Hugging Face's logo

facebook
/
pe-a-frame-large

Safetensors
pe_audio
Model card Files
xet
Community
1
eustlb HF Staff commited on
Commit
d054e74
·
verified ·
1 Parent(s): 97cb636

Upload model

Browse files
Files changed (2) hide show
  1. config.json +146 -32
  2. model.safetensors +2 -2
config.json CHANGED
@@ -1,35 +1,149 @@
1
  {
2
- "model_type": "pe_audio_frame",
3
- "audio_config": {
4
- "hidden_size": 1792,
5
- "intermediate_size": 4800,
6
- "num_hidden_layers": 28,
7
- "num_attention_heads": 14,
8
- "head_dim": 128,
9
- "num_key_value_heads": null,
10
- "hidden_act": "silu",
11
- "max_position_embeddings": 10000,
12
- "initializer_range": 0.02,
13
- "rms_norm_eps": 1e-05,
14
- "rope_parameters": {
15
- "rope_theta": 20000
16
- },
17
- "attention_bias": false,
18
- "attention_dropout": 0.0,
19
- "dac_config": {
20
- "encoder_hidden_size": 64,
21
- "downsampling_ratios": [
22
- 2,
23
- 8,
24
- 10,
25
- 12
26
- ],
27
- "decoder_hidden_size": 1536,
28
- "n_codebooks": 16,
29
- "codebook_size": 1024,
30
- "codebook_dim": 128,
31
- "quantizer_dropout": 0,
32
- "sampling_rate": 48000
33
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34
  }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
35
  }
 
1
  {
2
+ "architectures": [
3
+ "PeAudioFrameLevelModel"
4
+ ],
5
+ "audio_config": {
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "dac_config": {
9
+ "_name_or_path": "",
10
+ "add_cross_attention": false,
11
+ "architectures": null,
12
+ "bos_token_id": null,
13
+ "chunk_size_feed_forward": 0,
14
+ "codebook_dim": 128,
15
+ "codebook_loss_weight": 1.0,
16
+ "codebook_size": 1024,
17
+ "commitment_loss_weight": 0.25,
18
+ "cross_attention_hidden_size": null,
19
+ "decoder_hidden_size": 1536,
20
+ "decoder_start_token_id": null,
21
+ "downsampling_ratios": [
22
+ 2,
23
+ 8,
24
+ 10,
25
+ 12
26
+ ],
27
+ "dtype": null,
28
+ "encoder_hidden_size": 64,
29
+ "eos_token_id": null,
30
+ "finetuning_task": null,
31
+ "hidden_size": 1024,
32
+ "hop_length": 1920,
33
+ "id2label": {
34
+ "0": "LABEL_0",
35
+ "1": "LABEL_1"
36
+ },
37
+ "is_decoder": false,
38
+ "is_encoder_decoder": false,
39
+ "label2id": {
40
+ "LABEL_0": 0,
41
+ "LABEL_1": 1
42
+ },
43
+ "model_type": "dac",
44
+ "n_codebooks": 16,
45
+ "output_attentions": false,
46
+ "output_hidden_states": false,
47
+ "pad_token_id": null,
48
+ "prefix": null,
49
+ "problem_type": null,
50
+ "quantizer_dropout": 0,
51
+ "return_dict": true,
52
+ "sampling_rate": 48000,
53
+ "sep_token_id": null,
54
+ "task_specific_params": null,
55
+ "tie_encoder_decoder": false,
56
+ "tie_word_embeddings": true,
57
+ "tokenizer_class": null,
58
+ "upsampling_ratios": [
59
+ 12,
60
+ 10,
61
+ 8,
62
+ 2
63
+ ]
64
+ },
65
+ "head_dim": 128,
66
+ "hidden_act": "silu",
67
+ "hidden_size": 1792,
68
+ "initializer_range": 0.02,
69
+ "intermediate_size": 4800,
70
+ "max_position_embeddings": 10000,
71
+ "model_type": "pe_audio_encoder",
72
+ "num_attention_heads": 14,
73
+ "num_hidden_layers": 28,
74
+ "num_key_value_heads": 14,
75
+ "rms_norm_eps": 1e-05,
76
+ "rope_parameters": {
77
+ "rope_theta": 20000,
78
+ "rope_type": "default"
79
  }
80
+ },
81
+ "dtype": "float32",
82
+ "model_type": "pe_audio",
83
+ "text_config": {
84
+ "attention_bias": false,
85
+ "attention_dropout": 0.0,
86
+ "classifier_activation": "gelu",
87
+ "classifier_bias": false,
88
+ "classifier_dropout": 0.0,
89
+ "classifier_pooling": "cls",
90
+ "cls_token_id": 50281,
91
+ "decoder_bias": true,
92
+ "deterministic_flash_attn": false,
93
+ "embedding_dropout": 0.0,
94
+ "global_attn_every_n_layers": 3,
95
+ "hidden_activation": "gelu",
96
+ "hidden_size": 1024,
97
+ "initializer_cutoff_factor": 2.0,
98
+ "initializer_range": 0.02,
99
+ "intermediate_size": 2624,
100
+ "layer_types": [
101
+ "full_attention",
102
+ "sliding_attention",
103
+ "sliding_attention",
104
+ "full_attention",
105
+ "sliding_attention",
106
+ "sliding_attention",
107
+ "full_attention",
108
+ "sliding_attention",
109
+ "sliding_attention",
110
+ "full_attention",
111
+ "sliding_attention",
112
+ "sliding_attention",
113
+ "full_attention",
114
+ "sliding_attention",
115
+ "sliding_attention",
116
+ "full_attention",
117
+ "sliding_attention",
118
+ "sliding_attention",
119
+ "full_attention",
120
+ "sliding_attention",
121
+ "sliding_attention",
122
+ "full_attention"
123
+ ],
124
+ "local_attention": 128,
125
+ "max_position_embeddings": 8192,
126
+ "mlp_bias": false,
127
+ "mlp_dropout": 0.0,
128
+ "model_type": "modernbert",
129
+ "norm_bias": false,
130
+ "norm_eps": 1e-05,
131
+ "num_attention_heads": 16,
132
+ "num_hidden_layers": 22,
133
+ "repad_logits_with_grad": false,
134
+ "rope_parameters": {
135
+ "full_attention": {
136
+ "rope_theta": 160000.0,
137
+ "rope_type": "default"
138
+ },
139
+ "sliding_attention": {
140
+ "rope_theta": 10000.0,
141
+ "rope_type": "default"
142
+ }
143
+ },
144
+ "sparse_pred_ignore_index": -100,
145
+ "sparse_prediction": false,
146
+ "vocab_size": 50368
147
+ },
148
+ "transformers_version": "5.0.0.dev0"
149
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:09564eab252970a482be22fa119ddcc4f445baefcfd171becd73417548296558
3
- size 5826856520
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:779eb2e645bc986a3899d7b6195cd12b42ce57c0eb27b87494e336df746438db
3
+ size 5826331720