To achieve optimal results from large language models (LLMs), a multi-faceted approach is essential. Firstly, careful dataset selection and preparation are crucial to ensure the model learns relevant representations. Secondly, setting tuning through techniques like randomized search can significantly improve the model's ability to perform on unseen