Hugging Face's logo

facebook
/
pe-av-large

Safetensors
pe_audio_video
Model card Files
xet
Community
4
lematt1991 commited on
Commit
9c601f0
·
verified ·
1 Parent(s): 1326e4a

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +390 -180
  2. model.safetensors +2 -2
  3. preprocessor_config.json +7 -7
config.json CHANGED
@@ -1,7 +1,14 @@
1
  {
2
- "video_encoder": {
3
- "backbone": "PE-Core-L14-336",
4
- "backbone_checkpoint": null,
 
 
 
 
 
 
 
5
  "transformer": {
6
  "vocab_size": 151936,
7
  "max_position_embeddings": 10000,
@@ -90,86 +97,357 @@
90
  "use_bfloat16": false,
91
  "model_type": "qwen3",
92
  "output_attentions": false
93
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
94
  },
95
- "audio_codec": {
96
- "encoder_hidden_size": 64,
97
- "downsampling_ratios": [
98
- 2,
99
- 8,
100
- 10,
101
- 12
102
- ],
103
- "decoder_hidden_size": 1536,
104
- "n_codebooks": 16,
105
- "codebook_size": 1024,
106
- "codebook_dim": 128,
107
- "quantizer_dropout": 0,
108
- "sampling_rate": 48000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
109
  },
110
- "audio_encoder": {
111
- "vocab_size": 151936,
112
- "max_position_embeddings": 10000,
113
- "hidden_size": 1792,
114
- "intermediate_size": 4800,
115
- "num_hidden_layers": 28,
116
- "num_attention_heads": 14,
117
- "use_sliding_window": false,
118
- "sliding_window": null,
119
- "max_window_layers": 28,
120
- "num_key_value_heads": 14,
121
- "head_dim": 128,
122
- "hidden_act": "silu",
123
- "initializer_range": 0.02,
124
- "rms_norm_eps": 1e-05,
125
- "use_cache": true,
126
- "rope_theta": 20000,
127
- "rope_scaling": null,
128
- "attention_bias": false,
129
- "attention_dropout": 0.0,
130
- "layer_types": [
131
- "full_attention",
132
- "full_attention",
133
- "full_attention",
134
- "full_attention",
135
- "full_attention",
136
- "full_attention",
137
- "full_attention",
138
- "full_attention",
139
- "full_attention",
140
- "full_attention",
141
- "full_attention",
142
- "full_attention",
143
- "full_attention",
144
- "full_attention",
145
- "full_attention",
146
- "full_attention",
147
- "full_attention",
148
- "full_attention",
149
- "full_attention",
150
- "full_attention",
151
- "full_attention",
152
- "full_attention",
153
- "full_attention",
154
- "full_attention",
155
- "full_attention",
156
- "full_attention",
157
- "full_attention",
158
- "full_attention"
159
- ],
160
  "return_dict": true,
161
  "output_hidden_states": false,
162
  "torchscript": false,
163
- "dtype": null,
164
  "pruned_heads": {},
165
- "tie_word_embeddings": false,
166
  "chunk_size_feed_forward": 0,
167
  "is_encoder_decoder": false,
168
  "is_decoder": false,
169
  "cross_attention_hidden_size": null,
170
  "add_cross_attention": false,
171
  "tie_encoder_decoder": false,
172
- "architectures": null,
 
 
173
  "finetuning_task": null,
174
  "id2label": {
175
  "0": "LABEL_0",
@@ -183,10 +461,10 @@
183
  "problem_type": null,
184
  "tokenizer_class": null,
185
  "prefix": null,
186
- "bos_token_id": null,
187
- "pad_token_id": null,
188
- "eos_token_id": null,
189
- "sep_token_id": null,
190
  "decoder_start_token_id": null,
191
  "max_length": 20,
192
  "min_length": 0,
@@ -213,14 +491,47 @@
213
  "begin_suppress_tokens": null,
214
  "num_beam_groups": 1,
215
  "diversity_penalty": 0.0,
216
- "_name_or_path": "",
217
  "transformers_version": "4.57.0.dev0",
 
 
 
 
 
