Norm Tweaking: High-performance Low-bit Quantization of Large Language Models

TL;DR

This paper introduces 'norm tweaking', a plugin technique for post-training quantization that significantly improves low-bit quantization accuracy of large language models, enabling 2-bit quantization without accuracy loss.

Contribution

The paper proposes a novel norm tweaking method that enhances existing PTQ techniques, allowing high-precision low-bit quantization of large language models with minimal performance degradation.

Findings

Achieves 2-bit quantization accuracy comparable to float models on GLM-130B and OPT-66B.

Significantly outperforms existing PTQ methods in weight-only and joint quantization.

Demonstrates practical applicability for real-world deployment of large language models.

Abstract

As the size of large language models (LLMs) continues to grow, model compression without sacrificing accuracy has become a crucial challenge for deployment. While some quantization methods, such as GPTQ, have made progress in achieving acceptable 4-bit weight-only quantization, attempts at lower-bit quantization often result in severe performance degradation. In this paper, we introduce a technique called norm tweaking, which can be used as a plugin in current PTQ methods to achieve high precision while being cost-efficient. Our approach is inspired by the observation that rectifying the quantized activation distribution to match its float counterpart can readily restore accuracy for LLMs. To achieve this, we carefully design a tweaking strategy that includes calibration data generation and channel-wise distance constraint to update the weights of normalization layers for better generalization. We conduct extensive experiments on various datasets using several open-sourced LLMs. Our method demonstrates significant improvements in both weight-only quantization and joint quantization of weights and activations, surpassing existing PTQ methods. On GLM-130B and OPT-66B, our method even achieves the same level of accuracy at 2-bit quantization as their float ones. Our simple and effective approach makes it more practical for real-world applications.