ZHANGYUXUAN-zR commited on
Commit
8a1a121
·
verified ·
1 Parent(s): f55c930

Add files using upload-large-folder tool

Browse files
Files changed (50) hide show
  1. chat_template.jinja +257 -0
  2. config.json +290 -0
  3. generation_config.json +12 -0
  4. model-00001-of-00120.safetensors +3 -0
  5. model-00002-of-00120.safetensors +3 -0
  6. model-00006-of-00120.safetensors +3 -0
  7. model-00008-of-00120.safetensors +3 -0
  8. model-00012-of-00120.safetensors +3 -0
  9. model-00013-of-00120.safetensors +3 -0
  10. model-00017-of-00120.safetensors +3 -0
  11. model-00018-of-00120.safetensors +3 -0
  12. model-00019-of-00120.safetensors +3 -0
  13. model-00020-of-00120.safetensors +3 -0
  14. model-00022-of-00120.safetensors +3 -0
  15. model-00024-of-00120.safetensors +3 -0
  16. model-00025-of-00120.safetensors +3 -0
  17. model-00026-of-00120.safetensors +3 -0
  18. model-00029-of-00120.safetensors +3 -0
  19. model-00030-of-00120.safetensors +3 -0
  20. model-00031-of-00120.safetensors +3 -0
  21. model-00037-of-00120.safetensors +3 -0
  22. model-00038-of-00120.safetensors +3 -0
  23. model-00039-of-00120.safetensors +3 -0
  24. model-00042-of-00120.safetensors +3 -0
  25. model-00043-of-00120.safetensors +3 -0
  26. model-00044-of-00120.safetensors +3 -0
  27. model-00047-of-00120.safetensors +3 -0
  28. model-00049-of-00120.safetensors +3 -0
  29. model-00052-of-00120.safetensors +3 -0
  30. model-00053-of-00120.safetensors +3 -0
  31. model-00054-of-00120.safetensors +3 -0
  32. model-00055-of-00120.safetensors +3 -0
  33. model-00057-of-00120.safetensors +3 -0
  34. model-00061-of-00120.safetensors +3 -0
  35. model-00064-of-00120.safetensors +3 -0
  36. model-00067-of-00120.safetensors +3 -0
  37. model-00069-of-00120.safetensors +3 -0
  38. model-00071-of-00120.safetensors +3 -0
  39. model-00072-of-00120.safetensors +3 -0
  40. model-00073-of-00120.safetensors +3 -0
  41. model-00076-of-00120.safetensors +3 -0
  42. model-00080-of-00120.safetensors +3 -0
  43. model-00083-of-00120.safetensors +3 -0
  44. model-00100-of-00120.safetensors +3 -0
  45. model-00109-of-00120.safetensors +3 -0
  46. model-00110-of-00120.safetensors +3 -0
  47. model-00116-of-00120.safetensors +3 -0
  48. model.safetensors.index.json +0 -0
  49. processor_config.json +44 -0
  50. tokenizer_config.json +33 -0
chat_template.jinja ADDED
@@ -0,0 +1,257 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [gMASK]<sop>
2
+ {%- set effective_reasoning_effort = reasoning_effort if reasoning_effort is defined and reasoning_effort in ['low', 'high'] else 'max' -%}
3
+ {%- if effective_reasoning_effort is not none -%}<|system|>Reasoning Effort: {{ effective_reasoning_effort | capitalize }}{%- endif -%}
4
+ {%- set clear_thinking = clear_thinking if clear_thinking is defined else false -%}
5
+ {%- if tools -%}
6
+ {%- macro tool_to_json(tool) -%}
7
+ {%- set ns_tool = namespace(first=true) -%}
8
+ {{ '{' -}}
9
+ {%- for k, v in tool.items() -%}
10
+ {%- if k != 'defer_loading' and k != 'strict' -%}
11
+ {%- if not ns_tool.first -%}{{- ', ' -}}{%- endif -%}
12
+ {%- set ns_tool.first = false -%}
13
+ "{{ k }}": {{ v | tojson(ensure_ascii=False) }}
14
+ {%- endif -%}
15
+ {%- endfor -%}
16
+ {{- '}' -}}
17
+ {%- endmacro -%}
18
+ {%- macro tool_references_to_response(refs) -%}
19
+ {{- '<tool_response><tools>\n' -}}
20
+ {%- for tr in refs -%}
21
+ {%- for tool in tools -%}
22
+ {%- if 'function' in tool -%}
23
+ {%- set tool = tool['function'] -%}
24
+ {%- endif -%}
25
+ {%- if tool.name == tr.name -%}
26
+ {{- tool_to_json(tool) + '\n' -}}
27
+ {%- endif -%}
28
+ {%- endfor -%}
29
+ {%- endfor -%}
30
+ {{- '</tools></tool_response>' -}}
31
+ {%- endmacro -%}
32
+ <|system|>
33
+ # Tools
34
+
35
+ You may call one or more functions to assist with the user query.