218
  "tf_legacy_loss": false,
219
  "use_bfloat16": false,
220
- "model_type": "qwen3",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
221
  "output_attentions": false
222
  },
223
- "audio_video_encoder": {
224
  "vocab_size": 151936,
225
  "max_position_embeddings": 10000,
226
  "hidden_size": 1792,
@@ -311,107 +622,6 @@
311
  "model_type": "qwen3",
312
  "output_attentions": false
313
  },
314
- "text_encoder": {
315
- "nth_layer": 22,
316
- "return_dict": true,
317
- "output_hidden_states": false,
318
- "torchscript": false,
319
- "dtype": "float32",
320
- "pruned_heads": {},
321
- "tie_word_embeddings": true,
322
- "chunk_size_feed_forward": 0,
323
- "is_encoder_decoder": false,
324
- "is_decoder": false,
325
- "cross_attention_hidden_size": null,
326
- "add_cross_attention": false,
327
- "tie_encoder_decoder": false,
328
- "architectures": [
329
- "ModernBertForMaskedLM"
330
- ],
331
- "finetuning_task": null,
332
- "id2label": {
333
- "0": "LABEL_0",
334
- "1": "LABEL_1"
335
- },
336
- "label2id": {
337
- "LABEL_0": 0,
338
- "LABEL_1": 1
339
- },
340
- "task_specific_params": null,
341
- "problem_type": null,
342
- "tokenizer_class": null,
343
- "prefix": null,
344
- "bos_token_id": 50281,
345
- "pad_token_id": 50283,
346
- "eos_token_id": 50282,
347
- "sep_token_id": 50282,
348
- "decoder_start_token_id": null,
349
- "max_length": 20,
350
- "min_length": 0,
351
- "do_sample": false,
352
- "early_stopping": false,
353
- "num_beams": 1,
354
- "temperature": 1.0,
355
- "top_k": 50,
356
- "top_p": 1.0,
357
- "typical_p": 1.0,
358
- "repetition_penalty": 1.0,
359
- "length_penalty": 1.0,
360
- "no_repeat_ngram_size": 0,
361
- "encoder_no_repeat_ngram_size": 0,
362
- "bad_words_ids": null,
363
- "num_return_sequences": 1,
364
- "output_scores": false,
365
- "return_dict_in_generate": false,
366
- "forced_bos_token_id": null,
367
- "forced_eos_token_id": null,
368
- "remove_invalid_values": false,
369
- "exponential_decay_length_penalty": null,
370
- "suppress_tokens": null,
371
- "begin_suppress_tokens": null,
372
- "num_beam_groups": 1,
373
- "diversity_penalty": 0.0,
374
- "_name_or_path": "answerdotai/ModernBERT-large",
375
- "cls_token_id": 50281,
376
- "gradient_checkpointing": false,
377
- "layer_norm_eps": 1e-05,
378
- "model_type": "modernbert",
379
- "position_embedding_type": "absolute",
380
- "tf_legacy_loss": false,
381
- "use_bfloat16": false,
382
- "vocab_size": 50368,
383
- "max_position_embeddings": 8192,
384
- "hidden_size": 1024,
385
- "intermediate_size": 2624,
386
- "num_hidden_layers": 28,
387
- "num_attention_heads": 16,
388
- "initializer_range": 0.02,
389
- "initializer_cutoff_factor": 2.0,
390
- "norm_eps": 1e-05,
391
- "norm_bias": false,
392
- "global_rope_theta": 160000.0,
393
- "attention_bias": false,
394
- "attention_dropout": 0.0,
395
- "hidden_activation": "gelu",
396
- "global_attn_every_n_layers": 3,
397
- "local_attention": 128,
398
- "local_rope_theta": 10000.0,
399
- "embedding_dropout": 0.0,
400
- "mlp_bias": false,
401
- "mlp_dropout": 0.0,
402
- "decoder_bias": true,
403
- "classifier_pooling": "mean",
404
- "classifier_dropout": 0.0,
405
- "classifier_bias": false,
406
- "classifier_activation": "gelu",
407
- "deterministic_flash_attn": false,
408
- "sparse_prediction": false,
409
- "sparse_pred_ignore_index": -100,
410
- "repad_logits_with_grad": false,
411
- "output_attentions": false
412
- },
413
- "separate_text_heads": true,
414
  "output_dim": 1024,
415
- "contrastive_head_norm_type": "none",
416
- "fixed_len_video": false
417
  }
 
1
  {
2
+ "video_model": {
3
+ "clip_vision_model": {
4
+ "architecture": "vit_pe_core_large_patch14_336",
5
+ "do_pooling": true,
6
+ "global_pool": "map",
7
+ "initializer_range": 0.02,
8
+ "model_args": {},
9
+ "num_labels": 1024,
10
+ "model_type": "timm_wrapper"
11
+ },
12
  "transformer": {
13
  "vocab_size": 151936,
14
  "max_position_embeddings": 10000,
 
