Accelerating Sparse Autoencoder Training via Layer-Wise Transfer Learning in Large Language Models

被引：0

作者：

Ghilardi, Davide ^{[1
]}

Belotti, Federico ^{[1
]}

Molinari, Marco ^{[2
,4
]}

Lim, Jaehyuk ^{[2
,3
]}

机构：

[1] University of Milan-Bicocca, Italy

[2] LSE.AI

[3] University of Pennsylvania, United States

[4] London School of Economics, United Kingdom

来源：

BlackboxNLP 2024 - 7th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP - Proceedings of the Workshop | 2024年

关键词：

Compendex;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

Computational linguistics

引用

页码：530 / 550