Hugging Face's logo

trl-internal-testing
/
tiny-Glm5NextForConditionalGeneration

Image-Text-to-Text
Transformers
Safetensors
glm5_next
trl
conversational
fp8
Model card Files
xet
 Community
albertvillanova HF Staff commited on
Commit
6099d75
·
verified ·
1 Parent(s): 141fb62

Upload Glm5NextForConditionalGeneration

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. README.md +9 -0
  3. chat_template.jinja +257 -0
  4. config.json +1635 -0
  5. generation_config.json +11 -0
  6. model.safetensors +3 -0
  7. processor_config.json +67 -0
  8. tokenizer.json +3 -0
  9. tokenizer_config.json +35 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ 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
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - trl
5
+ ---
6
+
7
+ # Tiny Glm5NextForConditionalGeneration
8
+
9
+ This is a minimal model built for unit tests in the [TRL](https://github.com/huggingface/trl) library.
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
+ {%- elif content is not none -%}
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, res=[], mk=[], ek=[], tcs=[]) -%}
182
+ {%- for k in range(block_start, ns_blk.end + 1) -%}
183
+ {%- set m = messages[k] -%}
184
+ {%- if is_list_of_outputs(m) -%}
185
+ {%- for entry in m.content -%}
186
+ {%- set ns_chk.res = ns_chk.res + [id_of(entry)] -%}
187
+ {%- set ns_chk.mk = ns_chk.mk + [k] -%}
188
+ {%- set ns_chk.ek = ns_chk.ek + [loop.index0] -%}
189
+ {%- endfor -%}
190
+ {%- else -%}
191
+ {%- set ns_chk.res = ns_chk.res + [id_of(m)] -%}
192
+ {%- set ns_chk.mk = ns_chk.mk + [k] -%}
193
+ {%- set ns_chk.ek = ns_chk.ek + [-1] -%}
194
+ {%- endif -%}
195
+ {%- endfor -%}
196
+ {%- if not ns_a.tool_calls -%}
197
+ {%- set ns_chk.can_sort = false -%}
198
+ {%- else -%}
199
+ {%- for tc in ns_a.tool_calls -%}
200
+ {%- set ns_chk.tcs = ns_chk.tcs + [id_of(tc)] -%}
201
+ {%- endfor -%}
202
+ {%- if '' in ns_chk.res or '' in ns_chk.tcs -%}
203
+ {%- set ns_chk.can_sort = false -%}
204
+ {%- elif ns_chk.res | unique(case_sensitive=true) | list | length != ns_chk.res | length -%}
205
+ {%- set ns_chk.can_sort = false -%}
206
+ {%- elif ns_chk.tcs | unique(case_sensitive=true) | list | length != ns_chk.tcs | length -%}
207
+ {%- set ns_chk.can_sort = false -%}
208
+ {%- else -%}
209
+ {%- for eid in ns_chk.res -%}
210
+ {%- if eid not in ns_chk.tcs -%}
211
+ {%- set ns_chk.can_sort = false -%}
212
+ {%- break -%}
213
+ {%- endif -%}
214
+ {%- endfor -%}
215
+ {%- endif -%}
216
+ {%- endif -%}
217
+ {%- if ns_chk.can_sort -%}
218
+ {%- set ns_ord = namespace(res=[], tcs=[], j=0, pairs=[]) -%}
219
+ {%- for eid in ns_chk.res -%}
220
+ {%- set ns_ord.res = ns_ord.res + [{'id': eid, 'p': loop.index0}] -%}
221
+ {%- endfor -%}
222
+ {%- for tc_id in ns_chk.tcs -%}
223
+ {%- set ns_ord.tcs = ns_ord.tcs + [{'id': tc_id, 'i': loop.index0}] -%}
224
+ {%- endfor -%}
225
+ {%- set ns_ord.res = ns_ord.res | sort(attribute='id', case_sensitive=true) -%}
226
+ {%- for tc in ns_ord.tcs | sort(attribute='id', case_sensitive=true) -%}
227
+ {%- if ns_ord.j < ns_ord.res | length and ns_ord.res[ns_ord.j].id == tc.id -%}
228
+ {%- set ns_ord.pairs = ns_ord.pairs + [{'i': tc.i, 'p': ns_ord.res[ns_ord.j].p}] -%}
229
+ {%- set ns_ord.j = ns_ord.j + 1 -%}
230
+ {%- endif -%}
231
+ {%- endfor -%}
232
+ {%- for pair in ns_ord.pairs | sort(attribute='i') -%}
233
+ {%- set m = messages[ns_chk.mk[pair.p]] -%}
234
+ {%- if ns_chk.ek[pair.p] >= 0 -%}
235
+ {%- set entry = m.content[ns_chk.ek[pair.p]] -%}
236
+ {%- 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" -%}
237
+ {{- tool_references_to_response(entry.output) -}}
238
+ {%- else -%}
239
+ {{- tool_response(visible_text(entry.output)) -}}
240
+ {%- endif -%}
241
+ {%- else -%}
242
+ {{- render_tool_response(m) -}}
243
+ {%- endif -%}
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 +1,1635 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Glm5NextForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "image_end_token_id": 154831,
7
+ "image_start_token_id": 154830,
8
+ "image_token_id": 154854,
9
+ "model_type": "glm5_next",
10
+ "quantization_config": {
11
+ "activation_scheme": "dynamic",
12
+ "dequantize": false,
13
+ "modules_to_convert": null,
14
+ "modules_to_not_convert": [
15
+ "attn_mha",
16
+ "attn_mqa",
17
+ "dt_bias",
18
+ "hyper_connection",
19
+ "lm_head",
20
+ "mapping_proj",
21
+ "model.embed_tokens",
22
+ "model.layers.0.attn_hc.base",
23
+ "model.layers.0.attn_hc.fn",
24
+ "model.layers.0.attn_hc.scale",
25
+ "model.layers.0.ffn_hc.base",
26
+ "model.layers.0.ffn_hc.fn",
27
+ "model.layers.0.ffn_hc.scale",
28
+ "model.layers.0.input_layernorm",
29
+ "model.layers.0.post_attention_layernorm",
30
+ "model.layers.0.self_attn.forget_gate.A_log",
31
+ "model.layers.0.self_attn.b_proj",
32
+ "model.layers.0.self_attn.forget_gate.dt_bias",
33
+ "model.layers.0.self_attn.f_a_proj",
34
+ "model.layers.0.self_attn.f_b_proj",
35
+ "model.layers.0.self_attn.fused_qkvbfg_a_proj",
36
+ "model.layers.0.self_attn.g_a_proj",
37
+ "model.layers.0.self_attn.g_b_proj",
38
+ "model.layers.0.self_attn.k_conv1d",
39
+ "model.layers.0.self_attn.k_proj",
40
+ "model.layers.0.self_attn.o_norm",
41
+ "model.layers.0.self_attn.o_proj",
42
+ "model.layers.0.self_attn.q_conv1d",
43
+ "model.layers.0.self_attn.q_proj",
44
+ "model.layers.0.self_attn.qkv_proj",
45
+ "model.layers.0.self_attn.v_conv1d",
46
+ "model.layers.0.self_attn.v_proj",
47
+ "model.layers.1.attn_hc.base",
48
+ "model.layers.1.attn_hc.fn",
49
+ "model.layers.1.attn_hc.scale",
50
+ "model.layers.1.ffn_hc.base",
51
+ "model.layers.1.ffn_hc.fn",
52
+ "model.layers.1.ffn_hc.scale",
53
+ "model.layers.1.input_layernorm",
54
+ "model.layers.1.post_attention_layernorm",
55
+ "model.layers.1.self_attn.forget_gate.A_log",
56
+ "model.layers.1.self_attn.b_proj",
57
+ "model.layers.1.self_attn.forget_gate.dt_bias",
58
+ "model.layers.1.self_attn.f_a_proj",
59
+ "model.layers.1.self_attn.f_b_proj",
60
+ "model.layers.1.self_attn.fused_qkvbfg_a_proj",
61
+ "model.layers.1.self_attn.g_a_proj",
62
+ "model.layers.1.self_attn.g_b_proj",
63
+ "model.layers.1.self_attn.k_conv1d",
64
+ "model.layers.1.self_attn.k_proj",
65
+ "model.layers.1.self_attn.o_norm",
66
+ "model.layers.1.self_attn.o_proj",
67
+ "model.layers.1.self_attn.q_conv1d",
68
+ "model.layers.1.self_attn.q_proj",
69
+ "model.layers.1.self_attn.qkv_proj",
70
+ "model.layers.1.self_attn.v_conv1d",
71
+ "model.layers.1.self_attn.v_proj",
72
+ "model.layers.10.attn_hc.base",
73
+ "model.layers.10.attn_hc.fn",
74
+ "model.layers.10.attn_hc.scale",
75
+ "model.layers.10.ffn_hc.base",
76
+ "model.layers.10.ffn_hc.fn",
77
+ "model.layers.10.ffn_hc.scale",
78
+ "model.layers.10.input_layernorm",
79
+ "model.layers.10.mlp.gate",
80
+ "model.layers.10.mlp.gate.e_score_correction_bias",
81
+ "model.layers.10.post_attention_layernorm",
82
+ "model.layers.10.self_attn.forget_gate.A_log",
83
+ "model.layers.10.self_attn.b_proj",
84
+ "model.layers.10.self_attn.forget_gate.dt_bias",
85
+ "model.layers.10.self_attn.f_a_proj",
86
+ "model.layers.10.self_attn.f_b_proj",
87
+ "model.layers.10.self_attn.fused_qkvbfg_a_proj",
88
+ "model.layers.10.self_attn.g_a_proj",
89
+ "model.layers.10.self_attn.g_b_proj",
90
+ "model.layers.10.self_attn.k_conv1d",
91
+ "model.layers.10.self_attn.k_proj",
92
+ "model.layers.10.self_attn.o_norm",
93
+ "model.layers.10.self_attn.o_proj",
94
+ "model.layers.10.self_attn.q_conv1d",
95
+ "model.layers.10.self_attn.q_proj",
96
+ "model.layers.10.self_attn.qkv_proj",
97
+ "model.layers.10.self_attn.v_conv1d",
98
+ "model.layers.10.self_attn.v_proj",
99
+ "model.layers.11.attn_hc.base",
100
+ "model.layers.11.attn_hc.fn",
101
+ "model.layers.11.attn_hc.scale",
102
+ "model.layers.11.ffn_hc.base",
103
+ "model.layers.11.ffn_hc.fn",
104
+ "model.layers.11.ffn_hc.scale",
105
+ "model.layers.11.input_layernorm",
106
+ "model.layers.11.mlp.gate",
107
+ "model.layers.11.mlp.gate.e_score_correction_bias",
108
+ "model.layers.11.post_attention_layernorm",
109
+ "model.layers.11.self_attn.indexer.index_kpool_compress_ape",
110
+ "model.layers.11.self_attn.indexer.index_kpool_compress_gate",
111
+ "model.layers.11.self_attn.indexer.k_norm",
112
+ "model.layers.11.self_attn.indexer.k_norm.bias",
113
+ "model.layers.11.self_attn.indexer.weights_proj",
114
+ "model.layers.11.self_attn.indexer.wk",
115
+ "model.layers.11.self_attn.indexer.wq_b",
116
+ "model.layers.11.self_attn.kv_a_layernorm",
117
+ "model.layers.11.self_attn.kv_b_proj",
118
+ "model.layers.11.self_attn.q_a_layernorm",
119
+ "model.layers.12.attn_hc.base",
120
+ "model.layers.12.attn_hc.fn",
121
+ "model.layers.12.attn_hc.scale",
122
+ "model.layers.12.ffn_hc.base",
123
+ "model.layers.12.ffn_hc.fn",
124
+ "model.layers.12.ffn_hc.scale",
125
+ "model.layers.12.input_layernorm",
126
+ "model.layers.12.mlp.gate",
127
+ "model.layers.12.mlp.gate.e_score_correction_bias",
128
+ "model.layers.12.post_attention_layernorm",
129
+ "model.layers.12.self_attn.forget_gate.A_log",
130
+ "model.layers.12.self_attn.b_proj",
131
+ "model.layers.12.self_attn.forget_gate.dt_bias",
132
+ "model.layers.12.self_attn.f_a_proj",
133
+ "model.layers.12.self_attn.f_b_proj",
134
+ "model.layers.12.self_attn.fused_qkvbfg_a_proj",
