{
  "feature_dim": 1024,
  "embedding_dim": 300,
  "hidden_dim": 512,
  "vocab_size": 30522,
  "num_layers": 1,
  "dropout": 0.5,
  "architectures": ["modeling.ImageCaptioningModel"],
  "auto_map": {
    "AutoConfig": "modeling.ImageCaptioningConfig",
    "AutoModel": "modeling.ImageCaptioningModel"
  }
}