ovanerven commited on
Commit
a3db747
·
verified ·
1 Parent(s): c5edc12

Upload folder using huggingface_hub

Browse files
.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
chat_template.jinja ADDED
@@ -0,0 +1,344 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- macro format_parameters(properties, required) -%}
2
+ {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
3
+ {%- set ns = namespace(found_first=false) -%}
4
+ {%- for key, value in properties | dictsort -%}
5
+ {%- set add_comma = false -%}
6
+ {%- if key not in standard_keys -%}
7
+ {%- if ns.found_first %},{% endif -%}
8
+ {%- set ns.found_first = true -%}
9
+ {{ key }}:{
10
+ {%- if value['description'] -%}
11
+ description:<|"|>{{ value['description'] }}<|"|>
12
+ {%- set add_comma = true -%}
13
+ {%- endif -%}
14
+ {%- if value['type'] | upper == 'STRING' -%}
15
+ {%- if value['enum'] -%}
16
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
17
+ enum:{{ format_argument(value['enum']) }}
18
+ {%- endif -%}
19
+ {%- elif value['type'] | upper == 'ARRAY' -%}
20
+ {%- if value['items'] is mapping and value['items'] -%}
21
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
22
+ items:{
23
+ {%- set ns_items = namespace(found_first=false) -%}
24
+ {%- for item_key, item_value in value['items'] | dictsort -%}
25
+ {%- if item_value is not none -%}
26
+ {%- if ns_items.found_first %},{% endif -%}
27
+ {%- set ns_items.found_first = true -%}
28
+ {%- if item_key == 'properties' -%}
29
+ properties:{
30
+ {%- if item_value is mapping -%}
31
+ {{- format_parameters(item_value, value['items']['required'] | default([])) -}}
32
+ {%- endif -%}
33
+ }
34
+ {%- elif item_key == 'required' -%}
35
+ required:[
36
+ {%- for req_item in item_value -%}
37
+ <|"|>{{- req_item -}}<|"|>
38
+ {%- if not loop.last %},{% endif -%}
39
+ {%- endfor -%}
40
+ ]
41
+ {%- elif item_key == 'type' -%}
42
+ {%- if item_value is string -%}
43
+ type:{{ format_argument(item_value | upper) }}
44
+ {%- else -%}
45
+ type:{{ format_argument(item_value | map('upper') | list) }}
46
+ {%- endif -%}
47
+ {%- else -%}
48
+ {{ item_key }}:{{ format_argument(item_value) }}
49
+ {%- endif -%}
50
+ {%- endif -%}
51
+ {%- endfor -%}
52
+ }
53
+ {%- endif -%}
54
+ {%- endif -%}
55
+ {%- if value['nullable'] %}
56
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
57
+ nullable:true
58
+ {%- endif -%}
59
+ {%- if value['type'] | upper == 'OBJECT' -%}
60
+ {%- if value['properties'] is defined and value['properties'] is mapping -%}
61
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
62
+ properties:{
63
+ {{- format_parameters(value['properties'], value['required'] | default([])) -}}
64
+ }
65
+ {%- elif value is mapping -%}
66
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
67
+ properties:{
68
+ {{- format_parameters(value, value['required'] | default([])) -}}
69
+ }
70
+ {%- endif -%}
71
+ {%- if value['required'] -%}
72
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
73
+ required:[
74
+ {%- for item in value['required'] | default([]) -%}
75
+ <|"|>{{- item -}}<|"|>
76
+ {%- if not loop.last %},{% endif -%}
77
+ {%- endfor -%}
78
+ ]
79
+ {%- endif -%}
80
+ {%- endif -%}
81
+ {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
82
+ type:<|"|>{{ value['type'] | upper }}<|"|>}
83
+ {%- endif -%}
84
+ {%- endfor -%}
85
+ {%- endmacro -%}
86
+ {%- macro format_function_declaration(tool_data) -%}
87
+ declaration:{{- tool_data['function']['name'] -}}{description:<|"|>{{- tool_data['function']['description'] -}}<|"|>
88
+ {%- set params = tool_data['function']['parameters'] -%}
89
+ {%- if params -%}
90
+ ,parameters:{
91
+ {%- if params['properties'] -%}
92
+ properties:{ {{- format_parameters(params['properties'], params['required']) -}} },
93
+ {%- endif -%}
94
+ {%- if params['required'] -%}
95
+ required:[
96
+ {%- for item in params['required'] -%}
97
+ <|"|>{{- item -}}<|"|>
98
+ {{- ',' if not loop.last -}}
99
+ {%- endfor -%}
100
+ ],
101
+ {%- endif -%}
102
+ {%- if params['type'] -%}
103
+ type:<|"|>{{- params['type'] | upper -}}<|"|>}
104
+ {%- endif -%}
105
+ {%- endif -%}
106
+ {%- if 'response' in tool_data['function'] -%}
107
+ {%- set response_declaration = tool_data['function']['response'] -%}
108
+ ,response:{
109
+ {%- if response_declaration['description'] -%}
110
+ description:<|"|>{{- response_declaration['description'] -}}<|"|>,
111
+ {%- endif -%}
112
+ {%- if response_declaration['type'] | upper == 'OBJECT' -%}
113
+ type:<|"|>{{- response_declaration['type'] | upper -}}<|"|>}
114
+ {%- endif -%}
115
+ {%- endif -%}
116
+ }
117
+ {%- endmacro -%}
118
+ {%- macro format_argument(argument, escape_keys=True) -%}
119
+ {%- if argument is string -%}
120
+ {{- '<|"|>' + argument + '<|"|>' -}}
121
+ {%- elif argument is boolean -%}
122
+ {{- 'true' if argument else 'false' -}}
123
+ {%- elif argument is mapping -%}
124
+ {{- '{' -}}
125
+ {%- set ns = namespace(found_first=false) -%}
126
+ {%- for key, value in argument | dictsort -%}
127
+ {%- if ns.found_first %},{% endif -%}
128
+ {%- set ns.found_first = true -%}
129
+ {%- if escape_keys -%}
130
+ {{- '<|"|>' + key + '<|"|>' -}}
131
+ {%- else -%}
132
+ {{- key -}}
133
+ {%- endif -%}
134
+ :{{- format_argument(value, escape_keys=escape_keys) -}}
135
+ {%- endfor -%}
136
+ {{- '}' -}}
137
+ {%- elif argument is sequence -%}
138
+ {{- '[' -}}
139
+ {%- for item in argument -%}
140
+ {{- format_argument(item, escape_keys=escape_keys) -}}
141
+ {%- if not loop.last %},{% endif -%}
142
+ {%- endfor -%}
143
+ {{- ']' -}}
144
+ {%- else -%}
145
+ {{- argument -}}
146
+ {%- endif -%}
147
+ {%- endmacro -%}
148
+ {%- macro strip_thinking(text) -%}
149
+ {%- set ns = namespace(result='') -%}
150
+ {%- for part in text.split('<channel|>') -%}
151
+ {%- if '<|channel>' in part -%}
152
+ {%- set ns.result = ns.result + part.split('<|channel>')[0] -%}
153
+ {%- else -%}
154
+ {%- set ns.result = ns.result + part -%}
155
+ {%- endif -%}
156
+ {%- endfor -%}
157
+ {{- ns.result | trim -}}
158
+ {%- endmacro -%}
159
+
160
+ {%- macro format_tool_response_block(tool_name, response) -%}
161
+ {{- '<|tool_response>' -}}
162
+ {%- if response is mapping -%}
163
+ {{- 'response:' + tool_name + '{' -}}
164
+ {%- for key, value in response | dictsort -%}
165
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
166
+ {%- if not loop.last %},{% endif -%}
167
+ {%- endfor -%}
168
+ {{- '}' -}}
169
+ {%- else -%}
170
+ {{- 'response:' + tool_name + '{value:' + format_argument(response, escape_keys=False) + '}' -}}
171
+ {%- endif -%}
172
+ {{- '<tool_response|>' -}}
173
+ {%- endmacro -%}
174
+
175
+ {%- set ns = namespace(prev_message_type=None) -%}
176
+ {%- set loop_messages = messages -%}
177
+ {{- bos_token -}}
178
+ {#- Handle System/Tool Definitions Block -#}
179
+ {%- if (enable_thinking is defined and enable_thinking) or tools or messages[0]['role'] in ['system', 'developer'] -%}
180
+ {{- '<|turn>system\n' -}}
181
+
182
+ {#- Inject Thinking token at the very top of the FIRST system turn -#}
183
+ {%- if enable_thinking is defined and enable_thinking -%}
184
+ {{- '<|think|>\n' -}}
185
+ {%- set ns.prev_message_type = 'think' -%}
186
+ {%- endif -%}
187
+
188
+ {%- if messages[0]['role'] in ['system', 'developer'] -%}
189
+ {{- messages[0]['content'] | trim -}}
190
+ {%- set loop_messages = messages[1:] -%}
191
+ {%- endif -%}
192
+
193
+ {%- if tools -%}
194
+ {%- for tool in tools %}
195
+ {{- '<|tool>' -}}
196
+ {{- format_function_declaration(tool) | trim -}}
197
+ {{- '<tool|>' -}}
198
+ {%- endfor %}
199
+ {%- set ns.prev_message_type = 'tool' -%}
200
+ {%- endif -%}
201
+
202
+ {{- '<turn|>\n' -}}
203
+ {%- endif %}
204
+
205
+ {#- Pre-scan: find last user message index for reasoning guard -#}
206
+ {%- set ns_turn = namespace(last_user_idx=-1) -%}
207
+ {%- for i in range(loop_messages | length) -%}
208
+ {%- if loop_messages[i]['role'] == 'user' -%}
209
+ {%- set ns_turn.last_user_idx = i -%}
210
+ {%- endif -%}
211
+ {%- endfor -%}
212
+
213
+ {#- Loop through messages -#}
214
+ {%- for message in loop_messages -%}
215
+ {%- if message['role'] != 'tool' -%}
216
+ {%- set ns.prev_message_type = None -%}
217
+ {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%}
218
+ {#- Detect continuation: suppress duplicate <|turn>model when previous non-tool message was also assistant -#}
219
+ {%- set prev_nt = namespace(role=None, found=false) -%}
220
+ {%- if loop.index0 > 0 -%}
221
+ {%- for j in range(loop.index0 - 1, -1, -1) -%}
222
+ {%- if not prev_nt.found -%}
223
+ {%- if loop_messages[j]['role'] != 'tool' -%}
224
+ {%- set prev_nt.role = loop_messages[j]['role'] -%}
225
+ {%- set prev_nt.found = true -%}
226
+ {%- endif -%}
227
+ {%- endif -%}
228
+ {%- endfor -%}
229
+ {%- endif -%}
230
+ {%- set continue_same_model_turn = (role == 'model' and prev_nt.role == 'assistant') -%}
231
+ {%- if not continue_same_model_turn -%}
232
+ {{- '<|turn>' + role + '\n' }}
233
+ {%- endif -%}
234
+
235
+ {#- Render reasoning/reasoning_content as thinking channel -#}
236
+ {%- set thinking_text = message.get('reasoning') or message.get('reasoning_content') -%}
237
+ {%- if thinking_text and loop.index0 > ns_turn.last_user_idx and message.get('tool_calls') -%}
238
+ {{- '<|channel>thought\n' + thinking_text + '\n<channel|>' -}}
239
+ {%- endif -%}
240
+
241
+ {%- if message['tool_calls'] -%}
242
+ {%- for tool_call in message['tool_calls'] -%}
243
+ {%- set function = tool_call['function'] -%}
244
+ {{- '<|tool_call>call:' + function['name'] + '{' -}}
245
+ {%- if function['arguments'] is mapping -%}
246
+ {%- set ns_args = namespace(found_first=false) -%}
247
+ {%- for key, value in function['arguments'] | dictsort -%}
248
+ {%- if ns_args.found_first %},{% endif -%}
249
+ {%- set ns_args.found_first = true -%}
250
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
251
+ {%- endfor -%}
252
+ {%- elif function['arguments'] is string -%}
253
+ {{- function['arguments'] -}}
254
+ {%- endif -%}
255
+ {{- '}<tool_call|>' -}}
256
+ {%- endfor -%}
257
+ {%- set ns.prev_message_type = 'tool_call' -%}
258
+ {%- endif -%}
259
+
260
+ {%- set ns_tr_out = namespace(flag=false) -%}
261
+ {%- if message.get('tool_responses') -%}
262
+ {#- Legacy: tool_responses embedded on the assistant message (Google/Gemma native) -#}
263
+ {%- for tool_response in message['tool_responses'] -%}
264
+ {{- format_tool_response_block(tool_response['name'] | default('unknown'), tool_response['response']) -}}
265
+ {%- set ns_tr_out.flag = true -%}
266
+ {%- set ns.prev_message_type = 'tool_response' -%}
267
+ {%- endfor -%}
268
+ {%- elif message.get('tool_calls') -%}
269
+ {#- OpenAI Chat Completions: forward-scan consecutive role:tool messages -#}
270
+ {%- set ns_tool_scan = namespace(stopped=false) -%}
271
+ {%- for k in range(loop.index0 + 1, loop_messages | length) -%}
272
+ {%- if ns_tool_scan.stopped -%}
273
+ {%- elif loop_messages[k]['role'] != 'tool' -%}
274
+ {%- set ns_tool_scan.stopped = true -%}
275
+ {%- else -%}
276
+ {%- set follow = loop_messages[k] -%}
277
+ {#- Resolve tool_call_id to function name -#}
278