135
+ "model.layers.12.self_attn.g_a_proj",
136
+ "model.layers.12.self_attn.g_b_proj",
137
+ "model.layers.12.self_attn.k_conv1d",
138
+ "model.layers.12.self_attn.k_proj",
139
+ "model.layers.12.self_attn.o_norm",
140
+ "model.layers.12.self_attn.o_proj",
141
+ "model.layers.12.self_attn.q_conv1d",
142
+ "model.layers.12.self_attn.q_proj",
143
+ "model.layers.12.self_attn.qkv_proj",
144
+ "model.layers.12.self_attn.v_conv1d",
145
+ "model.layers.12.self_attn.v_proj",
146
+ "model.layers.13.attn_hc.base",
147
+ "model.layers.13.attn_hc.fn",
148
+ "model.layers.13.attn_hc.scale",
149
+ "model.layers.13.ffn_hc.base",
150
+ "model.layers.13.ffn_hc.fn",
151
+ "model.layers.13.ffn_hc.scale",
152
+ "model.layers.13.input_layernorm",
153
+ "model.layers.13.mlp.gate",
154
+ "model.layers.13.mlp.gate.e_score_correction_bias",
155
+ "model.layers.13.post_attention_layernorm",
156
+ "model.layers.13.self_attn.forget_gate.A_log",
157
+ "model.layers.13.self_attn.b_proj",
158
+ "model.layers.13.self_attn.forget_gate.dt_bias",
159
+ "model.layers.13.self_attn.f_a_proj",
160
+ "model.layers.13.self_attn.f_b_proj",
161
+ "model.layers.13.self_attn.fused_qkvbfg_a_proj",
162
+ "model.layers.13.self_attn.g_a_proj",
163
+ "model.layers.13.self_attn.g_b_proj",
164
+ "model.layers.13.self_attn.k_conv1d",
165
+ "model.layers.13.self_attn.k_proj",
166
+ "model.layers.13.self_attn.o_norm",
167
+ "model.layers.13.self_attn.o_proj",
168
+ "model.layers.13.self_attn.q_conv1d",
169
+ "model.layers.13.self_attn.q_proj",
170
+ "model.layers.13.self_attn.qkv_proj",
171
+ "model.layers.13.self_attn.v_conv1d",
172
+ "model.layers.13.self_attn.v_proj",
173
+ "model.layers.14.attn_hc.base",
174
+ "model.layers.14.attn_hc.fn",
175
+ "model.layers.14.attn_hc.scale",
176
+ "model.layers.14.ffn_hc.base",
177
+ "model.layers.14.ffn_hc.fn",
178
+ "model.layers.14.ffn_hc.scale",
179
+ "model.layers.14.input_layernorm",
180
+ "model.layers.14.mlp.gate",
181
+ "model.layers.14.mlp.gate.e_score_correction_bias",
182
+ "model.layers.14.post_attention_layernorm",
183
+ "model.layers.14.self_attn.forget_gate.A_log",
184
+ "model.layers.14.self_attn.b_proj",
185
+ "model.layers.14.self_attn.forget_gate.dt_bias",
186
+ "model.layers.14.self_attn.f_a_proj",
187
+ "model.layers.14.self_attn.f_b_proj",
188
+ "model.layers.14.self_attn.fused_qkvbfg_a_proj",
189
+ "model.layers.14.self_attn.g_a_proj",
190
+ "model.layers.14.self_attn.g_b_proj",
191
+ "model.layers.14.self_attn.k_conv1d",
192
+ "model.layers.14.self_attn.k_proj",
193
+ "model.layers.14.self_attn.o_norm",
194
+ "model.layers.14.self_attn.o_proj",
195
+ "model.layers.14.self_attn.q_conv1d",
196
+ "model.layers.14.self_attn.q_proj",
197
+ "model.layers.14.self_attn.qkv_proj",
198
+ "model.layers.14.self_attn.v_conv1d",
199
+ "model.layers.14.self_attn.v_proj",
200
+ "model.layers.15.attn_hc.base",
201
+ "model.layers.15.attn_hc.fn",
202
+ "model.layers.15.attn_hc.scale",
203
+ "model.layers.15.ffn_hc.base",
204
+ "model.layers.15.ffn_hc.fn",
205
+ "model.layers.15.ffn_hc.scale",
206
+ "model.layers.15.input_layernorm",
207
+ "model.layers.15.mlp.gate",
208
+ "model.layers.15.mlp.gate.e_score_correction_bias",
209
+ "model.layers.15.post_attention_layernorm",
210
+ "model.layers.15.self_attn.indexer.index_kpool_compress_ape",
211
+ "model.layers.15.self_attn.indexer.index_kpool_compress_gate",
212
+ "model.layers.15.self_attn.indexer.k_norm",
213
+ "model.layers.15.self_attn.indexer.k_norm.bias",
214
+ "model.layers.15.self_attn.indexer.weights_proj",
215
+ "model.layers.15.self_attn.indexer.wk",
216
+ "model.layers.15.self_attn.indexer.wq_b",
217
+ "model.layers.15.self_attn.kv_a_layernorm",
218
+ "model.layers.15.self_attn.kv_b_proj",
219
+ "model.layers.15.self_attn.q_a_layernorm",
220
+ "model.layers.16.attn_hc.base",
221
+ "model.layers.16.attn_hc.fn",
222
+ "model.layers.16.attn_hc.scale",
223
+ "model.layers.16.ffn_hc.base",
224
+ "model.layers.16.ffn_hc.fn",
225
+ "model.layers.16.ffn_hc.scale",
226
+ "model.layers.16.input_layernorm",
227
+ "model.layers.16.mlp.gate",
228
+ "model.layers.16.mlp.gate.e_score_correction_bias",
229
+ "model.layers.16.post_attention_layernorm",
230
+ "model.layers.16.self_attn.forget_gate.A_log",
231
+ "model.layers.16.self_attn.b_proj",
232
+ "model.layers.16.self_attn.forget_gate.dt_bias",
233
+ "model.layers.16.self_attn.f_a_proj",
234
+ "model.layers.16.self_attn.f_b_proj",
235
+ "model.layers.16.self_attn.fused_qkvbfg_a_proj",
236
+ "model.layers.16.self_attn.g_a_proj",
237
+ "model.layers.16.self_attn.g_b_proj",
238
+ "model.layers.16.self_attn.k_conv1d",
239
+ "model.layers.16.self_attn.k_proj",
240
+ "model.layers.16.self_attn.o_norm",
241
+ "model.layers.16.self_attn.o_proj",
242
+ "model.layers.16.self_attn.q_conv1d",
243
+ "model.layers.16.self_attn.q_proj",
244
+ "model.layers.16.self_attn.qkv_proj",
245
+ "model.layers.16.self_attn.v_conv1d",
246
+ "model.layers.16.self_attn.v_proj",
247
+ "model.layers.17.attn_hc.base",
248
+ "model.layers.17.attn_hc.fn",
249
+ "model.layers.17.attn_hc.scale",
250
+ "model.layers.17.ffn_hc.base",
251
+ "model.layers.17.ffn_hc.fn",
252
+ "model.layers.17.ffn_hc.scale",
253
+ "model.layers.17.input_layernorm",
254
+ "model.layers.17.mlp.gate",
255
+ "model.layers.17.mlp.gate.e_score_correction_bias",
256
+ "model.layers.17.post_attention_layernorm",
257
+ "model.layers.17.self_attn.forget_gate.A_log",
258
+ "model.layers.17.self_attn.b_proj",
259
+ "model.layers.17.self_attn.forget_gate.dt_bias",
260
+ "model.layers.17.self_attn.f_a_proj",
261
+ "model.layers.17.self_attn.f_b_proj",
262
+ "model.layers.17.self_attn.fused_qkvbfg_a_proj",
263
+ "model.layers.17.self_attn.g_a_proj",
264
+ "model.layers.17.self_attn.g_b_proj",
265
+ "model.layers.17.self_attn.k_conv1d",
266
+ "model.layers.17.self_attn.k_proj",
267
+ "model.layers.17.self_attn.o_norm",
268
+ "model.layers.17.self_attn.o_proj",
269
+ "model.layers.17.self_attn.q_conv1d",
270
+ "model.layers.17.self_attn.q_proj",
271
+ "model.layers.17.self_attn.qkv_proj",
272
+ "model.layers.17.self_attn.v_conv1d",
273
+ "model.layers.17.self_attn.v_proj",
274
+ "model.layers.18.attn_hc.base",
275
+ "model.layers.18.attn_hc.fn",
276
+ "model.layers.18.attn_hc.scale",
277
+ "model.layers.18.ffn_hc.base",
278
+ "model.layers.18.ffn_hc.fn",
279
+ "model.layers.18.ffn_hc.scale",
280
+ "model.layers.18.input_layernorm",
281
+ "model.layers.18.mlp.gate",
282
+ "model.layers.18.mlp.gate.e_score_correction_bias",
283
+ "model.layers.18.post_attention_layernorm",
284
+ "model.layers.18.self_attn.forget_gate.A_log",
285
+ "model.layers.18.self_attn.b_proj",
286
+ "model.layers.18.self_attn.forget_gate.dt_bias",
287
+ "model.layers.18.self_attn.f_a_proj",
288
+ "model.layers.18.self_attn.f_b_proj",
289
+ "model.layers.18.self_attn.fused_qkvbfg_a_proj",
290
+ "model.layers.18.self_attn.g_a_proj",
291
+ "model.layers.18.self_attn.g_b_proj",
292
+ "model.layers.18.self_attn.k_conv1d",
293
+ "model.layers.18.self_attn.k_proj",
294
+ "model.layers.18.self_attn.o_norm",
295
+ "model.layers.18.self_attn.o_proj",
296
+ "model.layers.18.self_attn.q_conv1d",
297
+ "model.layers.18.self_attn.q_proj",
298
+ "model.layers.18.self_attn.qkv_proj",
299
+ "model.layers.18.self_attn.v_conv1d",
300
+ "model.layers.18.self_attn.v_proj",
301
+ "model.layers.19.attn_hc.base",
302
+ "model.layers.19.attn_hc.fn",
303
+ "model.layers.19.attn_hc.scale",
304
+ "model.layers.19.ffn_hc.base",
305
+ "model.layers.19.ffn_hc.fn",
306
+ "model.layers.19.ffn_hc.scale",
307
+ "model.layers.19.input_layernorm",
308
+ "model.layers.19.mlp.gate",
309
+ "model.layers.19.mlp.gate.e_score_correction_bias",
310
+ "model.layers.19.post_attention_layernorm",
311
+ "model.layers.19.self_attn.indexer.index_kpool_compress_ape",
312
+ "model.layers.19.self_attn.indexer.index_kpool_compress_gate",
313
+ "model.layers.19.self_attn.indexer.k_norm",
314
+ "model.layers.19.self_attn.indexer.k_norm.bias",
315
+ "model.layers.19.self_attn.indexer.weights_proj",
316
+ "model.layers.19.self_attn.indexer.wk",
317
+ "model.layers.19.self_attn.indexer.wq_b",
318
+ "model.layers.19.self_attn.kv_a_layernorm",
319
+ "model.layers.19.self_attn.kv_b_proj",
320
+ "model.layers.19.self_attn.q_a_layernorm",
321
+ "model.layers.2.attn_hc.base",
322
+ "model.layers.2.attn_hc.fn",
323
+ "model.layers.2.attn_hc.scale",
324
+ "model.layers.2.ffn_hc.base",
325
+ "model.layers.2.ffn_hc.fn",
326
+ "model.layers.2.ffn_hc.scale",
327
+ "model.layers.2.input_layernorm",
328
+ "model.layers.2.post_attention_layernorm",
329
+ "model.layers.2.self_attn.forget_gate.A_log",
330
+ "model.layers.2.self_attn.b_proj",
331
+ "model.layers.2.self_attn.forget_gate.dt_bias",
332
+ "model.layers.2.self_attn.f_a_proj",
333
+ "model.layers.2.self_attn.f_b_proj",
334
+ "model.layers.2.self_attn.fused_qkvbfg_a_proj",
335
+ "model.layers.2.self_attn.g_a_proj",
336
+ "model.layers.2.self_attn.g_b_proj",
337
+ "model.layers.2.self_attn.k_conv1d",
338
+ "model.layers.2.self_attn.k_proj",
339
+ "model.layers.2.self_attn.o_norm",
340
+ "model.layers.2.self_attn.o_proj",
341
+ "model.layers.2.self_attn.q_conv1d",
342
+ "model.layers.2.self_attn.q_proj",
343
+ "model.layers.2.self_attn.qkv_proj",
344
+ "model.layers.2.self_attn.v_conv1d",
