qlora merge and load requires that base model isn't loaded in 4 or 8 bit

README.md

@@ -363,12 +363,18 @@ Pass the appropriate flag to the train command:
 
 ### Merge LORA to base
 
-Add below flag to train command above
+Add below flag to train command above (and using LoRA)
 
 ```bash
 --merge_lora --lora_model_dir="./completed-model"
 ```
 
+Add below flag to train command above (and using QLoRA)
+
+```bash
+--merge_lora --lora_model_dir="./completed-model" --load_in_8bit False --load_in_4bit False
+```
+
 ## Common Errors 🧰
 
 > Cuda out of memory