diff --git a/examples/dreambooth/train_dreambooth.py b/examples/dreambooth/train_dreambooth.py
index adc1359a5e56..447adeb9beee 100644
--- a/examples/dreambooth/train_dreambooth.py
+++ b/examples/dreambooth/train_dreambooth.py
@@ -159,6 +159,12 @@ def parse_args(input_args=None):
             " training using `--resume_from_checkpoint`."
         ),
     )
+    parser.add_argument(
+        "--generating_progress_steps",
+        type=int,
+        default=500,
+        help=("Save an image generated from the model every X steps."),
+    )
     parser.add_argument(
         "--resume_from_checkpoint",
         type=str,
@@ -416,6 +422,11 @@ def main(args):
 
     if args.seed is not None:
         set_seed(args.seed)
+        g = torch.Generator(device=accelerator.device.type)
+        g.manual_seed(args.seed)
+    else:
+        g = torch.Generator(device=accelerator.device.type)
+        g.manual_seed(42)
 
     if args.with_prior_preservation:
         class_images_dir = Path(args.class_data_dir)
@@ -739,6 +750,19 @@ def main(args):
                         accelerator.save_state(save_path)
                         logger.info(f"Saved state to {save_path}")
 
+                if global_step % args.generating_progress_steps == 0:
+                    if accelerator.is_main_process:
+                        # generate and save the image
+                        pipeline = DiffusionPipeline.from_pretrained(
+                            args.pretrained_model_name_or_path,
+                            unet=accelerator.unwrap_model(unet),
+                            text_encoder=accelerator.unwrap_model(text_encoder),
+                            torch_dtype=weight_dtype,
+                        ).to(accelerator.device.type)
+
+                        image = pipeline(args.instance_prompt, generator=g).images[0]
+                        image.save(f"{args.output_dir}/logs/image_{global_step}.png")
+
             logs = {"loss": loss.detach().item(), "lr": lr_scheduler.get_last_lr()[0]}
             progress_bar.set_postfix(**logs)
             accelerator.log(logs, step=global_step)