LLM Code Vulnerability Detection

This repository contains research done to try to utilize Large Language Models to detect and correct vulnerabilities in code. The lava directory contains information and code on how I extracted data for a conversational AI using the LAVA tool. The juliet directory contains information on how I extracted similar data from the Juliet Dataset. The llama2 directory contains information on how I downloaded and queried the LLaMA 2 13b parameter model locally. The extra directory has more information on various other turorials and code written to download and finetune LLMs.