0

Outlier Dimensions Encode Task-Specific Knowledge

Outlier dimensions in large language model representations persist through fine-tuning and can individually encode crucial task-specific knowledge, driving model performance.

Year
2023
Venue
arXiv 2023
Authors
3
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2310.17715v2ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

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.

Authors

3