345
+ "model.layers.2.self_attn.v_proj",
346
+ "model.layers.20.attn_hc.base",
347
+ "model.layers.20.attn_hc.fn",
348
+ "model.layers.20.attn_hc.scale",
349
+ "model.layers.20.ffn_hc.base",
350
+ "model.layers.20.ffn_hc.fn",
351
+ "model.layers.20.ffn_hc.scale",
352
+ "model.layers.20.input_layernorm",
353
+ "model.layers.20.mlp.gate",
354
+ "model.layers.20.mlp.gate.e_score_correction_bias",
355
+ "model.layers.20.post_attention_layernorm",
356
+ "model.layers.20.self_attn.forget_gate.A_log",
357
+ "model.layers.20.self_attn.b_proj",
358
+ "model.layers.20.self_attn.forget_gate.dt_bias",
359
+ "model.layers.20.self_attn.f_a_proj",
360
+ "model.layers.20.self_attn.f_b_proj",
361
+ "model.layers.20.self_attn.fused_qkvbfg_a_proj",
362
+ "model.layers.20.self_attn.g_a_proj",
363
+ "model.layers.20.self_attn.g_b_proj",
364
+ "model.layers.20.self_attn.k_conv1d",
365
+ "model.layers.20.self_attn.k_proj",
366
+ "model.layers.20.self_attn.o_norm",
367
+ "model.layers.20.self_attn.o_proj",
368
+ "model.layers.20.self_attn.q_conv1d",
369
+ "model.layers.20.self_attn.q_proj",
370
+ "model.layers.20.self_attn.qkv_proj",
371
+ "model.layers.20.self_attn.v_conv1d",
372
+ "model.layers.20.self_attn.v_proj",
373
+ "model.layers.21.attn_hc.base",
374
+ "model.layers.21.attn_hc.fn",
375
+ "model.layers.21.attn_hc.scale",
376
+ "model.layers.21.ffn_hc.base",
377
+ "model.layers.21.ffn_hc.fn",
378
+ "model.layers.21.ffn_hc.scale",
379
+ "model.layers.21.input_layernorm",
380
+ "model.layers.21.mlp.gate",
381
+ "model.layers.21.mlp.gate.e_score_correction_bias",
382
+ "model.layers.21.post_attention_layernorm",
383
+ "model.layers.21.self_attn.forget_gate.A_log",
384
+ "model.layers.21.self_attn.b_proj",
385
+ "model.layers.21.self_attn.forget_gate.dt_bias",
386
+ "model.layers.21.self_attn.f_a_proj",
387
+ "model.layers.21.self_attn.f_b_proj",
388
+ "model.layers.21.self_attn.fused_qkvbfg_a_proj",
389
+ "model.layers.21.self_attn.g_a_proj",
390
+ "model.layers.21.self_attn.g_b_proj",
391
+ "model.layers.21.self_attn.k_conv1d",
392
+ "model.layers.21.self_attn.k_proj",
393
+ "model.layers.21.self_attn.o_norm",
394
+ "model.layers.21.self_attn.o_proj",
395
+ "model.layers.21.self_attn.q_conv1d",
396
+ "model.layers.21.self_attn.q_proj",
397
+ "model.layers.21.self_attn.qkv_proj",
398
+ "model.layers.21.self_attn.v_conv1d",
399
+ "model.layers.21.self_attn.v_proj",
400
+ "model.layers.22.attn_hc.base",
401
+ "model.layers.22.attn_hc.fn",
402
+ "model.layers.22.attn_hc.scale",
403
+ "model.layers.22.ffn_hc.base",
404
+ "model.layers.22.ffn_hc.fn",
405
+ "model.layers.22.ffn_hc.scale",
406
+ "model.layers.22.input_layernorm",
407
+ "model.layers.22.mlp.gate",
408
+ "model.layers.22.mlp.gate.e_score_correction_bias",
409
+ "model.layers.22.post_attention_layernorm",
410
+ "model.layers.22.self_attn.forget_gate.A_log",
411
+ "model.layers.22.self_attn.b_proj",
412
+ "model.layers.22.self_attn.forget_gate.dt_bias",
413
+ "model.layers.22.self_attn.f_a_proj",
414
+ "model.layers.22.self_attn.f_b_proj",
415
+ "model.layers.22.self_attn.fused_qkvbfg_a_proj",
416
+ "model.layers.22.self_attn.g_a_proj",
417
+ "model.layers.22.self_attn.g_b_proj",
418
+ "model.layers.22.self_attn.k_conv1d",
419
+ "model.layers.22.self_attn.k_proj",
420
+ "model.layers.22.self_attn.o_norm",
421
+ "model.layers.22.self_attn.o_proj",
422
+ "model.layers.22.self_attn.q_conv1d",
423
+ "model.layers.22.self_attn.q_proj",
424
+ "model.layers.22.self_attn.qkv_proj",
425
+ "model.layers.22.self_attn.v_conv1d",
426
+ "model.layers.22.self_attn.v_proj",
427
+ "model.layers.23.attn_hc.base",
428
+ "model.layers.23.attn_hc.fn",
429
+ "model.layers.23.attn_hc.scale",
430
+ "model.layers.23.ffn_hc.base",
431
+ "model.layers.23.ffn_hc.fn",
432
+ "model.layers.23.ffn_hc.scale",
433
+ "model.layers.23.input_layernorm",
434
+ "model.layers.23.mlp.gate",
435
+ "model.layers.23.mlp.gate.e_score_correction_bias",
436
+ "model.layers.23.post_attention_layernorm",
437
+ "model.layers.23.self_attn.indexer.index_kpool_compress_ape",
438
+ "model.layers.23.self_attn.indexer.index_kpool_compress_gate",
439
+ "model.layers.23.self_attn.indexer.k_norm",
440
+ "model.layers.23.self_attn.indexer.k_norm.bias",
441
+ "model.layers.23.self_attn.indexer.weights_proj",
442
+ "model.layers.23.self_attn.indexer.wk",
443
+ "model.layers.23.self_attn.indexer.wq_b",
444
+ "model.layers.23.self_attn.kv_a_layernorm",
445
+ "model.layers.23.self_attn.kv_b_proj",
446
+ "model.layers.23.self_attn.q_a_layernorm",
447
+ "model.layers.24.attn_hc.base",
448
+ "model.layers.24.attn_hc.fn",
449
+ "model.layers.24.attn_hc.scale",
450
+ "model.layers.24.ffn_hc.base",
451
+ "model.layers.24.ffn_hc.fn",
452
+ "model.layers.24.ffn_hc.scale",
453
+ "model.layers.24.input_layernorm",
454
+ "model.layers.24.mlp.gate",
455
+ "model.layers.24.mlp.gate.e_score_correction_bias",
456
+ "model.layers.24.post_attention_layernorm",
457
+ "model.layers.24.self_attn.forget_gate.A_log",
458
+ "model.layers.24.self_attn.b_proj",
459
+ "model.layers.24.self_attn.forget_gate.dt_bias",
460
+ "model.layers.24.self_attn.f_a_proj",
461
+ "model.layers.24.self_attn.f_b_proj",
462
+ "model.layers.24.self_attn.fused_qkvbfg_a_proj",
463
+ "model.layers.24.self_attn.g_a_proj",
464
+ "model.layers.24.self_attn.g_b_proj",
465
+ "model.layers.24.self_attn.k_conv1d",
466
+ "model.layers.24.self_attn.k_proj",
467
+ "model.layers.24.self_attn.o_norm",
468
+ "model.layers.24.self_attn.o_proj",
469
+ "model.layers.24.self_attn.q_conv1d",
470
+ "model.layers.24.self_attn.q_proj",
471
+ "model.layers.24.self_attn.qkv_proj",
472
+ "model.layers.24.self_attn.v_conv1d",
473
+ "model.layers.24.self_attn.v_proj",
474
+ "model.layers.25.attn_hc.base",
475
+ "model.layers.25.attn_hc.fn",
476
+ "model.layers.25.attn_hc.scale",
477
+ "model.layers.25.ffn_hc.base",
478
+ "model.layers.25.ffn_hc.fn",
479
+ "model.layers.25.ffn_hc.scale",
480
+ "model.layers.25.input_layernorm",
481
+ "model.layers.25.mlp.gate",
482
+ "model.layers.25.mlp.gate.e_score_correction_bias",
483
+ "model.layers.25.post_attention_layernorm",
484
+ "model.layers.25.self_attn.forget_gate.A_log",
485
+ "model.layers.25.self_attn.b_proj",
486
+ "model.layers.25.self_attn.forget_gate.dt_bias",
487
+ "model.layers.25.self_attn.f_a_proj",
488
+ "model.layers.25.self_attn.f_b_proj",
489
+ "model.layers.25.self_attn.fused_qkvbfg_a_proj",
490
+ "model.layers.25.self_attn.g_a_proj",
491
+ "model.layers.25.self_attn.g_b_proj",
492
+ "model.layers.25.self_attn.k_conv1d",
493
+ "model.layers.25.self_attn.k_proj",
494
+ "model.layers.25.self_attn.o_norm",
495
+ "model.layers.25.self_attn.o_proj",
496
+ "model.layers.25.self_attn.q_conv1d",
497
+ "model.layers.25.self_attn.q_proj",
498
+ "model.layers.25.self_attn.qkv_proj",
499
+ "model.layers.25.self_attn.v_conv1d",
500
+ "model.layers.25.self_attn.v_proj",
501
+ "model.layers.26.attn_hc.base",
502
+ "model.layers.26.attn_hc.fn",
503
+ "model.layers.26.attn_hc.scale",
504
+ "model.layers.26.ffn_hc.base",
505
+ "model.layers.26.ffn_hc.fn",
506
+ "model.layers.26.ffn_hc.scale",
507
+ "model.layers.26.input_layernorm",
508
+ "model.layers.26.mlp.gate",
509
+ "model.layers.26.mlp.gate.e_score_correction_bias",
510
+ "model.layers.26.post_attention_layernorm",
511
+ "model.layers.26.self_attn.forget_gate.A_log",
512
+ "model.layers.26.self_attn.b_proj",
513
+ "model.layers.26.self_attn.forget_gate.dt_bias",
514
+ "model.layers.26.self_attn.f_a_proj",
515
+ "model.layers.26.self_attn.f_b_proj",
516
+ "model.layers.26.self_attn.fused_qkvbfg_a_proj",
517
+ "model.layers.26.self_attn.g_a_proj",
518
+ "model.layers.26.self_attn.g_b_proj",
519
+ "model.layers.26.self_attn.k_conv1d",
520
+ "model.layers.26.self_attn.k_proj",
521
+ "model.layers.26.self_attn.o_norm",
522
+ "model.layers.26.self_attn.o_proj",
523
+ "model.layers.26.self_attn.q_conv1d",
524
+ "model.layers.26.self_attn.q_proj",
525
+ "model.layers.26.self_attn.qkv_proj",
526
+ "model.layers.26.self_attn.v_conv1d",
527
+ "model.layers.26.self_attn.v_proj",
528
+ "model.layers.27.attn_hc.base",
529
+ "model.layers.27.attn_hc.fn",
530
+ "model.layers.27.attn_hc.scale",
531
+ "model.layers.27.ffn_hc.base",
532
+ "model.layers.27.ffn_hc.fn",
533
+ "model.layers.27.ffn_hc.scale",
534
+ "model.layers.27.input_layernorm",
535
+ "model.layers.27.mlp.gate",
536
+ "model.layers.27.mlp.gate.e_score_correction_bias",
537
+ "model.layers.27.post_attention_layernorm",
538
+ "model.layers.27.self_attn.indexer.index_kpool_compress_ape",
539
+ "model.layers.27.self_attn.indexer.index_kpool_compress_gate",
540
+ "model.layers.27.self_attn.indexer.k_norm",
541
+ "model.layers.27.self_attn.indexer.k_norm.bias",
542
+ "model.layers.27.self_attn.indexer.weights_proj",
543
+ "model.layers.27.self_attn.indexer.wk",
544
+ "model.layers.27.self_attn.indexer.wq_b",
545
+ "model.layers.27.self_attn.kv_a_layernorm",
546
+ "model.layers.27.self_attn.kv_b_proj",
547
+ "model.layers.27.self_attn.q_a_layernorm",
548
+ "model.layers.28.attn_hc.base",
549
+ "model.layers.28.attn_hc.fn",
550
+ "model.layers.28.attn_hc.scale",
551
+ "model.layers.28.ffn_hc.base",
552
+ "model.layers.28.ffn_hc.fn",
553
+ "model.layers.28.ffn_hc.scale",
554
+ "model.layers.28.input_layernorm",
555
