config.json

{
  "_name_or_path": "openai/whisper-large-v2",
  "activation_dropout": 0.0,
  "activation_function": "gelu",
  "apply_spec_augment": false,
  "architectures": [
    "WhisperForConditionalGeneration"
  ],
  "attention_dropout": 0.0,
  "begin_suppress_tokens": [
    186,
    49639
  ],
  "bos_token_id": 49639,
  "classifier_proj_size": 256,
  "d_model": 1280,
  "decoder_attention_heads": 20,
  "decoder_ffn_dim": 5120,
  "decoder_layerdrop": 0.0,
  "decoder_layers": 32,
  "decoder_start_token_id": 49640,
  "dropout": 0.0,
  "encoder_attention_heads": 20,
  "encoder_ffn_dim": 5120,
  "encoder_layerdrop": 0.0,
  "encoder_layers": 32,
  "eos_token_id": 49639,
  "forced_decoder_ids": [
    [
      1,
      49641
    ],
    [
      49741
    ],
    [
      49745
    ]
  ],
  "init_std": 0.02,
  "is_encoder_decoder": true,
  "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": 1500,
  "max_target_positions": 448,
  "median_filter_width": 7,
  "model_type": "whisper",
  "num_hidden_layers": 32,
  "num_mel_bins": 80,
  "pad_token_id": 49639,
  "scale_embedding": false,
  "suppress_tokens": [
    1,
    3,
    4,
    8,
    9,
    324,
    467,
    486,
    506,
    834,
    862,
    878,
    882,
    891,
    1305,
    1801,
    1929,
    2400,
    2566,
    3178,
    3185,
    3200,
    3461,
    3768,
    3883,
    4103,
    4580,
    6473,
    6533,
    7148,
    8901,
    10253,
    10747,
    11742,
    11835,
    12130,
    12359,
    13567,
    13924,
    14396,
    15019,
    15372,
    16292,
    16343,
    18081,
    18670,
    21357,
    22189,
    25749,
    25780,
    26052,
    27878,
    31206,
    31852,
    32019,
    36364,
    42314,
    46827,
    49255,
    49636,
    49640,
    49740,
    49741,
    49742,
    49743,
    49744
  ],
  "torch_dtype": "float32",
  "transformers_version": "4.35.2",
  "use_cache": true,
  "use_weighted_layer_sum": false,
  "vocab_size": 51247
}