97
  "use_bfloat16": false,
98
  "model_type": "qwen3",
99
  "output_attentions": false
100
+ },
101
+ "text_model": {
102
+ "return_dict": true,
103
+ "output_hidden_states": false,
104
+ "torchscript": false,
105
+ "dtype": "float32",
106
+ "pruned_heads": {},
107
+ "tie_word_embeddings": true,
108
+ "chunk_size_feed_forward": 0,
109
+ "is_encoder_decoder": false,
110
+ "is_decoder": false,
111
+ "cross_attention_hidden_size": null,
112
+ "add_cross_attention": false,
113
+ "tie_encoder_decoder": false,
114
+ "architectures": [
115
+ "ModernBertForMaskedLM"
116
+ ],
117
+ "finetuning_task": null,
118
+ "id2label": {
119
+ "0": "LABEL_0",
120
+ "1": "LABEL_1"
121
+ },
122
+ "label2id": {
123
+ "LABEL_0": 0,
124
+ "LABEL_1": 1
125
+ },
126
+ "task_specific_params": null,
127
+ "problem_type": null,
128
+ "tokenizer_class": null,
129
+ "prefix": null,
130
+ "bos_token_id": 50281,
131
+ "pad_token_id": 50283,
132
+ "eos_token_id": 50282,
133
+ "sep_token_id": 50282,
134
+ "decoder_start_token_id": null,
135
+ "max_length": 20,
136
+ "min_length": 0,
137
+ "do_sample": false,
138
+ "early_stopping": false,
139
+ "num_beams": 1,
140
+ "temperature": 1.0,
141
+ "top_k": 50,
142
+ "top_p": 1.0,
143
+ "typical_p": 1.0,
144
+ "repetition_penalty": 1.0,
145
+ "length_penalty": 1.0,
146
+ "no_repeat_ngram_size": 0,
147
+ "encoder_no_repeat_ngram_size": 0,
148
+ "bad_words_ids": null,
149
+ "num_return_sequences": 1,
150
+ "output_scores": false,
151
+ "return_dict_in_generate": false,
152
+ "forced_bos_token_id": null,
153
+ "forced_eos_token_id": null,
154
+ "remove_invalid_values": false,
155
+ "exponential_decay_length_penalty": null,
156
+ "suppress_tokens": null,
157
+ "begin_suppress_tokens": null,
158
+ "num_beam_groups": 1,
159
+ "diversity_penalty": 0.0,
160
+ "_name_or_path": "answerdotai/ModernBERT-large",
161
+ "transformers_version": "4.57.0.dev0",
162
+ "cls_token_id": 50281,
163
+ "gradient_checkpointing": false,
164
+ "layer_norm_eps": 1e-05,
165
+ "model_type": "modernbert",
166
+ "position_embedding_type": "absolute",
167
+ "tf_legacy_loss": false,
168
+ "use_bfloat16": false,
169
+ "vocab_size": 50368,
170
+ "max_position_embeddings": 8192,
171
+ "hidden_size": 1024,
172
+ "intermediate_size": 2624,
173
+ "num_hidden_layers": 28,
174
+ "num_attention_heads": 16,
175
+ "initializer_range": 0.02,
176
+ "initializer_cutoff_factor": 2.0,
177
+ "norm_eps": 1e-05,
178
+ "norm_bias": false,
179
+ "global_rope_theta": 160000.0,
180
+ "attention_bias": false,
181
+ "attention_dropout": 0.0,
182
+ "hidden_activation": "gelu",
183
+ "global_attn_every_n_layers": 3,
184
+ "local_attention": 128,
185
+ "local_rope_theta": 10000.0,
186
+ "embedding_dropout": 0.0,
187
+ "mlp_bias": false,
188
+ "mlp_dropout": 0.0,
189
+ "decoder_bias": true,
190
+ "classifier_pooling": "mean",
191
+ "classifier_dropout": 0.0,
192
+ "classifier_bias": false,
193
+ "classifier_activation": "gelu",
194
+ "deterministic_flash_attn": false,
195
+ "sparse_prediction": false,
196
+ "sparse_pred_ignore_index": -100,
197
+ "repad_logits_with_grad": false,
198
+ "output_attentions": false
199
+ },
200
+ "output_dim": 1024,
201
+ "fixed_len_video": false,
202
+ "nth_text_layer": 22
203
  },
204
+ "audio_model": {
205
+ "dac_vae_encoder": {
206
+ "encoder_hidden_size": 64,
207
+ "downsampling_ratios": [
208
+ 2,
209
+ 8,
210
+ 10,
211
+ 12
212
+ ],
213
+ "decoder_hidden_size": 1536,
214
+ "n_codebooks": 16,
215
+ "codebook_size": 1024,