+ "model.layers.28.mlp.gate",
556
+ "model.layers.28.mlp.gate.e_score_correction_bias",
557
+ "model.layers.28.post_attention_layernorm",
558
+ "model.layers.28.self_attn.forget_gate.A_log",
559
+ "model.layers.28.self_attn.b_proj",
560
+ "model.layers.28.self_attn.forget_gate.dt_bias",
561
+ "model.layers.28.self_attn.f_a_proj",
562
+ "model.layers.28.self_attn.f_b_proj",
563
+ "model.layers.28.self_attn.fused_qkvbfg_a_proj",
564
+ "model.layers.28.self_attn.g_a_proj",
565
+ "model.layers.28.self_attn.g_b_proj",
566
+ "model.layers.28.self_attn.k_conv1d",
567
+ "model.layers.28.self_attn.k_proj",
568
+ "model.layers.28.self_attn.o_norm",
569
+ "model.layers.28.self_attn.o_proj",
570
+ "model.layers.28.self_attn.q_conv1d",
571
+ "model.layers.28.self_attn.q_proj",
572
+ "model.layers.28.self_attn.qkv_proj",
573
+ "model.layers.28.self_attn.v_conv1d",
574
+ "model.layers.28.self_attn.v_proj",
575
+ "model.layers.29.attn_hc.base",
576
+ "model.layers.29.attn_hc.fn",
577
+ "model.layers.29.attn_hc.scale",
578
+ "model.layers.29.ffn_hc.base",
579
+ "model.layers.29.ffn_hc.fn",
580
+ "model.layers.29.ffn_hc.scale",
581
+ "model.layers.29.input_layernorm",
582
+ "model.layers.29.mlp.gate",
583
+ "model.layers.29.mlp.gate.e_score_correction_bias",
584
+ "model.layers.29.post_attention_layernorm",
585
+ "model.layers.29.self_attn.forget_gate.A_log",
586
+ "model.layers.29.self_attn.b_proj",
587
+ "model.layers.29.self_attn.forget_gate.dt_bias",
588
+ "model.layers.29.self_attn.f_a_proj",
589
+ "model.layers.29.self_attn.f_b_proj",
590
+ "model.layers.29.self_attn.fused_qkvbfg_a_proj",
591
+ "model.layers.29.self_attn.g_a_proj",
592
+ "model.layers.29.self_attn.g_b_proj",
593
+ "model.layers.29.self_attn.k_conv1d",
594
+ "model.layers.29.self_attn.k_proj",
595
+ "model.layers.29.self_attn.o_norm",
596
+ "model.layers.29.self_attn.o_proj",
597
+ "model.layers.29.self_attn.q_conv1d",
598
+ "model.layers.29.self_attn.q_proj",
599
+ "model.layers.29.self_attn.qkv_proj",
600
+ "model.layers.29.self_attn.v_conv1d",
601
+ "model.layers.29.self_attn.v_proj",
602
+ "model.layers.3.attn_hc.base",
603
+ "model.layers.3.attn_hc.fn",
604
+ "model.layers.3.attn_hc.scale",
605
+ "model.layers.3.ffn_hc.base",
606
+ "model.layers.3.ffn_hc.fn",
607
+ "model.layers.3.ffn_hc.scale",
608
+ "model.layers.3.input_layernorm",
609
+ "model.layers.3.mlp.gate",
610
+ "model.layers.3.mlp.gate.e_score_correction_bias",
611
+ "model.layers.3.post_attention_layernorm",
612
+ "model.layers.3.self_attn.indexer.index_kpool_compress_ape",
613
+ "model.layers.3.self_attn.indexer.index_kpool_compress_gate",
614
+ "model.layers.3.self_attn.indexer.k_norm",
615
+ "model.layers.3.self_attn.indexer.k_norm.bias",
616
+ "model.layers.3.self_attn.indexer.weights_proj",
617
+ "model.layers.3.self_attn.indexer.wk",
618
+ "model.layers.3.self_attn.indexer.wq_b",
619
+ "model.layers.3.self_attn.kv_a_layernorm",
620
+ "model.layers.3.self_attn.kv_b_proj",
621
+ "model.layers.3.self_attn.q_a_layernorm",
622
+ "model.layers.30.attn_hc.base",
623
+ "model.layers.30.attn_hc.fn",
624
+ "model.layers.30.attn_hc.scale",
625
+ "model.layers.30.ffn_hc.base",
626
+ "model.layers.30.ffn_hc.fn",
627
+ "model.layers.30.ffn_hc.scale",
628
+ "model.layers.30.input_layernorm",
629
+ "model.layers.30.mlp.gate",
630
+ "model.layers.30.mlp.gate.e_score_correction_bias",
631
+ "model.layers.30.post_attention_layernorm",
632
+ "model.layers.30.self_attn.forget_gate.A_log",
633
+ "model.layers.30.self_attn.b_proj",
634
+ "model.layers.30.self_attn.forget_gate.dt_bias",
635
+ "model.layers.30.self_attn.f_a_proj",
636
+ "model.layers.30.self_attn.f_b_proj",
637
+ "model.layers.30.self_attn.fused_qkvbfg_a_proj",
638
+ "model.layers.30.self_attn.g_a_proj",
639
+ "model.layers.30.self_attn.g_b_proj",
640
+ "model.layers.30.self_attn.k_conv1d",
641
+ "model.layers.30.self_attn.k_proj",
642
+ "model.layers.30.self_attn.o_norm",
643
+ "model.layers.30.self_attn.o_proj",
644
+ "model.layers.30.self_attn.q_conv1d",
645
+ "model.layers.30.self_attn.q_proj",
646
+ "model.layers.30.self_attn.qkv_proj",
647
+ "model.layers.30.self_attn.v_conv1d",
648
+ "model.layers.30.self_attn.v_proj",
649
+ "model.layers.31.attn_hc.base",
650
+ "model.layers.31.attn_hc.fn",
651
+ "model.layers.31.attn_hc.scale",
652
+ "model.layers.31.ffn_hc.base",
653
+ "model.layers.31.ffn_hc.fn",
654
+ "model.layers.31.ffn_hc.scale",
655
+ "model.layers.31.input_layernorm",
656
+ "model.layers.31.mlp.gate",
657
+ "model.layers.31.mlp.gate.e_score_correction_bias",
658
+ "model.layers.31.post_attention_layernorm",
659
+ "model.layers.31.self_attn.indexer.index_kpool_compress_ape",
660
+ "model.layers.31.self_attn.indexer.index_kpool_compress_gate",
661
+ "model.layers.31.self_attn.indexer.k_norm",
662
+ "model.layers.31.self_attn.indexer.k_norm.bias",
663
+ "model.layers.31.self_attn.indexer.weights_proj",
664
+ "model.layers.31.self_attn.indexer.wk",
665
+ "model.layers.31.self_attn.indexer.wq_b",
666
+ "model.layers.31.self_attn.kv_a_layernorm",
667
+ "model.layers.31.self_attn.kv_b_proj",
668
+ "model.layers.31.self_attn.q_a_layernorm",
669
+ "model.layers.32.attn_hc.base",
670
+ "model.layers.32.attn_hc.fn",
671
+ "model.layers.32.attn_hc.scale",
672
+ "model.layers.32.ffn_hc.base",
673
+ "model.layers.32.ffn_hc.fn",
674
+ "model.layers.32.ffn_hc.scale",
675
+ "model.layers.32.input_layernorm",
676
+ "model.layers.32.mlp.gate",
677
+ "model.layers.32.mlp.gate.e_score_correction_bias",
678
+ "model.layers.32.post_attention_layernorm",
679
+ "model.layers.32.self_attn.forget_gate.A_log",
680
+ "model.layers.32.self_attn.b_proj",
681
+ "model.layers.32.self_attn.forget_gate.dt_bias",
682
+ "model.layers.32.self_attn.f_a_proj",
683
+ "model.layers.32.self_attn.f_b_proj",
684
+ "model.layers.32.self_attn.fused_qkvbfg_a_proj",
685
+ "model.layers.32.self_attn.g_a_proj",
686
+ "model.layers.32.self_attn.g_b_proj",
687
+ "model.layers.32.self_attn.k_conv1d",
688
+ "model.layers.32.self_attn.k_proj",
689
+ "model.layers.32.self_attn.o_norm",
690
+ "model.layers.32.self_attn.o_proj",
691
+ "model.layers.32.self_attn.q_conv1d",
692
+ "model.layers.32.self_attn.q_proj",
693
+ "model.layers.32.self_attn.qkv_proj",
694
+ "model.layers.32.self_attn.v_conv1d",
695
+ "model.layers.32.self_attn.v_proj",
696
+ "model.layers.33.attn_hc.base",
697
+ "model.layers.33.attn_hc.fn",
698
+ "model.layers.33.attn_hc.scale",
699
+ "model.layers.33.ffn_hc.base",
700
+ "model.layers.33.ffn_hc.fn",
701
+ "model.layers.33.ffn_hc.scale",
702
+ "model.layers.33.input_layernorm",
703
+ "model.layers.33.mlp.gate",
704
+ "model.layers.33.mlp.gate.e_score_correction_bias",
705
+ "model.layers.33.post_attention_layernorm",
706
+ "model.layers.33.self_attn.forget_gate.A_log",
707
+ "model.layers.33.self_attn.b_proj",
708
+ "model.layers.33.self_attn.forget_gate.dt_bias",
709
+ "model.layers.33.self_attn.f_a_proj",
710
+ "model.layers.33.self_attn.f_b_proj",
711
+ "model.layers.33.self_attn.fused_qkvbfg_a_proj",
712
+ "model.layers.33.self_attn.g_a_proj",
713
+ "model.layers.33.self_attn.g_b_proj",
714
+ "model.layers.33.self_attn.k_conv1d",
715
+ "model.layers.33.self_attn.k_proj",
716
+ "model.layers.33.self_attn.o_norm",
717
+ "model.layers.33.self_attn.o_proj",
718
+ "model.layers.33.self_attn.q_conv1d",
719
+ "model.layers.33.self_attn.q_proj",
720
+ "model.layers.33.self_attn.qkv_proj",
721
+ "model.layers.33.self_attn.v_conv1d",
722
+ "model.layers.33.self_attn.v_proj",
723
+ "model.layers.34.attn_hc.base",
724
+ "model.layers.34.attn_hc.fn",
725
+ "model.layers.34.attn_hc.scale",
726
+ "model.layers.34.ffn_hc.base",
727
+ "model.layers.34.ffn_hc.fn",
728
+ "model.layers.34.ffn_hc.scale",
729
+ "model.layers.34.input_layernorm",
730
+ "model.layers.34.mlp.gate",
731
+ "model.layers.34.mlp.gate.e_score_correction_bias",
732
+ "model.layers.34.post_attention_layernorm",
733
+ "model.layers.34.self_attn.forget_gate.A_log",
734
+ "model.layers.34.self_attn.b_proj",
735
+ "model.layers.34.self_attn.forget_gate.dt_bias",
736
+ "model.layers.34.self_attn.f_a_proj",
737
+ "model.layers.34.self_attn.f_b_proj",
738
+ "model.layers.34.self_attn.fused_qkvbfg_a_proj",
739
+ "model.layers.34.self_attn.g_a_proj",
740
+ "model.layers.34.self_attn.g_b_proj",
741
+ "model.layers.34.self_attn.k_conv1d",
742
+ "model.layers.34.self_attn.k_proj",
743
+ "model.layers.34.self_attn.o_norm",
744
+ "model.layers.34.self_attn.o_proj",
745
+ "model.layers.34.self_attn.q_conv1d",
746
+ "model.layers.34.self_attn.q_proj",
747
+ "model.layers.34.self_attn.qkv_proj",
748
+ "model.layers.34.self_attn.v_conv1d",
749
+ "model.layers.34.self_attn.v_proj",
750
+ "model.layers.35.attn_hc.base",
751
+ "model.layers.35.attn_hc.fn",
752
+ "model.layers.35.attn_hc.scale",
753
+ "model.layers.35.ffn_hc.base",
754
+ "model.layers.35.ffn_hc.fn",
755
+ "model.layers.35.ffn_hc.scale",
756
+ "model.layers.35.input_layernorm",
757
+ "model.layers.35.mlp.gate",
758
+ "model.layers.35.mlp.gate.e_score_correction_bias",
759
+ "model.layers.35.post_attention_layernorm",
760
+ "model.layers.35.self_attn.indexer.index_kpool_compress_ape",
761
+ "model.layers.35.self_attn.indexer.index_kpool_compress_gate",
762
+ "model.layers.35.self_attn.indexer.k_norm",
763
