if medium_effort is defined else False %}
14
+ {%- set truncate_history_thinking = truncate_history_thinking if truncate_history_thinking is defined else True %}
15
+ {%- set ns = namespace(last_user_idx = -1) %}
16
+ {%- set loop_messages = messages %}
17
+ {%- for m in loop_messages %}
18
+ {%- if m["role"] == "user" %}
19
+ {%- set ns.last_user_idx = loop.index0 %}
20
+ {%- endif %}
21
+ {%- endfor %}
22
+ {%- if messages[0]["role"] == "system" %}
23
+ {%- set system_message = messages[0]["content"] %}
24
+ {%- set loop_messages = messages[1:] %}
25
+ {%- else %}
26
+ {%- set system_message = "" %}
27
+ {%- set loop_messages = messages %}
28
+ {%- endif %}
29
+ {%- if not tools is defined %}
30
+ {%- set tools = [] %}
31
+ {%- endif %}
32
+ {%- set ns = namespace(last_user_idx = -1) %}
33
+ {%- for m in loop_messages %}
34
+ {%- if m["role"] == "user" %}
35
+ {%- set ns.last_user_idx = loop.index0 %}
36
+ {%- endif %}
37
+ {%- endfor %}
38
+ {%- if system_message is defined %}
39
+ {{- "<|im_start|>system\n" + system_message }}
40
+ {%- else %}
41
+ {%- if tools is iterable and tools | length > 0 %}
42
+ {{- "<|im_start|>system\n" }}
43
+ {%- endif %}
44
+ {%- endif %}
45
+ {%- if tools is iterable and tools | length > 0 %}
46
+ {%- if system_message is defined and system_message | length > 0 %}
47
+ {{- "\n\n" }}
48
+ {%- endif %}
49
+ {{- "# Tools\n\nYou have access to the following functions:\n\n" }}
50
+ {{- "<tools>" }}
51
+ {%- for tool in tools %}
52
+ {%- if tool.function is defined %}
53
+ {%- set tool = tool.function %}
54
+ {%- endif %}
55
+ {{- "\n<function>\n<name>" ~ tool.name ~ "</name>" }}
56
+ {%- if tool.description is defined %}
57
+ {{- '\n<description>' ~ (tool.description | trim) ~ '</description>' }}
58
+ {%- endif %}
59
+ {{- '\n<parameters>' }}
60
+ {%- if tool.parameters is defined and tool.parameters is mapping and tool.parameters.properties is defined and tool.parameters.properties is mapping %}
61
+ {%- for param_name, param_fields in tool.parameters.properties|items %}
62
+ {{- '\n<parameter>' }}
63
+ {{- '\n<name>' ~ param_name ~ '</name>' }}
64
+ {%- if param_fields.type is defined %}
65
+ {{- '\n<type>' ~ (param_fields.type | string) ~ '</type>' }}
66
+ {%- endif %}
67
+ {%- if param_fields.description is defined %}
68
+ {{- '\n<description>' ~ (param_fields.description | trim) ~ '</description>' }}
69
+ {%- endif %}
70
+ {%- if param_fields.enum is defined %}
71
+ {{- '\n<enum>' ~ (param_fields.enum | tojson | safe) ~ '</enum>' }}
72
+ {%- endif %}
73
+ {%- set handled_keys = ['name', 'type', 'description', 'enum'] %}
74
+ {{- render_extra_keys(param_fields, handled_keys) }}
75
+ {{- '\n</parameter>' }}
76
+ {%- endfor %}
77
+ {%- endif %}
78
+ {% set handled_keys = ['type', 'properties', 'required'] %}
79
+ {{- render_extra_keys(tool.parameters, handled_keys) }}
80
+ {%- if tool.parameters is defined and tool.parameters.required is defined %}
81
+ {{- '\n<required>' ~ (tool.parameters.required | tojson | safe) ~ '</required>' }}
82
+ {%- endif %}
83
+ {{- '\n</parameters>' }}
84
+ {%- set handled_keys = ['type', 'name', 'description', 'parameters'] %}
85
+ {{- render_extra_keys(tool, handled_keys) }}
86
+ {{- '\n</function>' }}
87
+ {%- endfor %}
88
+ {{- "\n</tools>" }}
89
+ {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
90
+ {%- endif %}
91
+ {%- if system_message is defined %}
92
+ {{- '<|im_end|>\n' }}
93
+ {%- else %}
94
+ {%- if tools is iterable and tools | length > 0 %}
95
+ {{- '<|im_end|>\n' }}
96
+ {%- endif %}
97
+ {%- endif %}
98
+ {%- for message in loop_messages %}
99
+ {%- if message.role == "assistant" %}
100
+ {%- if message.reasoning_content is defined and message.reasoning_content is string and message.reasoning_content | trim | length > 0 %}
101
+ {%- set content = "<think>\n" ~ message.reasoning_content ~ "</think>" ~ (message.content | default('', true)) %}
102
+ {%- else %}
103
+ {%- set content = message.content | default('', true) %}
104
+ {%- if content is string -%}
105
+ {%- if '<think>' not in content and '</think>' not in content -%}
106
+ {%- set content = "<think></think>" ~ content -%}
107
+ {%- endif -%}
108
+ {%- else -%}
109
+ {%- set content = content -%}
110
+ {%- endif -%}
111
+ {%- endif %}
112
+ {%- if message.tool_calls is defined and message.tool_calls is iterable and message.tool_calls | length > 0 %}
113
+ {{- '<|im_start|>assistant\n' }}
114
+ {%- set include_content = not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
115
+ {%- if content is string and content | trim | length > 0 %}
116
+ {%- if include_content %}
117
+ {{- (content | trim) ~ '\n' -}}
118
+ {%- else %}
119
+ {%- set c = (content | string) %}
120
+ {%- if '</think>' in c %}
121
+ {%- set c = c.split('</think>')[-1] %}
122
+ {%- elif '<think>' in c %}
123
+ {%- set c = c.split('<think>')[0] %}
124
+ {%- endif %}
125
+ {%- set c = "<think></think>" ~ c %}
126
+ {%- if c | length > 0 %}
127
+ {{- c ~ '\n' -}}
128
+ {%- endif %}
129
+ {%- endif %}
130
+ {%- else %}
131
+ {{- "<think></think>" -}}
132
+ {%- endif %}
133
+ {%- for tool_call in message.tool_calls %}
134
+ {%- if tool_call.function is defined %}
135
+ {%- set tool_call = tool_call.function %}
136
+ {%- endif %}
137
+ {{- '<tool_call>\n<function=' ~ tool_call.name ~ '>\n' -}}
138
+ {%- if tool_call.arguments is defined %}
139
+ {%- for args_name, args_value in tool_call.arguments|items %}
140
+ {{- '<parameter=' ~ args_name ~ '>\n' -}}
141
+ {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
142
+ {{- args_value ~ '\n</parameter>\n' -}}
143
+ {%- endfor %}
144
+ {%- endif %}
145
+ {{- '</function>\n</tool_call>\n' -}}
146
+ {%- endfor %}
147
+ {{- '<|im_end|>\n' }}
148
+ {%- else %}
149
+ {%- if not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
150
+ {{- '<|im_start|>assistant\n' ~ (content | default('', true) | string | trim) ~ '<|im_end|>\n' }}
151
+ {%- else %}
152
+ {%- set c = (content | default('', true) | string) %}
153
+ {%- if '<think>' in c and '</think>' in c %}
154
+ {%- set c = "<think></think>" ~ c.split('</think>')[-1] %}
155
+ {%- endif %}
156
+ {%- set c = c | trim %}
157
+ {%- if c | length > 0 %}
158
+ {{- '<|im_start|>assistant\n' ~ c ~ '<|im_end|>\n' }}
159
+ {%- else %}
160
+ {{- '<|im_start|>assistant\n<|im_end|>\n' }}
161
+ {%- endif %}
162
+ {%- endif %}
163
+ {%- endif %}
164
+ {%- elif message.role == "user" or message.role == "system" %}
165
+ {{- '<|im_start|>' + message.role + '\n' }}
166
+ {%- set content = message.content | string %}
167
+ {%- if message.role == "user" and loop.index0 == ns.last_user_idx and medium_effort %}
168
+ {{- content + '\n\n{reasoning effort: efficient}' }}
169
+ {%- else %}
170
+ {{- content }}
171
+ {%- endif %}
172
+ {{- '<|im_end|>\n' }}
173
+ {%- elif message.role == "tool" %}
174
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
175
+ {{- '<|im_start|>user\n' }}
176
+ {%- endif %}
177
+ {{- '<tool_response>\n' }}
178
+ {{- message.content }}
179
+ {{- '\n</tool_response>\n' }}
180
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
181
+ {{- '<|im_end|>\n' }}
182
+ {%- elif loop.last %}
183
+ {{- '<|im_end|>\n' }}
184
+ {%- endif %}
185
+ {%- else %}
186
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>\n' }}
187
+ {%- endif %}
188
+ {%- endfor %}
189
+ {%- if add_generation_prompt %}
190
+ {%- if enable_thinking %}
191
+ {{- '<|im_start|>assistant\n<think>\n' }}
192
+ {%- else %}
193
+ {{- '<|im_start|>assistant\n<think></think>' }}
194
+ {%- endif %}
195
+ {%- endif %}
hf_quant_config.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00003-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.2.mixer.in_proj.input_scale": "model-00003-of-00113.safetensors",
7
+ "backbone.layers.2.mixer.in_proj.weight_scale": "model-00003-of-00113.safetensors",
8
+ "backbone.layers.2.mixer.out_proj.input_scale": "model-00003-of-00113.safetensors",
9
+ "backbone.layers.2.mixer.out_proj.weight_scale": "model-00003-of-00113.safetensors",
10
+ "backbone.layers.2.mixer.A_log": "model-00003-of-00113.safetensors",
11
+ "backbone.layers.2.mixer.D": "model-00003-of-00113.safetensors",
12
+ "backbone.layers.2.mixer.conv1d.bias": "model-00003-of-00113.safetensors",
13
+ "backbone.layers.2.mixer.conv1d.weight": "model-00003-of-00113.safetensors",
14
+ "backbone.layers.2.mixer.dt_bias": "model-00003-of-00113.safetensors",
15
+ "backbone.layers.2.mixer.norm.weight": "model-00003-of-00113.safetensors",
16
+ "backbone.layers.2.norm.weight": "model-00003-of-00113.safetensors",
17
+ "backbone.layers.2.mixer.in_proj.weight": "model-00003-of-00113.safetensors",
18
+ "backbone.layers.2.mixer.out_proj.weight": "model-00003-of-00113.safetensors"
19
+ }
20
+ }
model-00004-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00007-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.6.mixer.in_proj.input_scale": "model-00007-of-00113.safetensors",
7
+ "backbone.layers.6.mixer.in_proj.weight_scale": "model-00007-of-00113.safetensors",
8
+ "backbone.layers.6.mixer.out_proj.input_scale": "model-00007-of-00113.safetensors",
9
+ "backbone.layers.6.mixer.out_proj.weight_scale": "model-00007-of-00113.safetensors",
10