36
+
37
+ You are provided with function signatures within <tools></tools> XML tags:
38
+ <tools>
39
+ {% for tool in tools %}
40
+ {%- if 'function' in tool -%}
41
+ {%- set tool = tool['function'] -%}
42
+ {%- endif -%}
43
+ {% if tool.defer_loading is not defined or not tool.defer_loading %}
44
+ {{ tool_to_json(tool) }}
45
+ {% endif %}
46
+ {% endfor %}
47
+ </tools>
48
+
49
+ For each function call, output the function name and arguments within the following XML format:
50
+ <tool_call>{function-name}<arg_key>{arg-key-1}</arg_key><arg_value>{arg-value-1}</arg_value><arg_key>{arg-key-2}</arg_key><arg_value>{arg-value-2}</arg_value>...</tool_call>{%- endif -%}
51
+ {%- macro emit_image() -%}<|begin_of_image|><|image|><|end_of_image|>{%- endmacro -%}
52
+ {%- macro emit_video() -%}<|begin_of_video|><|video|><|end_of_video|>{%- endmacro -%}
53
+ {%- macro emit_audio() -%}<|begin_of_audio|><|end_of_audio|>{%- endmacro -%}
54
+ {%- macro visible_text(content) -%}
55
+ {%- if content is string -%}
56
+ {{- content -}}
57
+ {%- elif content is iterable and content is not mapping -%}
58
+ {%- for item in content -%}
59
+ {%- if item is mapping and item.type == 'text' -%}
60
+ {{- item.text -}}
61
+ {%- elif item is string -%}
62
+ {{- item -}}
63
+ {%- elif item is mapping and item.type in ['image', 'image_url'] -%}
64
+ {{- emit_image() -}}
65
+ {%- elif item is mapping and item.type in ['video', 'video_url'] -%}
66
+ {{- emit_video() -}}
67
+ {%- elif item is mapping and item.type in ['audio', 'audio_url', 'input_audio'] -%}
68
+ {{- emit_audio() -}}
69
+ {%- endif -%}
70
+ {%- endfor -%}
71
+ {%- else -%}
72
+ {{- content }}
73
+ {%- endif -%}
74
+ {%- endmacro -%}
75
+ {%- macro tool_response(text) -%}
76
+ {{- '<tool_response>' + text + '</tool_response>' -}}
77
+ {%- endmacro -%}
78
+ {%- macro render_tool_response(m) -%}
79
+ {%- if m.content is string -%}
80
+ {{- tool_response(m.content) -}}
81
+ {%- elif m.content and m.content is not mapping and m.content.0.type == "tool_reference" -%}
82
+ {{- tool_references_to_response(m.content) -}}
83
+ {%- elif is_list_of_outputs(m) -%}
84
+ {%- for tr in m.content -%}
85
+ {%- if tr.output is iterable and tr.output is not string and tr.output is not mapping and tr.output and tr.output.0.type == "tool_reference" -%}
86
+ {{- tool_references_to_response(tr.output) -}}
87
+ {%- else -%}
88
+ {{- tool_response(visible_text(tr.output)) -}}
89
+ {%- endif -%}
90
+ {%- endfor -%}
91
+ {%- else -%}
92
+ {{- tool_response(visible_text(m.content)) -}}
93
+ {%- endif -%}
94
+ {%- endmacro -%}
95
+ {%- macro id_of(obj) -%}
96
+ {%- if obj.tool_call_id -%}
97
+ {{- obj.tool_call_id -}}
98
+ {%- elif obj.id -%}
99
+ {{- obj.id -}}
100
+ {%- endif -%}
101
+ {%- endmacro -%}
102
+ {%- macro is_list_of_outputs(m) -%}
103
+ {%- if m.content and m.content.0.output is defined -%}1{%- endif -%}
104
+ {%- endmacro -%}
105
+ {%- macro has_dup_tool_result_id(lo, hi, target) -%}
106
+ {%- set ns_cnt = namespace(n=0) -%}
107
+ {%- for k in range(lo, hi + 1) -%}
108
+ {%- set m = messages[k] -%}
109
+ {%- if is_list_of_outputs(m) -%}
110
+ {%- for entry in m.content -%}
111
+ {%- if id_of(entry) == target -%}
112
+ {%- set ns_cnt.n = ns_cnt.n + 1 -%}
113
+ {%- endif -%}
114
+ {%- endfor -%}
115
+ {%- elif id_of(m) == target -%}
116
+ {%- set ns_cnt.n = ns_cnt.n + 1 -%}
