Document detail
ID

oai:arXiv.org:2312.05491

Topic
Computer Science - Computation and... Computer Science - Artificial Inte... I.2.7
Author
Miglani, Vivek Yang, Aobo Markosyan, Aram H. Garcia-Olano, Diego Kokhlikyan, Narine
Category

Computer Science

Year

2023

listing date

12/13/2023

Keywords
generative language
Metrics

Abstract

Captum is a comprehensive library for model explainability in PyTorch, offering a range of methods from the interpretability literature to enhance users' understanding of PyTorch models.

In this paper, we introduce new features in Captum that are specifically designed to analyze the behavior of generative language models.

We provide an overview of the available functionalities and example applications of their potential for understanding learned associations within generative language models.

Miglani, Vivek,Yang, Aobo,Markosyan, Aram H.,Garcia-Olano, Diego,Kokhlikyan, Narine, 2023, Using Captum to Explain Generative Language Models

Document

Open

Share

Source

Articles recommended by ES/IODE AI

AttentionLut: Attention Fusion-based Canonical Polyadic LUT for Real-time Image Enhancement
polyadic methods real-time feature module canonical image-adaptive enhancement luts image