216
+ "codebook_dim": 128,
217
+ "quantizer_dropout": 0,
218
+ "sampling_rate": 48000
219
+ },
220
+ "transformer": {
221
+ "vocab_size": 151936,
222
+ "max_position_embeddings": 10000,
223
+ "hidden_size": 1792,
224
+ "intermediate_size": 4800,
225
+ "num_hidden_layers": 28,
226
+ "num_attention_heads": 14,
227
+ "use_sliding_window": false,
228
+ "sliding_window": null,
229
+ "max_window_layers": 28,
230
+ "num_key_value_heads": 14,
231
+ "head_dim": 128,
232
+ "hidden_act": "silu",
233
+ "initializer_range": 0.02,
234
+ "rms_norm_eps": 1e-05,
235
+ "use_cache": true,
236
+ "rope_theta": 20000,
237
+ "rope_scaling": null,
238
+ "attention_bias": false,
239
+ "attention_dropout": 0.0,
240
+ "layer_types": [
241
+ "full_attention",
242
+ "full_attention",
243
+ "full_attention",
244
+ "full_attention",
245
+ "full_attention",
246
+ "full_attention",
247
+ "full_attention",
248
+ "full_attention",
249
+ "full_attention",
250
+ "full_attention",
251
+ "full_attention",
252
+ "full_attention",
253
+ "full_attention",
254
+ "full_attention",
255
+ "full_attention",
256
+ "full_attention",
257
+ "full_attention",
258
+ "full_attention",
259
+ "full_attention",
260
+ "full_attention",
261
+ "full_attention",
262
+ "full_attention",
263
+ "full_attention",
264
+ "full_attention",
265
+ "full_attention",
266
+ "full_attention",
267
+ "full_attention",
268
+ "full_attention"
269
+ ],
270
+ "return_dict": true,
271
+ "output_hidden_states": false,
272
+ "torchscript": false,
273
+ "dtype": null,
274
+ "pruned_heads": {},
275
+ "tie_word_embeddings": false,
276
+ "chunk_size_feed_forward": 0,
277
+ "is_encoder_decoder": false,
278
+ "is_decoder": false,
279
+ "cross_attention_hidden_size": null,
280
+ "add_cross_attention": false,
281
+ "tie_encoder_decoder": false,
282
+ "architectures": null,
283
+ "finetuning_task": null,
284
+ "id2label": {
285
+ "0": "LABEL_0",
286
+ "1": "LABEL_1"
287
+ },
288
+ "label2id": {
289
+ "LABEL_0": 0,
290
+ "LABEL_1": 1
291
+ },
292
+ "task_specific_params": null,
293
+ "problem_type": null,
294
+ "tokenizer_class": null,
295
+ "prefix": null,
296
+ "bos_token_id": null,
297
+ "pad_token_id": null,
298
+ "eos_token_id": null,
299
+ "sep_token_id": null,
300
+ "decoder_start_token_id": null,
301
+ "max_length": 20,
302
+ "min_length": 0,
303
+ "do_sample": false,
304
+ "early_stopping": false,
305
+ "num_beams": 1,
306
+ "temperature": 1.0,
307
+ "top_k": 50,
308
+ "top_p": 1.0,
309
+ "typical_p": 1.0,
310
+ "repetition_penalty": 1.0,
311
+ "length_penalty": 1.0,
312
+ "no_repeat_ngram_size": 0,
313
+ "encoder_no_repeat_ngram_size": 0,
314
+ "bad_words_ids": null,
315
+ "num_return_sequences": 1,
316
+ "output_scores": false,
317
+ "return_dict_in_generate": false,
318
+ "forced_bos_token_id": null,
319
+ "forced_eos_token_id": null,
320
+ "remove_invalid_values": false,
321
+ "exponential_decay_length_penalty": null,
322
+ "suppress_tokens": null,
323
+ "begin_suppress_tokens": null,
324
+ "num_beam_groups": 1,
325
+ "diversity_penalty": 0.0,
326
+ "_name_or_path": "",
327
+ "transformers_version": "4.57.0.dev0",
328
+ "tf_legacy_loss": false,
329
+ "use_bfloat16": false,
330
+ "model_type": "qwen3",
331
+ "output_attentions": false
332
+ },
333
+ "text_model": {
334
+ "return_dict": true,
335
+ "output_hidden_states": false,
336
+ "torchscript": false,
337
+ "dtype": "float32",
338
+ "pruned_heads": {},
339
+ "tie_word_embeddings": true,
340
+ "chunk_size_feed_forward": 0,
