diff --git a/library/train_util.py b/library/train_util.py
index 0fdbadc1..ffa099ef 100644
--- a/library/train_util.py
+++ b/library/train_util.py
@@ -1423,5 +1423,17 @@ def save_state_on_train_end(args: argparse.Namespace, accelerator):
   model_name = DEFAULT_LAST_OUTPUT_NAME if args.output_name is None else args.output_name
   accelerator.save_state(os.path.join(args.output_dir, LAST_STATE_NAME.format(model_name)))
 
+def generate_step_logs(args: argparse.Namespace, current_loss, avr_loss, lr_scheduler):
+    logs = {"loss/current": current_loss, "loss/average": avr_loss}
+
+    if args.network_train_unet_only:
+      logs["lr/unet"] = lr_scheduler.get_last_lr()[0]
+    elif args.network_train_text_encoder_only:
+      logs["lr/textencoder"] = lr_scheduler.get_last_lr()[0]
+    else:
+      logs["lr/textencoder"] = lr_scheduler.get_last_lr()[0]
+      logs["lr/unet"] = lr_scheduler.get_last_lr()[-1]
+
+    return logs
 
 # endregion
diff --git a/train_network.py b/train_network.py
index 8b4e008b..bd45d980 100644
--- a/train_network.py
+++ b/train_network.py
@@ -347,20 +347,21 @@ def train(args):
         global_step += 1
 
       current_loss = loss.detach().item()
-      if args.logging_dir is not None:
-        logs = {"loss": current_loss, "lr": lr_scheduler.get_last_lr()[0]}
-        accelerator.log(logs, step=global_step)
-
       loss_total += current_loss
       avr_loss = loss_total / (step+1)
       logs = {"loss": avr_loss}  # , "lr": lr_scheduler.get_last_lr()[0]}
       progress_bar.set_postfix(**logs)
 
+      if args.logging_dir is not None:
+        logs = train_util.generate_step_logs(args, current_loss, avr_loss, lr_scheduler)
+
+        accelerator.log(logs, step=global_step)
+
       if global_step >= args.max_train_steps:
         break
 
     if args.logging_dir is not None:
-      logs = {"epoch_loss": loss_total / len(train_dataloader)}
+      logs = {"loss/epoch": loss_total / len(train_dataloader)}
       accelerator.log(logs, step=epoch+1)
 
     accelerator.wait_for_everyone()