Large Language Models (LLMs) have impressive capabilities, but are also prone
to outputting falsehoods. Recent work has developed techniques for inferring
whether a LLM is telling the truth by training probes on the LLM's internal
activations. However, this line of work is controversial, with some authors
pointing out failures of these probes to generalize in basic ways, among other
conceptual issues. In this work, we curate high-quality datasets of true/false
statements and use them to study in detail the structure of LLM representations
of truth, drawing on three lines of evidence: 1. Visualizations of LLM
true/false statement representations, which reveal clear linear structure. 2.
Transfer experiments in which probes trained on one dataset generalize to
different datasets. 3. Causal evidence obtained by surgically intervening in a
LLM's forward pass, causing it to treat false statements as true and vice
versa. Overall, we present evidence that language models linearly represent the
truth or falsehood of factual statements. We also introduce a novel technique,
mass-mean probing, which generalizes better and is more causally implicated in
model outputs than other probing techniques.

大语言模型 (LLM) 可以输出令人印象深刻的结果，但也容易输出错误信息。最近的研究开发了一种通过对 LLM 的内部激活进行训练推断 LLM 是否在讲真话的技术。然而，这一研究领域颇具争议，一些作者指出该方法在一些基本方面无法推广，存在概念上的问题。在本研究中，我们精心策划了一组高质量的真 / 假语句数据集，并利用这些数据集详细研究了 LLM 对真实性的表示结构，从三个方面获得证据：1. 展示 LLM 真 / 假语句表示的可视化结果，揭示了明显的线性结构；2. 将在一个数据集上训练的推断器推广到不同数据集的转移实验；3. 通过对 LLM 前向传递进行手术干扰，使其将假语句视为真实或将真实语句视为假。总体而言，我们提出证据表明语言模型线性地表示事实陈述的真实性或虚假性。我们还介绍了一种新技术，质量均值推断法，比其他推断技术具有更好的推广性和更多地与模型输出相关。