341
+ "is_encoder_decoder": false,
342
+ "is_decoder": false,
343
+ "cross_attention_hidden_size": null,
344
+ "add_cross_attention": false,
345
+ "tie_encoder_decoder": false,
346
+ "architectures": [
347
+ "ModernBertForMaskedLM"
348
+ ],
349
+ "finetuning_task": null,
350
+ "id2label": {
351
+ "0": "LABEL_0",
352
+ "1": "LABEL_1"
353
+ },
354
+ "label2id": {
355
+ "LABEL_0": 0,
356
+ "LABEL_1": 1
357
+ },
358
+ "task_specific_params": null,
359
+ "problem_type": null,
360
+ "tokenizer_class": null,
361
+ "prefix": null,
362
+ "bos_token_id": 50281,
363
+ "pad_token_id": 50283,
364
+ "eos_token_id": 50282,
365
+ "sep_token_id": 50282,
366
+ "decoder_start_token_id": null,
367
+ "max_length": 20,
368
+ "min_length": 0,
369
+ "do_sample": false,
370
+ "early_stopping": false,
371
+ "num_beams": 1,
372
+ "temperature": 1.0,
373
+ "top_k": 50,
374
+ "top_p": 1.0,
375
+ "typical_p": 1.0,
376
+ "repetition_penalty": 1.0,
377
+ "length_penalty": 1.0,
378
+ "no_repeat_ngram_size": 0,
379
+ "encoder_no_repeat_ngram_size": 0,
380
+ "bad_words_ids": null,
381
+ "num_return_sequences": 1,
382
+ "output_scores": false,
383
+ "return_dict_in_generate": false,
384
+ "forced_bos_token_id": null,
385
+ "forced_eos_token_id": null,
386
+ "remove_invalid_values": false,
387
+ "exponential_decay_length_penalty": null,
388
+ "suppress_tokens": null,
389
+ "begin_suppress_tokens": null,
390
+ "num_beam_groups": 1,
391
+ "diversity_penalty": 0.0,
392
+ "_name_or_path": "answerdotai/ModernBERT-large",
393
+ "transformers_version": "4.57.0.dev0",
394
+ "cls_token_id": 50281,
395
+ "gradient_checkpointing": false,
396
+ "layer_norm_eps": 1e-05,
397
+ "model_type": "modernbert",
398
+ "position_embedding_type": "absolute",
399
+ "tf_legacy_loss": false,
400
+ "use_bfloat16": false,
401
+ "vocab_size": 50368,
402
+ "max_position_embeddings": 8192,
403
+ "hidden_size": 1024,
404
+ "intermediate_size": 2624,
405
+ "num_hidden_layers": 28,
406
+ "num_attention_heads": 16,
407
+ "initializer_range": 0.02,
408
+ "initializer_cutoff_factor": 2.0,
409
+ "norm_eps": 1e-05,
410
+ "norm_bias": false,
411
+ "global_rope_theta": 160000.0,
412
+ "attention_bias": false,
413
+ "attention_dropout": 0.0,
414
+ "hidden_activation": "gelu",
415
+ "global_attn_every_n_layers": 3,
416
+ "local_attention": 128,
417
+ "local_rope_theta": 10000.0,
418
+ "embedding_dropout": 0.0,
419
+ "mlp_bias": false,
420
+ "mlp_dropout": 0.0,
421
+ "decoder_bias": true,
422
+ "classifier_pooling": "mean",
423
+ "classifier_dropout": 0.0,
424
+ "classifier_bias": false,
425
+ "classifier_activation": "gelu",
426
+ "deterministic_flash_attn": false,
427
+ "sparse_prediction": false,
428
+ "sparse_pred_ignore_index": -100,
429
+ "repad_logits_with_grad": false,
430
+ "output_attentions": false
431
+ },
432
+ "output_dim": 1024,
433
+ "nth_text_layer": 22
434
  },
435
+ "text_model": {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
436
  "return_dict": true,
437
  "output_hidden_states": false,
438
  "torchscript": false,
439
+ "dtype": "float32",
440
  "pruned_heads": {},
441
+ "tie_word_embeddings": true,
442
  "chunk_size_feed_forward": 0,
443
  "is_encoder_decoder": false,
444
  "is_decoder": false,
445
  "cross_attention_hidden_size": null,
446
  "add_cross_attention": false,
447
  "tie_encoder_decoder": false,
448
+ "architectures": [
449
+ "ModernBertForMaskedLM"
450
+ ],
451
  "finetuning_task": null,
452
  "id2label": {
453
  "0": "LABEL_0",
 