+ "backbone.layers.6.mixer.A_log": "model-00007-of-00113.safetensors",
11
+ "backbone.layers.6.mixer.D": "model-00007-of-00113.safetensors",
12
+ "backbone.layers.6.mixer.conv1d.bias": "model-00007-of-00113.safetensors",
13
+ "backbone.layers.6.mixer.conv1d.weight": "model-00007-of-00113.safetensors",
14
+ "backbone.layers.6.mixer.dt_bias": "model-00007-of-00113.safetensors",
15
+ "backbone.layers.6.mixer.norm.weight": "model-00007-of-00113.safetensors",
16
+ "backbone.layers.6.norm.weight": "model-00007-of-00113.safetensors",
17
+ "backbone.layers.6.mixer.in_proj.weight": "model-00007-of-00113.safetensors",
18
+ "backbone.layers.6.mixer.out_proj.weight": "model-00007-of-00113.safetensors"
19
+ }
20
+ }
model-00008-of-00113.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 276840456
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.7.mixer.k_proj.k_scale": "model-00008-of-00113.safetensors",
7
+ "backbone.layers.7.mixer.v_proj.v_scale": "model-00008-of-00113.safetensors",
8
+ "backbone.layers.7.mixer.k_proj.weight": "model-00008-of-00113.safetensors",
9
+ "backbone.layers.7.mixer.o_proj.weight": "model-00008-of-00113.safetensors",
10
+ "backbone.layers.7.mixer.q_proj.weight": "model-00008-of-00113.safetensors",
11
+ "backbone.layers.7.mixer.v_proj.weight": "model-00008-of-00113.safetensors",
12
+ "backbone.layers.7.norm.weight": "model-00008-of-00113.safetensors"
13
+ }
14
+ }
model-00009-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00013-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00016-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00021-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.20.mixer.in_proj.input_scale": "model-00021-of-00113.safetensors",
7
+ "backbone.layers.20.mixer.in_proj.weight_scale": "model-00021-of-00113.safetensors",
8
+ "backbone.layers.20.mixer.out_proj.input_scale": "model-00021-of-00113.safetensors",
9
+ "backbone.layers.20.mixer.out_proj.weight_scale": "model-00021-of-00113.safetensors",
10
+ "backbone.layers.20.mixer.A_log": "model-00021-of-00113.safetensors",
11
+ "backbone.layers.20.mixer.D": "model-00021-of-00113.safetensors",
12
+ "backbone.layers.20.mixer.conv1d.bias": "model-00021-of-00113.safetensors",
13
+ "backbone.layers.20.mixer.conv1d.weight": "model-00021-of-00113.safetensors",
14
+ "backbone.layers.20.mixer.dt_bias": "model-00021-of-00113.safetensors",
15
+ "backbone.layers.20.mixer.norm.weight": "model-00021-of-00113.safetensors",
16
+ "backbone.layers.20.norm.weight": "model-00021-of-00113.safetensors",
17
+ "backbone.layers.20.mixer.in_proj.weight": "model-00021-of-00113.safetensors",
18
+ "backbone.layers.20.mixer.out_proj.weight": "model-00021-of-00113.safetensors"
19
+ }
20
+ }
model-00023-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.22.mixer.in_proj.input_scale": "model-00023-of-00113.safetensors",
7
+ "backbone.layers.22.mixer.in_proj.weight_scale": "model-00023-of-00113.safetensors",
8
+ "backbone.layers.22.mixer.out_proj.input_scale": "model-00023-of-00113.safetensors",
9
+ "backbone.layers.22.mixer.out_proj.weight_scale": "model-00023-of-00113.safetensors",
10
+ "backbone.layers.22.mixer.A_log": "model-00023-of-00113.safetensors",
11
+ "backbone.layers.22.mixer.D": "model-00023-of-00113.safetensors",
12
+ "backbone.layers.22.mixer.conv1d.bias": "model-00023-of-00113.safetensors",
13
+ "backbone.layers.22.mixer.conv1d.weight": "model-00023-of-00113.safetensors",
14
+ "backbone.layers.22.mixer.dt_bias": "model-00023-of-00113.safetensors",
15
+ "backbone.layers.22.mixer.norm.weight": "model-00023-of-00113.safetensors",
16
+ "backbone.layers.22.norm.weight": "model-00023-of-00113.safetensors",
17
+ "backbone.layers.22.mixer.in_proj.weight": "model-00023-of-00113.safetensors",
18
+ "backbone.layers.22.mixer.out_proj.weight": "model-00023-of-00113.safetensors"
19
+ }
20
+ }
model-00026-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.25.mixer.in_proj.input_scale": "model-00026-of-00113.safetensors",
7
+ "backbone.layers.25.mixer.in_proj.weight_scale": "model-00026-of-00113.safetensors",
8
+ "backbone.layers.25.mixer.out_proj.input_scale": "model-00026-of-00113.safetensors",
9
+ "backbone.layers.25.mixer.out_proj.weight_scale": "model-00026-of-00113.safetensors",
10
+ "backbone.layers.25.mixer.A_log": "model-00026-of-00113.safetensors",
11
+ "backbone.layers.25.mixer.D": "model-00026-of-00113.safetensors",
12
+ "backbone.layers.25.mixer.conv1d.bias": "model-00026-of-00113.safetensors",
13
+ "backbone.layers.25.mixer.conv1d.weight": "model-00026-of-00113.safetensors",
14
+ "backbone.layers.25.mixer.dt_bias": "model-00026-of-00113.safetensors",
15
+ "backbone.layers.25.mixer.norm.weight": "model-00026-of-00113.safetensors",
16
+ "backbone.layers.25.norm.weight": "model-00026-of-00113.safetensors",
17
+ "backbone.layers.25.mixer.in_proj.weight": "model-00026-of-00113.safetensors",
18
+ "backbone.layers.25.mixer.out_proj.weight": "model-00026-of-00113.safetensors"
19
+ }
20
+ }
model-00027-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00029-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00032-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.31.mixer.in_proj.input_scale": "model-00032-of-00113.safetensors",
7
+ "backbone.layers.31.mixer.in_proj.weight_scale": "model-00032-of-00113.safetensors",
8
+ "backbone.layers.31.mixer.out_proj.input_scale": "model-00032-of-00113.safetensors",
9
+ "backbone.layers.31.mixer.out_proj.weight_scale": "model-00032-of-00113.safetensors",
10
+ "backbone.layers.31.mixer.A_log": "model-00032-of-00113.safetensors",
11
+ "backbone.layers.31.mixer.D": "model-00032-of-00113.safetensors",
12
+ "backbone.layers.31.mixer.conv1d.bias": "model-00032-of-00113.safetensors",
13
+ "backbone.layers.31.mixer.conv1d.weight": "model-00032-of-00113.safetensors",
14
+ "backbone.layers.31.mixer.dt_bias": "model-00032-of-00113.safetensors",
15
+ "backbone.layers.31.mixer.norm.weight": "model-00032-of-00113.safetensors",
16
+ "backbone.layers.31.norm.weight": "model-00032-of-00113.safetensors",
17
+ "backbone.layers.31.mixer.in_proj.weight": "model-00032-of-00113.safetensors",
18
+ "backbone.layers.31.mixer.out_proj.weight": "model-00032-of-00113.safetensors"
19
+ }
20
+ }
model-00034-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00035-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.34.mixer.in_proj.input_scale": "model-00035-of-00113.safetensors",
7
+ "backbone.layers.34.mixer.in_proj.weight_scale": "model-00035-of-00113.safetensors",
8
+ "backbone.layers.34.mixer.out_proj.input_scale": "model-00035-of-00113.safetensors",
9
+ "backbone.layers.34.mixer.out_proj.weight_scale": "model-00035-of-00113.safetensors",
10
+ "backbone.layers.34.mixer.A_log": "model-00035-of-00113.safetensors",
11
+ "backbone.layers.34.mixer.D": "model-00035-of-00113.safetensors",
12
+ "backbone.layers.34.mixer.conv1d.bias": "model-00035-of-00113.safetensors",
13
+ "backbone.layers.34.mixer.conv1d.weight": "model-00035-of-00113.safetensors",
14
+ "backbone.layers.34.mixer.dt_bias": "model-00035-of-00113.safetensors",
15
+ "backbone.layers.34.mixer.norm.weight": "model-00035-of-00113.safetensors",
16
+ "backbone.layers.34.norm.weight": "model-00035-of-00113.safetensors",
17
+ "backbone.layers.34.mixer.in_proj.weight": "model-00035-of-00113.safetensors",
18
+ "backbone.layers.34.mixer.out_proj.weight": "model-00035-of-00113.safetensors"
19
+ }
20
+ }
model-00037-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.36.mixer.in_proj.input_scale": "model-00037-of-00113.safetensors",
7
+ "backbone.layers.36.mixer.in_proj.weight_scale": "model-00037-of-00113.safetensors",
8
+ "backbone.layers.36.mixer.out_proj.input_scale": "model-00037-of-00113.safetensors",
9
+ "backbone.layers.36.mixer.out_proj.weight_scale": "model-00037-of-00113.safetensors",
10
+ "backbone.layers.36.mixer.A_log": "model-00037-of-00113.safetensors",
11
+ "backbone.layers.36.mixer.D": "model-00037-of-00113.safetensors",
12
+ "backbone.layers.36.mixer.conv1d.bias": "model-00037-of-00113.safetensors",
13
+ "backbone.layers.36.mixer.conv1d.weight": "model-00037-of-00113.safetensors",
14
+ "backbone.layers.36.mixer.dt_bias": "model-00037-of-00113.safetensors",
15
+ "backbone.layers.36.mixer.norm.weight": "model-00037-of-00113.safetensors",
16
+ "backbone.layers.36.norm.weight": "model-00037-of-00113.safetensors",
17
+ "backbone.layers.36.mixer.in_proj.weight": "model-00037-of-00113.safetensors",
18
+ "backbone.layers.36.mixer.out_proj.weight": "model-00037-of-00113.safetensors"
19
+ }
20
+ }
model-00038-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00041-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00042-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.41.mixer.in_proj.input_scale": "model-00042-of-00113.safetensors",
7
+ "backbone.layers.41.mixer.in_proj.weight_scale": "model-00042-of-00113.safetensors",
8
+ "backbone.layers.41.mixer.out_proj.input_scale": "model-00042-of-00113.safetensors",
9
+ "backbone.layers.41.mixer.out_proj.weight_scale": "model-00042-of-00113.safetensors",
10
+ "backbone.layers.41.mixer.A_log": "model-00042-of-00113.safetensors",
11
+ "backbone.layers.41.mixer.D": "model-00042-of-00113.safetensors",