+ "model.layers.35.self_attn.indexer.k_norm.bias",
764
+ "model.layers.35.self_attn.indexer.weights_proj",
765
+ "model.layers.35.self_attn.indexer.wk",
766
+ "model.layers.35.self_attn.indexer.wq_b",
767
+ "model.layers.35.self_attn.kv_a_layernorm",
768
+ "model.layers.35.self_attn.kv_b_proj",
769
+ "model.layers.35.self_attn.q_a_layernorm",
770
+ "model.layers.36.attn_hc.base",
771
+ "model.layers.36.attn_hc.fn",
772
+ "model.layers.36.attn_hc.scale",
773
+ "model.layers.36.ffn_hc.base",
774
+ "model.layers.36.ffn_hc.fn",
775
+ "model.layers.36.ffn_hc.scale",
776
+ "model.layers.36.input_layernorm",
777
+ "model.layers.36.mlp.gate",
778
+ "model.layers.36.mlp.gate.e_score_correction_bias",
779
+ "model.layers.36.post_attention_layernorm",
780
+ "model.layers.36.self_attn.forget_gate.A_log",
781
+ "model.layers.36.self_attn.b_proj",
782
+ "model.layers.36.self_attn.forget_gate.dt_bias",
783
+ "model.layers.36.self_attn.f_a_proj",
784
+ "model.layers.36.self_attn.f_b_proj",
785
+ "model.layers.36.self_attn.fused_qkvbfg_a_proj",
786
+ "model.layers.36.self_attn.g_a_proj",
787
+ "model.layers.36.self_attn.g_b_proj",
788
+ "model.layers.36.self_attn.k_conv1d",
789
+ "model.layers.36.self_attn.k_proj",
790
+ "model.layers.36.self_attn.o_norm",
791
+ "model.layers.36.self_attn.o_proj",
792
+ "model.layers.36.self_attn.q_conv1d",
793
+ "model.layers.36.self_attn.q_proj",
794
+ "model.layers.36.self_attn.qkv_proj",
795
+ "model.layers.36.self_attn.v_conv1d",
796
+ "model.layers.36.self_attn.v_proj",
797
+ "model.layers.37.attn_hc.base",
798
+ "model.layers.37.attn_hc.fn",
799
+ "model.layers.37.attn_hc.scale",
800
+ "model.layers.37.ffn_hc.base",
801
+ "model.layers.37.ffn_hc.fn",
802
+ "model.layers.37.ffn_hc.scale",
803
+ "model.layers.37.input_layernorm",
804
+ "model.layers.37.mlp.gate",
805
+ "model.layers.37.mlp.gate.e_score_correction_bias",
806
+ "model.layers.37.post_attention_layernorm",
807
+ "model.layers.37.self_attn.forget_gate.A_log",
808
+ "model.layers.37.self_attn.b_proj",
809
+ "model.layers.37.self_attn.forget_gate.dt_bias",
810
+ "model.layers.37.self_attn.f_a_proj",
811
+ "model.layers.37.self_attn.f_b_proj",
812
+ "model.layers.37.self_attn.fused_qkvbfg_a_proj",
813
+ "model.layers.37.self_attn.g_a_proj",
814
+ "model.layers.37.self_attn.g_b_proj",
815
+ "model.layers.37.self_attn.k_conv1d",
816
+ "model.layers.37.self_attn.k_proj",
817
+ "model.layers.37.self_attn.o_norm",
818
+ "model.layers.37.self_attn.o_proj",
819
+ "model.layers.37.self_attn.q_conv1d",
820
+ "model.layers.37.self_attn.q_proj",
821
+ "model.layers.37.self_attn.qkv_proj",
822
+ "model.layers.37.self_attn.v_conv1d",
823
+ "model.layers.37.self_attn.v_proj",
824
+ "model.layers.38.attn_hc.base",
825
+ "model.layers.38.attn_hc.fn",
826
+ "model.layers.38.attn_hc.scale",
827
+ "model.layers.38.ffn_hc.base",
828
+ "model.layers.38.ffn_hc.fn",
829
+ "model.layers.38.ffn_hc.scale",
830
+ "model.layers.38.input_layernorm",
831
+ "model.layers.38.mlp.gate",
832
+ "model.layers.38.mlp.gate.e_score_correction_bias",
833
+ "model.layers.38.post_attention_layernorm",
834
+ "model.layers.38.self_attn.forget_gate.A_log",
835
+ "model.layers.38.self_attn.b_proj",
836
+ "model.layers.38.self_attn.forget_gate.dt_bias",
837
+ "model.layers.38.self_attn.f_a_proj",
838
+ "model.layers.38.self_attn.f_b_proj",
839
+ "model.layers.38.self_attn.fused_qkvbfg_a_proj",
840
+ "model.layers.38.self_attn.g_a_proj",
841
+ "model.layers.38.self_attn.g_b_proj",
842
+ "model.layers.38.self_attn.k_conv1d",
843
+ "model.layers.38.self_attn.k_proj",
844
+ "model.layers.38.self_attn.o_norm",
845
+ "model.layers.38.self_attn.o_proj",
846
+ "model.layers.38.self_attn.q_conv1d",
847
+ "model.layers.38.self_attn.q_proj",
848
+ "model.layers.38.self_attn.qkv_proj",
849
+ "model.layers.38.self_attn.v_conv1d",
850
+ "model.layers.38.self_attn.v_proj",
851
+ "model.layers.39.attn_hc.base",
852
+ "model.layers.39.attn_hc.fn",
853
+ "model.layers.39.attn_hc.scale",
854
+ "model.layers.39.ffn_hc.base",
855
+ "model.layers.39.ffn_hc.fn",
856
+ "model.layers.39.ffn_hc.scale",
857
+ "model.layers.39.input_layernorm",
858
+ "model.layers.39.mlp.gate",
859
+ "model.layers.39.mlp.gate.e_score_correction_bias",
860
+ "model.layers.39.post_attention_layernorm",
861
+ "model.layers.39.self_attn.indexer.index_kpool_compress_ape",
862
+ "model.layers.39.self_attn.indexer.index_kpool_compress_gate",
863
+ "model.layers.39.self_attn.indexer.k_norm",
864
+ "model.layers.39.self_attn.indexer.k_norm.bias",
865
+ "model.layers.39.self_attn.indexer.weights_proj",
866
+ "model.layers.39.self_attn.indexer.wk",
867
+ "model.layers.39.self_attn.indexer.wq_b",
868
+ "model.layers.39.self_attn.kv_a_layernorm",
869
+ "model.layers.39.self_attn.kv_b_proj",
870
+ "model.layers.39.self_attn.q_a_layernorm",
871
+ "model.layers.4.attn_hc.base",
872
+ "model.layers.4.attn_hc.fn",
873
+ "model.layers.4.attn_hc.scale",
874
+ "model.layers.4.ffn_hc.base",
875
+ "model.layers.4.ffn_hc.fn",
876
+ "model.layers.4.ffn_hc.scale",
877
+ "model.layers.4.input_layernorm",
878
+ "model.layers.4.mlp.gate",
879
+ "model.layers.4.mlp.gate.e_score_correction_bias",
880
+ "model.layers.4.post_attention_layernorm",
881
+ "model.layers.4.self_attn.forget_gate.A_log",
882
+ "model.layers.4.self_attn.b_proj",
883
+ "model.layers.4.self_attn.forget_gate.dt_bias",
884
+ "model.layers.4.self_attn.f_a_proj",
885
+ "model.layers.4.self_attn.f_b_proj",
886
+ "model.layers.4.self_attn.fused_qkvbfg_a_proj",
887
+ "model.layers.4.self_attn.g_a_proj",
888
+ "model.layers.4.self_attn.g_b_proj",
889
+ "model.layers.4.self_attn.k_conv1d",
890
+ "model.layers.4.self_attn.k_proj",
891
+ "model.layers.4.self_attn.o_norm",
892
+ "model.layers.4.self_attn.o_proj",
893
+ "model.layers.4.self_attn.q_conv1d",
894
+ "model.layers.4.self_attn.q_proj",
895
+ "model.layers.4.self_attn.qkv_proj",
896
+ "model.layers.4.self_attn.v_conv1d",
897
+ "model.layers.4.self_attn.v_proj",
898
+ "model.layers.40.attn_hc.base",
899
+ "model.layers.40.attn_hc.fn",
900
+ "model.layers.40.attn_hc.scale",
901
+ "model.layers.40.ffn_hc.base",
902
+ "model.layers.40.ffn_hc.fn",
903
+ "model.layers.40.ffn_hc.scale",
904
+ "model.layers.40.input_layernorm",
905
+ "model.layers.40.mlp.gate",
906
+ "model.layers.40.mlp.gate.e_score_correction_bias",
907
+ "model.layers.40.post_attention_layernorm",
908
+ "model.layers.40.self_attn.forget_gate.A_log",
909
+ "model.layers.40.self_attn.b_proj",
910
+ "model.layers.40.self_attn.forget_gate.dt_bias",
911
+ "model.layers.40.self_attn.f_a_proj",
912
+ "model.layers.40.self_attn.f_b_proj",
913
+ "model.layers.40.self_attn.fused_qkvbfg_a_proj",
914
+ "model.layers.40.self_attn.g_a_proj",
915
+ "model.layers.40.self_attn.g_b_proj",
916
+ "model.layers.40.self_attn.k_conv1d",
917
+ "model.layers.40.self_attn.k_proj",
918
+ "model.layers.40.self_attn.o_norm",
919
+ "model.layers.40.self_attn.o_proj",
920
+ "model.layers.40.self_attn.q_conv1d",
921
+ "model.layers.40.self_attn.q_proj",
922
+ "model.layers.40.self_attn.qkv_proj",
923
+ "model.layers.40.self_attn.v_conv1d",
924
+ "model.layers.40.self_attn.v_proj",
925
+ "model.layers.41.attn_hc.base",
926
+ "model.layers.41.attn_hc.fn",
927
+ "model.layers.41.attn_hc.scale",
928
+ "model.layers.41.ffn_hc.base",
929
+ "model.layers.41.ffn_hc.fn",
930
+ "model.layers.41.ffn_hc.scale",
931
+ "model.layers.41.input_layernorm",
932
+ "model.layers.41.mlp.gate",
933
+ "model.layers.41.mlp.gate.e_score_correction_bias",
934
+ "model.layers.41.post_attention_layernorm",
935
+ "model.layers.41.self_attn.forget_gate.A_log",
936
+ "model.layers.41.self_attn.b_proj",
937
+ "model.layers.41.self_attn.forget_gate.dt_bias",
938
+ "model.layers.41.self_attn.f_a_proj",
939
+ "model.layers.41.self_attn.f_b_proj",
940
+ "model.layers.41.self_attn.fused_qkvbfg_a_proj",
941
+ "model.layers.41.self_attn.g_a_proj",
942
+ "model.layers.41.self_attn.g_b_proj",
943
+ "model.layers.41.self_attn.k_conv1d",
944
+ "model.layers.41.self_attn.k_proj",
945
+ "model.layers.41.self_attn.o_norm",
946
+ "model.layers.41.self_attn.o_proj",
947
+ "model.layers.41.self_attn.q_conv1d",
948
+ "model.layers.41.self_attn.q_proj",
949
+ "model.layers.41.self_attn.qkv_proj",
950
+ "model.layers.41.self_attn.v_conv1d",
951
+ "model.layers.41.self_attn.v_proj",
952
+ "model.layers.42.attn_hc.base",
953
+ "model.layers.42.attn_hc.fn",
954
+ "model.layers.42.attn_hc.scale",
955
+ "model.layers.42.ffn_hc.base",
956
+ "model.layers.42.ffn_hc.fn",
957
+ "model.layers.42.ffn_hc.scale",
958
+ "model.layers.42.input_layernorm",
959
+ "model.layers.42.mlp.gate",
960
+ "model.layers.42.mlp.gate.e_score_correction_bias",
961
+ "model.layers.42.post_attention_layernorm",
962
+ "model.layers.42.self_attn.forget_gate.A_log",
963
+ "model.layers.42.self_attn.b_proj",
964
+ "model.layers.42.self_attn.forget_gate.dt_bias",
965
+ "model.layers.42.self_attn.f_a_proj",
966
+ "model.layers.42.self_attn.f_b_proj",
967
+ "model.layers.42.self_attn.fused_qkvbfg_a_proj",
968
+ "model.layers.42.self_attn.g_a_proj",
969
+ "model.layers.42.self_attn.g_b_proj",
970
+ "model.layers.42.self_attn.k_conv1d",
971
+ "model.layers.42.self_attn.k_proj",
972
+ "model.layers.42.self_attn.o_norm",
973
+ "model.layers.42.self_attn.o_proj",
974
