Alkd commited on
Commit
3c30d1c
·
verified ·
1 Parent(s): 04ce3c4

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +1700 -0
config.json ADDED
@@ -0,0 +1,1700 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ASRForConditionalGeneration"
4
+ ],
5
+ "model_type": "qwen3_asr",
6
+ "support_languages": [
7
+ "Chinese",
8
+ "English",
9
+ "Cantonese",
10
+ "Arabic",
11
+ "German",
12
+ "French",
13
+ "Spanish",
14
+ "Portuguese",
15
+ "Indonesian",
16
+ "Italian",
17
+ "Korean",
18
+ "Russian",
19
+ "Thai",
20
+ "Vietnamese",
21
+ "Japanese",
22
+ "Turkish",
23
+ "Hindi",
24
+ "Malay",
25
+ "Dutch",
26
+ "Swedish",
27
+ "Danish",
28
+ "Finnish",
29
+ "Polish",
30
+ "Czech",
31
+ "Filipino",
32
+ "Persian",
33
+ "Greek",
34
+ "Romanian",
35
+ "Hungarian",
36
+ "Macedonian"
37
+ ],
38
+ "thinker_config": {
39
+ "model_type": "qwen3_asr",
40
+ "architectures": [
41
+ "Qwen3ASRForConditionalGeneration"
42
+ ],
43
+ "audio_config": {
44
+ "_name_or_path": "",
45
+ "activation_dropout": 0,
46
+ "activation_function": "gelu",
47
+ "add_cross_attention": false,
48
+ "architectures": null,
49
+ "attention_dropout": 0,
50
+ "bad_words_ids": null,
51
+ "begin_suppress_tokens": null,
52
+ "bos_token_id": null,
53
+ "chunk_size_feed_forward": 0,
54
+ "conv_chunksize": 500,
55
+ "cross_attention_hidden_size": null,
56
+ "d_model": 1024,
57
+ "decoder_start_token_id": null,
58
+ "diversity_penalty": 0.0,
59
+ "do_sample": false,
60
+ "downsample_hidden_size": 480,
61
+ "dropout": 0,
62
+ "dtype": null,
63
+ "early_stopping": false,
64
+ "encoder_attention_heads": 16,
65
+ "encoder_ffn_dim": 4096,
66
+ "encoder_layers": 24,
67
+ "encoder_no_repeat_ngram_size": 0,
68
+ "eos_token_id": null,
69
+ "exponential_decay_length_penalty": null,
70
+ "finetuning_task": null,
71
+ "forced_bos_token_id": null,
72
+ "forced_eos_token_id": null,
73
+ "id2label": {
74
+ "0": "LABEL_0",
75
+ "1": "LABEL_1"
76
+ },
77
+ "initializer_range": 0.02,
78
+ "is_decoder": false,
79
+ "is_encoder_decoder": false,
80
+ "label2id": {
81
+ "LABEL_0": 0,
82
+ "LABEL_1": 1
83
+ },
84
+ "length_penalty": 1.0,
85
+ "max_length": 20,
86
+ "max_source_positions": 1500,
87
+ "min_length": 0,
88
+ "model_type": "qwen3_asr_audio_encoder",
89
+ "n_window": 50,
90
+ "n_window_infer": 800,
91
+ "no_repeat_ngram_size": 0,
92
+ "num_beam_groups": 1,
93
+ "num_beams": 1,
94
+ "num_hidden_layers": 24,
95
+ "num_mel_bins": 128,
96
+ "num_return_sequences": 1,
97
+ "output_attentions": false,
98
+ "output_dim": 2048,
99
+ "output_hidden_states": false,
100
+ "output_scores": false,
101
+ "pad_token_id": null,
102
+ "prefix": null,
103
+ "problem_type": null,
104
+ "pruned_heads": {},
105
+ "remove_invalid_values": false,
106
+ "repetition_penalty": 1.0,
107
+ "return_dict": true,
108
+ "return_dict_in_generate": false,
109
+ "scale_embedding": false,
110
+ "sep_token_id": null,
111
+ "suppress_tokens": null,
112
+ "task_specific_params": null,
113
+ "temperature": 1.0,
114
+ "tf_legacy_loss": false,
115
+ "tie_encoder_decoder": false,
116
+ "tie_word_embeddings": true,
117
+ "tokenizer_class": null,
118
+ "top_k": 50,
119
+ "top_p": 1.0,
120
+ "torchscript": false,
121
+ "typical_p": 1.0,
122
+ "use_bfloat16": false
123
+ },
124
+ "audio_end_token_id": 151670,
125
+ "audio_start_token_id": 151669,
126
+ "audio_token_id": 151676,
127
+ "dtype": "bfloat16",
128
+ "initializer_range": 0.02,
129
+ "text_config": {
130
+ "_name_or_path": "",
131
+ "add_cross_attention": false,
132
+ "architectures": null,
133
+ "attention_bias": false,
134
+ "attention_dropout": 0.0,
135
+ "bad_words_ids": null,
136
+ "begin_suppress_tokens": null,
137
+ "bos_token_id": null,
138
+ "chunk_size_feed_forward": 0,
139
+ "cross_attention_hidden_size": null,
140
+ "decoder_start_token_id": null,
141
+ "diversity_penalty": 0.0,
142
+ "do_sample": false,
143
+ "dtype": null,
144
+ "early_stopping": false,
145
+ "encoder_no_repeat_ngram_size": 0,
146
+ "eos_token_id": null,
147
+ "exponential_decay_length_penalty": null,
148
+ "finetuning_task": null,
149
+ "forced_bos_token_id": null,
150
+ "forced_eos_token_id": null,
151
+ "head_dim": 128,
152
+ "hidden_act": "silu",
153
+ "hidden_size": 2048,
154
+ "id2label": {
155
+ "0": "LABEL_0",
156
+ "1": "LABEL_1"
157
+ },
158
+ "initializer_range": 0.02,
159
+ "intermediate_size": 6144,
160
+ "is_decoder": false,
161
+ "is_encoder_decoder": false,
162
+ "label2id": {
163
+ "LABEL_0": 0,
164
+ "LABEL_1": 1
165
+ },
166
+ "length_penalty": 1.0,
167
+ "max_length": 20,
168
+ "max_position_embeddings": 65536,
169
+ "min_length": 0,
170
+ "model_type": "qwen3",
171
+ "no_repeat_ngram_size": 0,
172
+ "num_attention_heads": 16,
173
+ "num_beam_groups": 1,
174
+ "num_beams": 1,
175
+ "num_hidden_layers": 28,
176
+ "num_key_value_heads": 8,
177
+ "num_return_sequences": 1,
178
+ "output_attentions": false,
179
+ "output_hidden_states": false,
180
+ "output_scores": false,
181
+ "pad_token_id": null,
182
+ "prefix": null,
183
+ "problem_type": null,
184
+ "pruned_heads": {},
185
+ "remove_invalid_values": false,
186
+ "repetition_penalty": 1.0,
187
+ "return_dict": true,
188
+ "return_dict_in_generate": false,
189
+ "rms_norm_eps": 1e-06,
190