+ {%- set ns_tname = namespace(name=follow.get('name') | default('unknown')) -%}
279
+ {%- for tc in message['tool_calls'] -%}
280
+ {%- if tc.get('id') == follow.get('tool_call_id') -%}
281
+ {%- set ns_tname.name = tc['function']['name'] -%}
282
+ {%- endif -%}
283
+ {%- endfor -%}
284
+ {#- Handle content as string or content-parts array -#}
285
+ {%- set tool_body = follow.get('content') -%}
286
+ {%- if tool_body is string -%}
287
+ {{- format_tool_response_block(ns_tname.name, tool_body) -}}
288
+ {%- elif tool_body is sequence and tool_body is not string -%}
289
+ {%- set ns_txt = namespace(s='') -%}
290
+ {%- for part in tool_body -%}
291
+ {%- if part.get('type') == 'text' -%}
292
+ {%- set ns_txt.s = ns_txt.s + (part.get('text') | default('')) -%}
293
+ {%- endif -%}
294
+ {%- endfor -%}
295
+ {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
296
+ {%- else -%}
297
+ {{- format_tool_response_block(ns_tname.name, tool_body) -}}
298
+ {%- endif -%}
299
+ {%- set ns_tr_out.flag = true -%}
300
+ {%- set ns.prev_message_type = 'tool_response' -%}
301
+ {%- endif -%}
302
+ {%- endfor -%}
303
+ {%- endif -%}
304
+
305
+ {%- if message['content'] is string -%}
306
+ {%- if role == 'model' -%}
307
+ {{- strip_thinking(message['content']) -}}
308
+ {%- else -%}
309
+ {{- message['content'] | trim -}}
310
+ {%- endif -%}
311
+ {%- elif message['content'] is sequence -%}
312
+ {%- for item in message['content'] -%}
313
+ {%- if item['type'] == 'text' -%}
314
+ {%- if role == 'model' -%}
315
+ {{- strip_thinking(item['text']) -}}
316
+ {%- else -%}
317
+ {{- item['text'] | trim -}}
318
+ {%- endif -%}
319
+ {%- elif item['type'] == 'image' -%}
320
+ {{- '<|image|>' -}}
321
+ {%- set ns.prev_message_type = 'image' -%}
322
+ {%- elif item['type'] == 'audio' -%}
323
+ {{- '<|audio|>' -}}
324
+ {%- set ns.prev_message_type = 'audio' -%}
325
+ {%- elif item['type'] == 'video' -%}
326
+ {{- '<|video|>' -}}
327
+ {%- set ns.prev_message_type = 'video' -%}
328
+ {%- endif -%}
329
+ {%- endfor -%}
330
+ {%- endif -%}
331
+
332
+ {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%}
333
+ {{- '<|tool_response>' -}}
334
+ {%- elif not (ns_tr_out.flag and not message.get('content')) -%}
335
+ {{- '<turn|>\n' -}}
336
+ {%- endif -%}
337
+ {%- endif -%}
338
+ {%- endfor -%}
339
+
340
+ {%- if add_generation_prompt -%}
341
+ {%- if ns.prev_message_type != 'tool_response' and ns.prev_message_type != 'tool_call' -%}
342
+ {{- '<|turn>model\n' -}}
343
+ {%- endif -%}
344
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,191 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma4ForConditionalGeneration"
4
+ ],
5
+ "audio_config": {
6
+ "_name_or_path": "",
7
+ "architectures": null,
8
+ "attention_chunk_size": 12,
9
+ "attention_context_left": 13,
10
+ "attention_context_right": 0,
11
+ "attention_invalid_logits_value": -1000000000.0,
12
+ "attention_logit_cap": 50.0,
13
+ "chunk_size_feed_forward": 0,
14
+ "conv_kernel_size": 5,
15
+ "dtype": "bfloat16",
16
+ "gradient_clipping": 10000000000.0,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 1024,
19
+ "id2label": {
20
+ "0": "LABEL_0",
21
+ "1": "LABEL_1"
22
+ },
23
+ "initializer_range": 0.02,
24
+ "is_encoder_decoder": false,
25
+ "label2id": {
26
+ "LABEL_0": 0,
27
+ "LABEL_1": 1
28
+ },
29
+ "model_type": "gemma4_audio",
30
+ "num_attention_heads": 8,
31
+ "num_hidden_layers": 12,
32
+ "output_attentions": false,
33
+ "output_hidden_states": false,
34
+ "output_proj_dims": 1536,
35
+ "problem_type": null,
36
+ "residual_weight": 0.5,
37
+ "return_dict": true,
38
+ "rms_norm_eps": 1e-06,
39
+ "subsampling_conv_channels": [
40
+ 128,
41
+ 32
42
+ ],
43
+ "use_clipped_linears": true
44
+ },
45
+ "audio_token_id": 258881,
46
+ "boa_token_id": 256000,
47
+ "boi_token_id": 255999,
48
+ "dtype": "bfloat16",
49
+ "eoa_token_id": 258883,
50
+ "eoa_token_index": 258883,
51
+ "eoi_token_id": 258882,
52
+ "eos_token_id": [
53
+ 1,
54
+ 106
55
+ ],
56
+ "image_token_id": 258880,
57
+ "initializer_range": 0.02,
58
+ "model_type": "gemma4",
59
+ "text_config": {
60
+ "attention_bias": false,
61
+ "attention_dropout": 0.0,
62
+ "attention_k_eq_v": false,
63
+ "bos_token_id": 2,
64
+ "dtype": "bfloat16",
65
+ "enable_moe_block": false,
66
+ "eos_token_id": 1,
67
+ "expert_intermediate_size": null,
68
+ "final_logit_softcapping": 30.0,
69
+ "global_head_dim": 512,
70
+ "head_dim": 256,
71
+ "hidden_activation": "gelu_pytorch_tanh",
72
+ "hidden_size": 1536,
73
+ "hidden_size_per_layer_input": 256,
74
+ "initializer_range": 0.02,
75
+ "intermediate_size": 6144,
76
+ "layer_types": [
77
+ "sliding_attention",
78
+ "sliding_attention",
79
+ "sliding_attention",
80
+ "sliding_attention",
81
+ "full_attention",
82
+ "sliding_attention",
83
+ "sliding_attention",
84
+ "sliding_attention",
85
+ "sliding_attention",
86
+ "full_attention",
87
+ "sliding_attention",
88
+ "sliding_attention",
89
+ "sliding_attention",
90
+ "sliding_attention",
91
+ "full_attention",
92
+ "sliding_attention",
93
+ "sliding_attention",
94
+ "sliding_attention",
95
+ "sliding_attention",
96
+ "full_attention",
97
+ "sliding_attention",
98
+ "sliding_attention",
99
+ "sliding_attention",
100
+ "sliding_attention",
101
+ "full_attention",
102
+ "sliding_attention",
103
+ "sliding_attention",
104
+ "sliding_attention",
105
+ "sliding_attention",
106
+ "full_attention",
107
+ "sliding_attention",
108
+ "sliding_attention",
109
+ "sliding_attention",
110
+ "sliding_attention",
111
+ "full_attention"
112
+ ],
113
+ "max_position_embeddings": 131072,
114
+ "model_type": "gemma4_text",
115
+ "moe_intermediate_size": null,
116
+ "num_attention_heads": 8,
117
+ "num_experts": null,
118
+ "num_global_key_value_heads": null,
119
+ "num_hidden_layers": 35,
120
+ "num_key_value_heads": 1,
121
+ "num_kv_shared_layers": 20,
122
+ "pad_token_id": 0,
123
+ "rms_norm_eps": 1e-06,
124
+ "rope_parameters": {
125
+ "full_attention": {
126
+ "partial_rotary_factor": 0.25,
127
+ "rope_theta": 1000000.0,
128
+ "rope_type": "proportional"
129
+ },
130
+ "sliding_attention": {
131
+ "rope_theta": 10000.0,
132
+ "rope_type": "default"
133
+ }
134
+ },
135
+ "sliding_window": 512,
136
+ "tie_word_embeddings": true,
137
+ "top_k_experts": null,
138
+ "use_bidirectional_attention": null,
139
+ "use_cache": true,
140
+ "use_double_wide_mlp": true,
141
+ "vocab_size": 262144,
142
+ "vocab_size_per_layer_input": 262144
143
+ },
144
+ "tie_word_embeddings": true,
145
+ "transformers_version": "5.6.2",
146
+ "video_token_id": 258884,
147
+ "vision_config": {
148
+ "_name_or_path": "",
149
+ "architectures": null,
150
+ "attention_bias": false,
151
+ "attention_dropout": 0.0,
152
+ "chunk_size_feed_forward": 0,
153
+ "default_output_length": 280,
154
+ "dtype": "bfloat16",
155
+ "global_head_dim": 64,
156
+ "head_dim": 64,
157
+ "hidden_activation": "gelu_pytorch_tanh",
158
+ "hidden_size": 768,
159
+ "id2label": {
160
+ "0": "LABEL_0",
161
+ "1": "LABEL_1"
162
+ },
163
+ "initializer_range": 0.02,
164
+ "intermediate_size": 3072,
165
+ "is_encoder_decoder": false,
166
+ "label2id": {
167
+ "LABEL_0": 0,
168
+ "LABEL_1": 1
169
+ },
170
+ "max_position_embeddings": 131072,
171
+ "model_type": "gemma4_vision",
172
+ "num_attention_heads": 12,
173
+ "num_hidden_layers": 16,
174
+ "num_key_value_heads": 12,
175
+ "output_attentions": false,
176
+ "output_hidden_states": false,
177
+ "patch_size": 16,
178
+ "pooling_kernel_size": 3,
179
+ "position_embedding_size": 10240,
180
+ "problem_type": null,
181
+ "return_dict": true,
182
+ "rms_norm_eps": 1e-06,
183
+ "rope_parameters": {
184
+ "rope_theta": 100.0,
185
+ "rope_type": "default"
186
+ },
187
+ "standardize": false,
188
+ "use_clipped_linears": true
189
+ },
190
+ "vision_soft_tokens_per_image": 280
191
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69351ef8f41c3455cb110d54431ab23e63875754a0b7750bbf36f2fbb119b4d6
3
+ size 6991185214
quanto_qmap.json ADDED
@@ -0,0 +1,2114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model.vision_tower.patch_embedder.input_proj": {
3
+ "weights": "qint4",
4
+ "activations": "none"
5
+ },
6
+ "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear": {
7
+ "weights": "qint4",
8
+ "activations": "none"
9
+ },
10
+ "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear": {
11
+ "weights": "qint4",
12
+ "activations": "none"
13
+ },
14
+ "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear": {
15
+ "weights": "qint4",
16
+ "activations": "none"
17
+ },
18
+ "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear": {
19
+ "weights": "qint4",
20
+ "activations": "none"
21
+ },
22
+ "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear": {
23
+ "weights": "qint4",
24
+ "activations": "none"
25
+ },
26
+ "model.vision_tower.encoder.layers.0.mlp.up_proj.linear": {
27
+ "weights": "qint4",
28
+ "activations": "none"
29
+ },
30
+ "model.vision_tower.encoder.layers.0.mlp.down_proj.linear": {
31
+ "weights": "qint4",
32
+ "activations": "none"
33
+ },
34
+ "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear": {
35
+ "weights": "qint4",
36
+ "activations": "none"
37
+ },
38
+ "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear": {
39
+ "weights": "qint4",
40
+ "activations": "none"
41
+ },
42
+ "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear": {
43
+ "weights": "qint4",
44
+ "activations": "none"
45
+ },
46
+ "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear": {
47
+ "weights": "qint4",
48
+ "activations": "none"
49
+ },
50
+ "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear": {
51
+ "weights": "qint4",
52
+ "activations": "none"
53
+ },
54
+ "model.vision_tower.encoder.layers.1.mlp.up_proj.linear": {
55
+ "weights": "qint4",
56
+ "activations": "none"
57
+ },
58
+ "model.vision_tower.encoder.layers.1.mlp.down_proj.linear": {
59
+ "weights": "qint4",
60
+ "activations": "none"
61
+ },
62
+ "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear": {
63
+ "weights": "qint4",
64
+ "activations": "none"
65
+ },
66
+ "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear": {
67
+ "weights": "qint4",
68
+ "activations": "none"
69
+ },
70
+ "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear": {
71
+ "weights": "qint4",
72
+ "activations": "none"
73
+ },
74
+ "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear": {
75
+ "weights": "qint4",
76
+ "activations": "none"
77
+ },
78
+ "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear": {
79
+ "weights": "qint4",
80
+ "activations": "none"
81
+ },
82
+ "model.vision_tower.encoder.layers.2.mlp.up_proj.linear": {
83
+ "weights": "qint4",
84
+ "activations": "none"
85
+ },
86
+ "model.vision_tower.encoder.layers.2.mlp.down_proj.linear": {
87
+ "weights": "qint4",
88
+ "activations": "none"
89
+ },
90
+ "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear": {
91
+ "weights": "qint4",
92
+ "activations": "none"
93
+ },
94
+ "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear": {
95
+ "weights": "qint4",
96
+ "activations": "none"
97
+ },
98
+ "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear": {
99
+ "weights": "qint4",
100
+ "activations": "none"
101
+ },
102
