{
  "architectures": [
    "ImageCaptioningModel"
  ],
  "emb_size": 300,
  "model_type": "image_captioning_transformer",
  "num_heads": 6,
  "num_layers": 3,
  "torch_dtype": "float32",
  "transformers_version": "4.52.3",
  "vocab_size": 3478
}