Is It a Free Lunch for Removing Outliers during Pretraining?

arXiv:2402.12102v1 Announce Type: new
Abstract: With the growing size of large language models, the role of quantization becomes increasingly significant. However, outliers present in weights or activations notably influence the performance of quantized models. Recently, \citet{qtransformer} introduced a novel softmax function aimed at pretraining models in an outlier-free manner, thereby enhancing their suitability for quantization. Interestingly, we observed that such an approach leads to performance degradation in full precision. Building on this insight, we enhance the method by ensuring its …

abstract arxiv cs.ai cs.cl free function influence language language models large language large language models novel outlier outliers performance pretraining quantization role softmax type