Update README.md

README.md

@@ -15,7 +15,7 @@ Llama3-OGSQL-8B  was fine-tuned on the most recent and state of the art models (
 
 The model has been trained on more than 270 million tokens, ensuring robust performance and high accuracy in SQL generation tasks.
 
-- **Model type**: Mixture Of Experts (MoE)
+- **Model type**: Auto-regressive language model
 - **Language(s) (NLP)**: SQL (target language for generation)
 - **Finetuned from model**: Llama3-8B