+ "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear": {
103
+ "weights": "qint4",
104
+ "activations": "none"
105
+ },
106
+ "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear": {
107
+ "weights": "qint4",
108
+ "activations": "none"
109
+ },
110
+ "model.vision_tower.encoder.layers.3.mlp.up_proj.linear": {
111
+ "weights": "qint4",
112
+ "activations": "none"
113
+ },
114
+ "model.vision_tower.encoder.layers.3.mlp.down_proj.linear": {
115
+ "weights": "qint4",
116
+ "activations": "none"
117
+ },
118
+ "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear": {
119
+ "weights": "qint4",
120
+ "activations": "none"
121
+ },
122
+ "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear": {
123
+ "weights": "qint4",
124
+ "activations": "none"
125
+ },
126
+ "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear": {
127
+ "weights": "qint4",
128
+ "activations": "none"
129
+ },
130
+ "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear": {
131
+ "weights": "qint4",
132
+ "activations": "none"
133
+ },
134
+ "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear": {
135
+ "weights": "qint4",
136
+ "activations": "none"
137
+ },
138
+ "model.vision_tower.encoder.layers.4.mlp.up_proj.linear": {
139
+ "weights": "qint4",
140
+ "activations": "none"
141
+ },
142
+ "model.vision_tower.encoder.layers.4.mlp.down_proj.linear": {
143
+ "weights": "qint4",
144
+ "activations": "none"
145
+ },
146
+ "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear": {
147
+ "weights": "qint4",
148
+ "activations": "none"
149
+ },
150
+ "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear": {
151
+ "weights": "qint4",
152
+ "activations": "none"
153
+ },
154
+ "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear": {
155
+ "weights": "qint4",
156
+ "activations": "none"
157
+ },
158
+ "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear": {
159
+ "weights": "qint4",
160
+ "activations": "none"
161
+ },
162
+ "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear": {
163
+ "weights": "qint4",
164
+ "activations": "none"
165
+ },
166
+ "model.vision_tower.encoder.layers.5.mlp.up_proj.linear": {
167
+ "weights": "qint4",
168
+ "activations": "none"
169
+ },
170
+ "model.vision_tower.encoder.layers.5.mlp.down_proj.linear": {
171
+ "weights": "qint4",
172
+ "activations": "none"
173
+ },
174
+ "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear": {
175
+ "weights": "qint4",
176
+ "activations": "none"
177
+ },
178
+ "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear": {
179
+ "weights": "qint4",
180
+ "activations": "none"
181
+ },
182
+ "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear": {
183
+ "weights": "qint4",
184
+ "activations": "none"
185
+ },
186
+ "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear": {
187
+ "weights": "qint4",
188
+ "activations": "none"
189
+ },
190
+ "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear": {
191
+ "weights": "qint4",
192
+ "activations": "none"
193
+ },
194
+ "model.vision_tower.encoder.layers.6.mlp.up_proj.linear": {
195
+ "weights": "qint4",
196
+ "activations": "none"
197
+ },
198
+ "model.vision_tower.encoder.layers.6.mlp.down_proj.linear": {
199
+ "weights": "qint4",
200
+ "activations": "none"
201
+ },
202
+ "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear": {
203
+ "weights": "qint4",
204
+ "activations": "none"
205
+ },
206
+ "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear": {
207
+ "weights": "qint4",
208
+ "activations": "none"
209
+ },
210
+ "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear": {
211
+ "weights": "qint4",
212
+ "activations": "none"
213
+ },
214
+ "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear": {
215
+ "weights": "qint4",
216
+ "activations": "none"
217
+ },
218
+ "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear": {
219
+ "weights": "qint4",
220
+ "activations": "none"
221
+ },
222
+ "model.vision_tower.encoder.layers.7.mlp.up_proj.linear": {
223
+ "weights": "qint4",
224
+ "activations": "none"
225
+ },
226
+ "model.vision_tower.encoder.layers.7.mlp.down_proj.linear": {
227
+ "weights": "qint4",
228
+ "activations": "none"
229
+ },
230
+ "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear": {
231
+ "weights": "qint4",
232
+ "activations": "none"
233
+ },
234
+ "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear": {
235
+ "weights": "qint4",
236
+ "activations": "none"
237
+ },
238
+ "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear": {
239
+ "weights": "qint4",
240
+ "activations": "none"
241
+ },
242
+ "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear": {
243
+ "weights": "qint4",
244
+ "activations": "none"
245
+ },
246
+ "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear": {
247
+ "weights": "qint4",
248
+ "activations": "none"
249
+ },
250
+ "model.vision_tower.encoder.layers.8.mlp.up_proj.linear": {
251
+ "weights": "qint4",
252
+ "activations": "none"
253
+ },
254
+ "model.vision_tower.encoder.layers.8.mlp.down_proj.linear": {
255
+ "weights": "qint4",
256
+ "activations": "none"
257
+ },
258
+ "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear": {
259
+ "weights": "qint4",
260
+ "activations": "none"
261
+ },
262
+ "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear": {
263
+ "weights": "qint4",
264
+ "activations": "none"
265
+ },
266
+ "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear": {
267
+ "weights": "qint4",
268
+ "activations": "none"
269
+ },
270
+ "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear": {
271
+ "weights": "qint4",
272
+ "activations": "none"
273
+ },
274
+ "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear": {
275
+ "weights": "qint4",
276
+ "activations": "none"
277
+ },
278
+ "model.vision_tower.encoder.layers.9.mlp.up_proj.linear": {
279
+ "weights": "qint4",
280
+ "activations": "none"
281
+ },
282
+ "model.vision_tower.encoder.layers.9.mlp.down_proj.linear": {
283
+ "weights": "qint4",
284
+ "activations": "none"
285
+ },
286
+ "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear": {
287
+ "weights": "qint4",
288
+ "activations": "none"
289
+ },
290
+ "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear": {
291
+ "weights": "qint4",
292
+ "activations": "none"
293
+ },
294
+ "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear": {
295
+ "weights": "qint4",
296
+ "activations": "none"
297
+ },
298
+ "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear": {
299
+ "weights": "qint4",
300
+ "activations": "none"
301
+ },
302
+ "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear": {
303
+ "weights": "qint4",
304
+ "activations": "none"
305
+ },
306
+ "model.vision_tower.encoder.layers.10.mlp.up_proj.linear": {
307
+ "weights": "qint4",
308
+ "activations": "none"
309
+ },
310
+ "model.vision_tower.encoder.layers.10.mlp.down_proj.linear": {
311
+ "weights": "qint4",
312
+ "activations": "none"
313
+ },
314
+ "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear": {
315
+ "weights": "qint4",
316
+ "activations": "none"
317
+ },
318
+ "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear": {
319
+ "weights": "qint4",
320
+ "activations": "none"
321
+ },
322
+ "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear": {
323
+ "weights": "qint4",
324
+ "activations": "none"
325
+ },
326
+ "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear": {
327
+ "weights": "qint4",
328
+ "activations": "none"
329
+ },
330
+ "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear": {
331
+ "weights": "qint4",
332
+ "activations": "none"
333
+ },
334
+ "model.vision_tower.encoder.layers.11.mlp.up_proj.linear": {
335
+ "weights": "qint4",
336
+ "activations": "none"
337
+ },
338
+ "model.vision_tower.encoder.layers.11.mlp.down_proj.linear": {
339
+ "weights": "qint4",
340
+ "activations": "none"
341
+ },
342
+ "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear": {
343
+ "weights": "qint4",
344
+ "activations": "none"
345
+ },
346
+ "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear": {
347
+ "weights": "qint4",
348
+ "activations": "none"
349
+ },
350
+ "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear": {
351
+ "weights": "qint4",
352
+ "activations": "none"
353
+ },
354
+ "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear": {
355
+ "weights": "qint4",
356
+ "activations": "none"
357
+ },
358
+ "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear": {
359
+ "weights": "qint4",
360
+ "activations": "none"
361
+ },
362
+ "model.vision_tower.encoder.layers.12.mlp.up_proj.linear": {
363
+ "weights": "qint4",
364
+ "activations": "none"
365
+ },
366
+ "model.vision_tower.encoder.layers.12.mlp.down_proj.linear": {
367
+ "weights": "qint4",
368
+ "activations": "none"
369
+ },
370
+ "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear": {
371
+ "weights": "qint4",
372
+ "activations": "none"
373
+ },
374
+ "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear": {
375
+ "weights": "qint4",
376
+ "activations": "none"
377
+ },
378
+ "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear": {
379
+ "weights": "qint4",
380
+ "activations": "none"
381
+ },
382
+ "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear": {
383
+ "weights": "qint4",
384
+ "activations": "none"
385
+ },
386
+ "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear": {
387
+ "weights": "qint4",
388
+ "activations": "none"
389
+ },
390
+ "model.vision_tower.encoder.layers.13.mlp.up_proj.linear": {
391
+ "weights": "qint4",
392
+ "activations": "none"
393
+ },
394
+ "model.vision_tower.encoder.layers.13.mlp.down_proj.linear": {
395
+ "weights": "qint4",
396
+ "activations": "none"
397
+ },
398
+ "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear": {
399
+ "weights": "qint4",
400
+ "activations": "none"
401
+ },
402
+ "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear": {
403
+ "weights": "qint4",
404
+ "activations": "none"
405
+ },
406
+ "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear": {
407
+ "weights": "qint4",
408
+ "activations": "none"
409
+ },
410
+ "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear": {
411
+ "weights": "qint4",
412
+ "activations": "none"
413
+ },
414
+ "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear": {
415
+ "weights": "qint4",
416
+ "activations": "none"
417
+ },
418
+ "model.vision_tower.encoder.layers.14.mlp.up_proj.linear": {
419
+ "weights": "qint4",
420
+ "activations": "none"
421
+ },
422
+ "model.vision_tower.encoder.layers.14.mlp.down_proj.linear": {
423
+ "weights": "qint4",
424
+ "activations": "none"
425
+ },
426
+ "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear": {
427
+ "weights": "qint4",
428
+ "activations": "none"
429
+ },
430
+ "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear": {
431
+ "weights": "qint4",
432
+ "activations": "none"
433
+ },
434
+ "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear": {
435
+ "weights": "qint4",
436
+ "activations": "none"
437
+ },
438
+ "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear": {
439
+ "weights": "qint4",
440
+ "activations": "none"
441
+ },
442