12
+ "backbone.layers.41.mixer.conv1d.bias": "model-00042-of-00113.safetensors",
13
+ "backbone.layers.41.mixer.conv1d.weight": "model-00042-of-00113.safetensors",
14
+ "backbone.layers.41.mixer.dt_bias": "model-00042-of-00113.safetensors",
15
+ "backbone.layers.41.mixer.norm.weight": "model-00042-of-00113.safetensors",
16
+ "backbone.layers.41.norm.weight": "model-00042-of-00113.safetensors",
17
+ "backbone.layers.41.mixer.in_proj.weight": "model-00042-of-00113.safetensors",
18
+ "backbone.layers.41.mixer.out_proj.weight": "model-00042-of-00113.safetensors"
19
+ }
20
+ }
model-00044-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.43.mixer.in_proj.input_scale": "model-00044-of-00113.safetensors",
7
+ "backbone.layers.43.mixer.in_proj.weight_scale": "model-00044-of-00113.safetensors",
8
+ "backbone.layers.43.mixer.out_proj.input_scale": "model-00044-of-00113.safetensors",
9
+ "backbone.layers.43.mixer.out_proj.weight_scale": "model-00044-of-00113.safetensors",
10
+ "backbone.layers.43.mixer.A_log": "model-00044-of-00113.safetensors",
11
+ "backbone.layers.43.mixer.D": "model-00044-of-00113.safetensors",
12
+ "backbone.layers.43.mixer.conv1d.bias": "model-00044-of-00113.safetensors",
13
+ "backbone.layers.43.mixer.conv1d.weight": "model-00044-of-00113.safetensors",
14
+ "backbone.layers.43.mixer.dt_bias": "model-00044-of-00113.safetensors",
15
+ "backbone.layers.43.mixer.norm.weight": "model-00044-of-00113.safetensors",
16
+ "backbone.layers.43.norm.weight": "model-00044-of-00113.safetensors",
17
+ "backbone.layers.43.mixer.in_proj.weight": "model-00044-of-00113.safetensors",
18
+ "backbone.layers.43.mixer.out_proj.weight": "model-00044-of-00113.safetensors"
19
+ }
20
+ }
model-00045-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00046-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.45.mixer.in_proj.input_scale": "model-00046-of-00113.safetensors",
7
+ "backbone.layers.45.mixer.in_proj.weight_scale": "model-00046-of-00113.safetensors",
8
+ "backbone.layers.45.mixer.out_proj.input_scale": "model-00046-of-00113.safetensors",
9
+ "backbone.layers.45.mixer.out_proj.weight_scale": "model-00046-of-00113.safetensors",
10
+ "backbone.layers.45.mixer.A_log": "model-00046-of-00113.safetensors",
11
+ "backbone.layers.45.mixer.D": "model-00046-of-00113.safetensors",
12
+ "backbone.layers.45.mixer.conv1d.bias": "model-00046-of-00113.safetensors",
13
+ "backbone.layers.45.mixer.conv1d.weight": "model-00046-of-00113.safetensors",
14
+ "backbone.layers.45.mixer.dt_bias": "model-00046-of-00113.safetensors",
15
+ "backbone.layers.45.mixer.norm.weight": "model-00046-of-00113.safetensors",
16
+ "backbone.layers.45.norm.weight": "model-00046-of-00113.safetensors",
17
+ "backbone.layers.45.mixer.in_proj.weight": "model-00046-of-00113.safetensors",
18
+ "backbone.layers.45.mixer.out_proj.weight": "model-00046-of-00113.safetensors"
19
+ }
20
+ }
model-00047-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00052-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00054-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00055-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.54.mixer.in_proj.input_scale": "model-00055-of-00113.safetensors",
7
+ "backbone.layers.54.mixer.in_proj.weight_scale": "model-00055-of-00113.safetensors",
8
+ "backbone.layers.54.mixer.out_proj.input_scale": "model-00055-of-00113.safetensors",
9
+ "backbone.layers.54.mixer.out_proj.weight_scale": "model-00055-of-00113.safetensors",
10
+ "backbone.layers.54.mixer.A_log": "model-00055-of-00113.safetensors",
11
+ "backbone.layers.54.mixer.D": "model-00055-of-00113.safetensors",
12
+ "backbone.layers.54.mixer.conv1d.bias": "model-00055-of-00113.safetensors",
13
+ "backbone.layers.54.mixer.conv1d.weight": "model-00055-of-00113.safetensors",
14
+ "backbone.layers.54.mixer.dt_bias": "model-00055-of-00113.safetensors",
15
+ "backbone.layers.54.mixer.norm.weight": "model-00055-of-00113.safetensors",
16
+ "backbone.layers.54.norm.weight": "model-00055-of-00113.safetensors",
17
+ "backbone.layers.54.mixer.in_proj.weight": "model-00055-of-00113.safetensors",
18
+ "backbone.layers.54.mixer.out_proj.weight": "model-00055-of-00113.safetensors"
19
+ }
20
+ }
model-00056-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00057-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.56.mixer.in_proj.input_scale": "model-00057-of-00113.safetensors",
7
+ "backbone.layers.56.mixer.in_proj.weight_scale": "model-00057-of-00113.safetensors",
8
+ "backbone.layers.56.mixer.out_proj.input_scale": "model-00057-of-00113.safetensors",
9
+ "backbone.layers.56.mixer.out_proj.weight_scale": "model-00057-of-00113.safetensors",
10
+ "backbone.layers.56.mixer.A_log": "model-00057-of-00113.safetensors",
11
+ "backbone.layers.56.mixer.D": "model-00057-of-00113.safetensors",
12
+ "backbone.layers.56.mixer.conv1d.bias": "model-00057-of-00113.safetensors",
13
+ "backbone.layers.56.mixer.conv1d.weight": "model-00057-of-00113.safetensors",
14
+ "backbone.layers.56.mixer.dt_bias": "model-00057-of-00113.safetensors",
15
+ "backbone.layers.56.mixer.norm.weight": "model-00057-of-00113.safetensors",
16
+ "backbone.layers.56.norm.weight": "model-00057-of-00113.safetensors",
17
+ "backbone.layers.56.mixer.in_proj.weight": "model-00057-of-00113.safetensors",
18
+ "backbone.layers.56.mixer.out_proj.weight": "model-00057-of-00113.safetensors"
19
+ }
20
+ }
model-00063-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00065-of-00113.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 276840456
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.64.mixer.k_proj.k_scale": "model-00065-of-00113.safetensors",
7
+ "backbone.layers.64.mixer.v_proj.v_scale": "model-00065-of-00113.safetensors",
8
+ "backbone.layers.64.mixer.k_proj.weight": "model-00065-of-00113.safetensors",
9
+ "backbone.layers.64.mixer.o_proj.weight": "model-00065-of-00113.safetensors",
10
+ "backbone.layers.64.mixer.q_proj.weight": "model-00065-of-00113.safetensors",
11
+ "backbone.layers.64.mixer.v_proj.weight": "model-00065-of-00113.safetensors",
12
+ "backbone.layers.64.norm.weight": "model-00065-of-00113.safetensors"
13
+ }
14
+ }
model-00067-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.66.mixer.in_proj.input_scale": "model-00067-of-00113.safetensors",
7
+ "backbone.layers.66.mixer.in_proj.weight_scale": "model-00067-of-00113.safetensors",
8
+ "backbone.layers.66.mixer.out_proj.input_scale": "model-00067-of-00113.safetensors",
9
+ "backbone.layers.66.mixer.out_proj.weight_scale": "model-00067-of-00113.safetensors",
10
+ "backbone.layers.66.mixer.A_log": "model-00067-of-00113.safetensors",
11
+ "backbone.layers.66.mixer.D": "model-00067-of-00113.safetensors",
12
+ "backbone.layers.66.mixer.conv1d.bias": "model-00067-of-00113.safetensors",
13
+ "backbone.layers.66.mixer.conv1d.weight": "model-00067-of-00113.safetensors",
14
+ "backbone.layers.66.mixer.dt_bias": "model-00067-of-00113.safetensors",
15
+ "backbone.layers.66.mixer.norm.weight": "model-00067-of-00113.safetensors",
16
+ "backbone.layers.66.norm.weight": "model-00067-of-00113.safetensors",
17
+ "backbone.layers.66.mixer.in_proj.weight": "model-00067-of-00113.safetensors",
18
+ "backbone.layers.66.mixer.out_proj.weight": "model-00067-of-00113.safetensors"
19
+ }
20
+ }
model-00072-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00073-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.72.mixer.in_proj.input_scale": "model-00073-of-00113.safetensors",
7
+ "backbone.layers.72.mixer.in_proj.weight_scale": "model-00073-of-00113.safetensors",
8
+ "backbone.layers.72.mixer.out_proj.input_scale": "model-00073-of-00113.safetensors",
9
+ "backbone.layers.72.mixer.out_proj.weight_scale": "model-00073-of-00113.safetensors",
10
+ "backbone.layers.72.mixer.A_log": "model-00073-of-00113.safetensors",
11
+ "backbone.layers.72.mixer.D": "model-00073-of-00113.safetensors",
12
+ "backbone.layers.72.mixer.conv1d.bias": "model-00073-of-00113.safetensors",
13
+ "backbone.layers.72.mixer.conv1d.weight": "model-00073-of-00113.safetensors",
14
+ "backbone.layers.72.mixer.dt_bias": "model-00073-of-00113.safetensors",
15
+ "backbone.layers.72.mixer.norm.weight": "model-00073-of-00113.safetensors",
16
+ "backbone.layers.72.norm.weight": "model-00073-of-00113.safetensors",
17
+ "backbone.layers.72.mixer.in_proj.weight": "model-00073-of-00113.safetensors",
18
+ "backbone.layers.72.mixer.out_proj.weight": "model-00073-of-00113.safetensors"
19
+ }
20
+ }
model-00075-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00079-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00080-of-00113.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 421762576
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.79.mixer.in_proj.input_scale": "model-00080-of-00113.safetensors",
7
+ "backbone.layers.79.mixer.in_proj.weight_scale": "model-00080-of-00113.safetensors",
8
+ "backbone.layers.79.mixer.out_proj.input_scale": "model-00080-of-00113.safetensors",
9
+ "backbone.layers.79.mixer.out_proj.weight_scale": "model-00080-of-00113.safetensors",
10
+ "backbone.layers.79.mixer.A_log": "model-00080-of-00113.safetensors",
11