117
+ {%- endif -%}
118
+ {%- if ns_cnt.n > 1 -%}{%- break -%}{%- endif -%}
119
+ {%- endfor -%}
120
+ {%- if ns_cnt.n > 1 -%}1{%- endif -%}
121
+ {%- endmacro -%}
122
+ {%- macro tc_id_exists(tcs, target) -%}
123
+ {%- set ns_f = namespace(found=false) -%}
124
+ {%- for tc in tcs -%}
125
+ {%- if id_of(tc) == target -%}
126
+ {%- set ns_f.found = true -%}
127
+ {%- break -%}
128
+ {%- endif -%}
129
+ {%- endfor -%}
130
+ {%- if ns_f.found -%}1{%- endif -%}
131
+ {%- endmacro -%}
132
+ {%- set ns = namespace(last_user_index=-1) -%}
133
+ {%- for m in messages %}
134
+ {%- if m.role == 'user' %}
135
+ {%- set ns.last_user_index = loop.index0 -%}
136
+ {%- endif %}
137
+ {%- endfor %}
138
+ {%- for m in messages -%}
139
+ {%- if m.role == 'user' -%}<|user|>{{ visible_text(m.content) }}
140
+ {%- elif m.role == 'assistant' -%}
141
+ <|assistant|>
142
+ {%- set content = visible_text(m.content) %}
143
+ {%- if m.reasoning_content is string %}
144
+ {%- set reasoning_content = m.reasoning_content %}
145
+ {%- elif '</think>' in content %}
146
+ {%- set reasoning_content = content.split('</think>')[0].split('<think>')[-1] %}
147
+ {%- set content = content.split('</think>')[-1] %}
148
+ {%- endif %}
149
+ {%- if (not clear_thinking or loop.index0 > ns.last_user_index) and reasoning_content is defined -%}
150
+ {{ '<think>' + reasoning_content + '</think>'}}
151
+ {%- else -%}
152
+ {{ '<think></think>' }}
153
+ {%- endif -%}
154
+ {%- if content.strip() -%}
155
+ {{ content.strip() }}
156
+ {%- endif -%}
157
+ {% if m.tool_calls %}
158
+ {% for tc in m.tool_calls %}
159
+ {%- if tc.function %}
160
+ {%- set tc = tc.function %}
161
+ {%- endif %}
162
+ {{- '<tool_call>' + tc.name -}}
163
+ {% set _args = tc.arguments %}{% for k, v in _args.items() %}<arg_key>{{ k }}</arg_key><arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>{% endfor %}</tool_call>{% endfor %}
164
+ {% endif %}
165
+ {%- elif m.role == 'tool' -%}
166
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
167
+ {{- '<|observation|>' -}}
168
+ {%- set block_start = loop.index0 -%}
169
+ {%- set ns_blk = namespace(end=block_start) -%}
170
+ {%- for j in range(block_start, messages|length) -%}
171
+ {%- if messages[j].role == 'tool' -%}
172
+ {%- set ns_blk.end = j -%}
173
+ {%- else -%}
174
+ {%- break -%}
175
+ {%- endif -%}
176
+ {%- endfor -%}
177
+ {%- set ns_a = namespace(tool_calls=none) -%}
178
+ {%- if block_start > 0 and messages[block_start - 1].role == 'assistant' and messages[block_start - 1].tool_calls -%}
179
+ {%- set ns_a.tool_calls = messages[block_start - 1].tool_calls -%}
180
+ {%- endif -%}
181
+ {%- set ns_chk = namespace(can_sort=true) -%}
182
+ {%- if not ns_a.tool_calls -%}
183
+ {%- set ns_chk.can_sort = false -%}
184
+ {%- else -%}
185
+ {%- for k in range(block_start, ns_blk.end + 1) -%}
186
+ {%- set m = messages[k] -%}
187
+ {%- if is_list_of_outputs(m) -%}
188
+ {%- for entry in m.content -%}
189
+ {%- set eid = id_of(entry) -%}
190
+ {%- if not eid -%}
191
+ {%- set ns_chk.can_sort = false -%}
192
+ {%- elif has_dup_tool_result_id(block_start, ns_blk.end, eid) -%}
193
+ {%- set ns_chk.can_sort = false -%}
194
+ {%- elif not tc_id_exists(ns_a.tool_calls, eid) -%}
195
+ {%- set ns_chk.can_sort = false -%}
196
+ {%- endif -%}
197
+ {%- endfor -%}
198
+ {%- else -%}
199
+ {%- set tk_id = id_of(m) -%}
