Laura Kopf
@lkopf.bsky.social
PhD student in Interpretable Machine Learning at @tuberlin.bsky.social & @bifold.berlin
https://web.ml.tu-berlin.de/author/laura-kopf/
https://web.ml.tu-berlin.de/author/laura-kopf/
Reposted by Laura Kopf
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/2025.blackbo...
bsky.app/profile/nfel...
November 6, 2025 at 7:00 AM
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/2025.blackbo...
bsky.app/profile/nfel...
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
@tuberlin.bsky.social
@bifold.berlin
UMI Lab
@fraunhoferhhi.bsky.social
@unipotsdam.bsky.social
@leibnizatb.bsky.social
September 19, 2025 at 12:02 PM
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
@tuberlin.bsky.social
@bifold.berlin
UMI Lab
@fraunhoferhhi.bsky.social
@unipotsdam.bsky.social
@leibnizatb.bsky.social
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 🙏
September 19, 2025 at 12:02 PM
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 🙏
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)
@tuberlin.bsky.social
@bifold.berlin
UMI Lab
@fraunhoferhhi.bsky.social
@unipotsdam.bsky.social
@leibnizatb.bsky.social
(7/7)
June 19, 2025 at 3:18 PM
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)
@tuberlin.bsky.social
@bifold.berlin
UMI Lab
@fraunhoferhhi.bsky.social
@unipotsdam.bsky.social
@leibnizatb.bsky.social
(7/7)
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)
@nfel.bsky.social
@kirillbykov.bsky.social
@philinelb.bsky.social
Anna Hedström
Marina M.-C. Höhne
@eberleoliver.bsky.social
(6/7)
June 19, 2025 at 3:18 PM
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)
@nfel.bsky.social
@kirillbykov.bsky.social
@philinelb.bsky.social
Anna Hedström
Marina M.-C. Höhne
@eberleoliver.bsky.social
(6/7)
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)
June 19, 2025 at 3:18 PM
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)
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)
June 19, 2025 at 3:18 PM
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)
We benchmark PRISM across layers and architectures, showing how polysemanticity and interpretability shift through the model. (3/7)
June 19, 2025 at 3:18 PM
We benchmark PRISM across layers and architectures, showing how polysemanticity and interpretability shift through the model. (3/7)
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)
June 19, 2025 at 3:18 PM
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)
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.
@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.
December 13, 2024 at 2:48 AM
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.
@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.
Thanks for putting together this amazing list Margaret! I would love to be added if you still have space :)
December 12, 2024 at 8:24 AM
Thanks for putting together this amazing list Margaret! I would love to be added if you still have space :)
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
📄 Paper: arxiv.org/abs/2405.20331
💻 Code: github.com/lkopf/cosy
🔗 Poster: neurips.cc/virtual/2024...
#NeurIPS2024 #MechInterp #ExplainableAI #Interpretability
December 11, 2024 at 6:43 AM
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
📄 Paper: arxiv.org/abs/2405.20331
💻 Code: github.com/lkopf/cosy
🔗 Poster: neurips.cc/virtual/2024...
#NeurIPS2024 #MechInterp #ExplainableAI #Interpretability
Special thanks to our supporting institutions: UMI Lab, @xtraexer.bsky.social, @tuberline.bsky.social, Uni Potsdam, ATB Potsdam, and Fraunhofer Heinrich-Hertz-Institut.
December 11, 2024 at 6:43 AM
Special thanks to our supporting institutions: UMI Lab, @xtraexer.bsky.social, @tuberline.bsky.social, Uni Potsdam, ATB Potsdam, and Fraunhofer Heinrich-Hertz-Institut.
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.
December 11, 2024 at 6:43 AM
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.