Update README.md

README.md

@@ -19,9 +19,9 @@ The backbone is under `LoftQ/Llama-2-7b-hf-4bit-64rank` and LoRA adapters are un
 ## Model Info
 ### Backbone
 - Stored format: bitsandbytes nf4
-- Size: ~ 5 GiB
+- Size: ~ 4.2 GiB
 - Loaded format: bitsandbytes nf4
-- Size loaded on GPU: ~5 GiB
+- Size loaded on GPU: ~ 4.2 GiB
 
 ### LoRA adapters
 - rank: 64