Physical Sciences and Mathematics Commons^™

Security And Interpretability In Large Language Models, Lydia Danas

Undergraduate Honors Theses

Large Language Models (LLMs) have the capability to model long-term dependencies in sequences of tokens, and are consequently often utilized to generate text through language modeling. These capabilities are increasingly being used for code generation tasks; however, LLM-powered code generation tools such as GitHub's Copilot have been generating insecure code and thus pose a cybersecurity risk. To generate secure code we must first understand why LLMs are generating insecure code. This non-trivial task can be realized through interpretability methods, which investigate the hidden state of a neural network to explain model outputs. A new interpretability method is rationales, which obtains …