Outlier Dimensions Encode Task-Specific Knowledge

Autor:	Rudman, William, Chen, Catherine, Eickhoff, Carsten
Rok vydání:	2023
Předmět:	Computer Science - Computation and Language Computer Science - Artificial Intelligence
Druh dokumentu:	Working Paper
Popis:	Representations from large language models (LLMs) are known to be dominated by a small subset of dimensions with exceedingly high variance. Previous works have argued that although ablating these outlier dimensions in LLM representations hurts downstream performance, outlier dimensions are detrimental to the representational quality of embeddings. In this study, we investigate how fine-tuning impacts outlier dimensions and show that 1) outlier dimensions that occur in pre-training persist in fine-tuned models and 2) a single outlier dimension can complete downstream tasks with a minimal error rate. Our results suggest that outlier dimensions can encode crucial task-specific knowledge and that the value of a representation in a single outlier dimension drives downstream model decisions. Comment: Camera-ready version for EMNLP 2023
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2310.17715 Zobrazit plný text záznamu View this record from Arxiv