+ "rope_scaling": {
191
+ "interleaved": true,
192
+ "mrope_interleaved": true,
193
+ "mrope_section": [
194
+ 24,
195
+ 20,
196
+ 20
197
+ ],
198
+ "rope_type": "default",
199
+ "type": "default"
200
+ },
201
+ "rope_theta": 1000000,
202
+ "sep_token_id": null,
203
+ "suppress_tokens": null,
204
+ "task_specific_params": null,
205
+ "temperature": 1.0,
206
+ "tf_legacy_loss": false,
207
+ "tie_encoder_decoder": false,
208
+ "tie_word_embeddings": true,
209
+ "tokenizer_class": null,
210
+ "top_k": 50,
211
+ "top_p": 1.0,
212
+ "torchscript": false,
213
+ "typical_p": 1.0,
214
+ "use_bfloat16": false,
215
+ "use_cache": true,
216
+ "vocab_size": 151936
217
+ }
218
+ },
219
+ "transformers_version": "4.57.6",
220
+ "quantization": {
221
+ "group_size": 64,
222
+ "bits": 4,
223
+ "mode": "affine",
224
+ "audio_tower.conv_out": {
225
+ "bits": 8,
226
+ "group_size": 64,
227
+ "mode": "affine"
228
+ },
229
+ "audio_tower.layers.0.fc1": {
230
+ "bits": 8,
231
+ "group_size": 64,
232
+ "mode": "affine"
233
+ },
234
+ "audio_tower.layers.0.fc2": {
235
+ "bits": 8,
236
+ "group_size": 64,
237
+ "mode": "affine"
238
+ },
239
+ "audio_tower.layers.0.self_attn.k_proj": {
240
+ "bits": 8,
241
+ "group_size": 64,
242
+ "mode": "affine"
243
+ },
244
+ "audio_tower.layers.0.self_attn.out_proj": {
245
+ "bits": 8,
246
+ "group_size": 64,
247
+ "mode": "affine"
248
+ },
249
+ "audio_tower.layers.0.self_attn.q_proj": {
250
+ "bits": 8,
251
+ "group_size": 64,
252
+ "mode": "affine"
253
+ },
254
+ "audio_tower.layers.0.self_attn.v_proj": {
255
+ "bits": 8,
256
+ "group_size": 64,
257
+ "mode": "affine"
258
+ },
259
+ "audio_tower.layers.1.fc1": {
260
+ "bits": 8,
261
+ "group_size": 64,
262
+ "mode": "affine"
263
+ },
264
+ "audio_tower.layers.1.fc2": {
265
+ "bits": 8,
266
+ "group_size": 64,
267
+ "mode": "affine"
268
+ },
269
+ "audio_tower.layers.1.self_attn.k_proj": {
270
+ "bits": 8,
271
+ "group_size": 64,
272
+ "mode": "affine"
273
+ },
274
+ "audio_tower.layers.1.self_attn.out_proj": {
275
+ "bits": 8,
276
+ "group_size": 64,
277
+ "mode": "affine"
278
+ },
279
+ "audio_tower.layers.1.self_attn.q_proj": {
280
+ "bits": 8,
281
+ "group_size": 64,
282
+ "mode": "affine"
283
+ },
284
+ "audio_tower.layers.1.self_attn.v_proj": {
285
+ "bits": 8,
286
+ "group_size": 64,
287
+ "mode": "affine"
288
+ },
289
+ "audio_tower.layers.10.fc1": {
290
+ "bits": 8,
291
+ "group_size": 64,
292
+ "mode": "affine"
293
+ },
294
+ "audio_tower.layers.10.fc2": {
295
+ "bits": 8,
296
+ "group_size": 64,
297
+ "mode": "affine"
298
+ },
299
+ "audio_tower.layers.10.self_attn.k_proj": {
300
+ "bits": 8,
301
+ "group_size": 64,
302
+ "mode": "affine"
303
+ },
304
+ "audio_tower.layers.10.self_attn.out_proj": {
305
+ "bits": 8,
306
+ "group_size": 64,
307
+ "mode": "affine"
308
+ },
309
+ "audio_tower.layers.10.self_attn.q_proj": {
310
+ "bits": 8,
311
+ "group_size": 64,
312
+ "mode": "affine"
313
+ },
314
+ "audio_tower.layers.10.self_attn.v_proj": {
315
+ "bits": 8,
316
+ "group_size": 64,
317
+ "mode": "affine"
318
+ },
319
+ "audio_tower.layers.11.fc1": {
320
+ "bits": 8,
321
+ "group_size": 64,
322
+ "mode": "affine"
323
+ },
324
+ "audio_tower.layers.11.fc2": {
325
+ "bits": 8,
326
+ "group_size": 64,
327
+ "mode": "affine"
328
+ },
329
+ "audio_tower.layers.11.self_attn.k_proj": {
330
+ "bits": 8,
331
+ "group_size": 64,
332
+ "mode": "affine"
333
+ },
334
+ "audio_tower.layers.11.self_attn.out_proj": {
335
+ "bits": 8,
336
+ "group_size": 64,
337
+ "mode": "affine"
338
+ },
339
+ "audio_tower.layers.11.self_attn.q_proj": {
340
+ "bits": 8,
341
+ "group_size": 64,
342
+ "mode": "affine"
343
+ },
344
+ "audio_tower.layers.11.self_attn.v_proj": {
345
+ "bits": 8,
346
+ "group_size": 64,
347
+ "mode": "affine"
348
+ },
349
+ "audio_tower.layers.12.fc1": {
350
+ "bits": 8,
351
+ "group_size": 64,
352
+ "mode": "affine"
353
+ },
354
+ "audio_tower.layers.12.fc2": {
355
+ "bits": 8,
356
+ "group_size": 64,
357
+ "mode": "affine"
358
+ },
359
+ "audio_tower.layers.12.self_attn.k_proj": {
360
+ "bits": 8,
361
+ "group_size": 64,
362
+ "mode": "affine"
363
+ },
364
+ "audio_tower.layers.12.self_attn.out_proj": {
365
+ "bits": 8,
366
+ "group_size": 64,
367
+ "mode": "affine"
368
+ },
369
+ "audio_tower.layers.12.self_attn.q_proj": {
370
+ "bits": 8,
371
+ "group_size": 64,
372
+ "mode": "affine"
373
+ },
374
+ "audio_tower.layers.12.self_attn.v_proj": {
375
+ "bits": 8,
376
+ "group_size": 64,
377
+ "mode": "affine"
378
+ },
379
+ "audio_tower.layers.13.fc1": {
380
+ "bits": 8,
381
+ "group_size": 64,
382
+ "mode": "affine"
383
+ },
384
+ "audio_tower.layers.13.fc2": {
385
+ "bits": 8,
386
+ "group_size": 64,
387
+ "mode": "affine"
388
+ },
389
+ "audio_tower.layers.13.self_attn.k_proj": {
390
+ "bits": 8,
391
+ "group_size": 64,
392
+ "mode": "affine"
393
+ },
394
+ "audio_tower.layers.13.self_attn.out_proj": {
395
+ "bits": 8,
396
+ "group_size": 64,
397
+ "mode": "affine"
398
+ },
399
+ "audio_tower.layers.13.self_attn.q_proj": {
400
+ "bits": 8,
401
+ "group_size": 64,
402
+ "mode": "affine"
403
+ },
404
+ "audio_tower.layers.13.self_attn.v_proj": {
405
+ "bits": 8,
406
+ "group_size": 64,
407
+ "mode": "affine"
408
+ },
409
+ "audio_tower.layers.14.fc1": {