+ "model.layers.42.self_attn.q_conv1d",
975
+ "model.layers.42.self_attn.q_proj",
976
+ "model.layers.42.self_attn.qkv_proj",
977
+ "model.layers.42.self_attn.v_conv1d",
978
+ "model.layers.42.self_attn.v_proj",
979
+ "model.layers.43.attn_hc.base",
980
+ "model.layers.43.attn_hc.fn",
981
+ "model.layers.43.attn_hc.scale",
982
+ "model.layers.43.ffn_hc.base",
983
+ "model.layers.43.ffn_hc.fn",
984
+ "model.layers.43.ffn_hc.scale",
985
+ "model.layers.43.input_layernorm",
986
+ "model.layers.43.mlp.gate",
987
+ "model.layers.43.mlp.gate.e_score_correction_bias",
988
+ "model.layers.43.post_attention_layernorm",
989
+ "model.layers.43.self_attn.indexer.index_kpool_compress_ape",
990
+ "model.layers.43.self_attn.indexer.index_kpool_compress_gate",
991
+ "model.layers.43.self_attn.indexer.k_norm",
992
+ "model.layers.43.self_attn.indexer.k_norm.bias",
993
+ "model.layers.43.self_attn.indexer.weights_proj",
994
+ "model.layers.43.self_attn.indexer.wk",
995
+ "model.layers.43.self_attn.indexer.wq_b",
996
+ "model.layers.43.self_attn.kv_a_layernorm",
997
+ "model.layers.43.self_attn.kv_b_proj",
998
+ "model.layers.43.self_attn.q_a_layernorm",
999
+ "model.layers.44.attn_hc.base",
1000
+ "model.layers.44.attn_hc.fn",
1001
+ "model.layers.44.attn_hc.scale",
1002
+ "model.layers.44.ffn_hc.base",
1003
+ "model.layers.44.ffn_hc.fn",
1004
+ "model.layers.44.ffn_hc.scale",
1005
+ "model.layers.44.input_layernorm",
1006
+ "model.layers.44.mlp.gate",
1007
+ "model.layers.44.mlp.gate.e_score_correction_bias",
1008
+ "model.layers.44.post_attention_layernorm",
1009
+ "model.layers.44.self_attn.forget_gate.A_log",
1010
+ "model.layers.44.self_attn.b_proj",
1011
+ "model.layers.44.self_attn.forget_gate.dt_bias",
1012
+ "model.layers.44.self_attn.f_a_proj",
1013
+ "model.layers.44.self_attn.f_b_proj",
1014
+ "model.layers.44.self_attn.fused_qkvbfg_a_proj",
1015
+ "model.layers.44.self_attn.g_a_proj",
1016
+ "model.layers.44.self_attn.g_b_proj",
1017
+ "model.layers.44.self_attn.k_conv1d",
1018
+ "model.layers.44.self_attn.k_proj",
1019
+ "model.layers.44.self_attn.o_norm",
1020
+ "model.layers.44.self_attn.o_proj",
1021
+ "model.layers.44.self_attn.q_conv1d",
1022
+ "model.layers.44.self_attn.q_proj",
1023
+ "model.layers.44.self_attn.qkv_proj",
1024
+ "model.layers.44.self_attn.v_conv1d",
1025
+ "model.layers.44.self_attn.v_proj",
1026
+ "model.layers.45.eh_proj",
1027
+ "model.layers.45.enorm",
1028
+ "model.layers.45.hnorm",
1029
+ "model.layers.45.input_layernorm",
1030
+ "model.layers.45.mlp.gate",
1031
+ "model.layers.45.mlp.gate.e_score_correction_bias",
1032
+ "model.layers.45.post_attention_layernorm",
1033
+ "model.layers.45.self_attn.indexer.index_kpool_compress_ape",
1034
+ "model.layers.45.self_attn.indexer.index_kpool_compress_gate",
1035
+ "model.layers.45.self_attn.indexer.k_norm",
1036
+ "model.layers.45.self_attn.indexer.k_norm.bias",
1037
+ "model.layers.45.self_attn.indexer.weights_proj",
1038
+ "model.layers.45.self_attn.indexer.wk",
1039
+ "model.layers.45.self_attn.indexer.wq_b",
1040
+ "model.layers.45.self_attn.kv_a_layernorm",
1041
+ "model.layers.45.self_attn.kv_b_proj",
1042
+ "model.layers.45.self_attn.q_a_layernorm",
1043
+ "model.layers.45.shared_head.norm",
1044
+ "model.layers.5.attn_hc.base",
1045
+ "model.layers.5.attn_hc.fn",
1046
+ "model.layers.5.attn_hc.scale",
1047
+ "model.layers.5.ffn_hc.base",
1048
+ "model.layers.5.ffn_hc.fn",
1049
+ "model.layers.5.ffn_hc.scale",
1050
+ "model.layers.5.input_layernorm",
1051
+ "model.layers.5.mlp.gate",
1052
+ "model.layers.5.mlp.gate.e_score_correction_bias",
1053
+ "model.layers.5.post_attention_layernorm",
1054
+ "model.layers.5.self_attn.forget_gate.A_log",
1055
+ "model.layers.5.self_attn.b_proj",
1056
+ "model.layers.5.self_attn.forget_gate.dt_bias",
1057
+ "model.layers.5.self_attn.f_a_proj",
1058
+ "model.layers.5.self_attn.f_b_proj",
1059
+ "model.layers.5.self_attn.fused_qkvbfg_a_proj",
1060
+ "model.layers.5.self_attn.g_a_proj",
1061
+ "model.layers.5.self_attn.g_b_proj",
1062
+ "model.layers.5.self_attn.k_conv1d",
1063
+ "model.layers.5.self_attn.k_proj",
1064
+ "model.layers.5.self_attn.o_norm",
1065
+ "model.layers.5.self_attn.o_proj",
1066
+ "model.layers.5.self_attn.q_conv1d",
1067
+ "model.layers.5.self_attn.q_proj",
1068
+ "model.layers.5.self_attn.qkv_proj",
1069
+ "model.layers.5.self_attn.v_conv1d",
1070
+ "model.layers.5.self_attn.v_proj",
1071
+ "model.layers.6.attn_hc.base",
1072
+ "model.layers.6.attn_hc.fn",
1073
+ "model.layers.6.attn_hc.scale",
1074
+ "model.layers.6.ffn_hc.base",
1075
+ "model.layers.6.ffn_hc.fn",
1076
+ "model.layers.6.ffn_hc.scale",
1077
+ "model.layers.6.input_layernorm",
1078
+ "model.layers.6.mlp.gate",
1079
+ "model.layers.6.mlp.gate.e_score_correction_bias",
1080
+ "model.layers.6.post_attention_layernorm",
1081
+ "model.layers.6.self_attn.forget_gate.A_log",
1082
+ "model.layers.6.self_attn.b_proj",
1083
+ "model.layers.6.self_attn.forget_gate.dt_bias",
1084
+ "model.layers.6.self_attn.f_a_proj",
1085
+ "model.layers.6.self_attn.f_b_proj",
1086
+ "model.layers.6.self_attn.fused_qkvbfg_a_proj",
1087
+ "model.layers.6.self_attn.g_a_proj",
1088
+ "model.layers.6.self_attn.g_b_proj",
1089
+ "model.layers.6.self_attn.k_conv1d",
1090
+ "model.layers.6.self_attn.k_proj",
1091
+ "model.layers.6.self_attn.o_norm",
1092
+ "model.layers.6.self_attn.o_proj",
1093
+ "model.layers.6.self_attn.q_conv1d",
1094
+ "model.layers.6.self_attn.q_proj",
1095
+ "model.layers.6.self_attn.qkv_proj",
1096
+ "model.layers.6.self_attn.v_conv1d",
1097
+ "model.layers.6.self_attn.v_proj",
1098
+ "model.layers.7.attn_hc.base",
1099
+ "model.layers.7.attn_hc.fn",
1100
+ "model.layers.7.attn_hc.scale",
1101
+ "model.layers.7.ffn_hc.base",
1102
+ "model.layers.7.ffn_hc.fn",
1103
+ "model.layers.7.ffn_hc.scale",
1104
+ "model.layers.7.input_layernorm",
1105
+ "model.layers.7.mlp.gate",
1106
+ "model.layers.7.mlp.gate.e_score_correction_bias",
1107
+ "model.layers.7.post_attention_layernorm",
1108
+ "model.layers.7.self_attn.indexer.index_kpool_compress_ape",
1109
+ "model.layers.7.self_attn.indexer.index_kpool_compress_gate",
1110
+ "model.layers.7.self_attn.indexer.k_norm",
1111
+ "model.layers.7.self_attn.indexer.k_norm.bias",
1112
+ "model.layers.7.self_attn.indexer.weights_proj",
1113
+ "model.layers.7.self_attn.indexer.wk",
1114
+ "model.layers.7.self_attn.indexer.wq_b",
1115
+ "model.layers.7.self_attn.kv_a_layernorm",
1116
+ "model.layers.7.self_attn.kv_b_proj",
1117
+ "model.layers.7.self_attn.q_a_layernorm",
1118
+ "model.layers.8.attn_hc.base",
1119
+ "model.layers.8.attn_hc.fn",
1120
+ "model.layers.8.attn_hc.scale",
1121
+ "model.layers.8.ffn_hc.base",
1122
+ "model.layers.8.ffn_hc.fn",
1123
+ "model.layers.8.ffn_hc.scale",
1124
+ "model.layers.8.input_layernorm",
1125
+ "model.layers.8.mlp.gate",
1126
+ "model.layers.8.mlp.gate.e_score_correction_bias",
1127
+ "model.layers.8.post_attention_layernorm",
1128
+ "model.layers.8.self_attn.forget_gate.A_log",
1129
+ "model.layers.8.self_attn.b_proj",
1130
+ "model.layers.8.self_attn.forget_gate.dt_bias",
1131
+ "model.layers.8.self_attn.f_a_proj",
1132
+ "model.layers.8.self_attn.f_b_proj",
1133
+ "model.layers.8.self_attn.fused_qkvbfg_a_proj",
1134
+ "model.layers.8.self_attn.g_a_proj",
1135
+ "model.layers.8.self_attn.g_b_proj",
1136
+ "model.layers.8.self_attn.k_conv1d",
1137
+ "model.layers.8.self_attn.k_proj",
1138
+ "model.layers.8.self_attn.o_norm",
1139
+ "model.layers.8.self_attn.o_proj",
1140
+ "model.layers.8.self_attn.q_conv1d",
1141
+ "model.layers.8.self_attn.q_proj",
1142
+ "model.layers.8.self_attn.qkv_proj",
1143
+ "model.layers.8.self_attn.v_conv1d",
1144
+ "model.layers.8.self_attn.v_proj",
1145
+ "model.layers.9.attn_hc.base",
1146
+ "model.layers.9.attn_hc.fn",
1147
+ "model.layers.9.attn_hc.scale",
1148
+ "model.layers.9.ffn_hc.base",
1149
+ "model.layers.9.ffn_hc.fn",
1150
+ "model.layers.9.ffn_hc.scale",
1151
+ "model.layers.9.input_layernorm",
1152
+ "model.layers.9.mlp.gate",
1153
+ "model.layers.9.mlp.gate.e_score_correction_bias",
1154
+ "model.layers.9.post_attention_layernorm",
1155
+ "model.layers.9.self_attn.forget_gate.A_log",
1156
+ "model.layers.9.self_attn.b_proj",
1157
+ "model.layers.9.self_attn.forget_gate.dt_bias",
1158
+ "model.layers.9.self_attn.f_a_proj",
1159
+ "model.layers.9.self_attn.f_b_proj",
1160
+ "model.layers.9.self_attn.fused_qkvbfg_a_proj",
1161
+ "model.layers.9.self_attn.g_a_proj",
1162
+ "model.layers.9.self_attn.g_b_proj",
1163
+ "model.layers.9.self_attn.k_conv1d",
1164
+ "model.layers.9.self_attn.k_proj",
1165
+ "model.layers.9.self_attn.o_norm",
1166
+ "model.layers.9.self_attn.o_proj",
1167
+ "model.layers.9.self_attn.q_conv1d",
1168
+ "model.layers.9.self_attn.q_proj",
1169
+ "model.layers.9.self_attn.qkv_proj",
1170
+ "model.layers.9.self_attn.v_conv1d",
1171
+ "model.layers.9.self_attn.v_proj",
1172
+ "model.norm",
1173
+ "model.visual",
1174
+ "router",
1175
+ "visual",
1176
+ "visual.blocks.0.attn.k_norm",
1177
+ "visual.blocks.0.attn.proj",
1178
+ "visual.blocks.0.attn.proj.bias",
1179
+ "visual.blocks.0.attn.q_norm",
1180
+ "visual.blocks.0.attn.qkv",
1181
+ "visual.blocks.0.attn.qkv.bias",
1182
+ "visual.blocks.0.mlp.down_proj",
1183
+ "visual.blocks.0.mlp.down_proj.bias",
