Being probabilistic models, during inference large language models (LLMs) display rare events: behaviour that is far from typical but highly significant. By definition all rare events are hard to see, but the enormous scale of LLM usage means that events completely unobserved during development are likely to become prominent in deployment. Here we present an end-to-end framework for the systematic analysis of rare events in LLMs. We provide a practical implementation spanning theory, efficient generation strategies, probability estimation and error analysis, which we illustrate with concrete examples. We outline extensions and applications to other models and contexts, highlighting the generality of the concepts and techniques presented here.

翻译：作为概率模型，大语言模型（LLMs）在推理过程中会表现出罕见事件：即偏离典型行为但具有高度显著性的现象。根据定义，所有罕见事件都难以观测，但LLM使用的庞大规模意味着开发阶段完全未观察到的事件很可能在部署阶段变得突出。本文提出了一个用于系统分析LLM罕见事件的端到端框架。我们提供了一个涵盖理论、高效生成策略、概率估计与误差分析的实践方案，并通过具体案例进行阐释。最后，我们概述了该框架向其他模型和场景的扩展与应用，以突显本文所提出概念与技术的普适性。