410
+ "bits": 8,
411
+ "group_size": 64,
412
+ "mode": "affine"
413
+ },
414
+ "audio_tower.layers.14.fc2": {
415
+ "bits": 8,
416
+ "group_size": 64,
417
+ "mode": "affine"
418
+ },
419
+ "audio_tower.layers.14.self_attn.k_proj": {
420
+ "bits": 8,
421
+ "group_size": 64,
422
+ "mode": "affine"
423
+ },
424
+ "audio_tower.layers.14.self_attn.out_proj": {
425
+ "bits": 8,
426
+ "group_size": 64,
427
+ "mode": "affine"
428
+ },
429
+ "audio_tower.layers.14.self_attn.q_proj": {
430
+ "bits": 8,
431
+ "group_size": 64,
432
+ "mode": "affine"
433
+ },
434
+ "audio_tower.layers.14.self_attn.v_proj": {
435
+ "bits": 8,
436
+ "group_size": 64,
437
+ "mode": "affine"
438
+ },
439
+ "audio_tower.layers.15.fc1": {
440
+ "bits": 8,
441
+ "group_size": 64,
442
+ "mode": "affine"
443
+ },
444
+ "audio_tower.layers.15.fc2": {
445
+ "bits": 8,
446
+ "group_size": 64,
447
+ "mode": "affine"
448
+ },
449
+ "audio_tower.layers.15.self_attn.k_proj": {
450
+ "bits": 8,
451
+ "group_size": 64,
452
+ "mode": "affine"
453
+ },
454
+ "audio_tower.layers.15.self_attn.out_proj": {
455
+ "bits": 8,
456
+ "group_size": 64,
457
+ "mode": "affine"
458
+ },
459
+ "audio_tower.layers.15.self_attn.q_proj": {
460
+ "bits": 8,
461
+ "group_size": 64,
462
+ "mode": "affine"
463
+ },
464
+ "audio_tower.layers.15.self_attn.v_proj": {
465
+ "bits": 8,
466
+ "group_size": 64,
467
+ "mode": "affine"
468
+ },
469
+ "audio_tower.layers.16.fc1": {
470
+ "bits": 8,
471
+ "group_size": 64,
472
+ "mode": "affine"
473
+ },
474
+ "audio_tower.layers.16.fc2": {
475
+ "bits": 8,
476
+ "group_size": 64,
477
+ "mode": "affine"
478
+ },
479
+ "audio_tower.layers.16.self_attn.k_proj": {
480
+ "bits": 8,
481
+ "group_size": 64,
482
+ "mode": "affine"
483
+ },
484
+ "audio_tower.layers.16.self_attn.out_proj": {
485
+ "bits": 8,
486
+ "group_size": 64,
487
+ "mode": "affine"
488
+ },
489
+ "audio_tower.layers.16.self_attn.q_proj": {
490
+ "bits": 8,
491
+ "group_size": 64,
492
+ "mode": "affine"
493
+ },
494
+ "audio_tower.layers.16.self_attn.v_proj": {
495
+ "bits": 8,
496
+ "group_size": 64,
497
+ "mode": "affine"
498
+ },
499
+ "audio_tower.layers.17.fc1": {
500
+ "bits": 8,
501
+ "group_size": 64,
502
+ "mode": "affine"
503
+ },
504
+ "audio_tower.layers.17.fc2": {
505
+ "bits": 8,
506
+ "group_size": 64,
507
+ "mode": "affine"
508
+ },
509
+ "audio_tower.layers.17.self_attn.k_proj": {
510
+ "bits": 8,
511
+ "group_size": 64,
512
+ "mode": "affine"
513
+ },
514
+ "audio_tower.layers.17.self_attn.out_proj": {
515
+ "bits": 8,
516
+ "group_size": 64,
517
+ "mode": "affine"
518
+ },
519
+ "audio_tower.layers.17.self_attn.q_proj": {
520
+ "bits": 8,
521
+ "group_size": 64,
522
+ "mode": "affine"
523
+ },
524
+ "audio_tower.layers.17.self_attn.v_proj": {
525
+ "bits": 8,
526
+ "group_size": 64,
527
+ "mode": "affine"
528
+ },
529
+ "audio_tower.layers.18.fc1": {
530
+ "bits": 8,
531
+ "group_size": 64,
532
+ "mode": "affine"
533
+ },
534
+ "audio_tower.layers.18.fc2": {
535
+ "bits": 8,
536
+ "group_size": 64,
537
+ "mode": "affine"
538
+ },
539
+ "audio_tower.layers.18.self_attn.k_proj": {
540
+ "bits": 8,
541
+ "group_size": 64,
542
+ "mode": "affine"
543
+ },
544
+ "audio_tower.layers.18.self_attn.out_proj": {
545
+ "bits": 8,
546
+ "group_size": 64,
547
+ "mode": "affine"
548
+ },
549
+ "audio_tower.layers.18.self_attn.q_proj": {
550
+ "bits": 8,
551
+ "group_size": 64,
552
+ "mode": "affine"
553
+ },
554
+ "audio_tower.layers.18.self_attn.v_proj": {
555
+ "bits": 8,
556
+ "group_size": 64,
557
+ "mode": "affine"
558
+ },
559
+ "audio_tower.layers.19.fc1": {
560
+ "bits": 8,
561
+ "group_size": 64,
562
+ "mode": "affine"
563
+ },
564
+ "audio_tower.layers.19.fc2": {
565
+ "bits": 8,
566
+ "group_size": 64,
567
+ "mode": "affine"
568
+ },
569
+ "audio_tower.layers.19.self_attn.k_proj": {
570
+ "bits": 8,
571
+ "group_size": 64,
572
+ "mode": "affine"
573
+ },
574
+ "audio_tower.layers.19.self_attn.out_proj": {
575
+ "bits": 8,
576
+ "group_size": 64,
577
+ "mode": "affine"
578
+ },
579
+ "audio_tower.layers.19.self_attn.q_proj": {
580
+ "bits": 8,
581
+ "group_size": 64,
582
+ "mode": "affine"
583
+ },
584
+ "audio_tower.layers.19.self_attn.v_proj": {
585
+ "bits": 8,
586
+ "group_size": 64,
587
+ "mode": "affine"
588
+ },
589
+ "audio_tower.layers.2.fc1": {
590
+ "bits": 8,
591
+ "group_size": 64,
592
+ "mode": "affine"
593
+ },
594
+ "audio_tower.layers.2.fc2": {
595
+ "bits": 8,
596
+ "group_size": 64,
597
+ "mode": "affine"
598
+ },
599
+ "audio_tower.layers.2.self_attn.k_proj": {
600
+ "bits": 8,
601
+ "group_size": 64,
602
+ "mode": "affine"
603
+ },
604
+ "audio_tower.layers.2.self_attn.out_proj": {
605
+ "bits": 8,
606
+ "group_size": 64,
607
+ "mode": "affine"
608
+ },
609
+ "audio_tower.layers.2.self_attn.q_proj": {
610
+ "bits": 8,
611
+ "group_size": 64,
612
+ "mode": "affine"
613
+ },
614
+ "audio_tower.layers.2.self_attn.v_proj": {
615
+ "bits": 8,
616
+ "group_size": 64,
617
+ "mode": "affine"
618
+ },
619
+ "audio_tower.layers.20.fc1": {
620
+ "bits": 8,
621
+ "group_size": 64,
622
+ "mode": "affine"
623
+ },
624
+ "audio_tower.layers.20.fc2": {
625
+ "bits": 8,
626
+ "group_size": 64,
627
+ "mode": "affine"
628
+ },
629