+ "backbone.layers.79.mixer.D": "model-00080-of-00113.safetensors",
12
+ "backbone.layers.79.mixer.conv1d.bias": "model-00080-of-00113.safetensors",
13
+ "backbone.layers.79.mixer.conv1d.weight": "model-00080-of-00113.safetensors",
14
+ "backbone.layers.79.mixer.dt_bias": "model-00080-of-00113.safetensors",
15
+ "backbone.layers.79.mixer.norm.weight": "model-00080-of-00113.safetensors",
16
+ "backbone.layers.79.norm.weight": "model-00080-of-00113.safetensors",
17
+ "backbone.layers.79.mixer.in_proj.weight": "model-00080-of-00113.safetensors",
18
+ "backbone.layers.79.mixer.out_proj.weight": "model-00080-of-00113.safetensors"
19
+ }
20
+ }
model-00083-of-00113.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 276840456
4
+ },
5
+ "weight_map": {
6
+ "backbone.layers.82.mixer.k_proj.k_scale": "model-00083-of-00113.safetensors",
7
+ "backbone.layers.82.mixer.v_proj.v_scale": "model-00083-of-00113.safetensors",
8
+ "backbone.layers.82.mixer.k_proj.weight": "model-00083-of-00113.safetensors",
9
+ "backbone.layers.82.mixer.o_proj.weight": "model-00083-of-00113.safetensors",
10
+ "backbone.layers.82.mixer.q_proj.weight": "model-00083-of-00113.safetensors",
11
+ "backbone.layers.82.mixer.v_proj.weight": "model-00083-of-00113.safetensors",
12
+ "backbone.layers.82.norm.weight": "model-00083-of-00113.safetensors"
13
+ }
14
+ }
model-00084-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00086-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00095-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00100-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00102-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00108-of-00113.json ADDED
The diff for this file is too large to render. See raw diff
 
model-00109-of-00113.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 4991287296
4
+ },
5
+ "weight_map": {
6
+ "mtp.layers.0.eh_proj.weight": "model-00109-of-00113.safetensors",
7
+ "mtp.layers.0.enorm.weight": "model-00109-of-00113.safetensors",
8
+ "mtp.layers.0.hnorm.weight": "model-00109-of-00113.safetensors",
9
+ "mtp.layers.0.mixer.k_proj.weight": "model-00109-of-00113.safetensors",
10
+ "mtp.layers.0.mixer.o_proj.weight": "model-00109-of-00113.safetensors",
11
+ "mtp.layers.0.mixer.q_proj.weight": "model-00109-of-00113.safetensors",
12
+ "mtp.layers.0.mixer.v_proj.weight": "model-00109-of-00113.safetensors",
13
+ "mtp.layers.0.norm.weight": "model-00109-of-00113.safetensors",
14
+ "mtp.layers.1.final_layernorm.weight": "model-00109-of-00113.safetensors",
15
+ "mtp.layers.1.mixer.experts.0.down_proj.weight": "model-00109-of-00113.safetensors",
16
+ "mtp.layers.1.mixer.experts.0.up_proj.weight": "model-00109-of-00113.safetensors",
17
+ "mtp.layers.1.mixer.experts.1.down_proj.weight": "model-00109-of-00113.safetensors",
18
+ "mtp.layers.1.mixer.experts.1.up_proj.weight": "model-00109-of-00113.safetensors",
19
+ "mtp.layers.1.mixer.experts.10.down_proj.weight": "model-00109-of-00113.safetensors",
20
+ "mtp.layers.1.mixer.experts.10.up_proj.weight": "model-00109-of-00113.safetensors",
21
+ "mtp.layers.1.mixer.experts.100.down_proj.weight": "model-00109-of-00113.safetensors",
22
+ "mtp.layers.1.mixer.experts.100.up_proj.weight": "model-00109-of-00113.safetensors",
23
+ "mtp.layers.1.mixer.experts.101.down_proj.weight": "model-00109-of-00113.safetensors",
24
+ "mtp.layers.1.mixer.experts.101.up_proj.weight": "model-00109-of-00113.safetensors",
25
+ "mtp.layers.1.mixer.experts.102.down_proj.weight": "model-00109-of-00113.safetensors",
26
+ "mtp.layers.1.mixer.experts.102.up_proj.weight": "model-00109-of-00113.safetensors",
27
+ "mtp.layers.1.mixer.experts.103.down_proj.weight": "model-00109-of-00113.safetensors",
28
+ "mtp.layers.1.mixer.experts.103.up_proj.weight": "model-00109-of-00113.safetensors",
29
+ "mtp.layers.1.mixer.experts.104.down_proj.weight": "model-00109-of-00113.safetensors",
30
+ "mtp.layers.1.mixer.experts.104.up_proj.weight": "model-00109-of-00113.safetensors",
31
+ "mtp.layers.1.mixer.experts.105.down_proj.weight": "model-00109-of-00113.safetensors",
32
+ "mtp.layers.1.mixer.experts.105.up_proj.weight": "model-00109-of-00113.safetensors",
33
+ "mtp.layers.1.mixer.experts.106.down_proj.weight": "model-00109-of-00113.safetensors",
34
+ "mtp.layers.1.mixer.experts.106.up_proj.weight": "model-00109-of-00113.safetensors",
35
+ "mtp.layers.1.mixer.experts.107.down_proj.weight": "model-00109-of-00113.safetensors",
36
+ "mtp.layers.1.mixer.experts.107.up_proj.weight": "model-00109-of-00113.safetensors",
37
+ "mtp.layers.1.mixer.experts.108.down_proj.weight": "model-00109-of-00113.safetensors",
38
+ "mtp.layers.1.mixer.experts.108.up_proj.weight": "model-00109-of-00113.safetensors",
39
+ "mtp.layers.1.mixer.experts.109.down_proj.weight": "model-00109-of-00113.safetensors",
40
+ "mtp.layers.1.mixer.experts.109.up_proj.weight": "model-00109-of-00113.safetensors",
41
+ "mtp.layers.1.mixer.experts.11.down_proj.weight": "model-00109-of-00113.safetensors",
42
+ "mtp.layers.1.mixer.experts.11.up_proj.weight": "model-00109-of-00113.safetensors",
43
+ "mtp.layers.1.mixer.experts.110.down_proj.weight": "model-00109-of-00113.safetensors",
44
+ "mtp.layers.1.mixer.experts.110.up_proj.weight": "model-00109-of-00113.safetensors",
45
+ "mtp.layers.1.mixer.experts.111.down_proj.weight": "model-00109-of-00113.safetensors",
46
+ "mtp.layers.1.mixer.experts.111.up_proj.weight": "model-00109-of-00113.safetensors",
47
+ "mtp.layers.1.mixer.experts.112.down_proj.weight": "model-00109-of-00113.safetensors",
48
+ "mtp.layers.1.mixer.experts.112.up_proj.weight": "model-00109-of-00113.safetensors",
49
+ "mtp.layers.1.mixer.experts.113.down_proj.weight": "model-00109-of-00113.safetensors",
50
+ "mtp.layers.1.mixer.experts.113.up_proj.weight": "model-00109-of-00113.safetensors",
51
+ "mtp.layers.1.mixer.experts.114.down_proj.weight": "model-00109-of-00113.safetensors",
52
+ "mtp.layers.1.mixer.experts.114.up_proj.weight": "model-00109-of-00113.safetensors",
53
+ "mtp.layers.1.mixer.experts.115.down_proj.weight": "model-00109-of-00113.safetensors",
54
+ "mtp.layers.1.mixer.experts.115.up_proj.weight": "model-00109-of-00113.safetensors",
55
+ "mtp.layers.1.mixer.experts.116.down_proj.weight": "model-00109-of-00113.safetensors",
56
+ "mtp.layers.1.mixer.experts.116.up_proj.weight": "model-00109-of-00113.safetensors",
57
+ "mtp.layers.1.mixer.experts.117.down_proj.weight": "model-00109-of-00113.safetensors",
58
+ "mtp.layers.1.mixer.experts.117.up_proj.weight": "model-00109-of-00113.safetensors",
59
+ "mtp.layers.1.mixer.experts.118.down_proj.weight": "model-00109-of-00113.safetensors",
60
+ "mtp.layers.1.mixer.experts.118.up_proj.weight": "model-00109-of-00113.safetensors",
61
+ "mtp.layers.1.mixer.experts.119.down_proj.weight": "model-00109-of-00113.safetensors",
62
+ "mtp.layers.1.mixer.experts.119.up_proj.weight": "model-00109-of-00113.safetensors",
63
+ "mtp.layers.1.mixer.experts.12.down_proj.weight": "model-00109-of-00113.safetensors",
64
+ "mtp.layers.1.mixer.experts.12.up_proj.weight": "model-00109-of-00113.safetensors",
65
+ "mtp.layers.1.mixer.experts.120.down_proj.weight": "model-00109-of-00113.safetensors",
66
+ "mtp.layers.1.mixer.experts.120.up_proj.weight": "model-00109-of-00113.safetensors",
67
+ "mtp.layers.1.mixer.experts.121.down_proj.weight": "model-00109-of-00113.safetensors",
68
+ "mtp.layers.1.mixer.experts.121.up_proj.weight": "model-00109-of-00113.safetensors",
69
+ "mtp.layers.1.mixer.experts.122.down_proj.weight": "model-00109-of-00113.safetensors",
70
+ "mtp.layers.1.mixer.experts.122.up_proj.weight": "model-00109-of-00113.safetensors",
71
+ "mtp.layers.1.mixer.experts.123.down_proj.weight": "model-00109-of-00113.safetensors",
72
+ "mtp.layers.1.mixer.experts.123.up_proj.weight": "model-00109-of-00113.safetensors",
73
+ "mtp.layers.1.mixer.experts.124.down_proj.weight": "model-00109-of-00113.safetensors",
74
+ "mtp.layers.1.mixer.experts.124.up_proj.weight": "model-00109-of-00113.safetensors",
75
+ "mtp.layers.1.mixer.experts.125.down_proj.weight": "model-00109-of-00113.safetensors",
76
+ "mtp.layers.1.mixer.experts.125.up_proj.weight": "model-00109-of-00113.safetensors",
77
+ "mtp.layers.1.mixer.experts.126.down_proj.weight": "model-00109-of-00113.safetensors",
78
+ "mtp.layers.1.mixer.experts.126.up_proj.weight": "model-00109-of-00113.safetensors",
79
+ "mtp.layers.1.mixer.experts.127.down_proj.weight": "model-00109-of-00113.safetensors",
80
+ "mtp.layers.1.mixer.experts.127.up_proj.weight": "model-00109-of-00113.safetensors",
81
+ "mtp.layers.1.mixer.experts.128.down_proj.weight": "model-00109-of-00113.safetensors",
82
+ "mtp.layers.1.mixer.experts.128.up_proj.weight": "model-00109-of-00113.safetensors",
83
+ "mtp.layers.1.mixer.experts.129.down_proj.weight": "model-00109-of-00113.safetensors",
84
+ "mtp.layers.1.mixer.experts.129.up_proj.weight": "model-00109-of-00113.safetensors",
85
+ "mtp.layers.1.mixer.experts.13.down_proj.weight": "model-00109-of-00113.safetensors",
86
+ "mtp.layers.1.mixer.experts.13.up_proj.weight": "model-00109-of-00113.safetensors",
87
+ "mtp.layers.1.mixer.experts.130.down_proj.weight": "model-00109-of-00113.safetensors",
88
+ "mtp.layers.1.mixer.experts.130.up_proj.weight": "model-00109-of-00113.safetensors",
