Mitigating Outlier Activations in Low-Precision Fine-Tuning of Language Models
This addresses a specific bottleneck for deploying large language models more cost-effectively, representing an incremental improvement in low-precision fine-tuning techniques.
The paper tackles the problem of outlier activations degrading performance in low-precision fine-tuning of language models, proposing a method to represent outliers in 8-bit integers instead of FP16, which improves robustness and performance as demonstrated through experiments.
Low-precision fine-tuning of language models has gained prominence as a cost-effective and energy-efficient approach to deploying large-scale models in various applications. However, this approach is susceptible to the existence of outlier values in activation. The outlier values in the activation can negatively affect the performance of fine-tuning language models in the low-precision regime since they affect the scaling factor and thus make representing smaller values harder. This paper investigates techniques for mitigating outlier activation in low-precision integer fine-tuning of the language models. Our proposed novel approach enables us to represent the outlier activation values in 8-bit integers instead of floating-point (FP16) values. The benefit of using integers for outlier values is that it enables us to use operator tiling to avoid performing 16-bit integer matrix multiplication to address this problem effectively. We provide theoretical analysis and supporting experiments to demonstrate the effectiveness of our approach in improving the robustness and performance of low-precision fine-tuned language models.