Do different prompting methods yield a common task representation in language models?

Guy Davidson, Todd M. Gureckis, Brenden Lake, Adina Williams

Keywords: task representations; function vectors; in-context learning; instructions; mechanistic interpretability

TL;DR: Different ways of prompting the same task elicit different task representations in language models.

Abstract:

Demonstrations and instructions are two primary approaches for prompting language models to perform in-context learning (ICL) tasks.
Do identical tasks elicited in different ways result in similar representations of the task? An improved understanding of task representation mechanisms would offer interpretability insights and may aid in steering models. We study this through function vectors (FVs), recently proposed as a mechanism to extract few-shot ICL task representations. We generalize FVs to alternative task presentations, focusing on short textual instruction prompts, and successfully extract instruction function vectors that promote zero-shot task accuracy. We find evidence that demonstration- and instruction-based function vectors leverage different model components, and offer several controls to dissociate their contributions to task performance. Our results suggest that different task prompting forms do not induce a common task representation through FVs but elicit different, partly overlapping mechanisms. Our findings offer principled support to the practice of combining instructions and task demonstrations, imply challenges in universally monitoring task inference across presentation forms, and encourage further examinations of LLM task inference mechanisms.

Primary Area: Deep learning (e.g., architectures, generative models, optimization for deep networks, foundation models, LLMs)

Submission Number: 5256