89
+ "mtp.layers.1.mixer.experts.131.down_proj.weight": "model-00109-of-00113.safetensors",
90
+ "mtp.layers.1.mixer.experts.131.up_proj.weight": "model-00109-of-00113.safetensors",
91
+ "mtp.layers.1.mixer.experts.132.down_proj.weight": "model-00109-of-00113.safetensors",
92
+ "mtp.layers.1.mixer.experts.132.up_proj.weight": "model-00109-of-00113.safetensors",
93
+ "mtp.layers.1.mixer.experts.133.down_proj.weight": "model-00109-of-00113.safetensors",
94
+ "mtp.layers.1.mixer.experts.133.up_proj.weight": "model-00109-of-00113.safetensors",
95
+ "mtp.layers.1.mixer.experts.134.down_proj.weight": "model-00109-of-00113.safetensors",
96
+ "mtp.layers.1.mixer.experts.134.up_proj.weight": "model-00109-of-00113.safetensors",
97
+ "mtp.layers.1.mixer.experts.135.down_proj.weight": "model-00109-of-00113.safetensors",
98
+ "mtp.layers.1.mixer.experts.135.up_proj.weight": "model-00109-of-00113.safetensors",
99
+ "mtp.layers.1.mixer.experts.136.down_proj.weight": "model-00109-of-00113.safetensors",
100
+ "mtp.layers.1.mixer.experts.136.up_proj.weight": "model-00109-of-00113.safetensors",
101
+ "mtp.layers.1.mixer.experts.137.down_proj.weight": "model-00109-of-00113.safetensors",
102
+ "mtp.layers.1.mixer.experts.137.up_proj.weight": "model-00109-of-00113.safetensors",
103
+ "mtp.layers.1.mixer.experts.138.down_proj.weight": "model-00109-of-00113.safetensors",
104
+ "mtp.layers.1.mixer.experts.138.up_proj.weight": "model-00109-of-00113.safetensors",
105
+ "mtp.layers.1.mixer.experts.139.down_proj.weight": "model-00109-of-00113.safetensors",
106
+ "mtp.layers.1.mixer.experts.139.up_proj.weight": "model-00109-of-00113.safetensors",
107
+ "mtp.layers.1.mixer.experts.14.down_proj.weight": "model-00109-of-00113.safetensors",
108
+ "mtp.layers.1.mixer.experts.14.up_proj.weight": "model-00109-of-00113.safetensors",
109
+ "mtp.layers.1.mixer.experts.140.down_proj.weight": "model-00109-of-00113.safetensors",
110
+ "mtp.layers.1.mixer.experts.140.up_proj.weight": "model-00109-of-00113.safetensors",
111
+ "mtp.layers.1.mixer.experts.141.down_proj.weight": "model-00109-of-00113.safetensors",
112
+ "mtp.layers.1.mixer.experts.141.up_proj.weight": "model-00109-of-00113.safetensors",
113
+ "mtp.layers.1.mixer.experts.142.down_proj.weight": "model-00109-of-00113.safetensors",
114
+ "mtp.layers.1.mixer.experts.142.up_proj.weight": "model-00109-of-00113.safetensors",
115
+ "mtp.layers.1.mixer.experts.143.down_proj.weight": "model-00109-of-00113.safetensors",
116
+ "mtp.layers.1.mixer.experts.143.up_proj.weight": "model-00109-of-00113.safetensors",
117
+ "mtp.layers.1.mixer.experts.144.down_proj.weight": "model-00109-of-00113.safetensors",
118
+ "mtp.layers.1.mixer.experts.144.up_proj.weight": "model-00109-of-00113.safetensors",
119
+ "mtp.layers.1.mixer.experts.145.down_proj.weight": "model-00109-of-00113.safetensors",
120
+ "mtp.layers.1.mixer.experts.145.up_proj.weight": "model-00109-of-00113.safetensors",
121
+ "mtp.layers.1.mixer.experts.146.down_proj.weight": "model-00109-of-00113.safetensors",
122
+ "mtp.layers.1.mixer.experts.146.up_proj.weight": "model-00109-of-00113.safetensors",
123
+ "mtp.layers.1.mixer.experts.147.down_proj.weight": "model-00109-of-00113.safetensors",
124
+ "mtp.layers.1.mixer.experts.147.up_proj.weight": "model-00109-of-00113.safetensors",
125
+ "mtp.layers.1.mixer.experts.148.down_proj.weight": "model-00109-of-00113.safetensors",
126
+ "mtp.layers.1.mixer.experts.148.up_proj.weight": "model-00109-of-00113.safetensors",
127
+ "mtp.layers.1.mixer.experts.149.down_proj.weight": "model-00109-of-00113.safetensors",
128
+ "mtp.layers.1.mixer.experts.149.up_proj.weight": "model-00109-of-00113.safetensors",
129
+ "mtp.layers.1.mixer.experts.15.down_proj.weight": "model-00109-of-00113.safetensors",
130
+ "mtp.layers.1.mixer.experts.15.up_proj.weight": "model-00109-of-00113.safetensors",
131
+ "mtp.layers.1.mixer.experts.150.down_proj.weight": "model-00109-of-00113.safetensors",
132
+ "mtp.layers.1.mixer.experts.150.up_proj.weight": "model-00109-of-00113.safetensors",
133
+ "mtp.layers.1.mixer.experts.151.down_proj.weight": "model-00109-of-00113.safetensors",
134
+ "mtp.layers.1.mixer.experts.151.up_proj.weight": "model-00109-of-00113.safetensors",
135
+ "mtp.layers.1.mixer.experts.152.down_proj.weight": "model-00109-of-00113.safetensors",
136
+ "mtp.layers.1.mixer.experts.152.up_proj.weight": "model-00109-of-00113.safetensors",
137
+ "mtp.layers.1.mixer.experts.153.down_proj.weight": "model-00109-of-00113.safetensors",
138
+ "mtp.layers.1.mixer.experts.153.up_proj.weight": "model-00109-of-00113.safetensors",
139
+ "mtp.layers.1.mixer.experts.154.down_proj.weight": "model-00109-of-00113.safetensors",
140
+ "mtp.layers.1.mixer.experts.154.up_proj.weight": "model-00109-of-00113.safetensors",
141
+ "mtp.layers.1.mixer.experts.155.down_proj.weight": "model-00109-of-00113.safetensors",
142
+ "mtp.layers.1.mixer.experts.155.up_proj.weight": "model-00109-of-00113.safetensors",
143
+ "mtp.layers.1.mixer.experts.156.down_proj.weight": "model-00109-of-00113.safetensors",
144
+ "mtp.layers.1.mixer.experts.156.up_proj.weight": "model-00109-of-00113.safetensors",
145
+ "mtp.layers.1.mixer.experts.157.down_proj.weight": "model-00109-of-00113.safetensors",
146
+ "mtp.layers.1.mixer.experts.157.up_proj.weight": "model-00109-of-00113.safetensors",
147
+ "mtp.layers.1.mixer.experts.158.down_proj.weight": "model-00109-of-00113.safetensors",
148
+ "mtp.layers.1.mixer.experts.158.up_proj.weight": "model-00109-of-00113.safetensors",
149
+ "mtp.layers.1.mixer.experts.159.down_proj.weight": "model-00109-of-00113.safetensors",
150
+ "mtp.layers.1.mixer.experts.159.up_proj.weight": "model-00109-of-00113.safetensors",
151
+ "mtp.layers.1.mixer.experts.16.down_proj.weight": "model-00109-of-00113.safetensors",
152
+ "mtp.layers.1.mixer.experts.16.up_proj.weight": "model-00109-of-00113.safetensors",
153
+ "mtp.layers.1.mixer.experts.160.down_proj.weight": "model-00109-of-00113.safetensors",
154
+ "mtp.layers.1.mixer.experts.160.up_proj.weight": "model-00109-of-00113.safetensors",
155
+ "mtp.layers.1.mixer.experts.161.down_proj.weight": "model-00109-of-00113.safetensors",
156
+ "mtp.layers.1.mixer.experts.161.up_proj.weight": "model-00109-of-00113.safetensors",
157
+ "mtp.layers.1.mixer.experts.162.down_proj.weight": "model-00109-of-00113.safetensors",
158
+ "mtp.layers.1.mixer.experts.162.up_proj.weight": "model-00109-of-00113.safetensors",
159
+ "mtp.layers.1.mixer.experts.163.down_proj.weight": "model-00109-of-00113.safetensors",
160
+ "mtp.layers.1.mixer.experts.163.up_proj.weight": "model-00109-of-00113.safetensors",
161
+ "mtp.layers.1.mixer.experts.164.down_proj.weight": "model-00109-of-00113.safetensors",
162
+ "mtp.layers.1.mixer.experts.164.up_proj.weight": "model-00109-of-00113.safetensors",
163
+ "mtp.layers.1.mixer.experts.165.down_proj.weight": "model-00109-of-00113.safetensors",
164
+ "mtp.layers.1.mixer.experts.165.up_proj.weight": "model-00109-of-00113.safetensors",
165
+ "mtp.layers.1.mixer.experts.166.down_proj.weight": "model-00109-of-00113.safetensors",
166
+ "mtp.layers.1.mixer.experts.166.up_proj.weight": "model-00109-of-00113.safetensors",
167
+ "mtp.layers.1.mixer.experts.167.down_proj.weight": "model-00109-of-00113.safetensors",
168
+ "mtp.layers.1.mixer.experts.167.up_proj.weight": "model-00109-of-00113.safetensors",
169
+ "mtp.layers.1.mixer.experts.168.down_proj.weight": "model-00109-of-00113.safetensors",
170
+ "mtp.layers.1.mixer.experts.168.up_proj.weight": "model-00109-of-00113.safetensors",
171
+ "mtp.layers.1.mixer.experts.169.down_proj.weight": "model-00109-of-00113.safetensors",
172
+ "mtp.layers.1.mixer.experts.169.up_proj.weight": "model-00109-of-00113.safetensors",
173
+ "mtp.layers.1.mixer.experts.17.down_proj.weight": "model-00109-of-00113.safetensors",
174
+ "mtp.layers.1.mixer.experts.17.up_proj.weight": "model-00109-of-00113.safetensors",
175
+ "mtp.layers.1.mixer.experts.170.down_proj.weight": "model-00109-of-00113.safetensors",
176
+ "mtp.layers.1.mixer.experts.170.up_proj.weight": "model-00109-of-00113.safetensors",
177
+ "mtp.layers.1.mixer.experts.171.down_proj.weight": "model-00109-of-00113.safetensors",
178
+ "mtp.layers.1.mixer.experts.171.up_proj.weight": "model-00109-of-00113.safetensors",
179
+ "mtp.layers.1.mixer.experts.172.down_proj.weight": "model-00109-of-00113.safetensors",
180
+ "mtp.layers.1.mixer.experts.172.up_proj.weight": "model-00109-of-00113.safetensors",
181
+ "mtp.layers.1.mixer.experts.173.down_proj.weight": "model-00109-of-00113.safetensors",
182
+ "mtp.layers.1.mixer.experts.173.up_proj.weight": "model-00109-of-00113.safetensors",
183
+ "mtp.layers.1.mixer.experts.174.down_proj.weight": "model-00109-of-00113.safetensors",
184
+ "mtp.layers.1.mixer.experts.174.up_proj.weight": "model-00109-of-00113.safetensors",
185
+ "mtp.layers.1.mixer.experts.175.down_proj.weight": "model-00109-of-00113.safetensors",
186
+ "mtp.layers.1.mixer.experts.175.up_proj.weight": "model-00109-of-00113.safetensors",
187
+ "mtp.layers.1.mixer.experts.176.down_proj.weight": "model-00109-of-00113.safetensors",
188
