iiinova — where your digital self keeps evolving

Laura Kopf

@lkopf.bsky.social

369 followers 392 following 19 posts

PhD student in Interpretable Machine Learning at @tuberlin.bsky.social & @bifold.berlin web.ml.tu-berlin.de/author/laura-ko…

Posts Replies Media

Reposted by Laura Kopf

Explainable AI Berlin @xai-berlin.bsky.social · 29/06/2026

🔦 Past Paper Highlight: PRISM generates multi-concept descriptions of LLM features by clustering high-activation inputs and labeling each cluster. Two evaluation metrics measure the semantic diversity and faithfulness of the resulting descriptions. 🔗 proceedings.neurips.cc/paper_files/...

Reposted by Laura Kopf

Health NLP Lab @health-nlp.com · 21/01/2026

Last week, Dr. Nils Feldhus @nfel.bsky.social, postdoctoral researcher at @tuberlin.bsky.social and @bifold.berlin, visited our lab and presented his research during our weekly lab meeting.

Laura Kopf @lkopf.bsky.social · 02/12/2025

📄 openreview.net/forum?id=btJ... Special thanks to our co-authors @nfel.bsky.social @kirillbykov.bsky.social @philinelb.bsky.social Anna Hedström, and Marina Höhne who are not with us in person at the conference.

Capturing Polysemanticity with PRISM: A Multi-Concept Feature...

Automated interpretability research aims to identify concepts encoded in neural network features to enhance human understanding of model behavior. Within the context of large language models (LLMs)...

Laura Kopf @lkopf.bsky.social · 02/12/2025

I’m at #NeurIPS in San Diego this week! Come see our poster on feature interpretability. Find @eberleoliver.bsky.social and me at: 🪧Poster Session 1 @ Exhibit Hall C,D,E #1015 Wed 3 Dec, 11 am - 2 pm 🪧Poster @ Mech Interp Workshop Upper Level Room 30A-E Sun 7 Dec, 8 am - 5 pm

Reposted by Laura Kopf

Explainable AI Berlin @xai-berlin.bsky.social · 28/11/2025

🚀 Visit our #NeurIPS posters at @neuripsconf.bsky.social! Meet and interact with our authors at all locations — San Diego, Mexico City, and Copenhagen. Details in the thread. 👇👇👇

Reposted by Laura Kopf

Explainable AI Berlin @xai-berlin.bsky.social · 08/11/2025

Heading to the EMNLP BlackboxNLP Workshop this Sunday? Don’t miss @nfel.bsky.social and @lkopf.bsky.social poster on „Interpreting Language Models Through Concept Descriptions: A Survey“ aclanthology.org/2025.blackbo... #EMNLP #BlackboxNLP #XAI #Interpretapility

Reposted by Laura Kopf

Explainable AI Berlin @xai-berlin.bsky.social · 06/11/2025

We are grateful for the opportunity to present some of our work at the All Hands Meeting of the German AI Centers, hosted by @dfki.bsky.social in Saarbrücken. Andreas Lutz @eberleoliver.bsky.social Manuel Welte @lorenzlinhardt.bsky.social @lkopf.bsky.social #AI #XAI #Interpretability

Reposted by Laura Kopf

Nils Feldhus @nfel.bsky.social · 06/11/2025

Nov 9, @blackboxnlp.bsky.social , 11:00-12:00 @ Hall C – Interpreting Language Models Through Concept Descriptions: A Survey (Feldhus & Kopf) @lkopf.bsky.social 🗞️ aclanthology.org/2025.blackbo... bsky.app/profile/nfel...

aclanthology.org

Interpreting Language Models Through Concept Descriptions: A Survey

Nils Feldhus, Laura Kopf. Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP. 2025.

Reposted by Laura Kopf

Explainable AI Berlin @xai-berlin.bsky.social · 03/11/2025

This is the eXplainable AI research channel of the machine learning group of Prof. Klaus-Robert Müller at Technische Universität Berlin @tuberlin.bsky.social & BIFOLD @bifold.berlin. Let's connect! #XAI #ExplainableAI #MechInterp #MachineLearning #Interpretability

media.tenor.com

a black background with green text that says `` hello , world ''

ALT: a black background with green text that says `` hello , world ''

Reposted by Laura Kopf

Nils Feldhus @nfel.bsky.social · 02/10/2025

🔍 Are you curious about uncovering the underlying mechanisms and identifying the roles of model components (neurons, …) and abstractions (SAEs, …)? We provide the first survey of concept description generation and evaluation methods. Joint effort w/ @lkopf.bsky.social 📄 arxiv.org/abs/2510.01048

Overview of descriptions for model components (neurons, attention heads) and model abstractions (SAE features, circuits).

