config.json

{
  "_name_or_path": "jaranohaal/vit-base-violence-detection",
  "architectures": [
    "ViTForImageClassification"
  ],
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-12,
  "hidden_size": 768,
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "image_size": 224,
  "patch_size": 16,
  "num_channels": 3,
  "num_labels": 2,  # Assuming binary classification for violence detection
  "transformers_version": "4.41.0",
  "torch_dtype": "float32"
}