Saved in:
Bibliographic Details
Main Authors: Tan, Zhen, Dong, Daize, Zhao, Xinyu, Peng, Jie, Cheng, Yu, Chen, Tianlong
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.11030
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914871932616704
author Tan, Zhen
Dong, Daize
Zhao, Xinyu
Peng, Jie
Cheng, Yu
Chen, Tianlong
author_facet Tan, Zhen
Dong, Daize
Zhao, Xinyu
Peng, Jie
Cheng, Yu
Chen, Tianlong
contents 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. We will release our implementation and model weights upon acceptance.
format Preprint
id arxiv_https___arxiv_org_abs_2407_11030
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DLO: Dynamic Layer Operation for Efficient Vertical Scaling of LLMs
Tan, Zhen
Dong, Daize
Zhao, Xinyu
Peng, Jie
Cheng, Yu
Chen, Tianlong
Machine Learning
Artificial Intelligence
Computation and Language
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. We will release our implementation and model weights upon acceptance.
title DLO: Dynamic Layer Operation for Efficient Vertical Scaling of LLMs
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2407.11030