Laura Kopf @lkopf.bsky.social · 19/09/2025

Many thanks as well to the institutions that supported this research: @tuberlin.bsky.social @bifold.berlin UMI Lab @fraunhoferhhi.bsky.social @unipotsdam.bsky.social @leibnizatb.bsky.social

Laura Kopf @lkopf.bsky.social · 19/09/2025

I’m very grateful to my amazing collaborators @nfel.bsky.social, @kirillbykov.bsky.social, @philinelb.bsky.social, Anna Hedström, Marina M.-C. Höhne, and @eberleoliver.bsky.social 🙏

Laura Kopf @lkopf.bsky.social · 19/09/2025

Happy to share that our PRISM paper has been accepted at #NeurIPS2025 🎉 In this work, we introduce a multi-concept feature description framework that can identify and score polysemantic features. 📄 Paper: arxiv.org/abs/2506.15538 #NeurIPS #MechInterp #XAI

Laura Kopf @lkopf.bsky.social · 19/06/2025

Grateful to the institutions that supported this work: @tuberlin.bsky.social @bifold.berlin UMI Lab @fraunhoferhhi.bsky.social @unipotsdam.bsky.social @leibnizatb.bsky.social (7/7)

Laura Kopf @lkopf.bsky.social · 19/06/2025

Many thanks to my amazing co-authors: @nfel.bsky.social @kirillbykov.bsky.social @philinelb.bsky.social Anna Hedström Marina M.-C. Höhne @eberleoliver.bsky.social (6/7)

Laura Kopf @lkopf.bsky.social · 19/06/2025

Our results highlight that the PRISM framework not only provides multiple human interpretable descriptions for neurons but also aligns with the human interpretation of polysemanticity. (5/7)

Laura Kopf @lkopf.bsky.social · 19/06/2025

In exploring the concept space, we use PRISM to characterize more complex components, finding and interpreting patterns that specific attention heads or groups of neurons respond to. (4/7)

Laura Kopf @lkopf.bsky.social · 19/06/2025

We benchmark PRISM across layers and architectures, showing how polysemanticity and interpretability shift through the model. (3/7)

Laura Kopf @lkopf.bsky.social · 19/06/2025

PRISM samples sentences from the top percentile activation distribution, clusters them in embedding space, and uses an LLM to generate labels for each concept cluster. (2/7)

Laura Kopf @lkopf.bsky.social · 19/06/2025

🔍 When do neurons encode multiple concepts? We introduce PRISM, a framework for extracting multi-concept feature descriptions to better understand polysemanticity. 📄 Capturing Polysemanticity with PRISM: A Multi-Concept Feature Description Framework arxiv.org/abs/2506.15538 🧵 (1/7)

Laura Kopf @lkopf.bsky.social · 13/12/2024

Huge thanks to my incredible supervisor @kirillbykov.bsky.social, who laid the foundation for this project and provided brilliant guidance 🙏, and to @philinelb.bsky.social and Sebastian Lapuschkin, who unfortunately couldn’t be there.

Laura Kopf @lkopf.bsky.social · 13/12/2024

Still overwhelmed by the amazing response to our poster session at @neuripsconf.bsky.social with Anna Hedström and Marina Höhne! It was incredible to have such lively and inspiring discussions with brilliant people whose work I admire. ✨

Laura Kopf @lkopf.bsky.social · 12/12/2024

Thanks for putting together this amazing list Margaret! I would love to be added if you still have space :)

Laura Kopf @lkopf.bsky.social · 11/12/2024

Want to know more about CoSy? 📄 Paper: arxiv.org/abs/2405.20331 💻 Code: github.com/lkopf/cosy 🔗 Poster: neurips.cc/virtual/2024... #NeurIPS2024 #MechInterp #ExplainableAI #Interpretability

000

Laura Kopf @lkopf.bsky.social · 11/12/2024

Special thanks to our supporting institutions: UMI Lab, @xtraexer.bsky.social, @tuberline.bsky.social, Uni Potsdam, ATB Potsdam, and Fraunhofer Heinrich-Hertz-Institut.

100

Laura Kopf @lkopf.bsky.social · 11/12/2024

My co-authors Anna Hedström and Marina Höhne will also be at @neuripsconf.bsky.social. A big thank you to my other co-authors @kirillbykov.bsky.social, @philinelb.bsky.social and Sebastian Lapuschkin, who unfortunately couldn’t be there.

100

Laura Kopf @lkopf.bsky.social · 11/12/2024

I’ll be presenting our work at @neuripsconf.bsky.social in Vancouver! 🎉 Join me this Thursday, December 12th, in East Exhibit Hall A-C, Poster #3107, from 11 a.m. PST to 2 p.m. PST. I'll be discussing our paper “CoSy: Evaluating Textual Explanations of Neurons.”