Update config.json to match latest model

unet/config.json

@@ -1,6 +1,7 @@
 {
   "_class_name": "UNet3DConditionModel",
-  "_diffusers_version": "0.17.0.dev0",
+  "_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": [
@@ -9,7 +10,6 @@
     1280,
     1280
   ],
-  "center_input_sample": false,
   "cross_attention_dim": 768,
   "down_block_types": [
     "CrossAttnDownBlockInflated3D",
@@ -21,12 +21,11 @@
   "in_channels": 4,
   "layers_per_block": 2,
   "mid_block_scale_factor": 1,
-  "mid_block_type" : "UNetMidBlockInflated3DCrossAttn",
+  "mid_block_type": "UNetMidBlockInflated3DCrossAttn",
   "norm_eps": 1e-05,
   "norm_num_groups": 32,
-  "only_cross_attention": false,
+  "num_attention_heads": null,
   "out_channels": 4,
-  "resnet_time_scale_shift": "default",
   "sample_size": 64,
   "up_block_types": [
     "UpBlockInflated3D",
@@ -34,7 +33,5 @@
     "CrossAttnUpBlockInflated3D",
     "CrossAttnUpBlockInflated3D"
   ],
-  "upcast_attention": false,
-  "use_linear_projection": false,
   "use_temporal_transformer": false
-}
+}