200
+ {%- if not tk_id -%}
201
+ {%- set ns_chk.can_sort = false -%}
202
+ {%- elif has_dup_tool_result_id(block_start, ns_blk.end, tk_id) -%}
203
+ {%- set ns_chk.can_sort = false -%}
204
+ {%- elif not tc_id_exists(ns_a.tool_calls, tk_id) -%}
205
+ {%- set ns_chk.can_sort = false -%}
206
+ {%- endif -%}
207
+ {%- endif -%}
208
+ {%- endfor -%}
209
+ {%- for i in range(ns_a.tool_calls | length) -%}
210
+ {%- set tc_id = id_of(ns_a.tool_calls[i]) -%}
211
+ {%- if not tc_id -%}
212
+ {%- set ns_chk.can_sort = false -%}
213
+ {%- endif -%}
214
+ {%- for j in range(i + 1, ns_a.tool_calls | length) -%}
215
+ {%- if id_of(ns_a.tool_calls[j]) == tc_id -%}
216
+ {%- set ns_chk.can_sort = false -%}
217
+ {%- endif -%}
218
+ {%- endfor -%}
219
+ {%- endfor -%}
220
+ {%- endif -%}
221
+ {%- if ns_chk.can_sort -%}
222
+ {%- for tc in ns_a.tool_calls -%}
223
+ {%- set tc_id = id_of(tc) -%}
224
+ {%- for k in range(block_start, ns_blk.end + 1) -%}
225
+ {%- set m = messages[k] -%}
226
+ {%- if is_list_of_outputs(m) -%}
227
+ {%- for entry in m.content -%}
228
+ {%- set eid = id_of(entry) -%}
229
+ {%- if eid == tc_id -%}
230
+ {%- if entry.output is iterable and entry.output is not string and entry.output is not mapping and entry.output and entry.output.0.type == "tool_reference" -%}
231
+ {{- tool_references_to_response(entry.output) -}}
232
+ {%- else -%}
233
+ {{- tool_response(visible_text(entry.output)) -}}
234
+ {%- endif -%}
235
+ {%- endif -%}
236
+ {%- endfor -%}
237
+ {%- else -%}
238
+ {%- set tk_id = id_of(m) -%}
239
+ {%- if tk_id == tc_id -%}
240
+ {{- render_tool_response(m) -}}
241
+ {%- endif -%}
242
+ {%- endif -%}
243
+ {%- endfor -%}
244
+ {%- endfor -%}
245
+ {%- else -%}
246
+ {%- for k in range(block_start, ns_blk.end + 1) -%}
247
+ {{- render_tool_response(messages[k]) -}}
248
+ {%- endfor -%}
249
+ {%- endif -%}
250
+ {% endif -%}
251
+ {%- elif m.role == 'system' -%}
252
+ <|system|>{{ visible_text(m.content) }}
253
+ {%- endif -%}
254
+ {%- endfor -%}
255
+ {%- if add_generation_prompt -%}
256
+ <|assistant|>{{- '<think>' -}}
257
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,290 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Glm5NextForConditionalGeneration"
4
+ ],
5
+ "text_config": {
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": [
10
+ 154820,
11
+ 154827,
12
+ 154829
13
+ ],
14
+ "first_k_dense_replace": 3,
15
+ "hc_eps": 1e-06,
16
+ "hc_mult": 4,
17
+ "hc_sinkhorn_iters": 20,
18
+ "head_dim": 0,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 4096,
21
+ "index_head_dim": 128,
22
+ "index_kpool": 4,
23
+ "index_kpool_always_select_tail": true,
24
+ "index_kpool_compress": true,
25
+ "index_n_heads": 32,
26
+ "index_topk": 2048,
27
+ "index_share_for_mtp_iteration": true,
28
+ "indexer_rope_interleave": true,
29
+ "indexer_types": [
30
+ "full",
31
+ "full",
32
+ "full",
33
+ "full",
34
+ "full",
35
+ "full",
36
+ "full",
37
+ "full",
38
+ "full",
39
+ "full",
40
+ "full",
41
+ "full",
42
+ "full",
43
+ "full",
44
+ "full",
45
+ "full",
46
+ "full",
47
+ "full",
48
+ "full",
49
+ "full",
50
+ "full",
51
+ "full",
52
+ "full",
53
+ "full",
54
+ "full",
55
+ "full",
56
+ "full",
57
+ "full",
58
+ "full",
59
+ "full",
60
+ "full",
61
+ "full",
62
+ "full",
63
+ "full",
64
+ "full",
65
+ "full",
66
+ "full",
67
+ "full",
68
+ "full",
