config.json

{
  "_name_or_path": "google/vit-base-patch16-224-in21k",
  "architectures": [
    "ViTForImageClassification"
  ],
  "attention_probs_dropout_prob": 0.0,
  "encoder_stride": 16,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 768,
  "id2label": {
    "0": "bee",
    "1": "beetle",
    "10": "grasshopper",
    "11": "horse",
    "12": "lizard",
    "13": "mosquito",
    "14": "mouse",
    "15": "snake",
    "16": "spider",
    "17": "whale",
    "2": "bird",
    "3": "butterfly",
    "4": "camel",
    "5": "cat",
    "6": "caterpillar",
    "7": "crab",
    "8": "dog",
    "9": "fly"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "label2id": {
    "bee": "0",
    "beetle": "1",
    "bird": "2",
    "butterfly": "3",
    "camel": "4",
    "cat": "5",
    "caterpillar": "6",
    "crab": "7",
    "dog": "8",
    "fly": "9",
    "grasshopper": "10",
    "horse": "11",
    "lizard": "12",
    "mosquito": "13",
    "mouse": "14",
    "snake": "15",
    "spider": "16",
    "whale": "17"
  },
  "layer_norm_eps": 1e-12,
  "model_type": "vit",
  "num_attention_heads": 12,
  "num_channels": 3,
  "num_hidden_layers": 12,
  "patch_size": 16,
  "problem_type": "single_label_classification",
  "qkv_bias": true,
  "torch_dtype": "float32",
  "transformers_version": "4.24.0"
}