{
  "_name_or_path": "/data/wenhao/projects/xtuner/work_dirs/final_new_v/visual_encoder",
  "architectures": [
    "PikaSigVidEncoder"
  ],
  "attention_dropout": 0.0,
  "hidden_act": "gelu_pytorch_tanh",
  "hidden_size": 1152,
  "image_size": 384,
  "intermediate_size": 4304,
  "layer_norm_eps": 1e-06,
  "model_type": "siglip_vision_model",
  "num_attention_heads": 16,
  "num_channels": 3,
  "num_hidden_layers": 27,
  "patch_size": 14,
  "torch_dtype": "float32",
  "transformers_version": "4.39.2"
}