[2103.15949] Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors

Abstract:Transformer networks have revolutionized NLP representation learning since they were introduced. Though a great effort has been made to explain the representation in transformers, it is widely recognized that our understanding is not sufficient. One important reason is that there lack enough visualization tools for detailed analysis. In this paper, we propose to use dictionary learning to open up these "black boxes" as linear superpositions of transformer factors. Through visualization, we demonstrate the hierarchical semantic structures captured by the transformer factors, e.g., word-level polysemy disambiguation, sentence-level pattern formation, and long-range dependency. While some of these patterns confirm the conventional prior linguistic knowledge, the rest are relatively unexpected, which may provide new insights. We hope this visualization tool can bring further knowledge and a better understanding of how transformer networks work. The code is available at this https URL

[2103.15949] Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Computation and Language arXiv:2103.15949 (cs) [Submitted on 29 Mar 2021 ( v1 ), last revised 4 Apr 2023 (this version, v2)] Title: Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors Authors: Zeyu Yun , Yubei Chen , Bruno A Ol

related reading

• Towards Monosemanticity: Decomposing Language Models With Dictionary Learningtransformer-circuits.pub
• A Mathematical Framework for Transformer Circuitstransformer-circuits.pub
• The Illustrated Transformer – Jay Alammar – Visualizing machine learning one concept at a time.jalammar.github.io
• Transformer Circuits Threadtransformer-circuits.pub
• Transformer Explainer: LLM Transformer Model Visually Explainedpoloclub.github.io
• Transformers from Scratche2eml.school
• The Annotated Transformernlp.seas.harvard.edu
• Towards Monosemanticity: Decomposing Language Models With Dictionary Learningtransformer-circuits.pub
• Everything About Transformerskrupadave.com
• Transformer (deep learning) - Wikipediaen.wikipedia.org
• Circuits Updates - January 2024transformer-circuits.pub
• Interfaces for Explaining Transformer Language Models – Jay Alammar – Visualizing machine learning one concept at a time.jalammar.github.io