69
+ "full",
70
+ "full",
71
+ "full",
72
+ "full",
73
+ "full",
74
+ "full"
75
+ ],
76
+ "initializer_range": 0.02,
77
+ "intermediate_size": 12288,
78
+ "kv_lora_rank": 512,
79
+ "layer_types": [
80
+ "linear_attention",
81
+ "linear_attention",
82
+ "linear_attention",
83
+ "deepseek_sparse_attention",
84
+ "linear_attention",
85
+ "linear_attention",
86
+ "linear_attention",
87
+ "deepseek_sparse_attention",
88
+ "linear_attention",
89
+ "linear_attention",
90
+ "linear_attention",
91
+ "deepseek_sparse_attention",
92
+ "linear_attention",
93
+ "linear_attention",
94
+ "linear_attention",
95
+ "deepseek_sparse_attention",
96
+ "linear_attention",
97
+ "linear_attention",
98
+ "linear_attention",
99
+ "deepseek_sparse_attention",
100
+ "linear_attention",
101
+ "linear_attention",
102
+ "linear_attention",
103
+ "deepseek_sparse_attention",
104
+ "linear_attention",
105
+ "linear_attention",
106
+ "linear_attention",
107
+ "deepseek_sparse_attention",
108
+ "linear_attention",
109
+ "linear_attention",
110
+ "linear_attention",
111
+ "deepseek_sparse_attention",
112
+ "linear_attention",
113
+ "linear_attention",
114
+ "linear_attention",
115
+ "deepseek_sparse_attention",
116
+ "linear_attention",
117
+ "linear_attention",
118
+ "linear_attention",
119
+ "deepseek_sparse_attention",
120
+ "linear_attention",
121
+ "linear_attention",
122
+ "linear_attention",
123
+ "deepseek_sparse_attention",
124
+ "linear_attention"
125
+ ],
126
+ "linear_attn_config": {
127
+ "num_heads": 64,
128
+ "gate_lower_bound": -5.0,
129
+ "head_dim": 128,
130
+ "short_conv_kernel_size": 4,
131
+ "kda_layers": [
132
+ 0,
133
+ 1,
134
+ 2,
135
+ 4,
136
+ 5,
137
+ 6,
138
+ 8,
139
+ 9,
140
+ 10,
141
+ 12,
142
+ 13,
143
+ 14,
144
+ 16,
145
+ 17,
146
+ 18,
147
+ 20,
148
+ 21,
149
+ 22,
150
+ 24,
151
+ 25,
152
+ 26,
153
+ 28,
154
+ 29,
155
+ 30,
156
+ 32,
157
+ 33,
158
+ 34,
159
+ 36,
160
+ 37,
161
+ 38,
162
+ 40,
163
+ 41,
164
+ 42,
165
+ 44
166
+ ],
167
+ "full_attn_layers": [
168
+ 3,
169
+ 7,
170
+ 11,
171
+ 15,
172
+ 19,
173
+ 23,
174
+ 27,
175
+ 31,
176
+ 35,
177
+ 39,
178
+ 43
179
+ ]
180
+ },
181
+ "max_position_embeddings": 1048576,
182
+ "mhc": true,
183
+ "mla_use_nope": true,
184
+ "mlp_layer_types": [
185
+ "dense",
186
+ "dense",
187
+ "dense",
188
+ "sparse",
189
+ "sparse",
190
+ "sparse",
191
+ "sparse",
192
+ "sparse",
193
+ "sparse",
194
+ "sparse",
195
+ "sparse",
196
+ "sparse",
197
+ "sparse",
198
+ "sparse",
199
+ "sparse",
200
+ "sparse",
201
+ "sparse",
202
+ "sparse",
203
+ "sparse",
204
+ "sparse",
205
+ "sparse",
206
+ "sparse",
207
+ "sparse",
208
+ "sparse",
209
+ "sparse",
210
+ "sparse",
211
+ "sparse",
212
+ "sparse",
213
+ "sparse",
214
+ "sparse",
215
+ "sparse",
216
+ "sparse",
217
+ "sparse",
218
+ "sparse",
219
+ "sparse",
220
+ "sparse",
221
+ "sparse",
222
+ "sparse",
223
+ "sparse",
224
+ "sparse",
225
+ "sparse",
226
+ "sparse",
227
+ "sparse",
228
+ "sparse",
229
+ "sparse"
230
+ ],
231
+ "model_type": "glm5_next_text",
232
+ "moe_intermediate_size": 2048,
233
+ "moe_router_dtype": "float32",
234
+ "n_group": 1,
235
+ "n_routed_experts": 288,
236
+ "n_shared_experts": 1,
237
+ "norm_topk_prob": true,
238
+ "num_attention_heads": 64,
239
+ "num_experts_per_tok": 8,
240
+ "num_hidden_layers": 45,
241
+ "num_key_value_heads": 64,
242
+ "num_nextn_predict_layers": 1,
243
