config.json

{
  "architectures": [
    "Mapperatorinator"
  ],
  "backbone_config": {
    "_name_or_path": "openai/whisper-small",
    "activation_dropout": 0.0,
    "activation_function": "gelu",
    "apply_spec_augment": false,
    "architectures": [
      "WhisperForConditionalGeneration"
    ],
    "attention_bias": true,
    "attention_dropout": 0.0,
    "begin_suppress_tokens": null,
    "bos_token_id": 1,
    "classifier_proj_size": 256,
    "d_model": 768,
    "decoder_attention_heads": 12,
    "decoder_ffn_dim": 3072,
    "decoder_layerdrop": 0.0,
    "decoder_layers": 12,
    "decoder_start_token_id": 1,
    "deterministic_flash_attn": false,
    "dropout": 0.0,
    "dtype": "float32",
    "encoder_attention_heads": 12,
    "encoder_ffn_dim": 3072,
    "encoder_layerdrop": 0.0,
    "encoder_layers": 12,
    "eos_token_id": 2,
    "forced_decoder_ids": null,
    "global_attn_every_n_layers": 1,
    "global_rope_theta": 10000,
    "init_std": 0.02,
    "local_attention": 128,
    "local_rope_theta": 10000,
    "mask_feature_length": 10,
    "mask_feature_min_masks": 0,
    "mask_feature_prob": 0.0,
    "mask_time_length": 10,
    "mask_time_min_masks": 2,
    "mask_time_prob": 0.05,
    "max_length": 448,
    "max_source_positions": 1024,
    "max_target_positions": 2560,
    "median_filter_width": 7,
    "model_type": "whisper",
    "num_hidden_layers": 12,
    "num_mel_bins": 128,
    "pad_token_id": 0,
    "rope_scaling": {
      "factor": 1.0,
      "rope_type": "default"
    },
    "scale_embedding": false,
    "tie_word_embeddings": false,
    "use_cache": false,
    "use_weighted_layer_sum": false,
    "vocab_size": 4097
  },
  "backbone_model_name": "OliBomby/varwhisper-small",
  "bos_token_id": 1,
  "cond_dim": 128,
  "cond_size": 0,
  "decoder_start_token_id": 1,
  "disable_compile": true,
  "do_difficulty_embed": false,
  "do_mapper_embed": false,
  "do_sample": true,
  "do_song_position_embed": false,
  "do_style_embed": false,
  "dtype": "float32",
  "embed_decoder_input": true,
  "eos_token_id": 2,
  "f_max": 8000,
  "f_min": 20,
  "hidden_size": 768,
  "hop_length": 128,
  "init_std": 0.02,
  "input_features": true,
  "input_raw_wave": false,
  "is_encoder_decoder": true,
  "label_smoothing": 0.0,
  "max_length": null,
  "max_source_positions": 1024,
  "max_target_positions": 2560,
  "model_type": "mapperatorinator",
  "n_fft": 1024,
  "n_mels": 128,
  "num_attention_heads": 12,
  "num_classes": 0,
  "num_hidden_layers": 12,
  "num_mappers": 7605,
  "pad_mode": "reflect",
  "pad_token_id": 0,
  "project_encoder_input": false,
  "rhythm_token_end": 1647,
  "rhythm_token_start": 9,
  "rhythm_weight": 1.0,
  "sample_rate": 16000,
  "spectrogram_implementation": "torchaudio",
  "spectrogram_log_scale": true,
  "top_k": 0,
  "transformers_version": "4.57.3",
  "vocab_size": 4097,
  "vocab_size_in": 12505
}