1184
+ "visual.blocks.0.mlp.gate_proj",
1185
+ "visual.blocks.0.mlp.gate_proj.bias",
1186
+ "visual.blocks.0.mlp.up_proj",
1187
+ "visual.blocks.0.mlp.up_proj.bias",
1188
+ "visual.blocks.0.norm1",
1189
+ "visual.blocks.0.norm2",
1190
+ "visual.blocks.1.attn.k_norm",
1191
+ "visual.blocks.1.attn.proj",
1192
+ "visual.blocks.1.attn.proj.bias",
1193
+ "visual.blocks.1.attn.q_norm",
1194
+ "visual.blocks.1.attn.qkv",
1195
+ "visual.blocks.1.attn.qkv.bias",
1196
+ "visual.blocks.1.mlp.down_proj",
1197
+ "visual.blocks.1.mlp.down_proj.bias",
1198
+ "visual.blocks.1.mlp.gate_proj",
1199
+ "visual.blocks.1.mlp.gate_proj.bias",
1200
+ "visual.blocks.1.mlp.up_proj",
1201
+ "visual.blocks.1.mlp.up_proj.bias",
1202
+ "visual.blocks.1.norm1",
1203
+ "visual.blocks.1.norm2",
1204
+ "visual.blocks.10.attn.k_norm",
1205
+ "visual.blocks.10.attn.proj",
1206
+ "visual.blocks.10.attn.proj.bias",
1207
+ "visual.blocks.10.attn.q_norm",
1208
+ "visual.blocks.10.attn.qkv",
1209
+ "visual.blocks.10.attn.qkv.bias",
1210
+ "visual.blocks.10.mlp.down_proj",
1211
+ "visual.blocks.10.mlp.down_proj.bias",
1212
+ "visual.blocks.10.mlp.gate_proj",
1213
+ "visual.blocks.10.mlp.gate_proj.bias",
1214
+ "visual.blocks.10.mlp.up_proj",
1215
+ "visual.blocks.10.mlp.up_proj.bias",
1216
+ "visual.blocks.10.norm1",
1217
+ "visual.blocks.10.norm2",
1218
+ "visual.blocks.11.attn.k_norm",
1219
+ "visual.blocks.11.attn.proj",
1220
+ "visual.blocks.11.attn.proj.bias",
1221
+ "visual.blocks.11.attn.q_norm",
1222
+ "visual.blocks.11.attn.qkv",
1223
+ "visual.blocks.11.attn.qkv.bias",
1224
+ "visual.blocks.11.mlp.down_proj",
1225
+ "visual.blocks.11.mlp.down_proj.bias",
1226
+ "visual.blocks.11.mlp.gate_proj",
1227
+ "visual.blocks.11.mlp.gate_proj.bias",
1228
+ "visual.blocks.11.mlp.up_proj",
1229
+ "visual.blocks.11.mlp.up_proj.bias",
1230
+ "visual.blocks.11.norm1",
1231
+ "visual.blocks.11.norm2",
1232
+ "visual.blocks.12.attn.k_norm",
1233
+ "visual.blocks.12.attn.proj",
1234
+ "visual.blocks.12.attn.proj.bias",
1235
+ "visual.blocks.12.attn.q_norm",
1236
+ "visual.blocks.12.attn.qkv",
1237
+ "visual.blocks.12.attn.qkv.bias",
1238
+ "visual.blocks.12.mlp.down_proj",
1239
+ "visual.blocks.12.mlp.down_proj.bias",
1240
+ "visual.blocks.12.mlp.gate_proj",
1241
+ "visual.blocks.12.mlp.gate_proj.bias",
1242
+ "visual.blocks.12.mlp.up_proj",
1243
+ "visual.blocks.12.mlp.up_proj.bias",
1244
+ "visual.blocks.12.norm1",
1245
+ "visual.blocks.12.norm2",
1246
+ "visual.blocks.13.attn.k_norm",
1247
+ "visual.blocks.13.attn.proj",
1248
+ "visual.blocks.13.attn.proj.bias",
1249
+ "visual.blocks.13.attn.q_norm",
1250
+ "visual.blocks.13.attn.qkv",
1251
+ "visual.blocks.13.attn.qkv.bias",
1252
+ "visual.blocks.13.mlp.down_proj",
1253
+ "visual.blocks.13.mlp.down_proj.bias",
1254
+ "visual.blocks.13.mlp.gate_proj",
1255
+ "visual.blocks.13.mlp.gate_proj.bias",
1256
+ "visual.blocks.13.mlp.up_proj",
1257
+ "visual.blocks.13.mlp.up_proj.bias",
1258
+ "visual.blocks.13.norm1",
1259
+ "visual.blocks.13.norm2",
1260
+ "visual.blocks.14.attn.k_norm",
1261
+ "visual.blocks.14.attn.proj",
1262
+ "visual.blocks.14.attn.proj.bias",
1263
+ "visual.blocks.14.attn.q_norm",
1264
+ "visual.blocks.14.attn.qkv",
1265
+ "visual.blocks.14.attn.qkv.bias",
1266
+ "visual.blocks.14.mlp.down_proj",
1267
+ "visual.blocks.14.mlp.down_proj.bias",
1268
+ "visual.blocks.14.mlp.gate_proj",
1269
+ "visual.blocks.14.mlp.gate_proj.bias",
1270
+ "visual.blocks.14.mlp.up_proj",
1271
+ "visual.blocks.14.mlp.up_proj.bias",
1272
+ "visual.blocks.14.norm1",
1273
+ "visual.blocks.14.norm2",
1274
+ "visual.blocks.15.attn.k_norm",
1275
+ "visual.blocks.15.attn.proj",
1276
+ "visual.blocks.15.attn.proj.bias",
1277
+ "visual.blocks.15.attn.q_norm",
1278
+ "visual.blocks.15.attn.qkv",
1279
+ "visual.blocks.15.attn.qkv.bias",
1280
+ "visual.blocks.15.mlp.down_proj",
1281
+ "visual.blocks.15.mlp.down_proj.bias",
1282
+ "visual.blocks.15.mlp.gate_proj",
1283
+ "visual.blocks.15.mlp.gate_proj.bias",
1284
+ "visual.blocks.15.mlp.up_proj",
1285
+ "visual.blocks.15.mlp.up_proj.bias",
1286
+ "visual.blocks.15.norm1",
1287
+ "visual.blocks.15.norm2",
1288
+ "visual.blocks.16.attn.k_norm",
1289
+ "visual.blocks.16.attn.proj",
1290
+ "visual.blocks.16.attn.proj.bias",
1291
+ "visual.blocks.16.attn.q_norm",
1292
+ "visual.blocks.16.attn.qkv",
1293
+ "visual.blocks.16.attn.qkv.bias",
1294
+ "visual.blocks.16.mlp.down_proj",
1295
+ "visual.blocks.16.mlp.down_proj.bias",
1296
+ "visual.blocks.16.mlp.gate_proj",
1297
+ "visual.blocks.16.mlp.gate_proj.bias",
1298
+ "visual.blocks.16.mlp.up_proj",
1299
+ "visual.blocks.16.mlp.up_proj.bias",
1300
+ "visual.blocks.16.norm1",
1301
+ "visual.blocks.16.norm2",
1302
+ "visual.blocks.17.attn.k_norm",
1303
+ "visual.blocks.17.attn.proj",
1304
+ "visual.blocks.17.attn.proj.bias",
1305
+ "visual.blocks.17.attn.q_norm",
1306
+ "visual.blocks.17.attn.qkv",
1307
+ "visual.blocks.17.attn.qkv.bias",
1308
+ "visual.blocks.17.mlp.down_proj",
1309
+ "visual.blocks.17.mlp.down_proj.bias",
1310
+ "visual.blocks.17.mlp.gate_proj",
1311
+ "visual.blocks.17.mlp.gate_proj.bias",
1312
+ "visual.blocks.17.mlp.up_proj",
1313
+ "visual.blocks.17.mlp.up_proj.bias",
1314
+ "visual.blocks.17.norm1",
1315
+ "visual.blocks.17.norm2",
1316
+ "visual.blocks.18.attn.k_norm",
1317
+ "visual.blocks.18.attn.proj",
1318
+ "visual.blocks.18.attn.proj.bias",
1319
+ "visual.blocks.18.attn.q_norm",
1320
+ "visual.blocks.18.attn.qkv",
1321
+ "visual.blocks.18.attn.qkv.bias",
1322
+ "visual.blocks.18.mlp.down_proj",
1323
+ "visual.blocks.18.mlp.down_proj.bias",
1324
+ "visual.blocks.18.mlp.gate_proj",
1325
+ "visual.blocks.18.mlp.gate_proj.bias",
1326
+ "visual.blocks.18.mlp.up_proj",
1327
+ "visual.blocks.18.mlp.up_proj.bias",
1328
+ "visual.blocks.18.norm1",
1329
+ "visual.blocks.18.norm2",
1330
+ "visual.blocks.19.attn.k_norm",
1331
+ "visual.blocks.19.attn.proj",
1332
+ "visual.blocks.19.attn.proj.bias",
1333
+ "visual.blocks.19.attn.q_norm",
1334
+ "visual.blocks.19.attn.qkv",
1335
+ "visual.blocks.19.attn.qkv.bias",
1336
+ "visual.blocks.19.mlp.down_proj",
1337
+ "visual.blocks.19.mlp.down_proj.bias",
1338
+ "visual.blocks.19.mlp.gate_proj",
1339
+ "visual.blocks.19.mlp.gate_proj.bias",
1340
+ "visual.blocks.19.mlp.up_proj",
1341
+ "visual.blocks.19.mlp.up_proj.bias",
1342
+ "visual.blocks.19.norm1",
1343
+ "visual.blocks.19.norm2",
1344
+ "visual.blocks.2.attn.k_norm",
1345
+ "visual.blocks.2.attn.proj",
1346
+ "visual.blocks.2.attn.proj.bias",
1347
+ "visual.blocks.2.attn.q_norm",
1348
+ "visual.blocks.2.attn.qkv",
1349
+ "visual.blocks.2.attn.qkv.bias",
1350
+ "visual.blocks.2.mlp.down_proj",
1351
+ "visual.blocks.2.mlp.down_proj.bias",
1352
+ "visual.blocks.2.mlp.gate_proj",
1353
+ "visual.blocks.2.mlp.gate_proj.bias",
1354
+ "visual.blocks.2.mlp.up_proj",
1355
+ "visual.blocks.2.mlp.up_proj.bias",
1356
+ "visual.blocks.2.norm1",
1357
+ "visual.blocks.2.norm2",
1358
+ "visual.blocks.20.attn.k_norm",
1359
+ "visual.blocks.20.attn.proj",
1360
+ "visual.blocks.20.attn.proj.bias",
1361
+ "visual.blocks.20.attn.q_norm",
1362
+ "visual.blocks.20.attn.qkv",
1363
+ "visual.blocks.20.attn.qkv.bias",
1364
+ "visual.blocks.20.mlp.down_proj",
1365
+ "visual.blocks.20.mlp.down_proj.bias",
1366
+ "visual.blocks.20.mlp.gate_proj",
1367
+ "visual.blocks.20.mlp.gate_proj.bias",
1368
+ "visual.blocks.20.mlp.up_proj",
1369
+ "visual.blocks.20.mlp.up_proj.bias",
1370
+ "visual.blocks.20.norm1",
1371
+ "visual.blocks.20.norm2",
1372
+ "visual.blocks.21.attn.k_norm",
1373
+ "visual.blocks.21.attn.proj",
1374
+ "visual.blocks.21.attn.proj.bias",
1375
+ "visual.blocks.21.attn.q_norm",
1376
+ "visual.blocks.21.attn.qkv",
1377
+ "visual.blocks.21.attn.qkv.bias",
1378
+ "visual.blocks.21.mlp.down_proj",
1379
+ "visual.blocks.21.mlp.down_proj.bias",
1380
+ "visual.blocks.21.mlp.gate_proj",
1381
+ "visual.blocks.21.mlp.gate_proj.bias",
1382
+ "visual.blocks.21.mlp.up_proj",
1383
+ "visual.blocks.21.mlp.up_proj.bias",
1384
+ "visual.blocks.21.norm1",
1385
+ "visual.blocks.21.norm2",
1386
+ "visual.blocks.22.attn.k_norm",
1387
+ "visual.blocks.22.attn.proj",
1388
+ "visual.blocks.22.attn.proj.bias",
1389
+ "visual.blocks.22.attn.q_norm",
1390
+ "visual.blocks.22.attn.qkv",
1391
+ "visual.blocks.22.attn.qkv.bias",
1392
+ "visual.blocks.22.mlp.down_proj",
1393
+ "visual.blocks.22.mlp.down_proj.bias",
1394
+ "visual.blocks.22.mlp.gate_proj",
1395
+ "visual.blocks.22.mlp.gate_proj.bias",
1396
+ "visual.blocks.22.mlp.up_proj",
1397
+ "visual.blocks.22.mlp.up_proj.bias",
1398
+ "visual.blocks.22.norm1",
1399
+ "visual.blocks.22.norm2",
1400
+ "visual.blocks.23.attn.k_norm",
1401
+ "visual.blocks.23.attn.proj",
1402
+ "visual.blocks.23.attn.proj.bias",
1403
+ "visual.blocks.23.attn.q_norm",
1404
+ "visual.blocks.23.attn.qkv",
1405
+ "visual.blocks.23.attn.qkv.bias",
1406
+ "visual.blocks.23.mlp.down_proj",
1407
+ "visual.blocks.23.mlp.down_proj.bias",
1408
+ "visual.blocks.23.mlp.gate_proj",
1409
+ "visual.blocks.23.mlp.gate_proj.bias",
1410
+ "visual.blocks.23.mlp.up_proj",
1411
+ "visual.blocks.23.mlp.up_proj.bias",
1412
+ "visual.blocks.23.norm1",
1413
+ "visual.blocks.23.norm2",
1414
+ "visual.blocks.3.attn.k_norm",
1415
+ "visual.blocks.3.attn.proj",
1416
+ "visual.blocks.3.attn.proj.bias",
1417
