{
  "architectures": [
    "Qwen2_5_VisionTransformerPretrainedModel"
  ],
  "depth": 2,
  "dtype": "float32",
  "fullatt_block_indexes": [
    7,
    15,
    23,
    31
  ],
  "hidden_act": "silu",
  "hidden_size": 32,
  "in_channels": 3,
  "in_chans": 3,
  "initializer_range": 0.02,
  "intermediate_size": 32,
  "model_type": "qwen2_5_vl_vision",
  "num_heads": 4,
  "out_hidden_size": 32,
  "patch_size": 14,
  "rope_parameters": {
    "rope_theta": 10000.0,
    "rope_type": "axial"
  },
  "spatial_merge_size": 1,
  "spatial_patch_size": 14,
  "temporal_patch_size": 2,
  "tokens_per_second": 1,
  "transformers_version": "5.19.0.dev0",
  "window_size": 112
}