eustlb HF Staff commited on
Commit
1ad3c7d
·
verified ·
1 Parent(s): a2c6e79

Native LiquidCodecConfig (encoder/decoder sub-configs, huggingface/transformers#50393); weights untouched

Browse files
Files changed (1) hide show
  1. config.json +89 -26
config.json CHANGED
@@ -1,41 +1,104 @@
1
  {
2
- "transformers_version": "5.19.0.dev0",
3
  "architectures": [
4
  "LiquidCodecModel"
5
  ],
6
- "output_hidden_states": false,
7
- "return_dict": true,
8
- "dtype": "bfloat16",
9
- "chunk_size_feed_forward": 0,
10
- "is_encoder_decoder": false,
11
- "id2label": {
12
- "0": "LABEL_0",
13
- "1": "LABEL_1"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  },
15
- "label2id": {
16
- "LABEL_0": 0,
17
- "LABEL_1": 1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  },
19
- "problem_type": null,
20
- "encoder_dim": 512,
21
- "encoder_conv_dim": 256,
22
- "stft_n_fft": 1280,
23
- "stft_hop_length": 320,
24
- "stft_win_length": 1280,
25
- "stft_frame_stack": 6,
26
- "quantizer_dim": 256,
27
  "fsq_levels": [
28
  8,
29
  7,
30
  6,
31
  6
32
  ],
33
- "n_q": 12,
34
- "decoder_window": 30,
35
- "sampling_rate": 24000,
36
  "initializer_range": 0.02,
37
  "model_type": "liquid_codec",
38
- "sample_rate": 24000,
39
- "frame_samples": 1920,
40
- "output_attentions": false
 
41
  }
 
1
  {
 
2
  "architectures": [
3
  "LiquidCodecModel"
4
  ],
5
+ "decoder_config": {
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "conv_L_cache": 7,
9
+ "conv_bias": false,
10
+ "frame_stack": 6,
11
+ "head_dim": 64,
12
+ "hidden_size": 512,
13
+ "hop_length": 320,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 2304,
16
+ "layer_types": [
17
+ "conv",
18
+ "conv",
19
+ "sliding_attention",
20
+ "conv",
21
+ "sliding_attention",
22
+ "conv",
23
+ "sliding_attention",
24
+ "conv"
25
+ ],
26
+ "max_position_embeddings": 131072,
27
+ "model_type": "liquid_codec_decoder",
28
+ "n_fft": 1280,
29
+ "norm_eps": 1e-05,
30
+ "num_attention_heads": 8,
31
+ "num_hidden_layers": 8,
32
+ "num_key_value_heads": 2,
33
+ "quantizer_dim": 256,
34
+ "rope_parameters": {
35
+ "sliding_attention": {
36
+ "partial_rotary_factor": 1.0,
37
+ "rope_theta": 1000000.0,
38
+ "rope_type": "default"
39
+ }
40
+ },
41
+ "sliding_window": 30,
42
+ "use_attention_gate": true,
43
+ "use_cache": false,
44
+ "use_sandwich_norm": true,
45
+ "win_length": 1280
46
  },
47
+ "dtype": "bfloat16",
48
+ "encoder_config": {
49
+ "attention_bias": false,
50
+ "attention_dropout": 0.0,
51
+ "conv_L_cache": 7,
52
+ "conv_bias": false,
53
+ "conv_dim": 256,
54
+ "frame_stack": 6,
55
+ "head_dim": 64,
56
+ "hidden_act": "silu",
57
+ "hidden_size": 512,
58
+ "hop_length": 320,
59
+ "initializer_range": 0.02,
60
+ "intermediate_size": 2304,
61
+ "layer_types": [
62
+ "conv",
63
+ "conv",
64
+ "full_attention",
65
+ "conv",
66
+ "full_attention",
67
+ "conv",
68
+ "full_attention",
69
+ "conv"
70
+ ],
71
+ "max_position_embeddings": 131072,
72
+ "model_type": "liquid_codec_encoder",
73
+ "n_fft": 1280,
74
+ "norm_eps": 1e-05,
75
+ "num_attention_heads": 8,
76
+ "num_hidden_layers": 8,
77
+ "num_key_value_heads": 2,
78
+ "quantizer_dim": 256,
79
+ "rope_parameters": {
80
+ "full_attention": {
81
+ "partial_rotary_factor": 1.0,
82
+ "rope_theta": 1000000.0,
83
+ "rope_type": "default"
84
+ }
85
+ },
86
+ "sliding_window": null,
87
+ "use_attention_gate": true,
88
+ "use_cache": false,
89
+ "use_sandwich_norm": true,
90
+ "win_length": 1280
91
  },
 
 
 
 
 
 
 
 
92
  "fsq_levels": [
93
  8,
94
  7,
95
  6,
96
  6
97
  ],
 
 
 
98
  "initializer_range": 0.02,
99
  "model_type": "liquid_codec",
100
+ "n_q": 12,
101
+ "quantizer_dim": 256,
102
+ "sampling_rate": 24000,
103
+ "transformers_version": "5.23.0.dev0"
104
  }