qwen2_5_vit/config.json

{
  "_name_or_path": "/video_hy2/modelzoo/Qwen2-5-ViT-600m",
  "architectures": [
    "Qwen2_5_VisionTransformer"
  ],
  "auto_map": {
    "AutoConfig": "configuration_qwen2_5_vit.Qwen2_5_VLVisionConfig",
    "AutoModel": "qwen2_5_vit.Qwen2_5_VisionTransformer"
  },
  "depth": 32,
  "fullatt_block_indexes": [
    7,
    15,
    23,
    31
  ],
  "hidden_act": "silu",
  "hidden_size": 1280,
  "in_channels": 3,
  "in_chans": 3,
  "intermediate_size": 3456,
  "model_type": "qwen2_5_vit",
  "num_heads": 16,
  "out_hidden_size": 8192,
  "patch_size": 14,
  "spatial_merge_size": 2,
  "spatial_patch_size": 14,
  "temporal_patch_size": 2,
  "tokens_per_second": 2,
  "torch_dtype": "float32",
  "transformers_version": "4.46.3",
  "window_size": 112
}