+ "output_router_logits": false,
244
+ "pad_token_id": 154820,
245
+ "q_lora_rank": 1536,
246
+ "qk_head_dim": 256,
247
+ "qk_nope_head_dim": 256,
248
+ "qk_rope_head_dim": 0,
249
+ "rms_norm_eps": 1e-05,
250
+ "routed_scaling_factor": 2.5,
251
+ "router_aux_loss_coef": 0.001,
252
+ "scoring_func": "sigmoid",
253
+ "swiglu_limit": 10.0,
254
+ "tie_word_embeddings": false,
255
+ "topk_group": 1,
256
+ "topk_method": "noaux_tc",
257
+ "use_cache": true,
258
+ "v_head_dim": 256,
259
+ "vocab_size": 154880
260
+ },
261
+ "vision_config": {
262
+ "attention_bias": true,
263
+ "attention_dropout": 0.0,
264
+ "depth": 24,
265
+ "hidden_act": "silu",
266
+ "hidden_size": 1024,
267
+ "image_size": 448,
268
+ "in_channels": 3,
269
+ "initializer_range": 0.02,
270
+ "intermediate_size": 4096,
271
+ "model_type": "glm5_next_vision",
272
+ "num_heads": 16,
273
+ "out_hidden_size": 4096,
274
+ "patch_size": 14,
275
+ "projection_intermediate_size": 10240,
276
+ "rms_norm_eps": 1e-05,
277
+ "spatial_merge_size": 2,
278
+ "swiglu_limit": 10.0,
279
+ "temporal_patch_size": 2
280
+ },
281
+ "image_token_id": 154854,
282
+ "video_token_id": 154855,
283
+ "image_start_token_id": 154830,
284
+ "image_end_token_id": 154831,
285
+ "video_start_token_id": 154832,
286
+ "video_end_token_id": 154833,
287
+ "tie_word_embeddings": false,
288
+ "model_type": "glm5_next",
289
+ "transformers_version": "5.16.0"
290
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": [
4
+ 154820,
5
+ 154827,
6
+ 154829
7
+ ],
8
+ "pad_token_id": 154820,
9
+ "temperature": 1.0,
10
+ "top_p": 0.95,
11
+ "transformers_version": "5.16.0"
12
+ }
model-00001-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:235676346453d461485e6be4677d03efe7bc72fc6e2f63e82e8c289e89bdbbbc
3
+ size 5358221856
model-00002-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9a579db4b6673a12b42d1eea6c716b07cacd05eb9dc2055deb4a6b36a5c9547
3
+ size 5368754184
model-00006-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e19c52baa443dccf8ec49204d0f39e3cc54d9ad25487b6249974182c701b6787
3
+ size 5368754008
model-00008-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5be585f6548ea3188b4aa2af4578fa49c4be30f488909b104fac8c5afa16e25
3
+ size 5368754192
model-00012-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d87a818732bb1d148627e2859bea7032922f56cc9fded4d7a7adef972e56ad7
3
+ size 5362973008
model-00013-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98ab21dfd26d9f7eec1e5a191ff5530c4fc0f30be73964cdb5d873fcd5dd33df
3
+ size 5368754192
model-00017-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54fd5dc4368c77a58ae398a6981936ab22c14e3653401ca71c69a60df3b4b316
3
+ size 5368754024
model-00018-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96914257eda1ff08d39bf8bbe361846d6affc640a8aa70f8b95ba41f69de154c
3
+ size 5354098280
model-00019-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1d0ce28fd876355ca2bc79dd503ce77d6ae4594d8211695dc953fe9c219b48f
3
+ size 5368754184
model-00020-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c92e988143d426cc42ab8f956808851fe2662be2c6c9dc6705278a093f010559
3
+ size 5344548784
model-00022-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4eae929329f0b114dc9a7271dbe3584572abc26c62c185944e4da29faff41fe
3
+ size 5368754184
model-00024-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38445ed5d28886018629c414e08de37c160b27bd92eb572065cbfaf1318061cb