+ "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear": {
443
+ "weights": "qint4",
444
+ "activations": "none"
445
+ },
446
+ "model.vision_tower.encoder.layers.15.mlp.up_proj.linear": {
447
+ "weights": "qint4",
448
+ "activations": "none"
449
+ },
450
+ "model.vision_tower.encoder.layers.15.mlp.down_proj.linear": {
451
+ "weights": "qint4",
452
+ "activations": "none"
453
+ },
454
+ "model.language_model.layers.0.self_attn.q_proj": {
455
+ "weights": "qint4",
456
+ "activations": "none"
457
+ },
458
+ "model.language_model.layers.0.self_attn.k_proj": {
459
+ "weights": "qint4",
460
+ "activations": "none"
461
+ },
462
+ "model.language_model.layers.0.self_attn.v_proj": {
463
+ "weights": "qint4",
464
+ "activations": "none"
465
+ },
466
+ "model.language_model.layers.0.self_attn.o_proj": {
467
+ "weights": "qint4",
468
+ "activations": "none"
469
+ },
470
+ "model.language_model.layers.0.mlp.gate_proj": {
471
+ "weights": "qint4",
472
+ "activations": "none"
473
+ },
474
+ "model.language_model.layers.0.mlp.up_proj": {
475
+ "weights": "qint4",
476
+ "activations": "none"
477
+ },
478
+ "model.language_model.layers.0.mlp.down_proj": {
479
+ "weights": "qint4",
480
+ "activations": "none"
481
+ },
482
+ "model.language_model.layers.0.per_layer_input_gate": {
483
+ "weights": "qint4",
484
+ "activations": "none"
485
+ },
486
+ "model.language_model.layers.0.per_layer_projection": {
487
+ "weights": "qint4",
488
+ "activations": "none"
489
+ },
490
+ "model.language_model.layers.1.self_attn.q_proj": {
491
+ "weights": "qint4",
492
+ "activations": "none"
493
+ },
494
+ "model.language_model.layers.1.self_attn.k_proj": {
495
+ "weights": "qint4",
496
+ "activations": "none"
497
+ },
498
+ "model.language_model.layers.1.self_attn.v_proj": {
499
+ "weights": "qint4",
500
+ "activations": "none"
501
+ },
502
+ "model.language_model.layers.1.self_attn.o_proj": {
503
+ "weights": "qint4",
504
+ "activations": "none"
505
+ },
506
+ "model.language_model.layers.1.mlp.gate_proj": {
507
+ "weights": "qint4",
508
+ "activations": "none"
509
+ },
510
+ "model.language_model.layers.1.mlp.up_proj": {
511
+ "weights": "qint4",
512
+ "activations": "none"
513
+ },
514
+ "model.language_model.layers.1.mlp.down_proj": {
515
+ "weights": "qint4",
516
+ "activations": "none"
517
+ },
518
+ "model.language_model.layers.1.per_layer_input_gate": {
519
+ "weights": "qint4",
520
+ "activations": "none"
521
+ },
522
+ "model.language_model.layers.1.per_layer_projection": {
523
+ "weights": "qint4",
524
+ "activations": "none"
525
+ },
526
+ "model.language_model.layers.2.self_attn.q_proj": {
527
+ "weights": "qint4",
528
+ "activations": "none"
529
+ },
530
+ "model.language_model.layers.2.self_attn.k_proj": {
531
+ "weights": "qint4",
532
+ "activations": "none"
533
+ },
534
+ "model.language_model.layers.2.self_attn.v_proj": {
535
+ "weights": "qint4",
536
+ "activations": "none"
537
+ },
538
+ "model.language_model.layers.2.self_attn.o_proj": {
539
+ "weights": "qint4",
540
+ "activations": "none"
541
+ },
542
+ "model.language_model.layers.2.mlp.gate_proj": {
543
+ "weights": "qint4",
544
+ "activations": "none"
545
+ },
546
+ "model.language_model.layers.2.mlp.up_proj": {
547
+ "weights": "qint4",
548
+ "activations": "none"
549
+ },
550
+ "model.language_model.layers.2.mlp.down_proj": {
551
+ "weights": "qint4",
552
+ "activations": "none"
553
+ },
554
+ "model.language_model.layers.2.per_layer_input_gate": {
555
+ "weights": "qint4",
556
+ "activations": "none"
557
+ },
558
+ "model.language_model.layers.2.per_layer_projection": {
559
+ "weights": "qint4",
560
+ "activations": "none"
561
+ },
562
+ "model.language_model.layers.3.self_attn.q_proj": {
563
+ "weights": "qint4",
564
+ "activations": "none"
565
+ },
566
+ "model.language_model.layers.3.self_attn.k_proj": {
567
+ "weights": "qint4",
568
+ "activations": "none"
569
+ },
570
+ "model.language_model.layers.3.self_attn.v_proj": {
571
+ "weights": "qint4",
572
+ "activations": "none"
573
+ },
574
+ "model.language_model.layers.3.self_attn.o_proj": {
575
+ "weights": "qint4",
576
+ "activations": "none"
577
+ },
578
+ "model.language_model.layers.3.mlp.gate_proj": {
579
+ "weights": "qint4",
580
+ "activations": "none"
581
+ },
582
+ "model.language_model.layers.3.mlp.up_proj": {
583
+ "weights": "qint4",
584
+ "activations": "none"
585
+ },
586
+ "model.language_model.layers.3.mlp.down_proj": {
587
+ "weights": "qint4",
588
+ "activations": "none"
589
+ },
590
+ "model.language_model.layers.3.per_layer_input_gate": {
591
+ "weights": "qint4",
592
+ "activations": "none"
593
+ },
594
+ "model.language_model.layers.3.per_layer_projection": {
595
+ "weights": "qint4",
596
+ "activations": "none"
597
+ },
598
+ "model.language_model.layers.4.self_attn.q_proj": {
599
+ "weights": "qint4",
600
+ "activations": "none"
601
+ },
602
+ "model.language_model.layers.4.self_attn.k_proj": {
603
+ "weights": "qint4",
604
+ "activations": "none"
605
+ },
606
+ "model.language_model.layers.4.self_attn.v_proj": {
607
+ "weights": "qint4",
608
+ "activations": "none"
609
+ },
610
+ "model.language_model.layers.4.self_attn.o_proj": {
611
+ "weights": "qint4",
612
+ "activations": "none"
613
+ },
614
+ "model.language_model.layers.4.mlp.gate_proj": {
615
+ "weights": "qint4",
616
+ "activations": "none"
617
+ },
618
+ "model.language_model.layers.4.mlp.up_proj": {
619
+ "weights": "qint4",
620
+ "activations": "none"
621
+ },
622
+ "model.language_model.layers.4.mlp.down_proj": {
623
+ "weights": "qint4",
624
+ "activations": "none"
625
+ },
626
+ "model.language_model.layers.4.per_layer_input_gate": {
627
+ "weights": "qint4",
628
+ "activations": "none"
629
+ },
630
+ "model.language_model.layers.4.per_layer_projection": {
631
+ "weights": "qint4",
632
+ "activations": "none"
633
+ },
634
+ "model.language_model.layers.5.self_attn.q_proj": {
635
+ "weights": "qint4",
636
+ "activations": "none"
637
+ },
638
+ "model.language_model.layers.5.self_attn.k_proj": {
639
+ "weights": "qint4",
640
+ "activations": "none"
641
+ },
642
+ "model.language_model.layers.5.self_attn.v_proj": {
643
+ "weights": "qint4",
644
+ "activations": "none"
645
+ },
646
+ "model.language_model.layers.5.self_attn.o_proj": {
647
+ "weights": "qint4",
648
+ "activations": "none"
649
+ },
650
+ "model.language_model.layers.5.mlp.gate_proj": {
651
+ "weights": "qint4",
652
+ "activations": "none"
653
+ },
654
+ "model.language_model.layers.5.mlp.up_proj": {
655
+ "weights": "qint4",
656
+ "activations": "none"
657
+ },
658
+ "model.language_model.layers.5.mlp.down_proj": {
659
+ "weights": "qint4",
660
+ "activations": "none"
661
+ },
662
+ "model.language_model.layers.5.per_layer_input_gate": {
663
+ "weights": "qint4",
664
+ "activations": "none"
665
+ },
666
+ "model.language_model.layers.5.per_layer_projection": {
667
+ "weights": "qint4",
668
+ "activations": "none"
669
+ },
670
+ "model.language_model.layers.6.self_attn.q_proj": {
671
+ "weights": "qint4",
672
+ "activations": "none"
673
+ },
674
+ "model.language_model.layers.6.self_attn.k_proj": {
675
+ "weights": "qint4",
676
+ "activations": "none"
677
+ },
678
+ "model.language_model.layers.6.self_attn.v_proj": {
679
+ "weights": "qint4",
680
+ "activations": "none"
681
+ },
682
+ "model.language_model.layers.6.self_attn.o_proj": {
683
+ "weights": "qint4",
684
+ "activations": "none"
685
+ },
686
+ "model.language_model.layers.6.mlp.gate_proj": {
687
+ "weights": "qint4",
688
+ "activations": "none"
689
+ },
690
+ "model.language_model.layers.6.mlp.up_proj": {
691
+ "weights": "qint4",
692
+ "activations": "none"
693
+ },
694
+ "model.language_model.layers.6.mlp.down_proj": {
695
+ "weights": "qint4",
696
+ "activations": "none"
697
+ },
698
+ "model.language_model.layers.6.per_layer_input_gate": {
699
+ "weights": "qint4",
700
+ "activations": "none"
701
+ },
702
+ "model.language_model.layers.6.per_layer_projection": {
703
+ "weights": "qint4",
704
+ "activations": "none"
705
+ },
706
+ "model.language_model.layers.7.self_attn.q_proj": {
707
+ "weights": "qint4",
708
+ "activations": "none"
709
+ },
710
+ "model.language_model.layers.7.self_attn.k_proj": {
711
+ "weights": "qint4",
712
+ "activations": "none"
713
+ },
714
+ "model.language_model.layers.7.self_attn.v_proj": {
715
+ "weights": "qint4",
716
+ "activations": "none"
717
+ },
718
+ "model.language_model.layers.7.self_attn.o_proj": {
719
+ "weights": "qint4",
720
+ "activations": "none"
721
+ },
722
+ "model.language_model.layers.7.mlp.gate_proj": {
723
+ "weights": "qint4",
724
+ "activations": "none"
725
+ },
726
+ "model.language_model.layers.7.mlp.up_proj": {
727
+ "weights": "qint4",
728
+ "activations": "none"
729
+ },
730
+ "model.language_model.layers.7.mlp.down_proj": {
731
+ "weights": "qint4",
732
+ "activations": "none"
733
+ },
734
+ "model.language_model.layers.7.per_layer_input_gate": {
735
+ "weights": "qint4",
736
+ "activations": "none"
737
+ },
738
+ "model.language_model.layers.7.per_layer_projection": {
739
+ "weights": "qint4",
740
+ "activations": "none"
741
+ },
742
+ "model.language_model.layers.8.self_attn.q_proj": {
743
+ "weights": "qint4",
744
+ "activations": "none"
745
+ },
746
+ "model.language_model.layers.8.self_attn.k_proj": {
747
+ "weights": "qint4",
748
+ "activations": "none"
749
+ },
750
+ "model.language_model.layers.8.self_attn.v_proj": {
751
+ "weights": "qint4",
752
+ "activations": "none"
753
+ },
754
+ "model.language_model.layers.8.self_attn.o_proj": {
755
+ "weights": "qint4",
756
+ "activations": "none"
757
+ },
758
+ "model.language_model.layers.8.mlp.gate_proj": {
759
+ "weights": "qint4",
760
+ "activations": "none"
761
+ },
762
+ "model.language_model.layers.8.mlp.up_proj": {
763
+ "weights": "qint4",
764
+ "activations": "none"
765
+ },
766
+ "model.language_model.layers.8.mlp.down_proj": {
767
+ "weights": "qint4",
768
+ "activations": "none"
769
+ },
770
+ "model.language_model.layers.8.per_layer_input_gate": {
771
+ "weights": "qint4",
772
+ "activations": "none"
773
+ },
774
+ "model.language_model.layers.8.per_layer_projection": {
775
+ "weights": "qint4",
776
+ "activations": "none"
777
+ },
778
+ "model.language_model.layers.9.self_attn.q_proj": {
779
+ "weights": "qint4",
780
+ "activations": "none"
781
+ },
782
+ "model.language_model.layers.9.self_attn.k_proj": {
783
+ "weights": "qint4",
784
+ "activations": "none"
785
+ },
786
+ "model.language_model.layers.9.self_attn.v_proj": {
787
+ "weights": "qint4",
788
+ "activations": "none"
789
+ },
790
+ "model.language_model.layers.9.self_attn.o_proj": {
791
+ "weights": "qint4",
792
+ "activations": "none"
793
+ },
794
+ "model.language_model.layers.9.mlp.gate_proj": {
795
+ "weights": "qint4",
796
+ "activations": "none"
797
+ },
798
+ "model.language_model.layers.9.mlp.up_proj": {
799
+ "weights": "qint4",
800
+ "activations": "none"
801
+ },
802
+ "model.language_model.layers.9.mlp.down_proj": {
803
+ "weights": "qint4",
804
+ "activations": "none"
805
+ },
806
+ "model.language_model.layers.9.per_layer_input_gate": {
807
+ "weights": "qint4",
808
+ "activations": "none"
809
+ },
810
