As AI systems become more capable, we would like to enlist their help to supervise other AIs.
We experiment with methods for training a harmless AI assistant through self-improvement, …

We apply preference modeling and reinforcement learning from human feedback (RLHF) to
finetune language models to act as helpful and harmless assistants. We find this alignment …

We describe our early efforts to red team language models in order to simultaneously discover,
measure, and attempt to reduce their potentially harmful outputs. We make three main …

We study whether language models can evaluate the validity of their own claims and predict
which questions they will be able to answer correctly. We first show that larger models are …

"Induction heads" are attention heads that implement a simple algorithm to complete token
sequences like [A][B] ... [A] -> [B]. In this work, we present preliminary and indirect evidence …

Given the broad capabilities of large language models, it should be possible to work towards
a general-purpose, text-based assistant that is aligned with human values, meaning that it …

Large-scale pre-training has recently emerged as a technique for creating capable, general-purpose,
generative models such as GPT-3, Megatron-Turing NLG, Gopher, and many …

As language models (LMs) scale, they develop many novel behaviors, good and bad,
exacerbating the need to evaluate how they behave. Prior work creates evaluations with …

We test the hypothesis that language models trained with reinforcement learning from human
feedback (RLHF) have the capability to "morally self-correct" -- to avoid producing harmful …

Hypoxia within a tumor acts as a strong selective pressure that promotes angiogenesis,
invasion, and metastatic spread. In this study, we used immune competent bone marrow …