Interactions Across Blocks in Post-Training Quantization of Large Language Models

Khasmamad Shabanovi, Lukas Wiest, Vladimir Golkov, Daniel Cremers, Thomas Pfeil

Keywords: large language model, post training quantization, LLM, PTQ, weight quantization, deep learning, network compression, efficient inference

TL;DR: Post-training quantization tunes layers individually, assuming independence and ignoring downstream effects. We improve task performance by jointly optimizing multiple layers or using downstream errors.

Abstract: Post-training quantization is widely employed to reduce the computational demands of neural networks. Typically, individual substructures, such as layers or blocks of layers, are quantized with the objective of minimizing quantization errors in their pre-activations by fine-tuning the corresponding weights. Deriving this local objective from the global objective of minimizing task loss involves two key simplifications: assuming substructures are mutually independent and ignoring the knowledge of subsequent substructures as well as the task loss. In this work, we assess the effects of these simplifications on weight-only quantization of large language models. We introduce two multi-block fine-tuning strategies and compare them against the baseline of fine-tuning single transformer blocks. The first captures correlations of weights across blocks by jointly optimizing multiple quantized blocks. The second incorporates knowledge of subsequent blocks by minimizing the error in downstream pre-activations rather than focusing solely on the quantized block. Our findings indicate that the effectiveness of these methods depends on the specific network model, with no impact on some models but demonstrating significant benefits for others.

Submission Number: 114