@inproceedings{6cd2aca8af9b49768f2c96498ae9f10a,
title = "Can LLMs Facilitate Interpretation of Pre-trained Language Models?",
abstract = "Work done to uncover the knowledge encoded within pre-trained language models rely on annotated corpora or human-in-the-loop methods. However, these approaches are limited in terms of scalability and the scope of interpretation. We propose using a large language model, ChatGPT, as an annotator to enable fine-grained interpretation analysis of pre-trained language models. We discover latent concepts within pre-trained language models by applying agglomerative hierarchical clustering over contextualized representations and then annotate these concepts using ChatGPT. Our findings demonstrate that ChatGPT produces accurate and semantically richer annotations compared to human-annotated concepts. Additionally, we showcase how GPT-based annotations empower interpretation analysis methodologies of which we demonstrate two: probing frameworks and neuron interpretation. To facilitate further exploration and experimentation in the field, we make available a substantial Concept-Net dataset (TCN) comprising 39,000 annotated concepts.",
author = "Basel Mousi and Nadir Durrani and Fahim Dalvi",
note = "Publisher Copyright: {\textcopyright}2023 Association for Computational Linguistics.; 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP 2023 ; Conference date: 06-12-2023 Through 10-12-2023",
year = "2023",
language = "English",
series = "EMNLP 2023 - 2023 Conference on Empirical Methods in Natural Language Processing, Proceedings",
publisher = "Association for Computational Linguistics (ACL)",
pages = "3248--3268",
editor = "Houda Bouamor and Juan Pino and Kalika Bali",
booktitle = "EMNLP 2023 - 2023 Conference on Empirical Methods in Natural Language Processing, Proceedings",
address = "United States",
}