+ "mtp.layers.1.mixer.experts.176.up_proj.weight": "model-00109-of-00113.safetensors",
189
+ "mtp.layers.1.mixer.experts.177.down_proj.weight": "model-00109-of-00113.safetensors",
190
+ "mtp.layers.1.mixer.experts.177.up_proj.weight": "model-00109-of-00113.safetensors",
191
+ "mtp.layers.1.mixer.experts.178.down_proj.weight": "model-00109-of-00113.safetensors",
192
+ "mtp.layers.1.mixer.experts.178.up_proj.weight": "model-00109-of-00113.safetensors",
193
+ "mtp.layers.1.mixer.experts.179.down_proj.weight": "model-00109-of-00113.safetensors",
194
+ "mtp.layers.1.mixer.experts.179.up_proj.weight": "model-00109-of-00113.safetensors",
195
+ "mtp.layers.1.mixer.experts.18.down_proj.weight": "model-00109-of-00113.safetensors",
196
+ "mtp.layers.1.mixer.experts.18.up_proj.weight": "model-00109-of-00113.safetensors",
197
+ "mtp.layers.1.mixer.experts.180.down_proj.weight": "model-00109-of-00113.safetensors",
198
+ "mtp.layers.1.mixer.experts.180.up_proj.weight": "model-00109-of-00113.safetensors",
199
+ "mtp.layers.1.mixer.experts.181.down_proj.weight": "model-00109-of-00113.safetensors",
200
+ "mtp.layers.1.mixer.experts.181.up_proj.weight": "model-00109-of-00113.safetensors",
201
+ "mtp.layers.1.mixer.experts.182.down_proj.weight": "model-00109-of-00113.safetensors",
202
+ "mtp.layers.1.mixer.experts.182.up_proj.weight": "model-00109-of-00113.safetensors",
203
+ "mtp.layers.1.mixer.experts.183.down_proj.weight": "model-00109-of-00113.safetensors",
204
+ "mtp.layers.1.mixer.experts.183.up_proj.weight": "model-00109-of-00113.safetensors",
205
+ "mtp.layers.1.mixer.experts.184.down_proj.weight": "model-00109-of-00113.safetensors",
206
+ "mtp.layers.1.mixer.experts.184.up_proj.weight": "model-00109-of-00113.safetensors",
207
+ "mtp.layers.1.mixer.experts.185.down_proj.weight": "model-00109-of-00113.safetensors",
208
+ "mtp.layers.1.mixer.experts.185.up_proj.weight": "model-00109-of-00113.safetensors",
209
+ "mtp.layers.1.mixer.experts.186.down_proj.weight": "model-00109-of-00113.safetensors",
210
+ "mtp.layers.1.mixer.experts.186.up_proj.weight": "model-00109-of-00113.safetensors",
211
+ "mtp.layers.1.mixer.experts.187.down_proj.weight": "model-00109-of-00113.safetensors",
212
+ "mtp.layers.1.mixer.experts.187.up_proj.weight": "model-00109-of-00113.safetensors",
213
+ "mtp.layers.1.mixer.experts.188.down_proj.weight": "model-00109-of-00113.safetensors",
214
+ "mtp.layers.1.mixer.experts.188.up_proj.weight": "model-00109-of-00113.safetensors",
215
+ "mtp.layers.1.mixer.experts.189.down_proj.weight": "model-00109-of-00113.safetensors",
216
+ "mtp.layers.1.mixer.experts.189.up_proj.weight": "model-00109-of-00113.safetensors",
217
+ "mtp.layers.1.mixer.experts.19.down_proj.weight": "model-00109-of-00113.safetensors",
218
+ "mtp.layers.1.mixer.experts.19.up_proj.weight": "model-00109-of-00113.safetensors",
219
+ "mtp.layers.1.mixer.experts.190.down_proj.weight": "model-00109-of-00113.safetensors",
220
+ "mtp.layers.1.mixer.experts.190.up_proj.weight": "model-00109-of-00113.safetensors",
221
+ "mtp.layers.1.mixer.experts.191.down_proj.weight": "model-00109-of-00113.safetensors",
222
+ "mtp.layers.1.mixer.experts.191.up_proj.weight": "model-00109-of-00113.safetensors",
223
+ "mtp.layers.1.mixer.experts.192.down_proj.weight": "model-00109-of-00113.safetensors",
224
+ "mtp.layers.1.mixer.experts.192.up_proj.weight": "model-00109-of-00113.safetensors",
225
+ "mtp.layers.1.mixer.experts.193.down_proj.weight": "model-00109-of-00113.safetensors",
226
+ "mtp.layers.1.mixer.experts.193.up_proj.weight": "model-00109-of-00113.safetensors"
227
+ }
228
+ }
model-00112-of-00113.json ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 4991221760
4
+ },
5
+ "weight_map": {
6
+ "mtp.layers.1.mixer.experts.408.down_proj.weight": "model-00112-of-00113.safetensors",
7
+ "mtp.layers.1.mixer.experts.408.up_proj.weight": "model-00112-of-00113.safetensors",
8
+ "mtp.layers.1.mixer.experts.409.down_proj.weight": "model-00112-of-00113.safetensors",
9
+ "mtp.layers.1.mixer.experts.409.up_proj.weight": "model-00112-of-00113.safetensors",
10
+ "mtp.layers.1.mixer.experts.41.down_proj.weight": "model-00112-of-00113.safetensors",
11
+ "mtp.layers.1.mixer.experts.41.up_proj.weight": "model-00112-of-00113.safetensors",
12
+ "mtp.layers.1.mixer.experts.410.down_proj.weight": "model-00112-of-00113.safetensors",
13
+ "mtp.layers.1.mixer.experts.410.up_proj.weight": "model-00112-of-00113.safetensors",
14
+ "mtp.layers.1.mixer.experts.411.down_proj.weight": "model-00112-of-00113.safetensors",
15
+ "mtp.layers.1.mixer.experts.411.up_proj.weight": "model-00112-of-00113.safetensors",
16
+ "mtp.layers.1.mixer.experts.412.down_proj.weight": "model-00112-of-00113.safetensors",
17
+ "mtp.layers.1.mixer.experts.412.up_proj.weight": "model-00112-of-00113.safetensors",
18
+ "mtp.layers.1.mixer.experts.413.down_proj.weight": "model-00112-of-00113.safetensors",
19
+ "mtp.layers.1.mixer.experts.413.up_proj.weight": "model-00112-of-00113.safetensors",
20
+ "mtp.layers.1.mixer.experts.414.down_proj.weight": "model-00112-of-00113.safetensors",
21
+ "mtp.layers.1.mixer.experts.414.up_proj.weight": "model-00112-of-00113.safetensors",
22
+ "mtp.layers.1.mixer.experts.415.down_proj.weight": "model-00112-of-00113.safetensors",
23
+ "mtp.layers.1.mixer.experts.415.up_proj.weight": "model-00112-of-00113.safetensors",
24
+ "mtp.layers.1.mixer.experts.416.down_proj.weight": "model-00112-of-00113.safetensors",
25
+ "mtp.layers.1.mixer.experts.416.up_proj.weight": "model-00112-of-00113.safetensors",
26
+ "mtp.layers.1.mixer.experts.417.down_proj.weight": "model-00112-of-00113.safetensors",
27
+ "mtp.layers.1.mixer.experts.417.up_proj.weight": "model-00112-of-00113.safetensors",
28
+ "mtp.layers.1.mixer.experts.418.down_proj.weight": "model-00112-of-00113.safetensors",
29
+ "mtp.layers.1.mixer.experts.418.up_proj.weight": "model-00112-of-00113.safetensors",
30
+ "mtp.layers.1.mixer.experts.419.down_proj.weight": "model-00112-of-00113.safetensors",
31
+ "mtp.layers.1.mixer.experts.419.up_proj.weight": "model-00112-of-00113.safetensors",
32
+ "mtp.layers.1.mixer.experts.42.down_proj.weight": "model-00112-of-00113.safetensors",
33
+ "mtp.layers.1.mixer.experts.42.up_proj.weight": "model-00112-of-00113.safetensors",
34
+ "mtp.layers.1.mixer.experts.420.down_proj.weight": "model-00112-of-00113.safetensors",
35
+ "mtp.layers.1.mixer.experts.420.up_proj.weight": "model-00112-of-00113.safetensors",
36
+ "mtp.layers.1.mixer.experts.421.down_proj.weight": "model-00112-of-00113.safetensors",
37
+ "mtp.layers.1.mixer.experts.421.up_proj.weight": "model-00112-of-00113.safetensors",
38
+ "mtp.layers.1.mixer.experts.422.down_proj.weight": "model-00112-of-00113.safetensors",
39
+ "mtp.layers.1.mixer.experts.422.up_proj.weight": "model-00112-of-00113.safetensors",
40
+ "mtp.layers.1.mixer.experts.423.down_proj.weight": "model-00112-of-00113.safetensors",
41
+ "mtp.layers.1.mixer.experts.423.up_proj.weight": "model-00112-of-00113.safetensors",
42
+ "mtp.layers.1.mixer.experts.424.down_proj.weight": "model-00112-of-00113.safetensors",
43
+ "mtp.layers.1.mixer.experts.424.up_proj.weight": "model-00112-of-00113.safetensors",
44
+ "mtp.layers.1.mixer.experts.425.down_proj.weight": "model-00112-of-00113.safetensors",
45
+ "mtp.layers.1.mixer.experts.425.up_proj.weight": "model-00112-of-00113.safetensors",
46
+ "mtp.layers.1.mixer.experts.426.down_proj.weight": "model-00112-of-00113.safetensors",
47
+ "mtp.layers.1.mixer.experts.426.up_proj.weight": "model-00112-of-00113.safetensors",
48
+ "mtp.layers.1.mixer.experts.427.down_proj.weight": "model-00112-of-00113.safetensors",
49
+ "mtp.layers.1.mixer.experts.427.up_proj.weight": "model-00112-of-00113.safetensors",
50
+ "mtp.layers.1.mixer.experts.428.down_proj.weight": "model-00112-of-00113.safetensors",
51
+ "mtp.layers.1.mixer.experts.428.up_proj.weight": "model-00112-of-00113.safetensors",
52
+ "mtp.layers.1.mixer.experts.429.down_proj.weight": "model-00112-of-00113.safetensors",
53
+ "mtp.layers.1.mixer.experts.429.up_proj.weight": "model-00112-of-00113.safetensors",
54
+ "mtp.layers.1.mixer.experts.43.down_proj.weight": "model-00112-of-00113.safetensors",
55
+ "mtp.layers.1.mixer.experts.43.up_proj.weight": "model-00112-of-00113.safetensors",
56
+ "mtp.layers.1.mixer.experts.430.down_proj.weight": "model-00112-of-00113.safetensors",
57
+ "mtp.layers.1.mixer.experts.430.up_proj.weight": "model-00112-of-00113.safetensors",
58
+ "mtp.layers.1.mixer.experts.431.down_proj.weight": "model-00112-of-00113.safetensors",
59
+ "mtp.layers.1.mixer.experts.431.up_proj.weight": "model-00112-of-00113.safetensors",
60
+ "mtp.layers.1.mixer.experts.432.down_proj.weight": "model-00112-of-00113.safetensors",
61
+ "mtp.layers.1.mixer.experts.432.up_proj.weight": "model-00112-of-00113.safetensors",
62
+ "mtp.layers.1.mixer.experts.433.down_proj.weight": "model-00112-of-00113.safetensors",
63
+ "mtp.layers.1.mixer.experts.433.up_proj.weight": "model-00112-of-00113.safetensors",
64
+ "mtp.layers.1.mixer.experts.434.down_proj.weight": "model-00112-of-00113.safetensors",
65
+ "mtp.layers.1.mixer.experts.434.up_proj.weight": "model-00112-of-00113.safetensors",
66
