Title: Inseq: An Interpretability Toolkit for Sequence Generation Models
Authors: Gabriele Sarti, Nils Feldhus, Ludwig Sickert, Oskar van der Wal, Malvina Nissim, Arianna Bisazza
Published: 27th February 2023 (Monday) @ 16:45:50
Link: http://arxiv.org/abs/2302.13942v3
Abstract
Past work in natural language processing interpretability focused mainly on popular classification tasks while largely overlooking generation settings, partly due to a lack of dedicated tools. In this work, we introduce Inseq, a Python library to democratize access to interpretability analyses of sequence generation models. Inseq enables intuitive and optimized extraction of modelsâ internal information and feature importance scores for popular decoder-only and encoder-decoder Transformers architectures. We showcase its potential by adopting it to highlight gender biases in machine translation models and locate factual knowledge inside GPT-2. Thanks to its extensible interface supporting cutting-edge techniques such as contrastive feature attribution, Inseq can drive future advances in explainable natural language generation, centralizing good practices and enabling fair and reproducible model evaluations.