{
  "model_cfg": {
    "embed_dim": 1024,
    "vision_cfg": {
      "image_size": 224,
      "layers": [
        3,
        4,
        6,
        3
      ],
      "width": 64,
      "patch_size": null
    },
    "text_cfg": {
      "context_length": 77,
      "vocab_size": 49408,
      "width": 512,
      "heads": 8,
      "layers": 12
    }
  },
  "preprocess_cfg": {
    "mean": [
      0.0,
      0.0,
      0.0
    ],
    "std": [
      1.0,
      1.0,
      1.0
    ],
    "interpolation": "bicubic",
    "resize_mode": "shortest"
  }
}