End of training

unet/config.json

@@ -8,12 +8,10 @@
   "attention_head_dim": 8,
   "attention_type": "default",
   "block_out_channels": [
-    128,
-    128,
-    256,
-    256,
-    512,
-    512
+    320,
+    640,
+    1280,
+    1280
   ],
   "center_input_sample": false,
   "class_embed_type": null,
@@ -23,10 +21,8 @@
   "cross_attention_dim": 768,
   "cross_attention_norm": null,
   "down_block_types": [
-    "DownBlock2D",
-    "DownBlock2D",
-    "DownBlock2D",
-    "DownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
     "CrossAttnDownBlock2D",
     "DownBlock2D"
   ],
@@ -63,10 +59,8 @@
   "up_block_types": [
     "UpBlock2D",
     "CrossAttnUpBlock2D",
-    "UpBlock2D",
-    "UpBlock2D",
-    "UpBlock2D",
-    "UpBlock2D"
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D"
   ],
   "upcast_attention": false,
   "use_linear_projection": false

unet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce72cfc1d976c37ae091e52bad58e19cdf34776eccc05611252c0bc8ab339e7f
-size 574476604
+oid sha256:3fb6b8bd121af52a94c180287651f7339d87b098eaffafb790a6d72794679e67
+size 3438144492