+ "model.language_model.layers.9.per_layer_projection": {
811
+ "weights": "qint4",
812
+ "activations": "none"
813
+ },
814
+ "model.language_model.layers.10.self_attn.q_proj": {
815
+ "weights": "qint4",
816
+ "activations": "none"
817
+ },
818
+ "model.language_model.layers.10.self_attn.k_proj": {
819
+ "weights": "qint4",
820
+ "activations": "none"
821
+ },
822
+ "model.language_model.layers.10.self_attn.v_proj": {
823
+ "weights": "qint4",
824
+ "activations": "none"
825
+ },
826
+ "model.language_model.layers.10.self_attn.o_proj": {
827
+ "weights": "qint4",
828
+ "activations": "none"
829
+ },
830
+ "model.language_model.layers.10.mlp.gate_proj": {
831
+ "weights": "qint4",
832
+ "activations": "none"
833
+ },
834
+ "model.language_model.layers.10.mlp.up_proj": {
835
+ "weights": "qint4",
836
+ "activations": "none"
837
+ },
838
+ "model.language_model.layers.10.mlp.down_proj": {
839
+ "weights": "qint4",
840
+ "activations": "none"
841
+ },
842
+ "model.language_model.layers.10.per_layer_input_gate": {
843
+ "weights": "qint4",
844
+ "activations": "none"
845
+ },
846
+ "model.language_model.layers.10.per_layer_projection": {
847
+ "weights": "qint4",
848
+ "activations": "none"
849
+ },
850
+ "model.language_model.layers.11.self_attn.q_proj": {
851
+ "weights": "qint4",
852
+ "activations": "none"
853
+ },
854
+ "model.language_model.layers.11.self_attn.k_proj": {
855
+ "weights": "qint4",
856
+ "activations": "none"
857
+ },
858
+ "model.language_model.layers.11.self_attn.v_proj": {
859
+ "weights": "qint4",
860
+ "activations": "none"
861
+ },
862
+ "model.language_model.layers.11.self_attn.o_proj": {
863
+ "weights": "qint4",
864
+ "activations": "none"
865
+ },
866
+ "model.language_model.layers.11.mlp.gate_proj": {
867
+ "weights": "qint4",
868
+ "activations": "none"
869
+ },
870
+ "model.language_model.layers.11.mlp.up_proj": {
871
+ "weights": "qint4",
872
+ "activations": "none"
873
+ },
874
+ "model.language_model.layers.11.mlp.down_proj": {
875
+ "weights": "qint4",
876
+ "activations": "none"
877
+ },
878
+ "model.language_model.layers.11.per_layer_input_gate": {
879
+ "weights": "qint4",
880
+ "activations": "none"
881
+ },
882
+ "model.language_model.layers.11.per_layer_projection": {
883
+ "weights": "qint4",
884
+ "activations": "none"
885
+ },
886
+ "model.language_model.layers.12.self_attn.q_proj": {
887
+ "weights": "qint4",
888
+ "activations": "none"
889
+ },
890
+ "model.language_model.layers.12.self_attn.k_proj": {
891
+ "weights": "qint4",
892
+ "activations": "none"
893
+ },
894
+ "model.language_model.layers.12.self_attn.v_proj": {
895
+ "weights": "qint4",
896
+ "activations": "none"
897
+ },
898
+ "model.language_model.layers.12.self_attn.o_proj": {
899
+ "weights": "qint4",
900
+ "activations": "none"
901
+ },
902
+ "model.language_model.layers.12.mlp.gate_proj": {
903
+ "weights": "qint4",
904
+ "activations": "none"
905
+ },
906
+ "model.language_model.layers.12.mlp.up_proj": {
907
+ "weights": "qint4",
908
+ "activations": "none"
909
+ },
910
+ "model.language_model.layers.12.mlp.down_proj": {
911
+ "weights": "qint4",
912
+ "activations": "none"
913
+ },
914
+ "model.language_model.layers.12.per_layer_input_gate": {
915
+ "weights": "qint4",
916
+ "activations": "none"
917
+ },
918
+ "model.language_model.layers.12.per_layer_projection": {
919
+ "weights": "qint4",
920
+ "activations": "none"
921
+ },
922
+ "model.language_model.layers.13.self_attn.q_proj": {
923
+ "weights": "qint4",
924
+ "activations": "none"
925
+ },
926
+ "model.language_model.layers.13.self_attn.k_proj": {
927
+ "weights": "qint4",
928
+ "activations": "none"
929
+ },
930
+ "model.language_model.layers.13.self_attn.v_proj": {
931
+ "weights": "qint4",
932
+ "activations": "none"
933
+ },
934
+ "model.language_model.layers.13.self_attn.o_proj": {
935
+ "weights": "qint4",
936
+ "activations": "none"
937
+ },
938
+ "model.language_model.layers.13.mlp.gate_proj": {
939
+ "weights": "qint4",
940
+ "activations": "none"
941
+ },
942
+ "model.language_model.layers.13.mlp.up_proj": {
943
+ "weights": "qint4",
944
+ "activations": "none"
945
+ },
946
+ "model.language_model.layers.13.mlp.down_proj": {
947
+ "weights": "qint4",
948
+ "activations": "none"
949
+ },
950
+ "model.language_model.layers.13.per_layer_input_gate": {
951
+ "weights": "qint4",
952
+ "activations": "none"
953
+ },
954
+ "model.language_model.layers.13.per_layer_projection": {
955
+ "weights": "qint4",
956
+ "activations": "none"
957
+ },
958
+ "model.language_model.layers.14.self_attn.q_proj": {
959
+ "weights": "qint4",
960
+ "activations": "none"
961
+ },
962
+ "model.language_model.layers.14.self_attn.k_proj": {
963
+ "weights": "qint4",
964
+ "activations": "none"
965
+ },
966
+ "model.language_model.layers.14.self_attn.v_proj": {
967
+ "weights": "qint4",
968
+ "activations": "none"
969
+ },
970
+ "model.language_model.layers.14.self_attn.o_proj": {
971
+ "weights": "qint4",
972
+ "activations": "none"
973
+ },
974
+ "model.language_model.layers.14.mlp.gate_proj": {
975
+ "weights": "qint4",
976
+ "activations": "none"
977
+ },
978
+ "model.language_model.layers.14.mlp.up_proj": {
979
+ "weights": "qint4",
980
+ "activations": "none"
981
+ },
982
+ "model.language_model.layers.14.mlp.down_proj": {
983
+ "weights": "qint4",
984
+ "activations": "none"
985
+ },
986
+ "model.language_model.layers.14.per_layer_input_gate": {
987
+ "weights": "qint4",
988
+ "activations": "none"
989
+ },
990
+ "model.language_model.layers.14.per_layer_projection": {
991
+ "weights": "qint4",
992
+ "activations": "none"
993
+ },
994
+ "model.language_model.layers.15.self_attn.q_proj": {
995
+ "weights": "qint4",
996
+ "activations": "none"
997
+ },
998
+ "model.language_model.layers.15.self_attn.o_proj": {
999
+ "weights": "qint4",
1000
+ "activations": "none"
1001
+ },
1002
+ "model.language_model.layers.15.mlp.gate_proj": {
1003
+ "weights": "qint4",
1004
+ "activations": "none"
1005
+ },
1006
+ "model.language_model.layers.15.mlp.up_proj": {
1007
+ "weights": "qint4",
1008
+ "activations": "none"
1009
+ },
1010
+ "model.language_model.layers.15.mlp.down_proj": {
1011
+ "weights": "qint4",
1012
+ "activations": "none"
1013
+ },
1014
+ "model.language_model.layers.15.per_layer_input_gate": {
1015
+ "weights": "qint4",
1016
+ "activations": "none"
1017
+ },
1018
+ "model.language_model.layers.15.per_layer_projection": {
1019
+ "weights": "qint4",
1020
+ "activations": "none"
1021
+ },
1022
+ "model.language_model.layers.16.self_attn.q_proj": {
1023
+ "weights": "qint4",
1024
+ "activations": "none"
1025
+ },
1026
+ "model.language_model.layers.16.self_attn.o_proj": {
1027
+ "weights": "qint4",
1028
+ "activations": "none"
1029
+ },
1030
+ "model.language_model.layers.16.mlp.gate_proj": {
1031
+ "weights": "qint4",
1032
+ "activations": "none"
1033
+ },
1034
+ "model.language_model.layers.16.mlp.up_proj": {
1035
+ "weights": "qint4",
1036
+ "activations": "none"
1037
+ },
1038
+ "model.language_model.layers.16.mlp.down_proj": {
1039
+ "weights": "qint4",
1040
+ "activations": "none"
1041
+ },
1042
+ "model.language_model.layers.16.per_layer_input_gate": {
1043
+ "weights": "qint4",
1044
+ "activations": "none"
1045
+ },
1046
+ "model.language_model.layers.16.per_layer_projection": {
1047
+ "weights": "qint4",
1048
+ "activations": "none"
1049
+ },
1050
+ "model.language_model.layers.17.self_attn.q_proj": {
1051
+ "weights": "qint4",
1052
+ "activations": "none"
1053
+ },
1054
+ "model.language_model.layers.17.self_attn.o_proj": {
1055
+ "weights": "qint4",
1056
+ "activations": "none"
1057
+ },
1058
+ "model.language_model.layers.17.mlp.gate_proj": {
1059
+ "weights": "qint4",
1060
+ "activations": "none"
1061
+ },
1062
+ "model.language_model.layers.17.mlp.up_proj": {
1063
+ "weights": "qint4",
1064
+ "activations": "none"
1065
+ },
1066
+ "model.language_model.layers.17.mlp.down_proj": {
1067
+ "weights": "qint4",
1068
+ "activations": "none"
1069
+ },
1070
+ "model.language_model.layers.17.per_layer_input_gate": {
1071
+ "weights": "qint4",
1072
+ "activations": "none"
1073
+ },
1074
+ "model.language_model.layers.17.per_layer_projection": {
1075
+ "weights": "qint4",
1076
+ "activations": "none"
1077
+ },
1078
+ "model.language_model.layers.18.self_attn.q_proj": {
1079
+ "weights": "qint4",
1080
+ "activations": "none"
1081
+ },
1082
+ "model.language_model.layers.18.self_attn.o_proj": {
1083
+ "weights": "qint4",
1084
+ "activations": "none"
1085
+ },
1086
+ "model.language_model.layers.18.mlp.gate_proj": {
1087
+ "weights": "qint4",
1088
+ "activations": "none"
1089
+ },
1090
+ "model.language_model.layers.18.mlp.up_proj": {
1091
+ "weights": "qint4",
1092
+ "activations": "none"
1093
+ },
1094
+ "model.language_model.layers.18.mlp.down_proj": {
1095
+ "weights": "qint4",
1096
+ "activations": "none"
1097
+ },
1098
+ "model.language_model.layers.18.per_layer_input_gate": {
1099
+ "weights": "qint4",
1100
+ "activations": "none"
1101
+ },
1102
+ "model.language_model.layers.18.per_layer_projection": {
1103
+ "weights": "qint4",
1104
+ "activations": "none"
1105
+ },
1106
+ "model.language_model.layers.19.self_attn.q_proj": {
1107
+ "weights": "qint4",
1108
+ "activations": "none"
1109
+ },
1110
+ "model.language_model.layers.19.self_attn.o_proj": {
1111
+ "weights": "qint4",
1112
+ "activations": "none"
1113
+ },
1114
+ "model.language_model.layers.19.mlp.gate_proj": {
1115
+ "weights": "qint4",
1116
+ "activations": "none"
1117
+ },
1118
+ "model.language_model.layers.19.mlp.up_proj": {
1119
+ "weights": "qint4",
1120
+ "activations": "none"
1121
+ },
1122
+ "model.language_model.layers.19.mlp.down_proj": {
1123
+ "weights": "qint4",
1124
+ "activations": "none"
1125
+ },
1126
+ "model.language_model.layers.19.per_layer_input_gate": {
1127
+ "weights": "qint4",
1128
+ "activations": "none"
1129
+ },
1130
+ "model.language_model.layers.19.per_layer_projection": {
1131
+ "weights": "qint4",
1132
+ "activations": "none"
1133
+ },
1134
+ "model.language_model.layers.20.self_attn.q_proj": {
1135
+ "weights": "qint4",
1136
+ "activations": "none"
1137
+ },
1138
+ "model.language_model.layers.20.self_attn.o_proj": {
1139
+ "weights": "qint4",
1140
+ "activations": "none"
1141
+ },
1142
+ "model.language_model.layers.20.mlp.gate_proj": {
1143
+ "weights": "qint4",
1144
+ "activations": "none"
1145
+ },
1146
+ "model.language_model.layers.20.mlp.up_proj": {
1147
+ "weights": "qint4",
1148
+ "activations": "none"
1149
+ },
1150
+ "model.language_model.layers.20.mlp.down_proj": {
1151
+ "weights": "qint4",
1152
+ "activations": "none"
1153
+ },
1154
+ "model.language_model.layers.20.per_layer_input_gate": {
1155
+ "weights": "qint4",
1156
+ "activations": "none"
1157
+ },
1158
+ "model.language_model.layers.20.per_layer_projection": {
1159
+ "weights": "qint4",
1160
+ "activations": "none"
1161
+ },
1162
+ "model.language_model.layers.21.self_attn.q_proj": {
1163
+ "weights": "qint4",
1164
+ "activations": "none"
1165
+ },
1166
+ "model.language_model.layers.21.self_attn.o_proj": {
1167
+ "weights": "qint4",
1168
+ "activations": "none"
1169
+ },
1170