+ "mtp.layers.1.mixer.experts.435.down_proj.weight": "model-00112-of-00113.safetensors",
67
+ "mtp.layers.1.mixer.experts.435.up_proj.weight": "model-00112-of-00113.safetensors",
68
+ "mtp.layers.1.mixer.experts.436.down_proj.weight": "model-00112-of-00113.safetensors",
69
+ "mtp.layers.1.mixer.experts.436.up_proj.weight": "model-00112-of-00113.safetensors",
70
+ "mtp.layers.1.mixer.experts.437.down_proj.weight": "model-00112-of-00113.safetensors",
71
+ "mtp.layers.1.mixer.experts.437.up_proj.weight": "model-00112-of-00113.safetensors",
72
+ "mtp.layers.1.mixer.experts.438.down_proj.weight": "model-00112-of-00113.safetensors",
73
+ "mtp.layers.1.mixer.experts.438.up_proj.weight": "model-00112-of-00113.safetensors",
74
+ "mtp.layers.1.mixer.experts.439.down_proj.weight": "model-00112-of-00113.safetensors",
75
+ "mtp.layers.1.mixer.experts.439.up_proj.weight": "model-00112-of-00113.safetensors",
76
+ "mtp.layers.1.mixer.experts.44.down_proj.weight": "model-00112-of-00113.safetensors",
77
+ "mtp.layers.1.mixer.experts.44.up_proj.weight": "model-00112-of-00113.safetensors",
78
+ "mtp.layers.1.mixer.experts.440.down_proj.weight": "model-00112-of-00113.safetensors",
79
+ "mtp.layers.1.mixer.experts.440.up_proj.weight": "model-00112-of-00113.safetensors",
80
+ "mtp.layers.1.mixer.experts.441.down_proj.weight": "model-00112-of-00113.safetensors",
81
+ "mtp.layers.1.mixer.experts.441.up_proj.weight": "model-00112-of-00113.safetensors",
82
+ "mtp.layers.1.mixer.experts.442.down_proj.weight": "model-00112-of-00113.safetensors",
83
+ "mtp.layers.1.mixer.experts.442.up_proj.weight": "model-00112-of-00113.safetensors",
84
+ "mtp.layers.1.mixer.experts.443.down_proj.weight": "model-00112-of-00113.safetensors",
85
+ "mtp.layers.1.mixer.experts.443.up_proj.weight": "model-00112-of-00113.safetensors",
86
+ "mtp.layers.1.mixer.experts.444.down_proj.weight": "model-00112-of-00113.safetensors",
87
+ "mtp.layers.1.mixer.experts.444.up_proj.weight": "model-00112-of-00113.safetensors",
88
+ "mtp.layers.1.mixer.experts.445.down_proj.weight": "model-00112-of-00113.safetensors",
89
+ "mtp.layers.1.mixer.experts.445.up_proj.weight": "model-00112-of-00113.safetensors",
90
+ "mtp.layers.1.mixer.experts.446.down_proj.weight": "model-00112-of-00113.safetensors",
91
+ "mtp.layers.1.mixer.experts.446.up_proj.weight": "model-00112-of-00113.safetensors",
92
+ "mtp.layers.1.mixer.experts.447.down_proj.weight": "model-00112-of-00113.safetensors",
93
+ "mtp.layers.1.mixer.experts.447.up_proj.weight": "model-00112-of-00113.safetensors",
94
+ "mtp.layers.1.mixer.experts.448.down_proj.weight": "model-00112-of-00113.safetensors",
95
+ "mtp.layers.1.mixer.experts.448.up_proj.weight": "model-00112-of-00113.safetensors",
96
+ "mtp.layers.1.mixer.experts.449.down_proj.weight": "model-00112-of-00113.safetensors",
97
+ "mtp.layers.1.mixer.experts.449.up_proj.weight": "model-00112-of-00113.safetensors",
98
+ "mtp.layers.1.mixer.experts.45.down_proj.weight": "model-00112-of-00113.safetensors",
99
+ "mtp.layers.1.mixer.experts.45.up_proj.weight": "model-00112-of-00113.safetensors",
100
+ "mtp.layers.1.mixer.experts.450.down_proj.weight": "model-00112-of-00113.safetensors",
101
+ "mtp.layers.1.mixer.experts.450.up_proj.weight": "model-00112-of-00113.safetensors",
102
+ "mtp.layers.1.mixer.experts.451.down_proj.weight": "model-00112-of-00113.safetensors",
103
+ "mtp.layers.1.mixer.experts.451.up_proj.weight": "model-00112-of-00113.safetensors",
104
+ "mtp.layers.1.mixer.experts.452.down_proj.weight": "model-00112-of-00113.safetensors",
105
+ "mtp.layers.1.mixer.experts.452.up_proj.weight": "model-00112-of-00113.safetensors",
106
+ "mtp.layers.1.mixer.experts.453.down_proj.weight": "model-00112-of-00113.safetensors",
107
+ "mtp.layers.1.mixer.experts.453.up_proj.weight": "model-00112-of-00113.safetensors",
108
+ "mtp.layers.1.mixer.experts.454.down_proj.weight": "model-00112-of-00113.safetensors",
109
+ "mtp.layers.1.mixer.experts.454.up_proj.weight": "model-00112-of-00113.safetensors",
110
+ "mtp.layers.1.mixer.experts.455.down_proj.weight": "model-00112-of-00113.safetensors",
111
+ "mtp.layers.1.mixer.experts.455.up_proj.weight": "model-00112-of-00113.safetensors",
112
+ "mtp.layers.1.mixer.experts.456.down_proj.weight": "model-00112-of-00113.safetensors",
113
+ "mtp.layers.1.mixer.experts.456.up_proj.weight": "model-00112-of-00113.safetensors",
114
+ "mtp.layers.1.mixer.experts.457.down_proj.weight": "model-00112-of-00113.safetensors",
115
+ "mtp.layers.1.mixer.experts.457.up_proj.weight": "model-00112-of-00113.safetensors",
116
+ "mtp.layers.1.mixer.experts.458.down_proj.weight": "model-00112-of-00113.safetensors",
117
+ "mtp.layers.1.mixer.experts.458.up_proj.weight": "model-00112-of-00113.safetensors",
118
+ "mtp.layers.1.mixer.experts.459.down_proj.weight": "model-00112-of-00113.safetensors",
119
+ "mtp.layers.1.mixer.experts.459.up_proj.weight": "model-00112-of-00113.safetensors",
120
+ "mtp.layers.1.mixer.experts.46.down_proj.weight": "model-00112-of-00113.safetensors",
121
+ "mtp.layers.1.mixer.experts.46.up_proj.weight": "model-00112-of-00113.safetensors",
122
+ "mtp.layers.1.mixer.experts.460.down_proj.weight": "model-00112-of-00113.safetensors",
123
+ "mtp.layers.1.mixer.experts.460.up_proj.weight": "model-00112-of-00113.safetensors",
124
+ "mtp.layers.1.mixer.experts.461.down_proj.weight": "model-00112-of-00113.safetensors",
125
+ "mtp.layers.1.mixer.experts.461.up_proj.weight": "model-00112-of-00113.safetensors",
126
+ "mtp.layers.1.mixer.experts.462.down_proj.weight": "model-00112-of-00113.safetensors",
127
+ "mtp.layers.1.mixer.experts.462.up_proj.weight": "model-00112-of-00113.safetensors",
128
+ "mtp.layers.1.mixer.experts.463.down_proj.weight": "model-00112-of-00113.safetensors",
129
+ "mtp.layers.1.mixer.experts.463.up_proj.weight": "model-00112-of-00113.safetensors",
130
+ "mtp.layers.1.mixer.experts.464.down_proj.weight": "model-00112-of-00113.safetensors",
131
+ "mtp.layers.1.mixer.experts.464.up_proj.weight": "model-00112-of-00113.safetensors",
132
+ "mtp.layers.1.mixer.experts.465.down_proj.weight": "model-00112-of-00113.safetensors",
133
+ "mtp.layers.1.mixer.experts.465.up_proj.weight": "model-00112-of-00113.safetensors",
134
+ "mtp.layers.1.mixer.experts.466.down_proj.weight": "model-00112-of-00113.safetensors",
135
+ "mtp.layers.1.mixer.experts.466.up_proj.weight": "model-00112-of-00113.safetensors",
136
+ "mtp.layers.1.mixer.experts.467.down_proj.weight": "model-00112-of-00113.safetensors",
137
+ "mtp.layers.1.mixer.experts.467.up_proj.weight": "model-00112-of-00113.safetensors",
138
+ "mtp.layers.1.mixer.experts.468.down_proj.weight": "model-00112-of-00113.safetensors",
139
+ "mtp.layers.1.mixer.experts.468.up_proj.weight": "model-00112-of-00113.safetensors",
140
+ "mtp.layers.1.mixer.experts.469.down_proj.weight": "model-00112-of-00113.safetensors",
141
+ "mtp.layers.1.mixer.experts.469.up_proj.weight": "model-00112-of-00113.safetensors",
142
+ "mtp.layers.1.mixer.experts.47.down_proj.weight": "model-00112-of-00113.safetensors",
143
+ "mtp.layers.1.mixer.experts.47.up_proj.weight": "model-00112-of-00113.safetensors",
144
+ "mtp.layers.1.mixer.experts.470.down_proj.weight": "model-00112-of-00113.safetensors",
145
+ "mtp.layers.1.mixer.experts.470.up_proj.weight": "model-00112-of-00113.safetensors",
146
+ "mtp.layers.1.mixer.experts.471.down_proj.weight": "model-00112-of-00113.safetensors",
147
+ "mtp.layers.1.mixer.experts.471.up_proj.weight": "model-00112-of-00113.safetensors",
148
+ "mtp.layers.1.mixer.experts.472.down_proj.weight": "model-00112-of-00113.safetensors",
149
+ "mtp.layers.1.mixer.experts.472.up_proj.weight": "model-00112-of-00113.safetensors",
150
+ "mtp.layers.1.mixer.experts.473.down_proj.weight": "model-00112-of-00113.safetensors",
151
+ "mtp.layers.1.mixer.experts.473.up_proj.weight": "model-00112-of-00113.safetensors",
152
+ "mtp.layers.1.mixer.experts.474.down_proj.weight": "model-00112-of-00113.safetensors",
153
+ "mtp.layers.1.mixer.experts.474.up_proj.weight": "model-00112-of-00113.safetensors",
154
+ "mtp.layers.1.mixer.experts.475.down_proj.weight": "model-00112-of-00113.safetensors",
155
+ "mtp.layers.1.mixer.experts.475.up_proj.weight": "model-00112-of-00113.safetensors",
156
+ "mtp.layers.1.mixer.experts.476.down_proj.weight": "model-00112-of-00113.safetensors",
157
+ "mtp.layers.1.mixer.experts.476.up_proj.weight": "model-00112-of-00113.safetensors",
158
+ "mtp.layers.1.mixer.experts.477.down_proj.weight": "model-00112-of-00113.safetensors",
159
+ "mtp.layers.1.mixer.experts.477.up_proj.weight": "model-00112-of-00113.safetensors",
160
+ "mtp.layers.1.mixer.experts.478.down_proj.weight": "model-00112-of-00113.safetensors",
161
+ "mtp.layers.1.mixer.experts.478.up_proj.weight": "model-00112-of-00113.safetensors",
162
+ "mtp.layers.1.mixer.experts.479.down_proj.weight": "model-00112-of-00113.safetensors",
163
+ "mtp.layers.1.mixer.experts.479.up_proj.weight": "model-00112-of-00113.safetensors",
164
+ "mtp.layers.1.mixer.experts.48.down_proj.weight": "model-00112-of-00113.safetensors",
165
+ "mtp.layers.1.mixer.experts.48.up_proj.weight": "model-00112-of-00113.safetensors",
166
