DLO: Dynamic Layer Operation for Efficient Vertical Scaling of LLMs

Zhen Tan, Daize Dong, Xinyu Zhao, Jianing Cai, Jie Peng, Yu Cheng, Tianlong Chen

Track: Main paper track (up to 5 pages excluding references and appendix)

Keywords: Mixture of Expert, Large Language Models

TL;DR: Dynamic Layer Operation (DLO) optimizes large language models by dynamically expanding, activating, and skipping layers to improve both training efficiency and inference scalability without requiring pre-training modifications.

Abstract:

In this paper, we introduce Dynamic Layer Operations (DLO), a novel approach for vertically scaling transformer-based Large Language Models (LLMs) by dynamically expanding, activating, or skipping layers using a sophisticated routing policy based on layerwise feature similarity. Unlike traditional Mixture-of-Experts (MoE) methods that focus on extending the model width, our approach targets model depth, addressing the redundancy observed across layer representations for various input samples. Our framework is integrated with the Supervised Fine-Tuning (SFT) stage, eliminating the need for resource-intensive Continual Pre-Training (CPT). Experimental results demonstrate that DLO not only outperforms the original unscaled models but also achieves comparable results to densely expanded models with significantly improved efficiency. Our work offers a promising direction for building efficient yet powerful LLMs. Our code is available at {\url{https://github.com/DaizeDong/LLaMA-DLO.git}}.

Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.

Submission Number: 89