+ "audio_tower.layers.20.self_attn.k_proj": {
630
+ "bits": 8,
631
+ "group_size": 64,
632
+ "mode": "affine"
633
+ },
634
+ "audio_tower.layers.20.self_attn.out_proj": {
635
+ "bits": 8,
636
+ "group_size": 64,
637
+ "mode": "affine"
638
+ },
639
+ "audio_tower.layers.20.self_attn.q_proj": {
640
+ "bits": 8,
641
+ "group_size": 64,
642
+ "mode": "affine"
643
+ },
644
+ "audio_tower.layers.20.self_attn.v_proj": {
645
+ "bits": 8,
646
+ "group_size": 64,
647
+ "mode": "affine"
648
+ },
649
+ "audio_tower.layers.21.fc1": {
650
+ "bits": 8,
651
+ "group_size": 64,
652
+ "mode": "affine"
653
+ },
654
+ "audio_tower.layers.21.fc2": {
655
+ "bits": 8,
656
+ "group_size": 64,
657
+ "mode": "affine"
658
+ },
659
+ "audio_tower.layers.21.self_attn.k_proj": {
660
+ "bits": 8,
661
+ "group_size": 64,
662
+ "mode": "affine"
663
+ },
664
+ "audio_tower.layers.21.self_attn.out_proj": {
665
+ "bits": 8,
666
+ "group_size": 64,
667
+ "mode": "affine"
668
+ },
669
+ "audio_tower.layers.21.self_attn.q_proj": {
670
+ "bits": 8,
671
+ "group_size": 64,
672
+ "mode": "affine"
673
+ },
674
+ "audio_tower.layers.21.self_attn.v_proj": {
675
+ "bits": 8,
676
+ "group_size": 64,
677
+ "mode": "affine"
678
+ },
679
+ "audio_tower.layers.22.fc1": {
680
+ "bits": 8,
681
+ "group_size": 64,
682
+ "mode": "affine"
683
+ },
684
+ "audio_tower.layers.22.fc2": {
685
+ "bits": 8,
686
+ "group_size": 64,
687
+ "mode": "affine"
688
+ },
689
+ "audio_tower.layers.22.self_attn.k_proj": {
690
+ "bits": 8,
691
+ "group_size": 64,
692
+ "mode": "affine"
693
+ },
694
+ "audio_tower.layers.22.self_attn.out_proj": {
695
+ "bits": 8,
696
+ "group_size": 64,
697
+ "mode": "affine"
698
+ },
699
+ "audio_tower.layers.22.self_attn.q_proj": {
700
+ "bits": 8,
701
+ "group_size": 64,
702
+ "mode": "affine"
703
+ },
704
+ "audio_tower.layers.22.self_attn.v_proj": {
705
+ "bits": 8,
706
+ "group_size": 64,
707
+ "mode": "affine"
708
+ },
709
+ "audio_tower.layers.23.fc1": {
710
+ "bits": 8,
711
+ "group_size": 64,
712
+ "mode": "affine"
713
+ },
714
+ "audio_tower.layers.23.fc2": {
715
+ "bits": 8,
716
+ "group_size": 64,
717
+ "mode": "affine"
718
+ },
719
+ "audio_tower.layers.23.self_attn.k_proj": {
720
+ "bits": 8,
721
+ "group_size": 64,
722
+ "mode": "affine"
723
+ },
724
+ "audio_tower.layers.23.self_attn.out_proj": {
725
+ "bits": 8,
726
+ "group_size": 64,
727
+ "mode": "affine"
728
+ },
729
+ "audio_tower.layers.23.self_attn.q_proj": {
730
+ "bits": 8,
731
+ "group_size": 64,
732
+ "mode": "affine"
733
+ },
734
+ "audio_tower.layers.23.self_attn.v_proj": {
735
+ "bits": 8,
736
+ "group_size": 64,
737
+ "mode": "affine"
738
+ },
739
+ "audio_tower.layers.3.fc1": {
740
+ "bits": 8,
741
+ "group_size": 64,
742
+ "mode": "affine"
743
+ },
744
+ "audio_tower.layers.3.fc2": {
745
+ "bits": 8,
746
+ "group_size": 64,
747
+ "mode": "affine"
748
+ },
749
+ "audio_tower.layers.3.self_attn.k_proj": {
750
+ "bits": 8,
751
+ "group_size": 64,
752
+ "mode": "affine"
753
+ },
754
+ "audio_tower.layers.3.self_attn.out_proj": {
755
+ "bits": 8,
756
+ "group_size": 64,
757
+ "mode": "affine"
758
+ },
759
+ "audio_tower.layers.3.self_attn.q_proj": {
760
+ "bits": 8,
761
+ "group_size": 64,
762
+ "mode": "affine"
763
+ },
764
+ "audio_tower.layers.3.self_attn.v_proj": {
765
+ "bits": 8,
766
+ "group_size": 64,
767
+ "mode": "affine"
768
+ },
769
+ "audio_tower.layers.4.fc1": {
770
+ "bits": 8,
771
+ "group_size": 64,
772
+ "mode": "affine"
773
+ },
774
+ "audio_tower.layers.4.fc2": {
775
+ "bits": 8,
776
+ "group_size": 64,
777
+ "mode": "affine"
778
+ },
779
+ "audio_tower.layers.4.self_attn.k_proj": {
780
+ "bits": 8,
781
+ "group_size": 64,
782
+ "mode": "affine"
783
+ },
784
+ "audio_tower.layers.4.self_attn.out_proj": {
785
+ "bits": 8,
786
+ "group_size": 64,
787
+ "mode": "affine"
788
+ },
789
+ "audio_tower.layers.4.self_attn.q_proj": {
790
+ "bits": 8,
791
+ "group_size": 64,
792
+ "mode": "affine"
793
+ },
794
+ "audio_tower.layers.4.self_attn.v_proj": {
795
+ "bits": 8,
796
+ "group_size": 64,
797
+ "mode": "affine"
798
+ },
799
+ "audio_tower.layers.5.fc1": {
800
+ "bits": 8,
801
+ "group_size": 64,
802
+ "mode": "affine"
803
+ },
804
+ "audio_tower.layers.5.fc2": {
805
+ "bits": 8,
806
+ "group_size": 64,
807
+ "mode": "affine"
808
+ },
809
+ "audio_tower.layers.5.self_attn.k_proj": {
810
+ "bits": 8,
811
+ "group_size": 64,
812
+ "mode": "affine"
813
+ },
814
+ "audio_tower.layers.5.self_attn.out_proj": {
815
+ "bits": 8,
816
+ "group_size": 64,
817
+ "mode": "affine"
818
+ },
819
+ "audio_tower.layers.5.self_attn.q_proj": {
820
+ "bits": 8,
821
+ "group_size": 64,
822
+ "mode": "affine"
823
+ },
824
+ "audio_tower.layers.5.self_attn.v_proj": {
825
+ "bits": 8,
826
+ "group_size": 64,
827
+ "mode": "affine"
828
+ },
829
+ "audio_tower.layers.6.fc1": {
830
+ "bits": 8,
831
+ "group_size": 64,
832
+ "mode": "affine"
833
+ },
834
+ "audio_tower.layers.6.fc2": {
835
+ "bits": 8,
836
+ "group_size": 64,
837
+ "mode": "affine"
838
+ },
839
+ "audio_tower.layers.6.self_attn.k_proj": {
840
+ "bits": 8,
841
+ "group_size": 64,
842
+ "mode": "affine"
843
+ },
844
+ "audio_tower.layers.6.self_attn.out_proj": {
845
+ "bits": 8,
846
+ "group_size": 64,
847
+ "mode": "affine"
