{
  "_class_name": "UNet3DConditionModel",
  "_diffusers_version": "0.22.0.dev0",
  "_name_or_path": "/root/.cache/huggingface/hub/models--Tune-A-Video-library--df-cpt-mo-di-bear-guitar/snapshots/50f6abec0190a7dfa07f015aa20ea9d0b9ef96e6/unet",
  "act_fn": "silu",
  "attention_head_dim": 8,
  "block_out_channels": [
    320,
    640,
    1280,
    1280
  ],
  "cross_attention_dim": 768,
  "down_block_types": [
    "CrossAttnDownBlockInflated3D",
    "CrossAttnDownBlockInflated3D",
    "CrossAttnDownBlockInflated3D",
    "DownBlockInflated3D"
  ],
  "downsample_padding": 1,
  "in_channels": 4,
  "layers_per_block": 2,
  "mid_block_scale_factor": 1,
  "mid_block_type": "UNetMidBlockInflated3DCrossAttn",
  "norm_eps": 1e-05,
  "norm_num_groups": 32,
  "num_attention_heads": null,
  "out_channels": 4,
  "sample_size": 64,
  "up_block_types": [
    "UpBlockInflated3D",
    "CrossAttnUpBlockInflated3D",
    "CrossAttnUpBlockInflated3D",
    "CrossAttnUpBlockInflated3D"
  ],
  "use_temporal_transformer": false
}