+ "model.language_model.layers.21.mlp.gate_proj": {
1171
+ "weights": "qint4",
1172
+ "activations": "none"
1173
+ },
1174
+ "model.language_model.layers.21.mlp.up_proj": {
1175
+ "weights": "qint4",
1176
+ "activations": "none"
1177
+ },
1178
+ "model.language_model.layers.21.mlp.down_proj": {
1179
+ "weights": "qint4",
1180
+ "activations": "none"
1181
+ },
1182
+ "model.language_model.layers.21.per_layer_input_gate": {
1183
+ "weights": "qint4",
1184
+ "activations": "none"
1185
+ },
1186
+ "model.language_model.layers.21.per_layer_projection": {
1187
+ "weights": "qint4",
1188
+ "activations": "none"
1189
+ },
1190
+ "model.language_model.layers.22.self_attn.q_proj": {
1191
+ "weights": "qint4",
1192
+ "activations": "none"
1193
+ },
1194
+ "model.language_model.layers.22.self_attn.o_proj": {
1195
+ "weights": "qint4",
1196
+ "activations": "none"
1197
+ },
1198
+ "model.language_model.layers.22.mlp.gate_proj": {
1199
+ "weights": "qint4",
1200
+ "activations": "none"
1201
+ },
1202
+ "model.language_model.layers.22.mlp.up_proj": {
1203
+ "weights": "qint4",
1204
+ "activations": "none"
1205
+ },
1206
+ "model.language_model.layers.22.mlp.down_proj": {
1207
+ "weights": "qint4",
1208
+ "activations": "none"
1209
+ },
1210
+ "model.language_model.layers.22.per_layer_input_gate": {
1211
+ "weights": "qint4",
1212
+ "activations": "none"
1213
+ },
1214
+ "model.language_model.layers.22.per_layer_projection": {
1215
+ "weights": "qint4",
1216
+ "activations": "none"
1217
+ },
1218
+ "model.language_model.layers.23.self_attn.q_proj": {
1219
+ "weights": "qint4",
1220
+ "activations": "none"
1221
+ },
1222
+ "model.language_model.layers.23.self_attn.o_proj": {
1223
+ "weights": "qint4",
1224
+ "activations": "none"
1225
+ },
1226
+ "model.language_model.layers.23.mlp.gate_proj": {
1227
+ "weights": "qint4",
1228
+ "activations": "none"
1229
+ },
1230
+ "model.language_model.layers.23.mlp.up_proj": {
1231
+ "weights": "qint4",
1232
+ "activations": "none"
1233
+ },
1234
+ "model.language_model.layers.23.mlp.down_proj": {
1235
+ "weights": "qint4",
1236
+ "activations": "none"
1237
+ },
1238
+ "model.language_model.layers.23.per_layer_input_gate": {
1239
+ "weights": "qint4",
1240
+ "activations": "none"
1241
+ },
1242
+ "model.language_model.layers.23.per_layer_projection": {
1243
+ "weights": "qint4",
1244
+ "activations": "none"
1245
+ },
1246
+ "model.language_model.layers.24.self_attn.q_proj": {
1247
+ "weights": "qint4",
1248
+ "activations": "none"
1249
+ },
1250
+ "model.language_model.layers.24.self_attn.o_proj": {
1251
+ "weights": "qint4",
1252
+ "activations": "none"
1253
+ },
1254
+ "model.language_model.layers.24.mlp.gate_proj": {
1255
+ "weights": "qint4",
1256
+ "activations": "none"
1257
+ },
1258
+ "model.language_model.layers.24.mlp.up_proj": {
1259
+ "weights": "qint4",
1260
+ "activations": "none"
1261
+ },
1262
+ "model.language_model.layers.24.mlp.down_proj": {
1263
+ "weights": "qint4",
1264
+ "activations": "none"
1265
+ },
1266
+ "model.language_model.layers.24.per_layer_input_gate": {
1267
+ "weights": "qint4",
1268
+ "activations": "none"
1269
+ },
1270
+ "model.language_model.layers.24.per_layer_projection": {
1271
+ "weights": "qint4",
1272
+ "activations": "none"
1273
+ },
1274
+ "model.language_model.layers.25.self_attn.q_proj": {
1275
+ "weights": "qint4",
1276
+ "activations": "none"
1277
+ },
1278
+ "model.language_model.layers.25.self_attn.o_proj": {
1279
+ "weights": "qint4",
1280
+ "activations": "none"
1281
+ },
1282
+ "model.language_model.layers.25.mlp.gate_proj": {
1283
+ "weights": "qint4",
1284
+ "activations": "none"
1285
+ },
1286
+ "model.language_model.layers.25.mlp.up_proj": {
1287
+ "weights": "qint4",
1288
+ "activations": "none"
1289
+ },
1290
+ "model.language_model.layers.25.mlp.down_proj": {
1291
+ "weights": "qint4",
1292
+ "activations": "none"
1293
+ },
1294
+ "model.language_model.layers.25.per_layer_input_gate": {
1295
+ "weights": "qint4",
1296
+ "activations": "none"
1297
+ },
1298
+ "model.language_model.layers.25.per_layer_projection": {
1299
+ "weights": "qint4",
1300
+ "activations": "none"
1301
+ },
1302
+ "model.language_model.layers.26.self_attn.q_proj": {
1303
+ "weights": "qint4",
1304
+ "activations": "none"
1305
+ },
1306
+ "model.language_model.layers.26.self_attn.o_proj": {
1307
+ "weights": "qint4",
1308
+ "activations": "none"
1309
+ },
1310
+ "model.language_model.layers.26.mlp.gate_proj": {
1311
+ "weights": "qint4",
1312
+ "activations": "none"
1313
+ },
1314
+ "model.language_model.layers.26.mlp.up_proj": {
1315
+ "weights": "qint4",
1316
+ "activations": "none"
1317
+ },
1318
+ "model.language_model.layers.26.mlp.down_proj": {
1319
+ "weights": "qint4",
1320
+ "activations": "none"
1321
+ },
1322
+ "model.language_model.layers.26.per_layer_input_gate": {
1323
+ "weights": "qint4",
1324
+ "activations": "none"
1325
+ },
1326
+ "model.language_model.layers.26.per_layer_projection": {
1327
+ "weights": "qint4",
1328
+ "activations": "none"
1329
+ },
1330
+ "model.language_model.layers.27.self_attn.q_proj": {
1331
+ "weights": "qint4",
1332
+ "activations": "none"
1333
+ },
1334
+ "model.language_model.layers.27.self_attn.o_proj": {
1335
+ "weights": "qint4",
1336
+ "activations": "none"
1337
+ },
1338
+ "model.language_model.layers.27.mlp.gate_proj": {
1339
+ "weights": "qint4",
1340
+ "activations": "none"
1341
+ },
1342
+ "model.language_model.layers.27.mlp.up_proj": {
1343
+ "weights": "qint4",
1344
+ "activations": "none"
1345
+ },
1346
+ "model.language_model.layers.27.mlp.down_proj": {
1347
+ "weights": "qint4",
1348
+ "activations": "none"
1349
+ },
1350
+ "model.language_model.layers.27.per_layer_input_gate": {
1351
+ "weights": "qint4",
1352
+ "activations": "none"
1353
+ },
1354
+ "model.language_model.layers.27.per_layer_projection": {
1355
+ "weights": "qint4",
1356
+ "activations": "none"
1357
+ },
1358
+ "model.language_model.layers.28.self_attn.q_proj": {
1359
+ "weights": "qint4",
1360
+ "activations": "none"
1361
+ },
1362
+ "model.language_model.layers.28.self_attn.o_proj": {
1363
+ "weights": "qint4",
1364
+ "activations": "none"
1365
+ },
1366
+ "model.language_model.layers.28.mlp.gate_proj": {
1367
+ "weights": "qint4",
1368
+ "activations": "none"
1369
+ },
1370
+ "model.language_model.layers.28.mlp.up_proj": {
1371
+ "weights": "qint4",
1372
+ "activations": "none"
1373
+ },
1374
+ "model.language_model.layers.28.mlp.down_proj": {
1375
+ "weights": "qint4",
1376
+ "activations": "none"
1377
+ },
1378
+ "model.language_model.layers.28.per_layer_input_gate": {
1379
+ "weights": "qint4",
1380
+ "activations": "none"
1381
+ },
1382
+ "model.language_model.layers.28.per_layer_projection": {
1383
+ "weights": "qint4",
1384
+ "activations": "none"
1385
+ },
1386
+ "model.language_model.layers.29.self_attn.q_proj": {
1387
+ "weights": "qint4",
1388
+ "activations": "none"
1389
+ },
1390
+ "model.language_model.layers.29.self_attn.o_proj": {
1391
+ "weights": "qint4",
1392
+ "activations": "none"
1393
+ },
1394
+ "model.language_model.layers.29.mlp.gate_proj": {
1395
+ "weights": "qint4",
1396
+ "activations": "none"
1397
+ },
1398
+ "model.language_model.layers.29.mlp.up_proj": {
1399
+ "weights": "qint4",
1400
+ "activations": "none"
1401
+ },
1402
+ "model.language_model.layers.29.mlp.down_proj": {
1403
+ "weights": "qint4",
1404
+ "activations": "none"
1405
+ },
1406
+ "model.language_model.layers.29.per_layer_input_gate": {
1407
+ "weights": "qint4",
1408
+ "activations": "none"
1409
+ },
1410
+ "model.language_model.layers.29.per_layer_projection": {
1411
+ "weights": "qint4",
1412
+ "activations": "none"
1413
+ },
1414
+ "model.language_model.layers.30.self_attn.q_proj": {
1415
+ "weights": "qint4",
1416
+ "activations": "none"
1417
+ },
1418
+ "model.language_model.layers.30.self_attn.o_proj": {
1419
+ "weights": "qint4",
1420
+ "activations": "none"
1421
+ },
1422
+ "model.language_model.layers.30.mlp.gate_proj": {
1423
+ "weights": "qint4",
1424
+ "activations": "none"
1425
+ },
1426
+ "model.language_model.layers.30.mlp.up_proj": {
1427
+ "weights": "qint4",
1428
+ "activations": "none"
1429
+ },
1430
+ "model.language_model.layers.30.mlp.down_proj": {
1431
+ "weights": "qint4",
1432
+ "activations": "none"
1433
+ },
1434
+ "model.language_model.layers.30.per_layer_input_gate": {
1435
+ "weights": "qint4",
1436
+ "activations": "none"
1437
+ },
1438
+ "model.language_model.layers.30.per_layer_projection": {
1439
+ "weights": "qint4",
1440
+ "activations": "none"
1441
+ },
1442
+ "model.language_model.layers.31.self_attn.q_proj": {
1443
+ "weights": "qint4",
1444
+ "activations": "none"
1445
+ },
1446
+ "model.language_model.layers.31.self_attn.o_proj": {
1447
+ "weights": "qint4",
1448
+ "activations": "none"
1449
+ },
1450
+ "model.language_model.layers.31.mlp.gate_proj": {
1451
+ "weights": "qint4",
1452
+ "activations": "none"
1453
+ },
1454
+ "model.language_model.layers.31.mlp.up_proj": {
1455
+ "weights": "qint4",
1456
+ "activations": "none"
1457
+ },
1458
+ "model.language_model.layers.31.mlp.down_proj": {
1459
+ "weights": "qint4",
1460
+ "activations": "none"
1461
+ },
1462
+ "model.language_model.layers.31.per_layer_input_gate": {
1463
+ "weights": "qint4",
1464
+ "activations": "none"
1465
+ },
1466
+ "model.language_model.layers.31.per_layer_projection": {
1467
+ "weights": "qint4",
1468
+ "activations": "none"
1469
+ },
1470
+ "model.language_model.layers.32.self_attn.q_proj": {
1471
+ "weights": "qint4",
1472
+ "activations": "none"
1473
+ },
1474
+ "model.language_model.layers.32.self_attn.o_proj": {
1475
+ "weights": "qint4",
1476
+ "activations": "none"
1477
+ },
1478
+ "model.language_model.layers.32.mlp.gate_proj": {
1479
+ "weights": "qint4",
1480
+ "activations": "none"
1481
+ },
1482
+ "model.language_model.layers.32.mlp.up_proj": {
1483
+ "weights": "qint4",
1484
+ "activations": "none"
1485
+ },
1486
+ "model.language_model.layers.32.mlp.down_proj": {
1487
+ "weights": "qint4",
1488
+ "activations": "none"
1489
+ },
1490
+ "model.language_model.layers.32.per_layer_input_gate": {
1491
+ "weights": "qint4",
1492
+ "activations": "none"
1493
+ },
1494
+ "model.language_model.layers.32.per_layer_projection": {
1495
+ "weights": "qint4",
1496
+ "activations": "none"
1497
+ },
1498
+ "model.language_model.layers.33.self_attn.q_proj": {
1499
+ "weights": "qint4",
1500
+ "activations": "none"
1501
+ },
1502
+ "model.language_model.layers.33.self_attn.o_proj": {
1503
+ "weights": "qint4",
1504
+ "activations": "none"
1505
+ },
1506
+ "model.language_model.layers.33.mlp.gate_proj": {
1507
+ "weights": "qint4",
1508
+ "activations": "none"
1509
+ },
1510
+ "model.language_model.layers.33.mlp.up_proj": {
1511
+ "weights": "qint4",
1512
+ "activations": "none"
1513
+ },
1514
+ "model.language_model.layers.33.mlp.down_proj": {
1515
+ "weights": "qint4",
1516
+ "activations": "none"
1517
+ },
1518
+ "model.language_model.layers.33.per_layer_input_gate": {
1519
+ "weights": "qint4",
1520
+ "activations": "none"
1521
+ },
1522
+ "model.language_model.layers.33.per_layer_projection": {
1523
+ "weights": "qint4",
1524
+ "activations": "none"
