The purpose of instruction tuning is enabling zero-shot performance, but instruction tuning has also been shown to improve chain-of-thought reasoning and value alignment (Si et al., 2023). Here we consider the impact on $\textit{consistency}$, i.e., the sensitivity of language models to small perturbations in the input. We compare 10 instruction-tuned LLaMA models to the original LLaMA-7b model and show that almost across-the-board they become more consistent, both in terms of their representations and their predictions in zero-shot and downstream tasks. We explain these improvements through mechanistic analyses of factual recall.

中文翻译：

---

指令调整是否会使法学硕士更加一致？

指令调优的目的是实现零样本性能，但指令调优也被证明可以改善思想链推理和值对齐（Si 等人，2023）。这里我们考虑对$\textit{consistency}$的影响，即语言模型对输入中的小扰动的敏感性。我们将 10 个指令调整的 LLaMA 模型与原始 LLaMA-7b 模型进行了比较，结果表明，它们几乎全面变得更加一致，无论是在零样本和下游任务中的表示和预测方面。我们通过事实回忆的机械分析来解释这些改进。