This model is finetuned on a base llama-2-7b model with datasets b-mc2/sql-create-context Just for comparison the performance between basemodel and finetuned model