1525
+ },
1526
+ "model.language_model.layers.34.self_attn.q_proj": {
1527
+ "weights": "qint4",
1528
+ "activations": "none"
1529
+ },
1530
+ "model.language_model.layers.34.self_attn.o_proj": {
1531
+ "weights": "qint4",
1532
+ "activations": "none"
1533
+ },
1534
+ "model.language_model.layers.34.mlp.gate_proj": {
1535
+ "weights": "qint4",
1536
+ "activations": "none"
1537
+ },
1538
+ "model.language_model.layers.34.mlp.up_proj": {
1539
+ "weights": "qint4",
1540
+ "activations": "none"
1541
+ },
1542
+ "model.language_model.layers.34.mlp.down_proj": {
1543
+ "weights": "qint4",
1544
+ "activations": "none"
1545
+ },
1546
+ "model.language_model.layers.34.per_layer_input_gate": {
1547
+ "weights": "qint4",
1548
+ "activations": "none"
1549
+ },
1550
+ "model.language_model.layers.34.per_layer_projection": {
1551
+ "weights": "qint4",
1552
+ "activations": "none"
1553
+ },
1554
+ "model.language_model.per_layer_model_projection": {
1555
+ "weights": "qint4",
1556
+ "activations": "none"
1557
+ },
1558
+ "model.audio_tower.subsample_conv_projection.layer0.conv": {
1559
+ "weights": "qint4",
1560
+ "activations": "none"
1561
+ },
1562
+ "model.audio_tower.subsample_conv_projection.layer1.conv": {
1563
+ "weights": "qint4",
1564
+ "activations": "none"
1565
+ },
1566
+ "model.audio_tower.subsample_conv_projection.input_proj_linear": {
1567
+ "weights": "qint4",
1568
+ "activations": "none"
1569
+ },
1570
+ "model.audio_tower.layers.0.feed_forward1.ffw_layer_1.linear": {
1571
+ "weights": "qint4",
1572
+ "activations": "none"
1573
+ },
1574
+ "model.audio_tower.layers.0.feed_forward1.ffw_layer_2.linear": {
1575
+ "weights": "qint4",
1576
+ "activations": "none"
1577
+ },
1578
+ "model.audio_tower.layers.0.feed_forward2.ffw_layer_1.linear": {
1579
+ "weights": "qint4",
1580
+ "activations": "none"
1581
+ },
1582
+ "model.audio_tower.layers.0.feed_forward2.ffw_layer_2.linear": {
1583
+ "weights": "qint4",
1584
+ "activations": "none"
1585
+ },
1586
+ "model.audio_tower.layers.0.self_attn.q_proj.linear": {
1587
+ "weights": "qint4",
1588
+ "activations": "none"
1589
+ },
1590
+ "model.audio_tower.layers.0.self_attn.k_proj.linear": {
1591
+ "weights": "qint4",
1592
+ "activations": "none"
1593
+ },
1594
+ "model.audio_tower.layers.0.self_attn.v_proj.linear": {
1595
+ "weights": "qint4",
1596
+ "activations": "none"
1597
+ },
1598
+ "model.audio_tower.layers.0.self_attn.post.linear": {
1599
+ "weights": "qint4",
1600
+ "activations": "none"
1601
+ },
1602
+ "model.audio_tower.layers.0.self_attn.relative_k_proj": {
1603
+ "weights": "qint4",
1604
+ "activations": "none"
1605
+ },
1606
+ "model.audio_tower.layers.0.lconv1d.linear_start.linear": {
1607
+ "weights": "qint4",
1608
+ "activations": "none"
1609
+ },
1610
+ "model.audio_tower.layers.0.lconv1d.linear_end.linear": {
1611
+ "weights": "qint4",
1612
+ "activations": "none"
1613
+ },
1614
+ "model.audio_tower.layers.1.feed_forward1.ffw_layer_1.linear": {
1615
+ "weights": "qint4",
1616
+ "activations": "none"
1617
+ },
1618
+ "model.audio_tower.layers.1.feed_forward1.ffw_layer_2.linear": {
1619
+ "weights": "qint4",
1620
+ "activations": "none"
1621
+ },
1622
+ "model.audio_tower.layers.1.feed_forward2.ffw_layer_1.linear": {
1623
+ "weights": "qint4",
1624
+ "activations": "none"
1625
+ },
1626
+ "model.audio_tower.layers.1.feed_forward2.ffw_layer_2.linear": {
1627
+ "weights": "qint4",
1628
+ "activations": "none"
1629
+ },
1630
+ "model.audio_tower.layers.1.self_attn.q_proj.linear": {
1631
+ "weights": "qint4",
1632
+ "activations": "none"
1633
+ },
1634
+ "model.audio_tower.layers.1.self_attn.k_proj.linear": {
1635
+ "weights": "qint4",
1636
+ "activations": "none"
1637
+ },
1638
+ "model.audio_tower.layers.1.self_attn.v_proj.linear": {
1639
+ "weights": "qint4",
1640
+ "activations": "none"
1641
+ },
1642
+ "model.audio_tower.layers.1.self_attn.post.linear": {
1643
+ "weights": "qint4",
1644
+ "activations": "none"
1645
+ },
1646
+ "model.audio_tower.layers.1.self_attn.relative_k_proj": {
1647
+ "weights": "qint4",
1648
+ "activations": "none"
1649
+ },
1650
+ "model.audio_tower.layers.1.lconv1d.linear_start.linear": {
1651
+ "weights": "qint4",
1652
+ "activations": "none"
1653
+ },
1654
+ "model.audio_tower.layers.1.lconv1d.linear_end.linear": {
1655
+ "weights": "qint4",
1656
+ "activations": "none"
1657
+ },
1658
+ "model.audio_tower.layers.2.feed_forward1.ffw_layer_1.linear": {
1659
+ "weights": "qint4",
1660
+ "activations": "none"
1661
+ },
1662
+ "model.audio_tower.layers.2.feed_forward1.ffw_layer_2.linear": {
1663
+ "weights": "qint4",
1664
+ "activations": "none"
1665
+ },
1666
+ "model.audio_tower.layers.2.feed_forward2.ffw_layer_1.linear": {
1667
+ "weights": "qint4",
1668
+ "activations": "none"
1669
+ },
1670
+ "model.audio_tower.layers.2.feed_forward2.ffw_layer_2.linear": {
1671
+ "weights": "qint4",
1672
+ "activations": "none"
1673
+ },
1674
+ "model.audio_tower.layers.2.self_attn.q_proj.linear": {
1675
+ "weights": "qint4",
1676
+ "activations": "none"
1677
+ },
1678
+ "model.audio_tower.layers.2.self_attn.k_proj.linear": {
1679
+ "weights": "qint4",
1680
+ "activations": "none"
1681
+ },
1682
+ "model.audio_tower.layers.2.self_attn.v_proj.linear": {
1683
+ "weights": "qint4",
1684
+ "activations": "none"
1685
+ },
1686
+ "model.audio_tower.layers.2.self_attn.post.linear": {
1687
+ "weights": "qint4",
1688
+ "activations": "none"
1689
+ },
1690
+ "model.audio_tower.layers.2.self_attn.relative_k_proj": {
1691
+ "weights": "qint4",
1692
+ "activations": "none"
1693
+ },
1694
+ "model.audio_tower.layers.2.lconv1d.linear_start.linear": {
1695
+ "weights": "qint4",
1696
+ "activations": "none"
1697
+ },
1698
+ "model.audio_tower.layers.2.lconv1d.linear_end.linear": {
1699
+ "weights": "qint4",
1700
+ "activations": "none"
1701
+ },
1702
+ "model.audio_tower.layers.3.feed_forward1.ffw_layer_1.linear": {
1703
+ "weights": "qint4",
1704
+ "activations": "none"
1705
+ },
1706
+ "model.audio_tower.layers.3.feed_forward1.ffw_layer_2.linear": {
1707
+ "weights": "qint4",
1708
+ "activations": "none"
1709
+ },
1710
+ "model.audio_tower.layers.3.feed_forward2.ffw_layer_1.linear": {
1711
+ "weights": "qint4",
1712
+ "activations": "none"
1713
+ },
1714
+ "model.audio_tower.layers.3.feed_forward2.ffw_layer_2.linear": {
1715
+ "weights": "qint4",
1716
+ "activations": "none"
1717
+ },
1718
+ "model.audio_tower.layers.3.self_attn.q_proj.linear": {
1719
+ "weights": "qint4",
1720
+ "activations": "none"
1721
+ },
1722
+ "model.audio_tower.layers.3.self_attn.k_proj.linear": {
1723
+ "weights": "qint4",
1724
+ "activations": "none"
1725
+ },
1726
+ "model.audio_tower.layers.3.self_attn.v_proj.linear": {
1727
+ "weights": "qint4",
1728
+ "activations": "none"
1729
+ },
1730
+ "model.audio_tower.layers.3.self_attn.post.linear": {
1731
+ "weights": "qint4",
1732
+ "activations": "none"
1733
+ },
1734
+ "model.audio_tower.layers.3.self_attn.relative_k_proj": {
1735
+ "weights": "qint4",
1736
+ "activations": "none"
1737
+ },
1738
+ "model.audio_tower.layers.3.lconv1d.linear_start.linear": {
1739
+ "weights": "qint4",
1740
+ "activations": "none"
1741
+ },
1742
+ "model.audio_tower.layers.3.lconv1d.linear_end.linear": {
1743
+ "weights": "qint4",
1744
+ "activations": "none"
1745
+ },
1746
+ "model.audio_tower.layers.4.feed_forward1.ffw_layer_1.linear": {
1747
+ "weights": "qint4",
1748
+ "activations": "none"
1749
+ },
1750
+ "model.audio_tower.layers.4.feed_forward1.ffw_layer_2.linear": {
1751
+ "weights": "qint4",
1752
+ "activations": "none"
1753
+ },
1754
+ "model.audio_tower.layers.4.feed_forward2.ffw_layer_1.linear": {
1755
+ "weights": "qint4",
1756
+ "activations": "none"
1757
+ },
1758
+ "model.audio_tower.layers.4.feed_forward2.ffw_layer_2.linear": {
1759
+ "weights": "qint4",
1760
+ "activations": "none"
1761
+ },
1762
+ "model.audio_tower.layers.4.self_attn.q_proj.linear": {
1763
+ "weights": "qint4",
1764
+ "activations": "none"
1765
+ },
1766
+ "model.audio_tower.layers.4.self_attn.k_proj.linear": {
1767
+ "weights": "qint4",
1768
+ "activations": "none"
1769
+ },
1770
+ "model.audio_tower.layers.4.self_attn.v_proj.linear": {
1771
+ "weights": "qint4",
1772
+ "activations": "none"
1773
+ },
1774
+ "model.audio_tower.layers.4.self_attn.post.linear": {
1775
+ "weights": "qint4",
1776
+ "activations": "none"
1777
+ },
1778
+ "model.audio_tower.layers.4.self_attn.relative_k_proj": {
1779
+ "weights": "qint4",
1780
+ "activations": "none"
1781
+ },
1782
+ "model.audio_tower.layers.4.lconv1d.linear_start.linear": {
1783
+ "weights": "qint4",
1784
+ "activations": "none"
1785
+ },
1786
+ "model.audio_tower.layers.4.lconv1d.linear_end.linear": {
1787
+ "weights": "qint4",
1788
+ "activations": "none"
1789
+ },
1790
+ "model.audio_tower.layers.5.feed_forward1.ffw_layer_1.linear": {
1791
+ "weights": "qint4",
1792
+ "activations": "none"
1793
+ },
1794
+ "model.audio_tower.layers.5.feed_forward1.ffw_layer_2.linear": {
1795
+ "weights": "qint4",
1796
+ "activations": "none"
1797
+ },
1798
+ "model.audio_tower.layers.5.feed_forward2.ffw_layer_1.linear": {
1799
+ "weights": "qint4",
1800
+ "activations": "none"
1801
+ },
1802
+ "model.audio_tower.layers.5.feed_forward2.ffw_layer_2.linear": {
1803
+ "weights": "qint4",
1804
+ "activations": "none"
1805
+ },
1806
+ "model.audio_tower.layers.5.self_attn.q_proj.linear": {
1807
+ "weights": "qint4",
1808
+ "activations": "none"
1809
+ },
1810
+ "model.audio_tower.layers.5.self_attn.k_proj.linear": {
1811
+ "weights": "qint4",
1812
+ "activations": "none"
1813
+ },
1814
+ "model.audio_tower.layers.5.self_attn.v_proj.linear": {
1815
+ "weights": "qint4",
1816
+ "activations": "none"
1817
+ },
1818
+ "model.audio_tower.layers.5.self_attn.post.linear": {
1819
+ "weights": "qint4",
1820
+ "activations": "none"
1821
+ },
1822
+ "model.audio_tower.layers.5.self_attn.relative_k_proj": {
1823
+ "weights": "qint4",
1824
+ "activations": "none"
1825
+ },
1826
+ "model.audio_tower.layers.5.lconv1d.linear_start.linear": {
1827
+ "weights": "qint4",
1828
+ "activations": "none"
1829
+ },
1830
+ "model.audio_tower.layers.5.lconv1d.linear_end.linear": {
1831
+ "weights": "qint4",
1832
+ "activations": "none"
1833
+ },
1834
+ "model.audio_tower.layers.6.feed_forward1.ffw_layer_1.linear": {
1835
+ "weights": "qint4",
1836
+ "activations": "none"
1837
+ },
1838
+ "model.audio_tower.layers.6.feed_forward1.ffw_layer_2.linear": {
1839
+ "weights": "qint4",
1840
+ "activations": "none"
1841
+ },
1842
+ "model.audio_tower.layers.6.feed_forward2.ffw_layer_1.linear": {
1843
+ "weights": "qint4",
1844
+ "activations": "none"
1845
+ },
1846
+ "model.audio_tower.layers.6.feed_forward2.ffw_layer_2.linear": {
1847
+ "weights": "qint4",
1848
+ "activations": "none"
1849
+ },
1850
+ "model.audio_tower.layers.6.self_attn.q_proj.linear": {
1851
+ "weights": "qint4",
1852
+ "activations": "none"
1853
+ },
1854
+ "model.audio_tower.layers.6.self_attn.k_proj.linear": {
1855
+ "weights": "qint4",
1856
+ "activations": "none"
1857
+ },
1858
+ "model.audio_tower.layers.6.self_attn.v_proj.linear": {
1859
+ "weights": "qint4",
1860
+ "activations": "none"
1861
+ },