3
+ size 5368754184
model-00025-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69d1f2d1d57fe3176bfe1140e61db929c1f433ec402d896260fc19b2bd63a095
3
+ size 5368754120
model-00026-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6edab6ba31fe288596296d03616de97b3a5049816ed9e87028818f956872b1b
3
+ size 5362973064
model-00029-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56ef332e7e31bc0c0412c442a343d57e6a9db52679235712f040f5e8209a4845
3
+ size 5362731784
model-00030-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf27db4a2ec987f8e40ce13de72b376efd96ae54391faf87e26a81b64f8ee09f
3
+ size 5368754192
model-00031-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bca28a2651d2ff2fc0e91f67ac278b8191e8409d58d5a6d91edc4aba7c8828bc
3
+ size 5368754000
model-00037-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad04a34fa52e3862324c242c7475c7d1ae04e8d42c286d7d132b67b59d72da27
3
+ size 5362973016
model-00038-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5719298b0c75ec1781644d0f5e84d769676663abd75639eb6516fe4776b6a411
3
+ size 5368754184
model-00039-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a511ae0fb183da66bf6e63d39cddb5aeefd64bbae3652cde2f528439429ba6c6
3
+ size 5368754088
model-00042-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8248acd87af1ef78341c480e73e600e93586010d35bfc0c1e8ccb8c0f5a58fc
3
+ size 5368754008
model-00043-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66a0a4e0c9294e74569583c76114ad9894cb02b3146fbd56a5e9a0b95a427e2a
3
+ size 5362973160
model-00044-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28cc935cd8fdc66623fead64d5f9b89e9313bb2c9f16854b74a456a726bd757c
3
+ size 5368754192
model-00047-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5a3c9c94f788dd2e5bc69c593b7ed79c2b05dc9c1b350d71950ca96eef59a74
3
+ size 5368754176
model-00049-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b3f0f9e948a2053b5e465efaa10d69a55a2a786755db1a91b0bb54052217c3f
3
+ size 5368754192
model-00052-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18258b8b5c2f6d25a78b0e45a46654bcb499644931b69f461eb0ae985c4f0c32
3
+ size 5368754184
model-00053-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8132f67d7280f9397c5559d4e1efd760b7a9684defe9e3b3fd9e6c543e6ba395
3
+ size 5368754032
model-00054-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7de5a17d08e9b51327624acc51fcf6c284a6aeef698aefa6c2c48e09f6843c3
3
+ size 5362973152
model-00055-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cefe677413f92a30d0afb1d9d6d0ed5b14ee805ebde2efc6a45e103a34c605a
3
+ size 5368754184
model-00057-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:096a5ea11d64dcb97e2c13dfb5f584f1ab3bf43b7eeb1a024c1499df1154fb31
3
+ size 5353688600
model-00061-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14ce4edaf61c436b19d8ee0e7745de2c886dd2117b78f342f98dbeb9c5e9ab9d
3
+ size 5368754120
model-00064-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4abd3f2959cf1a7f0bff08800d6963c5dfe6ec1b867619ee8d541e1cfe4d0e23
3
+ size 5368754040
model-00067-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33c37366ea184bcb7b59e4b30bd927c1613ec1e199bd09db989bb7cd26a6a4ce
3
+ size 5368753968
model-00069-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e9b358eb8198bc45b198182e8b06d1455017bd3369a4a07bc62e5ff9869a0f5
