{
  "architecture": "SemBridgeGenerator",
  "audio_end_token_id": 102,
  "audio_start_token_id": 101,
  "base_lm_layers": 24,
  "cfm": {
    "inference_cfg_rate": 2.0,
    "noise_cond_prob_range": [
      0.0,
      0.0
    ],
    "noise_cond_scale": 0.0,
    "ratio_r_neq_t_range": [
      0.25,
      0.75
    ],
    "reg_loss_type": "l1",
    "sigma_min": 1e-06,
    "solver": "euler",
    "t_scheduler": "log-norm",
    "training_cfg_rate": 0.1
  },
  "dim_model_base": 256,
  "dit_hidden_size": 1024,
  "dit_intermediate_size": 4096,
  "dit_kv_channels": 64,
  "dit_mean_mode": false,
  "dit_num_heads": 16,
  "dit_num_key_value_heads": 2,
  "dit_num_layers": 8,
  "encoder_hidden_size": 1024,
  "encoder_intermediate_size": 4096,
  "encoder_kv_channels": 64,
  "encoder_num_heads": 16,
  "encoder_num_key_value_heads": 2,
  "encoder_num_layers": 8,
  "feat_dim": 64,
  "hidden_size": 1024,
  "intermediate_size": 4096,
  "lm_kv_channels": 64,
  "max_length": 8192,
  "max_position_embeddings": 32768,
  "merged_residual_layers": 8,
  "num_attention_heads": 16,
  "num_hidden_layers": 32,
  "num_key_value_heads": 2,
  "patch_size": 2,
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "long_factor": [
      1.0004360675811768,
      1.0668443441390991,
      1.1631425619125366,
      1.3025742769241333,
      1.5040205717086792,
      1.7941505908966064,
      2.2101221084594727,
      2.802666664123535,
      3.6389970779418945,
      4.804192543029785,
      6.39855432510376,
      8.527148246765137,
      11.277542114257812,
      14.684998512268066,
      18.69317054748535,
      23.13019371032715,
      27.72362518310547,
      32.1606559753418,
      36.168827056884766,
      39.57627868652344,
      42.32667541503906,
      44.45526885986328,
      46.04962921142578,
      47.21482849121094,
      48.05115509033203,
      48.64370346069336,
      49.05967712402344,
      49.34980392456055,
      49.551246643066406,
      49.69068145751953,
      49.78697967529297,
      49.85338592529297
    ],
    "original_max_position_embeddings": 32768,
    "short_factor": [
      1.0004360675811768,
      1.0668443441390991,
      1.1631425619125366,
      1.3025742769241333,
      1.5040205717086792,
      1.7941505908966064,
      2.2101221084594727,
      2.802666664123535,
      3.6389970779418945,
      4.804192543029785,
      6.39855432510376,
      8.527148246765137,
      11.277542114257812,
      14.684998512268066,
      18.69317054748535,
      23.13019371032715,
      27.72362518310547,
      32.1606559753418,
      36.168827056884766,
      39.57627868652344,
      42.32667541503906,
      44.45526885986328,
      46.04962921142578,
      47.21482849121094,
      48.05115509033203,
      48.64370346069336,
      49.05967712402344,
      49.34980392456055,
      49.551246643066406,
      49.69068145751953,
      49.78697967529297,
      49.85338592529297
    ],
    "type": "longrope"
  },
  "rope_theta": 10000.0,
  "scale_depth": 1.4,
  "scale_emb": 12.0,
  "semantic_anchor_layer": 24,
  "semantic_head_enabled": true,
  "semantic_vocab_size": 16384,
  "singing_task_token_id": 74079,
  "speech_task_token_id": 74078,
  "text_vocab_size": 74080,
  "use_mup": false
}