5 accepted papers
Can we train LLMs to *introspect*, i.e. to faithfully describe their own behaviors in natural language? Prior work has shown some, limited, success. However, it is difficult to scale introspection training due to a lack of ground-truth labels. In this work, we study an approach to introspection trai…
_Latent-space_ monitoring techniques have shown promise as defenses against LLM attacks. These defenses act as scanners to detect harmful activations before they lead to undesirable actions. This prompts the question: can models execute harmful behavior _via inconspicuous latent states_? Here, we st…
Methods for knowledge editing and unlearning in large language models seek to edit or remove undesirable knowledge or capabilities without compromising general language modeling performance. This work investigates how mechanistic interpretability---which, in part, aims to identify model components (…
*Alignment Faking in Large Language Models* presented a demonstration of Claude 3 Opus and Claude 3.5 Sonnet selectively complying with a helpful-only training objective to prevent modification of their behavior outside of training. We expand this analysis to 25 models and find that only 5 (Claude 3…
Transformers demonstrate impressive performance on a range of reasoning benchmarks. To evaluate the degree to which these abilities are a result of actual reasoning, existing work has focused on developing sophisticated benchmarks for behavioral studies. However, these studies do not provide insight…