+ "mtp.layers.1.mixer.experts.480.down_proj.weight": "model-00112-of-00113.safetensors",
167
+ "mtp.layers.1.mixer.experts.480.up_proj.weight": "model-00112-of-00113.safetensors",
168
+ "mtp.layers.1.mixer.experts.481.down_proj.weight": "model-00112-of-00113.safetensors",
169
+ "mtp.layers.1.mixer.experts.481.up_proj.weight": "model-00112-of-00113.safetensors",
170
+ "mtp.layers.1.mixer.experts.482.down_proj.weight": "model-00112-of-00113.safetensors",
171
+ "mtp.layers.1.mixer.experts.482.up_proj.weight": "model-00112-of-00113.safetensors",
172
+ "mtp.layers.1.mixer.experts.483.down_proj.weight": "model-00112-of-00113.safetensors",
173
+ "mtp.layers.1.mixer.experts.483.up_proj.weight": "model-00112-of-00113.safetensors",
174
+ "mtp.layers.1.mixer.experts.484.down_proj.weight": "model-00112-of-00113.safetensors",
175
+ "mtp.layers.1.mixer.experts.484.up_proj.weight": "model-00112-of-00113.safetensors",
176
+ "mtp.layers.1.mixer.experts.485.down_proj.weight": "model-00112-of-00113.safetensors",
177
+ "mtp.layers.1.mixer.experts.485.up_proj.weight": "model-00112-of-00113.safetensors",
178
+ "mtp.layers.1.mixer.experts.486.down_proj.weight": "model-00112-of-00113.safetensors",
179
+ "mtp.layers.1.mixer.experts.486.up_proj.weight": "model-00112-of-00113.safetensors",
180
+ "mtp.layers.1.mixer.experts.487.down_proj.weight": "model-00112-of-00113.safetensors",
181
+ "mtp.layers.1.mixer.experts.487.up_proj.weight": "model-00112-of-00113.safetensors",
182
+ "mtp.layers.1.mixer.experts.488.down_proj.weight": "model-00112-of-00113.safetensors",
183
+ "mtp.layers.1.mixer.experts.488.up_proj.weight": "model-00112-of-00113.safetensors",
184
+ "mtp.layers.1.mixer.experts.489.down_proj.weight": "model-00112-of-00113.safetensors",
185
+ "mtp.layers.1.mixer.experts.489.up_proj.weight": "model-00112-of-00113.safetensors",
186
+ "mtp.layers.1.mixer.experts.49.down_proj.weight": "model-00112-of-00113.safetensors",
187
+ "mtp.layers.1.mixer.experts.49.up_proj.weight": "model-00112-of-00113.safetensors",
188
+ "mtp.layers.1.mixer.experts.490.down_proj.weight": "model-00112-of-00113.safetensors",
189
+ "mtp.layers.1.mixer.experts.490.up_proj.weight": "model-00112-of-00113.safetensors",
190
+ "mtp.layers.1.mixer.experts.491.down_proj.weight": "model-00112-of-00113.safetensors",
191
+ "mtp.layers.1.mixer.experts.491.up_proj.weight": "model-00112-of-00113.safetensors",
192
+ "mtp.layers.1.mixer.experts.492.down_proj.weight": "model-00112-of-00113.safetensors",
193
+ "mtp.layers.1.mixer.experts.492.up_proj.weight": "model-00112-of-00113.safetensors",
194
+ "mtp.layers.1.mixer.experts.493.down_proj.weight": "model-00112-of-00113.safetensors",
195
+ "mtp.layers.1.mixer.experts.493.up_proj.weight": "model-00112-of-00113.safetensors",
196
+ "mtp.layers.1.mixer.experts.494.down_proj.weight": "model-00112-of-00113.safetensors",
197
+ "mtp.layers.1.mixer.experts.494.up_proj.weight": "model-00112-of-00113.safetensors",
198
+ "mtp.layers.1.mixer.experts.495.down_proj.weight": "model-00112-of-00113.safetensors",
199
+ "mtp.layers.1.mixer.experts.495.up_proj.weight": "model-00112-of-00113.safetensors",
200
+ "mtp.layers.1.mixer.experts.496.down_proj.weight": "model-00112-of-00113.safetensors",
201
+ "mtp.layers.1.mixer.experts.496.up_proj.weight": "model-00112-of-00113.safetensors",
202
+ "mtp.layers.1.mixer.experts.497.down_proj.weight": "model-00112-of-00113.safetensors",
203
+ "mtp.layers.1.mixer.experts.497.up_proj.weight": "model-00112-of-00113.safetensors",
204
+ "mtp.layers.1.mixer.experts.498.down_proj.weight": "model-00112-of-00113.safetensors",
205
+ "mtp.layers.1.mixer.experts.498.up_proj.weight": "model-00112-of-00113.safetensors",
206
+ "mtp.layers.1.mixer.experts.499.down_proj.weight": "model-00112-of-00113.safetensors",
207
+ "mtp.layers.1.mixer.experts.499.up_proj.weight": "model-00112-of-00113.safetensors",
208
+ "mtp.layers.1.mixer.experts.5.down_proj.weight": "model-00112-of-00113.safetensors",
209
+ "mtp.layers.1.mixer.experts.5.up_proj.weight": "model-00112-of-00113.safetensors",
210
+ "mtp.layers.1.mixer.experts.50.down_proj.weight": "model-00112-of-00113.safetensors",
211
+ "mtp.layers.1.mixer.experts.50.up_proj.weight": "model-00112-of-00113.safetensors",
212
+ "mtp.layers.1.mixer.experts.500.down_proj.weight": "model-00112-of-00113.safetensors",
213
+ "mtp.layers.1.mixer.experts.500.up_proj.weight": "model-00112-of-00113.safetensors",
214
+ "mtp.layers.1.mixer.experts.501.down_proj.weight": "model-00112-of-00113.safetensors",
215
+ "mtp.layers.1.mixer.experts.501.up_proj.weight": "model-00112-of-00113.safetensors",
216
+ "mtp.layers.1.mixer.experts.502.down_proj.weight": "model-00112-of-00113.safetensors",
217
+ "mtp.layers.1.mixer.experts.502.up_proj.weight": "model-00112-of-00113.safetensors",
218
+ "mtp.layers.1.mixer.experts.503.down_proj.weight": "model-00112-of-00113.safetensors",
219
+ "mtp.layers.1.mixer.experts.503.up_proj.weight": "model-00112-of-00113.safetensors",
220
+ "mtp.layers.1.mixer.experts.504.down_proj.weight": "model-00112-of-00113.safetensors",
221
+ "mtp.layers.1.mixer.experts.504.up_proj.weight": "model-00112-of-00113.safetensors",
222
+ "mtp.layers.1.mixer.experts.505.down_proj.weight": "model-00112-of-00113.safetensors",
223
+ "mtp.layers.1.mixer.experts.505.up_proj.weight": "model-00112-of-00113.safetensors",
224
+ "mtp.layers.1.mixer.experts.506.down_proj.weight": "model-00112-of-00113.safetensors",
225
+ "mtp.layers.1.mixer.experts.506.up_proj.weight": "model-00112-of-00113.safetensors",
226
+ "mtp.layers.1.mixer.experts.507.down_proj.weight": "model-00112-of-00113.safetensors",
227
+ "mtp.layers.1.mixer.experts.507.up_proj.weight": "model-00112-of-00113.safetensors",
228
+ "mtp.layers.1.mixer.experts.508.down_proj.weight": "model-00112-of-00113.safetensors",
229
+ "mtp.layers.1.mixer.experts.508.up_proj.weight": "model-00112-of-00113.safetensors",
230
+ "mtp.layers.1.mixer.experts.509.down_proj.weight": "model-00112-of-00113.safetensors",
231
+ "mtp.layers.1.mixer.experts.509.up_proj.weight": "model-00112-of-00113.safetensors",
232
+ "mtp.layers.1.mixer.experts.51.down_proj.weight": "model-00112-of-00113.safetensors",
233
+ "mtp.layers.1.mixer.experts.51.up_proj.weight": "model-00112-of-00113.safetensors",
234
+ "mtp.layers.1.mixer.experts.510.down_proj.weight": "model-00112-of-00113.safetensors",
235
+ "mtp.layers.1.mixer.experts.510.up_proj.weight": "model-00112-of-00113.safetensors",
236
+ "mtp.layers.1.mixer.experts.511.down_proj.weight": "model-00112-of-00113.safetensors",
237
+ "mtp.layers.1.mixer.experts.511.up_proj.weight": "model-00112-of-00113.safetensors",
238
+ "mtp.layers.1.mixer.experts.52.down_proj.weight": "model-00112-of-00113.safetensors",
239
+ "mtp.layers.1.mixer.experts.52.up_proj.weight": "model-00112-of-00113.safetensors",
240
+ "mtp.layers.1.mixer.experts.53.down_proj.weight": "model-00112-of-00113.safetensors",
241
+ "mtp.layers.1.mixer.experts.53.up_proj.weight": "model-00112-of-00113.safetensors",
242
+ "mtp.layers.1.mixer.experts.54.down_proj.weight": "model-00112-of-00113.safetensors",
243
+ "mtp.layers.1.mixer.experts.54.up_proj.weight": "model-00112-of-00113.safetensors"
244
+ }
245
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|im_end|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|im_end|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
ultra_v3_reasoning_parser.py ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from vllm.reasoning.abs_reasoning_parsers import ReasoningParserManager
2
+ from vllm.reasoning.deepseek_r1_reasoning_parser import DeepSeekR1ReasoningParser
3
+
4
+
5
+ @ReasoningParserManager.register_module("ultra_v3")
6
+ class UltraV3ReasoningParser(DeepSeekR1ReasoningParser):
7
+ def extract_reasoning(self, model_output, request):
8
+ reasoning_content, final_content = super().extract_reasoning(
9
+ model_output, request
10
+ )
11
+ if (
12
+ hasattr(request, "chat_template_kwargs")
13
+ and request.chat_template_kwargs
14
+ and (
15
+ request.chat_template_kwargs.get("enable_thinking") is False
16
+ or request.chat_template_kwargs.get("force_nonempty_content") is True
17
+ )
18
+ and final_content is None
19
+ ):
20
+ """
21
+ The original `deepseek_r1` reasoning parser this inherits from will automatically put everything in the reasoning content when it cannot parse out reasoning. This was fine for the DeepSeek R1 model that was not intended to be used without reasoning.
22
+ 1. Since the Nemotron 3 Nano and Super both have thinking off modes modulated by "enable_thinking=false" in the chat template kwargs, this change instead which will properly place the content in cases where there is no thinking enabled via config.
23
+ 2. There are rare cases where the model will output only reasoning without an end-think token `</think>` (e.g. reasoning exceeds max length), which results in empty content returned. End users may want to unilaterally avoid such cases and always have a content response even if the model does not finish its reasoning.
24
+ """
25
+ # Put all nonempty content into the content, rather than return content
26
+ reasoning_content, final_content = None, reasoning_content
27
+
28
+ return reasoning_content, final_content