+ "visual.blocks.3.attn.q_norm",
1418
+ "visual.blocks.3.attn.qkv",
1419
+ "visual.blocks.3.attn.qkv.bias",
1420
+ "visual.blocks.3.mlp.down_proj",
1421
+ "visual.blocks.3.mlp.down_proj.bias",
1422
+ "visual.blocks.3.mlp.gate_proj",
1423
+ "visual.blocks.3.mlp.gate_proj.bias",
1424
+ "visual.blocks.3.mlp.up_proj",
1425
+ "visual.blocks.3.mlp.up_proj.bias",
1426
+ "visual.blocks.3.norm1",
1427
+ "visual.blocks.3.norm2",
1428
+ "visual.blocks.4.attn.k_norm",
1429
+ "visual.blocks.4.attn.proj",
1430
+ "visual.blocks.4.attn.proj.bias",
1431
+ "visual.blocks.4.attn.q_norm",
1432
+ "visual.blocks.4.attn.qkv",
1433
+ "visual.blocks.4.attn.qkv.bias",
1434
+ "visual.blocks.4.mlp.down_proj",
1435
+ "visual.blocks.4.mlp.down_proj.bias",
1436
+ "visual.blocks.4.mlp.gate_proj",
1437
+ "visual.blocks.4.mlp.gate_proj.bias",
1438
+ "visual.blocks.4.mlp.up_proj",
1439
+ "visual.blocks.4.mlp.up_proj.bias",
1440
+ "visual.blocks.4.norm1",
1441
+ "visual.blocks.4.norm2",
1442
+ "visual.blocks.5.attn.k_norm",
1443
+ "visual.blocks.5.attn.proj",
1444
+ "visual.blocks.5.attn.proj.bias",
1445
+ "visual.blocks.5.attn.q_norm",
1446
+ "visual.blocks.5.attn.qkv",
1447
+ "visual.blocks.5.attn.qkv.bias",
1448
+ "visual.blocks.5.mlp.down_proj",
1449
+ "visual.blocks.5.mlp.down_proj.bias",
1450
+ "visual.blocks.5.mlp.gate_proj",
1451
+ "visual.blocks.5.mlp.gate_proj.bias",
1452
+ "visual.blocks.5.mlp.up_proj",
1453
+ "visual.blocks.5.mlp.up_proj.bias",
1454
+ "visual.blocks.5.norm1",
1455
+ "visual.blocks.5.norm2",
1456
+ "visual.blocks.6.attn.k_norm",
1457
+ "visual.blocks.6.attn.proj",
1458
+ "visual.blocks.6.attn.proj.bias",
1459
+ "visual.blocks.6.attn.q_norm",
1460
+ "visual.blocks.6.attn.qkv",
1461
+ "visual.blocks.6.attn.qkv.bias",
1462
+ "visual.blocks.6.mlp.down_proj",
1463
+ "visual.blocks.6.mlp.down_proj.bias",
1464
+ "visual.blocks.6.mlp.gate_proj",
1465
+ "visual.blocks.6.mlp.gate_proj.bias",
1466
+ "visual.blocks.6.mlp.up_proj",
1467
+ "visual.blocks.6.mlp.up_proj.bias",
1468
+ "visual.blocks.6.norm1",
1469
+ "visual.blocks.6.norm2",
1470
+ "visual.blocks.7.attn.k_norm",
1471
+ "visual.blocks.7.attn.proj",
1472
+ "visual.blocks.7.attn.proj.bias",
1473
+ "visual.blocks.7.attn.q_norm",
1474
+ "visual.blocks.7.attn.qkv",
1475
+ "visual.blocks.7.attn.qkv.bias",
1476
+ "visual.blocks.7.mlp.down_proj",
1477
+ "visual.blocks.7.mlp.down_proj.bias",
1478
+ "visual.blocks.7.mlp.gate_proj",
1479
+ "visual.blocks.7.mlp.gate_proj.bias",
1480
+ "visual.blocks.7.mlp.up_proj",
1481
+ "visual.blocks.7.mlp.up_proj.bias",
1482
+ "visual.blocks.7.norm1",
1483
+ "visual.blocks.7.norm2",
1484
+ "visual.blocks.8.attn.k_norm",
1485
+ "visual.blocks.8.attn.proj",
1486
+ "visual.blocks.8.attn.proj.bias",
1487
+ "visual.blocks.8.attn.q_norm",
1488
+ "visual.blocks.8.attn.qkv",
1489
+ "visual.blocks.8.attn.qkv.bias",
1490
+ "visual.blocks.8.mlp.down_proj",
1491
+ "visual.blocks.8.mlp.down_proj.bias",
1492
+ "visual.blocks.8.mlp.gate_proj",
1493
+ "visual.blocks.8.mlp.gate_proj.bias",
1494
+ "visual.blocks.8.mlp.up_proj",
1495
+ "visual.blocks.8.mlp.up_proj.bias",
1496
+ "visual.blocks.8.norm1",
1497
+ "visual.blocks.8.norm2",
1498
+ "visual.blocks.9.attn.k_norm",
1499
+ "visual.blocks.9.attn.proj",
1500
+ "visual.blocks.9.attn.proj.bias",
1501
+ "visual.blocks.9.attn.q_norm",
1502
+ "visual.blocks.9.attn.qkv",
1503
+ "visual.blocks.9.attn.qkv.bias",
1504
+ "visual.blocks.9.mlp.down_proj",
1505
+ "visual.blocks.9.mlp.down_proj.bias",
1506
+ "visual.blocks.9.mlp.gate_proj",
1507
+ "visual.blocks.9.mlp.gate_proj.bias",
1508
+ "visual.blocks.9.mlp.up_proj",
1509
+ "visual.blocks.9.mlp.up_proj.bias",
1510
+ "visual.blocks.9.norm1",
1511
+ "visual.blocks.9.norm2",
1512
+ "visual.downsample",
1513
+ "visual.downsample.bias",
1514
+ "visual.merger.down_proj",
1515
+ "visual.merger.gate_proj",
1516
+ "visual.merger.post_projection_norm",
1517
+ "visual.merger.post_projection_norm.bias",
1518
+ "visual.merger.proj",
1519
+ "visual.merger.up_proj",
1520
+ "visual.patch_embed.proj",
1521
+ "visual.patch_embed.proj.bias",
1522
+ "visual.post_layernorm",
1523
+ "weights_proj"
1524
+ ],
1525
+ "quant_method": "fp8",
1526
+ "scale_fmt": "float",
1527
+ "weight_block_size": [
1528
+ 32,
1529
+ 32
1530
+ ]
1531
+ },
1532
+ "text_config": {
1533
+ "attention_bias": false,
1534
+ "attention_dropout": 0.0,
1535
+ "bos_token_id": null,
1536
+ "dtype": "bfloat16",
1537
+ "eos_token_id": [
1538
+ 154820,
1539
+ 154827,
1540
+ 154829
1541
+ ],
1542
+ "hc_eps": 1e-06,
1543
+ "hc_mult": 4,
1544
+ "hc_sinkhorn_iters": 20,
1545
+ "head_dim": 0,
1546
+ "hidden_act": "silu",
1547
+ "hidden_size": 32,
1548
+ "index_head_dim": 4,
1549
+ "index_kpool": 4,
1550
+ "index_kpool_always_select_tail": true,
1551
+ "index_n_heads": 2,
1552
+ "index_topk": 8,
1553
+ "indexer_types": [
1554
+ "full",
1555
+ "full",
1556
+ "full",
1557
+ "full"
1558
+ ],
1559
+ "initializer_range": 0.02,
1560
+ "intermediate_size": 32,
1561
+ "kv_lora_rank": 32,
1562
+ "layer_types": [
1563
+ "linear_attention",
1564
+ "linear_attention",
1565
+ "linear_attention",
1566
+ "deepseek_sparse_attention"
1567
+ ],
1568
+ "linear_conv_kernel_dim": 4,
1569
+ "linear_head_dim": 8,
1570
+ "linear_lower_bound": -5.0,
1571
+ "linear_num_heads": 4,
1572
+ "max_position_embeddings": 1048576,
1573
+ "mlp_layer_types": [
1574
+ "dense",
1575
+ "dense",
1576
+ "dense",
1577
+ "sparse"
1578
+ ],
1579
+ "model_type": "glm5_next_text",
1580
+ "moe_intermediate_size": 64,
1581
+ "n_group": 1,
1582
+ "n_routed_experts": 4,
1583
+ "n_shared_experts": 1,
1584
+ "norm_topk_prob": true,
1585
+ "num_attention_heads": 4,
1586
+ "num_experts_per_tok": 2,
1587
+ "num_hidden_layers": 4,
1588
+ "num_key_value_heads": 4,
1589
+ "output_router_logits": false,
1590
+ "pad_token_id": 154820,
1591
+ "q_lora_rank": 32,
1592
+ "qk_head_dim": 8,
1593
+ "qk_nope_head_dim": 8,
1594
+ "qk_rope_head_dim": 0,
1595
+ "rms_norm_eps": 1e-05,
1596
+ "routed_scaling_factor": 2.5,
1597
+ "router_aux_loss_coef": 0.001,
1598
+ "swiglu_limit": 10.0,
1599
+ "tie_word_embeddings": false,
1600
+ "topk_group": 1,
1601
+ "use_cache": true,
1602
+ "v_head_dim": 8,
1603
+ "vocab_size": 154880
1604
+ },
1605
+ "tie_word_embeddings": false,
1606
+ "transformers_version": "5.17.0",
1607
+ "video_end_token_id": 154833,
1608
+ "video_start_token_id": 154832,
1609
+ "video_token_id": 154855,
1610
+ "vision_config": {
1611
+ "attention_bias": true,
1612
+ "attention_dropout": 0.0,
1613
+ "depth": 2,
1614
+ "dtype": "bfloat16",
1615
+ "hidden_act": "silu",
1616
+ "hidden_size": 16,
1617
+ "image_size": 448,
1618
+ "in_channels": 3,
1619
+ "initializer_range": 0.02,
1620
+ "intermediate_size": 32,
1621
+ "model_type": "glm5_next_vision",
1622
+ "num_heads": 4,
1623
+ "out_hidden_size": 32,
1624
+ "patch_size": 14,
1625
+ "projection_intermediate_size": 32,
1626
+ "rms_norm_eps": 1e-05,
1627
+ "rope_parameters": {
1628
+ "rope_theta": 10000.0,
1629
+ "rope_type": "axial"
1630
+ },
1631
+ "spatial_merge_size": 2,
1632
+ "swiglu_limit": 10.0,
1633
+ "temporal_patch_size": 2
1634
+ }
1635
+ }
generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "transformers_version": "5.17.0"
11
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3039383183d86258b73dad4e73b6a425fed8655007d5aea94aebfaad42f4193
3
+ size 20043924
processor_config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "do_convert_rgb": true,
4
+ "do_normalize": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.48145466,
9
+ 0.4578275,
10
+ 0.40821073
11
+ ],
12
+ "image_processor_type": "Glm5NextImageProcessor",
13
+ "image_std": [
14
+ 0.26862954,
15
+ 0.26130258,
16
+ 0.27577711
17
+ ],
18
+ "max_image_tokens": 8000,
19
+ "merge_size": 2,
20
+ "min_image_tokens": 16,
21
+ "patch_expand_factor": 1,
22
+ "patch_size": 14,
23
+ "resample": 3,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "longest_edge": 1
27
+ },
28
+ "temporal_patch_size": 2
29
+ },
30
+ "processor_class": "Glm5NextProcessor",
31
+ "video_processor": {
32
+ "do_convert_rgb": true,
33
+ "do_normalize": true,
34
+ "do_rescale": true,
35
+ "do_resize": true,
36
+ "do_sample_frames": true,
37
+ "fps": 2,
38
+ "image_mean": [
39
+ 0.48145466,
40
+ 0.4578275,
41
+ 0.40821073
42
+ ],
43
+ "image_std": [
44
+ 0.26862954,
45
+ 0.26130258,
46
+ 0.27577711
47
+ ],
48
+ "max_frames": 2048,
49
+ "max_image_size": {
50
+ "longest_edge": 47040000
51
+ },
52
+ "max_image_tokens": 240000,
53
+ "merge_size": 2,
54
+ "min_image_tokens": 16,
55
+ "num_frames": 16,
56
+ "patch_expand_factor": 1,
57
+ "patch_size": 14,
58
+ "resample": 3,
59
+ "rescale_factor": 0.00392156862745098,
60
+ "return_metadata": false,
61
+ "size": {
62
+ "longest_edge": 1
63
+ },
64
+ "temporal_patch_size": 2,
65
+ "video_processor_type": "Glm5NextVideoProcessor"
66
+ }
67
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f2005e8bb3fa61c8b16cb0177e09eae30726d3658e0302da790ca27b5fbfd6b
3
+ size 20217609
tokenizer_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
27
+ "local_files_only": false,
28
+ "model_max_length": 1048576,
29
+ "model_specific_special_tokens": {},
30
+ "pad_token": "<|endoftext|>",
31
+ "padding_side": "left",
32
+ "processor_class": "Glm5NextProcessor",
33
+ "remove_space": false,
34
+ "tokenizer_class": "TokenizersBackend"
35
+ }