848
+ },
849
+ "audio_tower.layers.6.self_attn.q_proj": {
850
+ "bits": 8,
851
+ "group_size": 64,
852
+ "mode": "affine"
853
+ },
854
+ "audio_tower.layers.6.self_attn.v_proj": {
855
+ "bits": 8,
856
+ "group_size": 64,
857
+ "mode": "affine"
858
+ },
859
+ "audio_tower.layers.7.fc1": {
860
+ "bits": 8,
861
+ "group_size": 64,
862
+ "mode": "affine"
863
+ },
864
+ "audio_tower.layers.7.fc2": {
865
+ "bits": 8,
866
+ "group_size": 64,
867
+ "mode": "affine"
868
+ },
869
+ "audio_tower.layers.7.self_attn.k_proj": {
870
+ "bits": 8,
871
+ "group_size": 64,
872
+ "mode": "affine"
873
+ },
874
+ "audio_tower.layers.7.self_attn.out_proj": {
875
+ "bits": 8,
876
+ "group_size": 64,
877
+ "mode": "affine"
878
+ },
879
+ "audio_tower.layers.7.self_attn.q_proj": {
880
+ "bits": 8,
881
+ "group_size": 64,
882
+ "mode": "affine"
883
+ },
884
+ "audio_tower.layers.7.self_attn.v_proj": {
885
+ "bits": 8,
886
+ "group_size": 64,
887
+ "mode": "affine"
888
+ },
889
+ "audio_tower.layers.8.fc1": {
890
+ "bits": 8,
891
+ "group_size": 64,
892
+ "mode": "affine"
893
+ },
894
+ "audio_tower.layers.8.fc2": {
895
+ "bits": 8,
896
+ "group_size": 64,
897
+ "mode": "affine"
898
+ },
899
+ "audio_tower.layers.8.self_attn.k_proj": {
900
+ "bits": 8,
901
+ "group_size": 64,
902
+ "mode": "affine"
903
+ },
904
+ "audio_tower.layers.8.self_attn.out_proj": {
905
+ "bits": 8,
906
+ "group_size": 64,
907
+ "mode": "affine"
908
+ },
909
+ "audio_tower.layers.8.self_attn.q_proj": {
910
+ "bits": 8,
911
+ "group_size": 64,
912
+ "mode": "affine"
913
+ },
914
+ "audio_tower.layers.8.self_attn.v_proj": {
915
+ "bits": 8,
916
+ "group_size": 64,
917
+ "mode": "affine"
918
+ },
919
+ "audio_tower.layers.9.fc1": {
920
+ "bits": 8,
921
+ "group_size": 64,
922
+ "mode": "affine"
923
+ },
924
+ "audio_tower.layers.9.fc2": {
925
+ "bits": 8,
926
+ "group_size": 64,
927
+ "mode": "affine"
928
+ },
929
+ "audio_tower.layers.9.self_attn.k_proj": {
930
+ "bits": 8,
931
+ "group_size": 64,
932
+ "mode": "affine"
933
+ },
934
+ "audio_tower.layers.9.self_attn.out_proj": {
935
+ "bits": 8,
936
+ "group_size": 64,
937
+ "mode": "affine"
938
+ },
939
+ "audio_tower.layers.9.self_attn.q_proj": {
940
+ "bits": 8,
941
+ "group_size": 64,
942
+ "mode": "affine"
943
+ },
944
+ "audio_tower.layers.9.self_attn.v_proj": {
945
+ "bits": 8,
946
+ "group_size": 64,
947
+ "mode": "affine"
948
+ },
949
+ "audio_tower.proj1": {
950
+ "bits": 8,
951
+ "group_size": 64,
952
+ "mode": "affine"
953
+ },
954
+ "audio_tower.proj2": {
955
+ "bits": 8,
956
+ "group_size": 64,
957
+ "mode": "affine"
958
+ }
959
+ },
960
+ "quantization_config": {
961
+ "group_size": 64,
962
+ "bits": 4,
963
+ "mode": "affine",
964
+ "audio_tower.conv_out": {
965
+ "bits": 8,
966
+ "group_size": 64,
967
+ "mode": "affine"
968
+ },
969
+ "audio_tower.layers.0.fc1": {
970
+ "bits": 8,
971
+ "group_size": 64,
972
+ "mode": "affine"
973
+ },
974
+ "audio_tower.layers.0.fc2": {
975
+ "bits": 8,
976
+ "group_size": 64,
977
+ "mode": "affine"
978
+ },
979
+ "audio_tower.layers.0.self_attn.k_proj": {
980
+ "bits": 8,
981
+ "group_size": 64,
982
+ "mode": "affine"
983
+ },
984
+ "audio_tower.layers.0.self_attn.out_proj": {
985
+ "bits": 8,
986
+ "group_size": 64,
987
+ "mode": "affine"
988
+ },
989
+ "audio_tower.layers.0.self_attn.q_proj": {
990
+ "bits": 8,
991
+ "group_size": 64,
992
+ "mode": "affine"
993
+ },
994
+ "audio_tower.layers.0.self_attn.v_proj": {
995
+ "bits": 8,
996
+ "group_size": 64,
997
+ "mode": "affine"
998
+ },
999
+ "audio_tower.layers.1.fc1": {
1000
+ "bits": 8,
1001
+ "group_size": 64,
1002
+ "mode": "affine"
1003
+ },
1004
+ "audio_tower.layers.1.fc2": {
1005
+ "bits": 8,
1006
+ "group_size": 64,
1007
+ "mode": "affine"
1008
+ },
1009
+ "audio_tower.layers.1.self_attn.k_proj": {
1010
+ "bits": 8,
1011
+ "group_size": 64,
1012
+ "mode": "affine"
1013
+ },
1014
+ "audio_tower.layers.1.self_attn.out_proj": {
1015
+ "bits": 8,
1016
+ "group_size": 64,
1017
+ "mode": "affine"
1018
+ },
1019
+ "audio_tower.layers.1.self_attn.q_proj": {
1020
+ "bits": 8,
1021
+ "group_size": 64,
1022
+ "mode": "affine"
1023
+ },
1024
+ "audio_tower.layers.1.self_attn.v_proj": {
1025
+ "bits": 8,
1026
+ "group_size": 64,
1027
+ "mode": "affine"
1028
+ },
1029
+ "audio_tower.layers.10.fc1": {
1030
+ "bits": 8,
1031
+ "group_size": 64,
1032
+ "mode": "affine"
1033
+ },
1034
+ "audio_tower.layers.10.fc2": {
1035
+ "bits": 8,
1036
+ "group_size": 64,
1037
+ "mode": "affine"
1038
+ },
1039
+ "audio_tower.layers.10.self_attn.k_proj": {
1040
+ "bits": 8,
1041
+ "group_size": 64,
1042
+ "mode": "affine"
1043
+ },
1044
+ "audio_tower.layers.10.self_attn.out_proj": {
1045
+ "bits": 8,
1046
+ "group_size": 64,
1047
+ "mode": "affine"
1048
+ },
1049
+ "audio_tower.layers.10.self_attn.q_proj": {
1050
+ "bits": 8,
1051
+ "group_size": 64,
1052
+ "mode": "affine"
1053
+ },
1054
+ "audio_tower.layers.10.self_attn.v_proj": {
1055
+ "bits": 8,
1056
+ "group_size": 64,
1057
+ "mode": "affine"
1058
+ },
1059
+ "audio_tower.layers.11.fc1": {
1060
+ "bits": 8,
1061
+ "group_size": 64,
1062
+ "mode": "affine"
1063
+ },
1064
+ "audio_tower.layers.11.fc2": {
1065
+ "bits": 8,
1066
+ "group_size": 64,
1067
+ "mode": "affine"