1862
+ "model.audio_tower.layers.6.self_attn.post.linear": {
1863
+ "weights": "qint4",
1864
+ "activations": "none"
1865
+ },
1866
+ "model.audio_tower.layers.6.self_attn.relative_k_proj": {
1867
+ "weights": "qint4",
1868
+ "activations": "none"
1869
+ },
1870
+ "model.audio_tower.layers.6.lconv1d.linear_start.linear": {
1871
+ "weights": "qint4",
1872
+ "activations": "none"
1873
+ },
1874
+ "model.audio_tower.layers.6.lconv1d.linear_end.linear": {
1875
+ "weights": "qint4",
1876
+ "activations": "none"
1877
+ },
1878
+ "model.audio_tower.layers.7.feed_forward1.ffw_layer_1.linear": {
1879
+ "weights": "qint4",
1880
+ "activations": "none"
1881
+ },
1882
+ "model.audio_tower.layers.7.feed_forward1.ffw_layer_2.linear": {
1883
+ "weights": "qint4",
1884
+ "activations": "none"
1885
+ },
1886
+ "model.audio_tower.layers.7.feed_forward2.ffw_layer_1.linear": {
1887
+ "weights": "qint4",
1888
+ "activations": "none"
1889
+ },
1890
+ "model.audio_tower.layers.7.feed_forward2.ffw_layer_2.linear": {
1891
+ "weights": "qint4",
1892
+ "activations": "none"
1893
+ },
1894
+ "model.audio_tower.layers.7.self_attn.q_proj.linear": {
1895
+ "weights": "qint4",
1896
+ "activations": "none"
1897
+ },
1898
+ "model.audio_tower.layers.7.self_attn.k_proj.linear": {
1899
+ "weights": "qint4",
1900
+ "activations": "none"
1901
+ },
1902
+ "model.audio_tower.layers.7.self_attn.v_proj.linear": {
1903
+ "weights": "qint4",
1904
+ "activations": "none"
1905
+ },
1906
+ "model.audio_tower.layers.7.self_attn.post.linear": {
1907
+ "weights": "qint4",
1908
+ "activations": "none"
1909
+ },
1910
+ "model.audio_tower.layers.7.self_attn.relative_k_proj": {
1911
+ "weights": "qint4",
1912
+ "activations": "none"
1913
+ },
1914
+ "model.audio_tower.layers.7.lconv1d.linear_start.linear": {
1915
+ "weights": "qint4",
1916
+ "activations": "none"
1917
+ },
1918
+ "model.audio_tower.layers.7.lconv1d.linear_end.linear": {
1919
+ "weights": "qint4",
1920
+ "activations": "none"
1921
+ },
1922
+ "model.audio_tower.layers.8.feed_forward1.ffw_layer_1.linear": {
1923
+ "weights": "qint4",
1924
+ "activations": "none"
1925
+ },
1926
+ "model.audio_tower.layers.8.feed_forward1.ffw_layer_2.linear": {
1927
+ "weights": "qint4",
1928
+ "activations": "none"
1929
+ },
1930
+ "model.audio_tower.layers.8.feed_forward2.ffw_layer_1.linear": {
1931
+ "weights": "qint4",
1932
+ "activations": "none"
1933
+ },
1934
+ "model.audio_tower.layers.8.feed_forward2.ffw_layer_2.linear": {
1935
+ "weights": "qint4",
1936
+ "activations": "none"
1937
+ },
1938
+ "model.audio_tower.layers.8.self_attn.q_proj.linear": {
1939
+ "weights": "qint4",
1940
+ "activations": "none"
1941
+ },
1942
+ "model.audio_tower.layers.8.self_attn.k_proj.linear": {
1943
+ "weights": "qint4",
1944
+ "activations": "none"
1945
+ },
1946
+ "model.audio_tower.layers.8.self_attn.v_proj.linear": {
1947
+ "weights": "qint4",
1948
+ "activations": "none"
1949
+ },
1950
+ "model.audio_tower.layers.8.self_attn.post.linear": {
1951
+ "weights": "qint4",
1952
+ "activations": "none"
1953
+ },
1954
+ "model.audio_tower.layers.8.self_attn.relative_k_proj": {
1955
+ "weights": "qint4",
1956
+ "activations": "none"
1957
+ },
1958
+ "model.audio_tower.layers.8.lconv1d.linear_start.linear": {
1959
+ "weights": "qint4",
1960
+ "activations": "none"
1961
+ },
1962
+ "model.audio_tower.layers.8.lconv1d.linear_end.linear": {
1963
+ "weights": "qint4",
1964
+ "activations": "none"
1965
+ },
1966
+ "model.audio_tower.layers.9.feed_forward1.ffw_layer_1.linear": {
1967
+ "weights": "qint4",
1968
+ "activations": "none"
1969
+ },
1970
+ "model.audio_tower.layers.9.feed_forward1.ffw_layer_2.linear": {
1971
+ "weights": "qint4",
1972
+ "activations": "none"
1973
+ },
1974
+ "model.audio_tower.layers.9.feed_forward2.ffw_layer_1.linear": {
1975
+ "weights": "qint4",
1976
+ "activations": "none"
1977
+ },
1978
+ "model.audio_tower.layers.9.feed_forward2.ffw_layer_2.linear": {
1979
+ "weights": "qint4",
1980
+ "activations": "none"
1981
+ },
1982
+ "model.audio_tower.layers.9.self_attn.q_proj.linear": {
1983
+ "weights": "qint4",
1984
+ "activations": "none"
1985
+ },
1986
+ "model.audio_tower.layers.9.self_attn.k_proj.linear": {
1987
+ "weights": "qint4",
1988
+ "activations": "none"
1989
+ },
1990
+ "model.audio_tower.layers.9.self_attn.v_proj.linear": {
1991
+ "weights": "qint4",
1992
+ "activations": "none"
1993
+ },
1994
+ "model.audio_tower.layers.9.self_attn.post.linear": {
1995
+ "weights": "qint4",
1996
+ "activations": "none"
1997
+ },
1998
+ "model.audio_tower.layers.9.self_attn.relative_k_proj": {
1999
+ "weights": "qint4",
2000
+ "activations": "none"
2001
+ },
2002
+ "model.audio_tower.layers.9.lconv1d.linear_start.linear": {
2003
+ "weights": "qint4",
2004
+ "activations": "none"
2005
+ },
2006
+ "model.audio_tower.layers.9.lconv1d.linear_end.linear": {
2007
+ "weights": "qint4",
2008
+ "activations": "none"
2009
+ },
2010
+ "model.audio_tower.layers.10.feed_forward1.ffw_layer_1.linear": {
2011
+ "weights": "qint4",
2012
+ "activations": "none"
2013
+ },
2014
+ "model.audio_tower.layers.10.feed_forward1.ffw_layer_2.linear": {
2015
+ "weights": "qint4",
2016
+ "activations": "none"
2017
+ },
2018
+ "model.audio_tower.layers.10.feed_forward2.ffw_layer_1.linear": {
2019
+ "weights": "qint4",
2020
+ "activations": "none"
2021
+ },
2022
+ "model.audio_tower.layers.10.feed_forward2.ffw_layer_2.linear": {
2023
+ "weights": "qint4",
2024
+ "activations": "none"
2025
+ },
2026
+ "model.audio_tower.layers.10.self_attn.q_proj.linear": {
2027
+ "weights": "qint4",
2028
+ "activations": "none"
2029
+ },
2030
+ "model.audio_tower.layers.10.self_attn.k_proj.linear": {
2031
+ "weights": "qint4",
2032
+ "activations": "none"
2033
+ },
2034
+ "model.audio_tower.layers.10.self_attn.v_proj.linear": {
2035
+ "weights": "qint4",
2036
+ "activations": "none"
2037
+ },
2038
+ "model.audio_tower.layers.10.self_attn.post.linear": {
2039
+ "weights": "qint4",
2040
+ "activations": "none"
2041
+ },
2042
+ "model.audio_tower.layers.10.self_attn.relative_k_proj": {
2043
+ "weights": "qint4",
2044
+ "activations": "none"
2045
+ },
2046
+ "model.audio_tower.layers.10.lconv1d.linear_start.linear": {
2047
+ "weights": "qint4",
2048
+ "activations": "none"
2049
+ },
2050
+ "model.audio_tower.layers.10.lconv1d.linear_end.linear": {
2051
+ "weights": "qint4",
2052
+ "activations": "none"
2053
+ },
2054
+ "model.audio_tower.layers.11.feed_forward1.ffw_layer_1.linear": {
2055
+ "weights": "qint4",
2056
+ "activations": "none"
2057
+ },
2058
+ "model.audio_tower.layers.11.feed_forward1.ffw_layer_2.linear": {
2059
+ "weights": "qint4",
2060
+ "activations": "none"
2061
+ },
2062
+ "model.audio_tower.layers.11.feed_forward2.ffw_layer_1.linear": {
2063
+ "weights": "qint4",
2064
+ "activations": "none"
2065
+ },
2066
+ "model.audio_tower.layers.11.feed_forward2.ffw_layer_2.linear": {
2067
+ "weights": "qint4",
2068
+ "activations": "none"
2069
+ },
2070
+ "model.audio_tower.layers.11.self_attn.q_proj.linear": {
2071
+ "weights": "qint4",
2072
+ "activations": "none"
2073
+ },
2074
+ "model.audio_tower.layers.11.self_attn.k_proj.linear": {
2075
+ "weights": "qint4",
2076
+ "activations": "none"
2077
+ },
2078
+ "model.audio_tower.layers.11.self_attn.v_proj.linear": {
2079
+ "weights": "qint4",
2080
+ "activations": "none"
2081
+ },
2082
+ "model.audio_tower.layers.11.self_attn.post.linear": {
2083
+ "weights": "qint4",
2084
+ "activations": "none"
2085
+ },
2086
+ "model.audio_tower.layers.11.self_attn.relative_k_proj": {
2087
+ "weights": "qint4",
2088
+ "activations": "none"
2089
+ },
2090
+ "model.audio_tower.layers.11.lconv1d.linear_start.linear": {
2091
+ "weights": "qint4",
2092
+ "activations": "none"
2093
+ },
2094
+ "model.audio_tower.layers.11.lconv1d.linear_end.linear": {
2095
+ "weights": "qint4",
2096
+ "activations": "none"
2097
+ },
2098
+ "model.audio_tower.output_proj": {
2099
+ "weights": "qint4",
2100
+ "activations": "none"
2101
+ },
2102
+ "model.embed_vision.embedding_projection": {
2103
+ "weights": "qint4",
2104
+ "activations": "none"
2105
+ },
2106
+ "model.embed_audio.embedding_projection": {
2107
+ "weights": "qint4",
2108
+ "activations": "none"
2109
+ },
2110
+ "lm_head": {
2111
+ "weights": "qint4",
2112
+ "activations": "none"
2113
+ }
2114
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc8d3a0ce36466ccc1278bf987df5f71db1719b9ca6b4118264f45cb627bfe0f
3
+ size 32169626
tokenizer_config.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio_token": "<|audio|>",
3
+ "backend": "tokenizers",
4
+ "boa_token": "<|audio>",
5
+ "boi_token": "<|image>",
6
+ "bos_token": "<bos>",
7
+ "eoa_token": "<audio|>",
8
+ "eoc_token": "<channel|>",
9
+ "eoi_token": "<image|>",
10
+ "eos_token": "<eos>",
11
+ "eot_token": "<turn|>",
12
+ "escape_token": "<|\"|>",
13
+ "etc_token": "<tool_call|>",
14
+ "etd_token": "<tool|>",
15
+ "etr_token": "<tool_response|>",
16
+ "extra_special_tokens": [
17
+ "<|video|>"
18
+ ],
19
+ "image_token": "<|image|>",
20
+ "is_local": true,
21
+ "local_files_only": false,
22
+ "mask_token": "<mask>",
23
+ "model_max_length": 1000000000000000019884624838656,
24
+ "model_specific_special_tokens": {
25
+ "audio_token": "<|audio|>",
26
+ "boa_token": "<|audio>",
27
+ "boi_token": "<|image>",
28
+ "eoa_token": "<audio|>",
29
+ "eoc_token": "<channel|>",
30
+ "eoi_token": "<image|>",
31
+ "eot_token": "<turn|>",
32
+ "escape_token": "<|\"|>",
33
+ "etc_token": "<tool_call|>",
34
+ "etd_token": "<tool|>",
35
+ "etr_token": "<tool_response|>",
36
+ "image_token": "<|image|>",
37
+ "soc_token": "<|channel>",
38
+ "sot_token": "<|turn>",
39
+ "stc_token": "<|tool_call>",
40
+ "std_token": "<|tool>",
41
+ "str_token": "<|tool_response>",
42
+ "think_token": "<|think|>"
43
+ },
44
+ "pad_token": "<pad>",
45
+ "padding_side": "left",
46
+ "processor_class": "Gemma4Processor",
47
+ "response_schema": {
48
+ "properties": {
49
+ "content": {
50
+ "type": "string"
51
+ },
52
+ "role": {
53
+ "const": "assistant"
54
+ },
55
+ "thinking": {
56
+ "type": "string"
57
+ },
58
+ "tool_calls": {
59
+ "items": {
60
+ "properties": {
61
+ "function": {
62
+ "properties": {
63
+ "arguments": {
64
+ "additionalProperties": {},
65
+ "type": "object",
66
+ "x-parser": "gemma4-tool-call"
67
+ },
68
+ "name": {
69
+ "type": "string"
70
+ }
71
+ },
72
+ "type": "object",
73
+ "x-regex": "call\\:(?P<name>\\w+)(?P<arguments>\\{.*\\})"
74
+ },
75
+ "type": {
76
+ "const": "function"
77
+ }
78
+ },
79
+ "type": "object"
80
+ },
81
+ "type": "array",
82
+ "x-regex-iterator": "<\\|tool_call>(.*?)<tool_call\\|>"
83
+ }
84
+ },
85
+ "type": "object",
86
+ "x-regex": "(\\<\\|channel\\>thought\\n(?P<thinking>.*?)\\<channel\\|\\>)?(?P<tool_calls>\\<\\|tool_call\\>.*\\<tool_call\\|\\>)?(?P<content>(?:(?!\\<turn\\|\\>)(?!\\<\\|tool_response\\>).)+)?(?:\\<turn\\|\\>|\\<\\|tool_response\\>)?"
87
+ },
88
+ "soc_token": "<|channel>",
89
+ "sot_token": "<|turn>",
90
+ "stc_token": "<|tool_call>",
91
+ "std_token": "<|tool>",
92
+ "str_token": "<|tool_response>",
93
+ "think_token": "<|think|>",
94
+ "tokenizer_class": "GemmaTokenizer",
95
+ "unk_token": "<unk>"
96
+ }