Systematic Weight Evaluation for Pruning Large Language Models: Enhancing Performance and Sustainability

TL;DR

This paper introduces a systematic weight evaluation method for pruning large language models, aiming to improve efficiency and sustainability without sacrificing model performance.

Contribution

It proposes a novel approach that monitors weight importance over training to optimize pruning, balancing model size reduction with performance preservation.

Findings

Moderate pruning improves efficiency and reduces loss.

Excessive pruning significantly degrades model performance.

Monitoring weight evolution enables sustainable model development.

Abstract

The exponential growth of large language models (LLMs) like ChatGPT has revolutionized artificial intelligence, offering unprecedented capabilities in natural language processing. However, the extensive computational resources required for training these models have significant environmental implications, including high carbon emissions, energy consumption, and water usage. This research presents a novel approach to LLM pruning, focusing on the systematic evaluation of individual weight importance throughout the training process. By monitoring parameter evolution over time, we propose a method that effectively reduces model size without compromising performance. Extensive experiments with both a scaled-down LLM and a large multimodal model reveal that moderate pruning enhances efficiency and reduces loss, while excessive pruning drastically deteriorates model performance. These findings highlight the critical need for optimized AI models to ensure sustainable development, balancing technological advancement with environmental responsibility.