1068
+ },
1069
+ "audio_tower.layers.11.self_attn.k_proj": {
1070
+ "bits": 8,
1071
+ "group_size": 64,
1072
+ "mode": "affine"
1073
+ },
1074
+ "audio_tower.layers.11.self_attn.out_proj": {
1075
+ "bits": 8,
1076
+ "group_size": 64,
1077
+ "mode": "affine"
1078
+ },
1079
+ "audio_tower.layers.11.self_attn.q_proj": {
1080
+ "bits": 8,
1081
+ "group_size": 64,
1082
+ "mode": "affine"
1083
+ },
1084
+ "audio_tower.layers.11.self_attn.v_proj": {
1085
+ "bits": 8,
1086
+ "group_size": 64,
1087
+ "mode": "affine"
1088
+ },
1089
+ "audio_tower.layers.12.fc1": {
1090
+ "bits": 8,
1091
+ "group_size": 64,
1092
+ "mode": "affine"
1093
+ },
1094
+ "audio_tower.layers.12.fc2": {
1095
+ "bits": 8,
1096
+ "group_size": 64,
1097
+ "mode": "affine"
1098
+ },
1099
+ "audio_tower.layers.12.self_attn.k_proj": {
1100
+ "bits": 8,
1101
+ "group_size": 64,
1102
+ "mode": "affine"
1103
+ },
1104
+ "audio_tower.layers.12.self_attn.out_proj": {
1105
+ "bits": 8,
1106
+ "group_size": 64,
1107
+ "mode": "affine"
1108
+ },
1109
+ "audio_tower.layers.12.self_attn.q_proj": {
1110
+ "bits": 8,
1111
+ "group_size": 64,
1112
+ "mode": "affine"
1113
+ },
1114
+ "audio_tower.layers.12.self_attn.v_proj": {
1115
+ "bits": 8,
1116
+ "group_size": 64,
1117
+ "mode": "affine"
1118
+ },
1119
+ "audio_tower.layers.13.fc1": {
1120
+ "bits": 8,
1121
+ "group_size": 64,
1122
+ "mode": "affine"
1123
+ },
1124
+ "audio_tower.layers.13.fc2": {
1125
+ "bits": 8,
1126
+ "group_size": 64,
1127
+ "mode": "affine"
1128
+ },
1129
+ "audio_tower.layers.13.self_attn.k_proj": {
1130
+ "bits": 8,
1131
+ "group_size": 64,
1132
+ "mode": "affine"
1133
+ },
1134
+ "audio_tower.layers.13.self_attn.out_proj": {
1135
+ "bits": 8,
1136
+ "group_size": 64,
1137
+ "mode": "affine"
1138
+ },
1139
+ "audio_tower.layers.13.self_attn.q_proj": {
1140
+ "bits": 8,
1141
+ "group_size": 64,
1142
+ "mode": "affine"
1143
+ },
1144
+ "audio_tower.layers.13.self_attn.v_proj": {
1145
+ "bits": 8,
1146
+ "group_size": 64,
1147
+ "mode": "affine"
1148
+ },
1149
+ "audio_tower.layers.14.fc1": {
1150
+ "bits": 8,
1151
+ "group_size": 64,
1152
+ "mode": "affine"
1153
+ },
1154
+ "audio_tower.layers.14.fc2": {
1155
+ "bits": 8,
1156
+ "group_size": 64,
1157
+ "mode": "affine"
1158
+ },
1159
+ "audio_tower.layers.14.self_attn.k_proj": {
1160
+ "bits": 8,
1161
+ "group_size": 64,
1162
+ "mode": "affine"
1163
+ },
1164
+ "audio_tower.layers.14.self_attn.out_proj": {
1165
+ "bits": 8,
1166
+ "group_size": 64,
1167
+ "mode": "affine"
1168
+ },
1169
+ "audio_tower.layers.14.self_attn.q_proj": {
1170
+ "bits": 8,
1171
+ "group_size": 64,
1172
+ "mode": "affine"
1173
+ },
1174
+ "audio_tower.layers.14.self_attn.v_proj": {
1175
+ "bits": 8,
1176
+ "group_size": 64,
1177
+ "mode": "affine"
1178
+ },
1179
+ "audio_tower.layers.15.fc1": {
1180
+ "bits": 8,
1181
+ "group_size": 64,
1182
+ "mode": "affine"
1183
+ },
1184
+ "audio_tower.layers.15.fc2": {
1185
+ "bits": 8,
1186
+ "group_size": 64,
1187
+ "mode": "affine"
1188
+ },
1189
+ "audio_tower.layers.15.self_attn.k_proj": {
1190
+ "bits": 8,
1191
+ "group_size": 64,
1192
+ "mode": "affine"
1193
+ },
1194
+ "audio_tower.layers.15.self_attn.out_proj": {
1195
+ "bits": 8,
1196
+ "group_size": 64,
1197
+ "mode": "affine"
1198
+ },
1199
+ "audio_tower.layers.15.self_attn.q_proj": {
1200
+ "bits": 8,
1201
+ "group_size": 64,
1202
+ "mode": "affine"
1203
+ },
1204
+ "audio_tower.layers.15.self_attn.v_proj": {
1205
+ "bits": 8,
1206
+ "group_size": 64,
1207
+ "mode": "affine"
1208
+ },
1209
+ "audio_tower.layers.16.fc1": {
1210
+ "bits": 8,
1211
+ "group_size": 64,
1212
+ "mode": "affine"
1213
+ },
1214
+ "audio_tower.layers.16.fc2": {
1215
+ "bits": 8,
1216
+ "group_size": 64,
1217
+ "mode": "affine"
1218
+ },
1219
+ "audio_tower.layers.16.self_attn.k_proj": {
1220
+ "bits": 8,
1221
+ "group_size": 64,
1222
+ "mode": "affine"
1223
+ },
1224
+ "audio_tower.layers.16.self_attn.out_proj": {
1225
+ "bits": 8,
1226
+ "group_size": 64,
1227
+ "mode": "affine"
1228
+ },
1229
+ "audio_tower.layers.16.self_attn.q_proj": {
1230
+ "bits": 8,
1231
+ "group_size": 64,
1232
+ "mode": "affine"
1233
+ },
1234
+ "audio_tower.layers.16.self_attn.v_proj": {
1235
+ "bits": 8,
1236
+ "group_size": 64,
1237
+ "mode": "affine"
1238
+ },
1239
+ "audio_tower.layers.17.fc1": {
1240
+ "bits": 8,
1241
+ "group_size": 64,
1242
+ "mode": "affine"
1243
+ },
1244
+ "audio_tower.layers.17.fc2": {
1245
+ "bits": 8,
1246
+ "group_size": 64,
1247
+ "mode": "affine"
1248
+ },
1249
+ "audio_tower.layers.17.self_attn.k_proj": {
1250
+ "bits": 8,
1251
+ "group_size": 64,
1252
+ "mode": "affine"
1253
+ },
1254
+ "audio_tower.layers.17.self_attn.out_proj": {
1255
+ "bits": 8,
1256
+ "group_size": 64,
1257
+ "mode": "affine"
1258
+ },
1259
+ "audio_tower.layers.17.self_attn.q_proj": {
1260
+ "bits": 8,
1261
+ "group_size": 64,
1262
+ "mode": "affine"
1263
+ },
1264
+ "audio_tower.layers.17.self_attn.v_proj": {
1265
+ "bits": 8,
1266
+ "group_size": 64,
1267
+ "mode": "affine"
1268
+ },
1269
+ "audio_tower.layers.18.fc1": {
1270
+ "bits": 8,
1271
+ "group_size": 64,
1272
+ "mode": "affine"
1273
+ },
1274
+ "audio_tower.layers.18.fc2": {
1275
+ "bits": 8,
1276
+ "group_size": 64,
1277
