Text this: Finetuning Large Language Models for Vulnerability Detection