File size: 2,107 Bytes
64807c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
{
  "model_type": "haru",
  "architectures": [
    "HaruForConditionalGeneration"
  ],
  "dtype": "bfloat16",
  "sample_rate": 24000,
  "frame_samples": 3200,
  "frame_rate": 7.5,
  "latent_dim": 64,
  "pad_token_id": 248076,
  "eop_token_id": 248071,
  "lm": {
    "model_type": "qwen3_5_text",
    "vocab_size": 248320,
    "hidden_size": 32,
    "intermediate_size": 64,
    "num_hidden_layers": 8,
    "num_attention_heads": 4,
    "num_key_value_heads": 1,
    "head_dim": 16,
    "hidden_act": "silu",
    "max_position_embeddings": 2048,
    "rms_norm_eps": 1e-06,
    "attention_bias": false,
    "attention_dropout": 0.0,
    "attn_output_gate": true,
    "full_attention_interval": 4,
    "layer_types": [
      "linear_attention",
      "linear_attention",
      "linear_attention",
      "full_attention",
      "linear_attention",
      "linear_attention",
      "linear_attention",
      "full_attention"
    ],
    "linear_conv_kernel_dim": 4,
    "linear_key_head_dim": 16,
    "linear_value_head_dim": 16,
    "linear_num_key_heads": 2,
    "linear_num_value_heads": 4,
    "mamba_ssm_dtype": "float32",
    "partial_rotary_factor": 0.25,
    "rope_parameters": {
      "mrope_interleaved": true,
      "mrope_section": [
        11,
        11,
        10
      ],
      "partial_rotary_factor": 0.25,
      "rope_theta": 10000000,
      "rope_type": "default"
    },
    "tie_word_embeddings": true,
    "initializer_range": 0.02,
    "eos_token_id": 248044
  },
  "codec": {
    "latent_dim": 64,
    "base_channels": 1,
    "ratios": [
      8,
      5,
      5,
      4,
      2,
      2
    ],
    "encoder_depths": [
      1,
      1,
      1,
      1,
      1,
      1,
      1
    ],
    "decoder_depths": [
      1,
      1,
      1,
      1,
      1,
      1,
      1
    ],
    "norm_eps": 1e-05,
    "latent_std": 0.5
  },
  "head": {
    "hidden_size": 32,
    "latent_dim": 64,
    "num_layers": 1,
    "ffn_ratio": 3.0,
    "norm_eps": 1e-05,
    "interval_embed": 7.0,
    "num_steps": 1,
    "guidance_scale": 1.5,
    "guidance_range": [
      1.0,
      2.0
    ]
  }
}