+ "mode": "affine"
1278
+ },
1279
+ "audio_tower.layers.18.self_attn.k_proj": {
1280
+ "bits": 8,
1281
+ "group_size": 64,
1282
+ "mode": "affine"
1283
+ },
1284
+ "audio_tower.layers.18.self_attn.out_proj": {
1285
+ "bits": 8,
1286
+ "group_size": 64,
1287
+ "mode": "affine"
1288
+ },
1289
+ "audio_tower.layers.18.self_attn.q_proj": {
1290
+ "bits": 8,
1291
+ "group_size": 64,
1292
+ "mode": "affine"
1293
+ },
1294
+ "audio_tower.layers.18.self_attn.v_proj": {
1295
+ "bits": 8,
1296
+ "group_size": 64,
1297
+ "mode": "affine"
1298
+ },
1299
+ "audio_tower.layers.19.fc1": {
1300
+ "bits": 8,
1301
+ "group_size": 64,
1302
+ "mode": "affine"
1303
+ },
1304
+ "audio_tower.layers.19.fc2": {
1305
+ "bits": 8,
1306
+ "group_size": 64,
1307
+ "mode": "affine"
1308
+ },
1309
+ "audio_tower.layers.19.self_attn.k_proj": {
1310
+ "bits": 8,
1311
+ "group_size": 64,
1312
+ "mode": "affine"
1313
+ },
1314
+ "audio_tower.layers.19.self_attn.out_proj": {
1315
+ "bits": 8,
1316
+ "group_size": 64,
1317
+ "mode": "affine"
1318
+ },
1319
+ "audio_tower.layers.19.self_attn.q_proj": {
1320
+ "bits": 8,
1321
+ "group_size": 64,
1322
+ "mode": "affine"
1323
+ },
1324
+ "audio_tower.layers.19.self_attn.v_proj": {
1325
+ "bits": 8,
1326
+ "group_size": 64,
1327
+ "mode": "affine"
1328
+ },
1329
+ "audio_tower.layers.2.fc1": {
1330
+ "bits": 8,
1331
+ "group_size": 64,
1332
+ "mode": "affine"
1333
+ },
1334
+ "audio_tower.layers.2.fc2": {
1335
+ "bits": 8,
1336
+ "group_size": 64,
1337
+ "mode": "affine"
1338
+ },
1339
+ "audio_tower.layers.2.self_attn.k_proj": {
1340
+ "bits": 8,
1341
+ "group_size": 64,
1342
+ "mode": "affine"
1343
+ },
1344
+ "audio_tower.layers.2.self_attn.out_proj": {
1345
+ "bits": 8,
1346
+ "group_size": 64,
1347
+ "mode": "affine"
1348
+ },
1349
+ "audio_tower.layers.2.self_attn.q_proj": {
1350
+ "bits": 8,
1351
+ "group_size": 64,
1352
+ "mode": "affine"
1353
+ },
1354
+ "audio_tower.layers.2.self_attn.v_proj": {
1355
+ "bits": 8,
1356
+ "group_size": 64,
1357
+ "mode": "affine"
1358
+ },
1359
+ "audio_tower.layers.20.fc1": {
1360
+ "bits": 8,
1361
+ "group_size": 64,
1362
+ "mode": "affine"
1363
+ },
1364
+ "audio_tower.layers.20.fc2": {
1365
+ "bits": 8,
1366
+ "group_size": 64,
1367
+ "mode": "affine"
1368
+ },
1369
+ "audio_tower.layers.20.self_attn.k_proj": {
1370
+ "bits": 8,
1371
+ "group_size": 64,
1372
+ "mode": "affine"
1373
+ },
1374
+ "audio_tower.layers.20.self_attn.out_proj": {
1375
+ "bits": 8,
1376
+ "group_size": 64,
1377
+ "mode": "affine"
1378
+ },
1379
+ "audio_tower.layers.20.self_attn.q_proj": {
1380
+ "bits": 8,
1381
+ "group_size": 64,
1382
+ "mode": "affine"
1383
+ },
1384
+ "audio_tower.layers.20.self_attn.v_proj": {
1385
+ "bits": 8,
1386
+ "group_size": 64,
1387
+ "mode": "affine"
1388
+ },
1389
+ "audio_tower.layers.21.fc1": {
1390
+ "bits": 8,
1391
+ "group_size": 64,
1392
+ "mode": "affine"
1393
+ },
1394
+ "audio_tower.layers.21.fc2": {
1395
+ "bits": 8,
1396
+ "group_size": 64,
1397
+ "mode": "affine"
1398
+ },
1399
+ "audio_tower.layers.21.self_attn.k_proj": {
1400
+ "bits": 8,
1401
+ "group_size": 64,
1402
+ "mode": "affine"
1403
+ },
1404
+ "audio_tower.layers.21.self_attn.out_proj": {
1405
+ "bits": 8,
1406
+ "group_size": 64,
1407
+ "mode": "affine"
1408
+ },
1409
+ "audio_tower.layers.21.self_attn.q_proj": {
1410
+ "bits": 8,
1411
+ "group_size": 64,
1412
+ "mode": "affine"
1413
+ },
1414
+ "audio_tower.layers.21.self_attn.v_proj": {
1415
+ "bits": 8,
1416
+ "group_size": 64,
1417
+ "mode": "affine"
1418
+ },
1419
+ "audio_tower.layers.22.fc1": {
1420
+ "bits": 8,
1421
+ "group_size": 64,
1422
+ "mode": "affine"
1423
+ },
1424
+ "audio_tower.layers.22.fc2": {
1425
+ "bits": 8,
1426
+ "group_size": 64,
1427
+ "mode": "affine"
1428
+ },
1429
+ "audio_tower.layers.22.self_attn.k_proj": {
1430
+ "bits": 8,
1431
+ "group_size": 64,
1432
+ "mode": "affine"
1433
+ },
1434
+ "audio_tower.layers.22.self_attn.out_proj": {
1435
+ "bits": 8,
1436
+ "group_size": 64,
1437
+ "mode": "affine"
1438
+ },
1439
+ "audio_tower.layers.22.self_attn.q_proj": {
1440
+ "bits": 8,
1441
+ "group_size": 64,
1442
+ "mode": "affine"
1443
+ },
1444
+ "audio_tower.layers.22.self_attn.v_proj": {
1445
+ "bits": 8,
1446
+ "group_size": 64,
1447
+ "mode": "affine"
1448
+ },
1449
+ "audio_tower.layers.23.fc1": {
1450
+ "bits": 8,
1451
+ "group_size": 64,
1452
+ "mode": "affine"
1453
+ },
1454
+ "audio_tower.layers.23.fc2": {
1455
+ "bits": 8,
1456
+ "group_size": 64,
1457
+ "mode": "affine"
1458
+ },
1459
+ "audio_tower.layers.23.self_attn.k_proj": {
1460
+ "bits": 8,
1461
+ "group_size": 64,
1462
+ "mode": "affine"
1463
+ },
1464
+ "audio_tower.layers.23.self_attn.out_proj": {
1465
+ "bits": 8,
1466
+ "group_size": 64,
1467
+ "mode": "affine"
1468
+ },
1469
+ "audio_tower.layers.23.self_attn.q_proj": {
1470
+ "bits": 8,
1471
+ "group_size": 64,
1472
+ "mode": "affine"
1473
+ },
1474
+ "audio_tower.layers.23.self_attn.v_proj": {
1475
+ "bits": 8,
1476
+ "group_size": 64,
1477
+ "mode": "affine"
1478
+ },
1479
+ "audio_tower.layers.3.fc1": {
1480
+ "bits": 8,
1481
+ "group_size": 64,
1482
+ "mode": "affine"
1483
+ },
1484
+ "audio_tower.layers.3.fc2": {
1485
+ "bits": 8,
1486
+ "group_size": 64,
1487
