Layer by Layer: Uncovering Where Multi-Task Learning Happens in Instruction-Tuned Large Language Models

TL;DR

This paper investigates how instruction tuning affects the encoding of task-specific knowledge in large language models across over 60 NLP tasks, revealing layer-specific transitions from general to task-oriented representations.

Contribution

It provides a detailed analysis of where and how instruction tuning modifies task-specific information in LLMs, advancing understanding of multi-task learning mechanisms.

Findings

Some tasks are already encoded pre-training.

Instruction tuning enhances task-specific representations.

Identifies layers where models shift from general to task-specific features.

Abstract

Fine-tuning pre-trained large language models (LLMs) on a diverse array of tasks has become a common approach for building models that can solve various natural language processing (NLP) tasks. However, where and to what extent these models retain task-specific knowledge remains largely unexplored. This study investigates the task-specific information encoded in pre-trained LLMs and the effects of instruction tuning on their representations across a diverse set of over 60 NLP tasks. We use a set of matrix analysis tools to examine the differences between the way pre-trained and instruction-tuned LLMs store task-specific information. Our findings reveal that while some tasks are already encoded within the pre-trained LLMs, others greatly benefit from instruction tuning. Additionally, we pinpointed the layers in which the model transitions from high-level general representations to more task-oriented representations. This finding extends our understanding of the governing mechanisms of LLMs and facilitates future research in the fields of parameter-efficient transfer learning and multi-task learning.