Update README.md

README.md

@@ -50,8 +50,12 @@ from transformers import LLaMATokenizer, LLaMAForCausalLM, GenerationConfig
 tokenizer = LLaMATokenizer.from_pretrained("cosimoiaia/Loquace-12B")
 model = LLaMAForCausalLM.from_pretrained(
     "cosimoiaia/Loquace-12B",
-    load_in_8bit=True,
+    load_in_4bit=True,
     device_map="auto",
+    quantization_config=BitsAndBytesConfig(
+      load_in_4bit=True,
+      llm_int8_has_fp16_weight=False
+    )
 )
 ```