3
+ size 5368754184
model-00071-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec869b56c2529d9ee2a75ec94848befd079dc881f42db91fdfee2d6d0a18de35
3
+ size 5368754192
model-00072-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66d61dc7290b132fda24816effbd9cbd56728e5acc191d81ada581bca6edd512
3
+ size 5368754136
model-00073-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed0381afb06c0a8b6437109df32161ba1294a5a2f3ba09cbe8c46f288c8474bd
3
+ size 5362973048
model-00076-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d1c0d9ba59e4b8e56ce1a29fe2f0e64e0d37660af359e7e552ad429f2b7f0dd
3
+ size 5354098248
model-00080-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5806c66fdad6a0c5233d18b75ad2d5a7017ed40473fd9f921cf8a617766c441
3
+ size 5368754192
model-00083-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddd86158a56434d2f9288bb9441a8679d418b12cc025f6cce436cb857095a709
3
+ size 5368754152
model-00100-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43c41b78f85cc6ebf9d84ecec8f4eaf922e09f0c30c90e693042c4d77913a8bf
3
+ size 5368754008
model-00109-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff79a0d3f8a641e74eafc78b91251118c5db278de3cb5545bccbfdb60fc1c0b7
3
+ size 5362972744
model-00110-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73e4c5569de0765184f294017fc4270cd6ec4077ba5c5a45debcefd556966d1b
3
+ size 5368753872
model-00116-of-00120.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91cecd3d34582fcaff0056be2ee95b01a0a27abfe7f15ef55a758664e03f13c9
3
+ size 5368753872
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
processor_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "do_rescale": true,
4
+ "patch_expand_factor": 1,
5
+ "merge_size": 2,
6
+ "image_mean": [
7
+ 0.48145466,
8
+ 0.4578275,
9
+ 0.40821073
10
+ ],
11
+ "image_std": [
12
+ 0.26862954,
13
+ 0.26130258,
14
+ 0.27577711
15
+ ],
16
+ "temporal_patch_size": 2,
17
+ "patch_size": 14,
18
+ "min_image_tokens": 16,
19
+ "max_image_tokens": 8000,
20
+ "image_processor_type": "Glm5NextImageProcessor"
21
+ },
22
+ "video_processor": {
23
+ "do_rescale": true,
24
+ "video_processor_type": "Glm5NextVideoProcessor",
25
+ "patch_expand_factor": 1,
26
+ "merge_size": 2,
27
+ "image_mean": [
28
+ 0.48145466,
29
+ 0.4578275,
30
+ 0.40821073
31
+ ],
32
+ "image_std": [
33
+ 0.26862954,
34
+ 0.26130258,
35
+ 0.27577711
36
+ ],
37
+ "temporal_patch_size": 2,
38
+ "patch_size": 14,
39
+ "min_image_tokens": 16,
40
+ "max_image_tokens": 240000,
41
+ "fps": 2
42
+ },
43
+ "processor_class": "Glm5NextProcessor"
44
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": false,
4
+ "do_lower_case": false,
5
+ "eos_token": "<|endoftext|>",
6
+ "extra_special_tokens": [
7
+ "<|endoftext|>",
8
+ "[MASK]",
9
+ "[gMASK]",
10
+ "[sMASK]",
11
+ "<sop>",
12
+ "<eop>",
13
+ "<|system|>",
14
+ "<|user|>",
15
+ "<|assistant|>",
16
+ "<|observation|>",
17
+ "<|begin_of_image|>",
18
+ "<|end_of_image|>",
19
+ "<|begin_of_video|>",
20
+ "<|end_of_video|>",
21
+ "<|begin_of_audio|>",
22
+ "<|end_of_audio|>",
23
+ "<|begin_of_transcription|>",
24
+ "<|end_of_transcription|>"
25
+ ],
26
+ "is_local": true,
27
+ "model_max_length": 1048576,
28
+ "model_specific_special_tokens": {},
29
+ "pad_token": "<|endoftext|>",
30
+ "padding_side": "left",
31
+ "remove_space": false,
32
+ "tokenizer_class": "TokenizersBackend"
33
+ }