config.json

{
  "card": 2048,
  "n_q": 32,
  "dep_q": 16,
  "delays": [
    0,
    0,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    0,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2,
    2
  ],
  "dim": 2560,
  "text_card": 48000,
  "existing_text_padding_id": 3,
  "num_heads": 20,
  "num_layers": 24,
  "hidden_scale": 4.125,
  "causal": true,
  "layer_scale": null,
  "context": 1500,
  "max_period": 100000.0,
  "gating": "silu",
  "norm": "rms_norm_f32",
  "positional_embedding": "rope",
  "depformer_dim": 1024,
  "depformer_num_heads": 16,
  "depformer_num_layers": 4,
  "depformer_dim_feedforward": 3072,
  "depformer_multi_linear": true,
  "depformer_pos_emb": "none",
  "depformer_weights_per_step": true,
  "depformer_low_rank_embeddings": 128,
  "conditioners": {
    "description": {
      "type": "lut",
      "lut": {
        "n_bins": 31,
        "dim": 16,
        "tokenizer": "noop",
        "possible_values": [
          "very_bad",
          "bad",
          "neutral",
          "good",
          "very_good"
        ]
      }
    }
  },
  "fuser": {
    "cross_attention_pos_emb": false,
    "cross_attention_pos_emb_scale": 1,
    "sum": [
      "description"
    ],
    "prepend": [],
    "cross": []
  },
  "cross_attention": [],
  "model_id": {
    "sig": "ccef4858",
    "epoch": 200
  },
  "depformer_weights_per_step_schedule": [
    0,
    1,
    2,
    3,
    4,
    5,
    6,
    7,
    8,
    8,
    8,
    8,
    8,
    8,
    8,
    8
  ],
  "model_type": "hibiki",
  "lm_gen_config": {
    "temp": 0.8,
    "temp_text": 0.8,
    "top_k": 250,
    "top_k_text": 50
  },
  "mimi_name": "mimi-pytorch-e351c8d8@125.safetensors",
  "tokenizer_name": "tokenizer_spm_48k_multi6_2.model",
  "moshi_name": "hibiki-pytorch-ccef4858@200.safetensors"
}