461
  "problem_type": null,
462
  "tokenizer_class": null,
463
  "prefix": null,
464
+ "bos_token_id": 50281,
465
+ "pad_token_id": 50283,
466
+ "eos_token_id": 50282,
467
+ "sep_token_id": 50282,
468
  "decoder_start_token_id": null,
469
  "max_length": 20,
470
  "min_length": 0,
 
491
  "begin_suppress_tokens": null,
492
  "num_beam_groups": 1,
493
  "diversity_penalty": 0.0,
494
+ "_name_or_path": "answerdotai/ModernBERT-large",
495
  "transformers_version": "4.57.0.dev0",
496
+ "cls_token_id": 50281,
497
+ "gradient_checkpointing": false,
498
+ "layer_norm_eps": 1e-05,
499
+ "model_type": "modernbert",
500
+ "position_embedding_type": "absolute",
501
  "tf_legacy_loss": false,
502
  "use_bfloat16": false,
503
+ "vocab_size": 50368,
504
+ "max_position_embeddings": 8192,
505
+ "hidden_size": 1024,
506
+ "intermediate_size": 2624,
507
+ "num_hidden_layers": 28,
508
+ "num_attention_heads": 16,
509
+ "initializer_range": 0.02,
510
+ "initializer_cutoff_factor": 2.0,
511
+ "norm_eps": 1e-05,
512
+ "norm_bias": false,
513
+ "global_rope_theta": 160000.0,
514
+ "attention_bias": false,
515
+ "attention_dropout": 0.0,
516
+ "hidden_activation": "gelu",
517
+ "global_attn_every_n_layers": 3,
518
+ "local_attention": 128,
519
+ "local_rope_theta": 10000.0,
520
+ "embedding_dropout": 0.0,
521
+ "mlp_bias": false,
522
+ "mlp_dropout": 0.0,
523
+ "decoder_bias": true,
524
+ "classifier_pooling": "mean",
525
+ "classifier_dropout": 0.0,
526
+ "classifier_bias": false,
527
+ "classifier_activation": "gelu",
528
+ "deterministic_flash_attn": false,
529
+ "sparse_prediction": false,
530
+ "sparse_pred_ignore_index": -100,
531
+ "repad_logits_with_grad": false,
532
  "output_attentions": false
533
  },
534
+ "transformer": {
535
  "vocab_size": 151936,
536
  "max_position_embeddings": 10000,
537
  "hidden_size": 1792,
 
622
  "model_type": "qwen3",
623
  "output_attentions": false
624
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
625
  "output_dim": 1024,
626
+ "nth_text_layer": 22
 
627
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:108e38c62cbeabcd56961ea85c217da733db84907f81b59f61598b2b90de3704
3
- size 10627032564
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da71d199c9222e045334d1bdd2ceba7a1bd57fa0072949f035f1d49971538d92
3
+ size 9234167264
preprocessor_config.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "feature_extractor_type": "PerceptionEncoderAVFeatureExtractor",
3
- "feature_size": 1,
4
- "hop_length": 1920,
5
- "padding_side": "right",
6
- "padding_value": 0.0,
7
- "return_attention_mask": true,
8
- "sampling_rate": 48000
9
  }
 
1
  {
2
+ "feature_extractor_type": "PerceptionEncoderAudioVideoFeatureExtractor",
3
+ "feature_size": 1,
4
+ "hop_length": 1920,
5
+ "padding_side": "right",
6
+ "padding_value": 0.0,
7
+ "return_attention_mask": true,
8
+ "sampling_rate": 48000
9
  }