vit_config.json

{
    "hidden_size": 64,
    "image_size": 980,
    "intermediate_size": 128,
    "model_type": "siglip_vision_model",
    "num_attention_heads": 2,
    "num_hidden_layers": 1,
    "patch_size": 14,
    "num_channels": 3
}