Update README.md

README.md

@@ -19,14 +19,10 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig,
 
 base_model = "bertin-project/bertin-gpt-j-6B-alpaca"
 tokenizer = AutoTokenizer.from_pretrained(base_model)
-model = AutoModelForCausalLM.from_pretrained(
-    base_model,
-    load_in_8bit=True,
-    device_map="auto",
-)
+model = AutoModelForCausalLM.from_pretrained(base_model)
 ```
 
-Until `PEFT` is fully supported in Hugginface's pipelines, for generation we can either consolidate the LoRA weights into the LLaMA model weights, or use the adapter's `generate()` method. Remember that the prompt still needs the English template:
+For generation, we can either use `pipeline()` or the model's `.generate()` method. Remember that the prompt needs a **Spanish** template:
 
 ```python
 # Generate responses