+ "mode": "affine"
1488
+ },
1489
+ "audio_tower.layers.3.self_attn.k_proj": {
1490
+ "bits": 8,
1491
+ "group_size": 64,
1492
+ "mode": "affine"
1493
+ },
1494
+ "audio_tower.layers.3.self_attn.out_proj": {
1495
+ "bits": 8,
1496
+ "group_size": 64,
1497
+ "mode": "affine"
1498
+ },
1499
+ "audio_tower.layers.3.self_attn.q_proj": {
1500
+ "bits": 8,
1501
+ "group_size": 64,
1502
+ "mode": "affine"
1503
+ },
1504
+ "audio_tower.layers.3.self_attn.v_proj": {
1505
+ "bits": 8,
1506
+ "group_size": 64,
1507
+ "mode": "affine"
1508
+ },
1509
+ "audio_tower.layers.4.fc1": {
1510
+ "bits": 8,
1511
+ "group_size": 64,
1512
+ "mode": "affine"
1513
+ },
1514
+ "audio_tower.layers.4.fc2": {
1515
+ "bits": 8,
1516
+ "group_size": 64,
1517
+ "mode": "affine"
1518
+ },
1519
+ "audio_tower.layers.4.self_attn.k_proj": {
1520
+ "bits": 8,
1521
+ "group_size": 64,
1522
+ "mode": "affine"
1523
+ },
1524
+ "audio_tower.layers.4.self_attn.out_proj": {
1525
+ "bits": 8,
1526
+ "group_size": 64,
1527
+ "mode": "affine"
1528
+ },
1529
+ "audio_tower.layers.4.self_attn.q_proj": {
1530
+ "bits": 8,
1531
+ "group_size": 64,
1532
+ "mode": "affine"
1533
+ },
1534
+ "audio_tower.layers.4.self_attn.v_proj": {
1535
+ "bits": 8,
1536
+ "group_size": 64,
1537
+ "mode": "affine"
1538
+ },
1539
+ "audio_tower.layers.5.fc1": {
1540
+ "bits": 8,
1541
+ "group_size": 64,
1542
+ "mode": "affine"
1543
+ },
1544
+ "audio_tower.layers.5.fc2": {
1545
+ "bits": 8,
1546
+ "group_size": 64,
1547
+ "mode": "affine"
1548
+ },
1549
+ "audio_tower.layers.5.self_attn.k_proj": {
1550
+ "bits": 8,
1551
+ "group_size": 64,
1552
+ "mode": "affine"
1553
+ },
1554
+ "audio_tower.layers.5.self_attn.out_proj": {
1555
+ "bits": 8,
1556
+ "group_size": 64,
1557
+ "mode": "affine"
1558
+ },
1559
+ "audio_tower.layers.5.self_attn.q_proj": {
1560
+ "bits": 8,
1561
+ "group_size": 64,
1562
+ "mode": "affine"
1563
+ },
1564
+ "audio_tower.layers.5.self_attn.v_proj": {
1565
+ "bits": 8,
1566
+ "group_size": 64,
1567
+ "mode": "affine"
1568
+ },
1569
+ "audio_tower.layers.6.fc1": {
1570
+ "bits": 8,
1571
+ "group_size": 64,
1572
+ "mode": "affine"
1573
+ },
1574
+ "audio_tower.layers.6.fc2": {
1575
+ "bits": 8,
1576
+ "group_size": 64,
1577
+ "mode": "affine"
1578
+ },
1579
+ "audio_tower.layers.6.self_attn.k_proj": {
1580
+ "bits": 8,
1581
+ "group_size": 64,
1582
+ "mode": "affine"
1583
+ },
1584
+ "audio_tower.layers.6.self_attn.out_proj": {
1585
+ "bits": 8,
1586
+ "group_size": 64,
1587
+ "mode": "affine"
1588
+ },
1589
+ "audio_tower.layers.6.self_attn.q_proj": {
1590
+ "bits": 8,
1591
+ "group_size": 64,
1592
+ "mode": "affine"
1593
+ },
1594
+ "audio_tower.layers.6.self_attn.v_proj": {
1595
+ "bits": 8,
1596
+ "group_size": 64,
1597
+ "mode": "affine"
1598
+ },
1599
+ "audio_tower.layers.7.fc1": {
1600
+ "bits": 8,
1601
+ "group_size": 64,
1602
+ "mode": "affine"
1603
+ },
1604
+ "audio_tower.layers.7.fc2": {
1605
+ "bits": 8,
1606
+ "group_size": 64,
1607
+ "mode": "affine"
1608
+ },
1609
+ "audio_tower.layers.7.self_attn.k_proj": {
1610
+ "bits": 8,
1611
+ "group_size": 64,
1612
+ "mode": "affine"
1613
+ },
1614
+ "audio_tower.layers.7.self_attn.out_proj": {
1615
+ "bits": 8,
1616
+ "group_size": 64,
1617
+ "mode": "affine"
1618
+ },
1619
+ "audio_tower.layers.7.self_attn.q_proj": {
1620
+ "bits": 8,
1621
+ "group_size": 64,
1622
+ "mode": "affine"
1623
+ },
1624
+ "audio_tower.layers.7.self_attn.v_proj": {
1625
+ "bits": 8,
1626
+ "group_size": 64,
1627
+ "mode": "affine"
1628
+ },
1629
+ "audio_tower.layers.8.fc1": {
1630
+ "bits": 8,
1631
+ "group_size": 64,
1632
+ "mode": "affine"
1633
+ },
1634
+ "audio_tower.layers.8.fc2": {
1635
+ "bits": 8,
1636
+ "group_size": 64,
1637
+ "mode": "affine"
1638
+ },
1639
+ "audio_tower.layers.8.self_attn.k_proj": {
1640
+ "bits": 8,
1641
+ "group_size": 64,
1642
+ "mode": "affine"
1643
+ },
1644
+ "audio_tower.layers.8.self_attn.out_proj": {
1645
+ "bits": 8,
1646
+ "group_size": 64,
1647
+ "mode": "affine"
1648
+ },
1649
+ "audio_tower.layers.8.self_attn.q_proj": {
1650
+ "bits": 8,
1651
+ "group_size": 64,
1652
+ "mode": "affine"
1653
+ },
1654
+ "audio_tower.layers.8.self_attn.v_proj": {
1655
+ "bits": 8,
1656
+ "group_size": 64,
1657
+ "mode": "affine"
1658
+ },
1659
+ "audio_tower.layers.9.fc1": {
1660
+ "bits": 8,
1661
+ "group_size": 64,
1662
+ "mode": "affine"
1663
+ },
1664
+ "audio_tower.layers.9.fc2": {
1665
+ "bits": 8,
1666
+ "group_size": 64,
1667
+ "mode": "affine"
1668
+ },
1669
+ "audio_tower.layers.9.self_attn.k_proj": {
1670
+ "bits": 8,
1671
+ "group_size": 64,
1672
+ "mode": "affine"
1673
+ },
1674
+ "audio_tower.layers.9.self_attn.out_proj": {
1675
+ "bits": 8,
1676
+ "group_size": 64,
1677
+ "mode": "affine"
1678
+ },
1679
+ "audio_tower.layers.9.self_attn.q_proj": {
1680
+ "bits": 8,
1681
+ "group_size": 64,
1682
+ "mode": "affine"
1683
+ },
1684
+ "audio_tower.layers.9.self_attn.v_proj": {
1685
+ "bits": 8,
1686
+ "group_size": 64,
1687
+ "mode": "affine"
1688
+ },
1689
+ "audio_tower.proj1": {
1690
+ "bits": 8,
1691
+ "group_size": 64,
1692
+ "mode": "affine"
1693
+ },
1694
+ "audio_tower.proj2": {
1695
+ "bits": 8,
1696
+ "group_size": 64,
1697
+ "mode": "affine"
1698
+ }
1699
+ }
1700
+ }