56,966 blogs · [ { "id": "01a08219-ddaf-7384-bcb4-f2bf9b4d86eb", "title": "Characterizing interference weights in a tiny language model", "url": "https://transformer-circuits.pub/2026/interference_effectiveness_helpfulness/index.html", "published_at": "2026-08-21T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9b9a02e3", "title": "Verbalizable Representations Form a Global Workspace in Language Models", "url": "https://transformer-circuits.pub/2026/workspace/index.html", "published_at": "2026-07-06T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9c652cef", "title": "Circuits Updates — June 2026", "url": "https://transformer-circuits.pub/2026/june-update/index.html", "published_at": "2026-06-30T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9c993ee1", "title": "Circuits Updates — May 2026", "url": "https://transformer-circuits.pub/2026/may-update/index.html", "published_at": "2026-06-01T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9d2cecbc", "title": "Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activations", "url": "https://transformer-circuits.pub/2026/nla/index.html", "published_at": "2026-05-07T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9e167dc9", "title": "HeadVis", "url": "https://transformer-circuits.pub/2026/headvis/index.html", "published_at": "2026-05-04T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9efe9c5b", "title": "Emotion Concepts and their Function in a Large Language Model", "url": "https://transformer-circuits.pub/2026/emotions/index.html", "published_at": "2026-04-02T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9f566b7c", "title": "Circuits Cross-Post — Activation Oracles", "url": "https://alignment.anthropic.com/2025/activation-oracles/", "published_at": "2025-12-19T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9f5ac8de", "title": "Circuits Updates — November 2025", "url": "https://transformer-circuits.pub/2025/november-update/index.html", "published_at": "2025-11-26T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bf9fcb8858", "title": "Emergent Introspective Awareness in Large Language Models", "url": "https://transformer-circuits.pub/2025/introspection/index.html", "published_at": "2025-10-29T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa0bd8d59", "title": "Circuits Updates — October 2025", "url": "https://transformer-circuits.pub/2025/october-update/index.html", "published_at": "2025-10-24T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa181642d", "title": "When Models Manipulate Manifolds: The Geometry of a Counting Task", "url": "https://transformer-circuits.pub/2025/linebreaks/index.html", "published_at": "2025-10-21T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa192056d", "title": "Circuits Updates — September 2025", "url": "https://transformer-circuits.pub/2025/september-update/index.html", "published_at": "2025-09-29T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa27e5398", "title": "Circuits Updates — August 2025", "url": "https://transformer-circuits.pub/2025/august-update/index.html", "published_at": "2025-08-28T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa33a1787", "title": "A Toy Model of Mechanistic (Un)Faithfulness", "url": "https://transformer-circuits.pub/2025/faithfulness-toy-model/index.html", "published_at": "2025-08-07T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa3692b14", "title": "Tracing Attention Computation Through Feature Interactions", "url": "https://transformer-circuits.pub/2025/attention-qk/index.html", "published_at": "2025-07-31T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa44364d5", "title": "A Toy Model of Interference Weights", "url": "https://transformer-circuits.pub/2025/interference-weights/index.html", "published_at": "2025-07-29T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa5282ec6", "title": "Sparse mixtures of linear transforms", "url": "https://transformer-circuits.pub/2025/bulk-update/index.html", "published_at": "2025-07-25T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa605ad6f", "title": "Circuits Updates — July 2025", "url": "https://transformer-circuits.pub/2025/july-update/index.html", "published_at": "2025-07-25T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa66afcdb", "title": "Automated Auditing", "url": "https://alignment.anthropic.com/2025/automated-auditing/", "published_at": "2025-07-24T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa6d32d17", "title": "Circuits Updates — April 2025", "url": "https://transformer-circuits.pub/2025/april-update/index.html", "published_at": "2025-04-29T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa7929f34", "title": "Progress on Attention", "url": "https://transformer-circuits.pub/2025/attention-update/index.html", "published_at": "2025-04-28T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa82d6e49", "title": "On the Biology of a Large Language Model", "url": "https://transformer-circuits.pub/2025/attribution-graphs/biology.html", "published_at": "2025-03-27T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa83a8a4f", "title": "Circuit Tracing: Revealing Computational Graphs in Language Models", "url": "https://transformer-circuits.pub/2025/attribution-graphs/methods.html", "published_at": "2025-03-27T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa86f2ce8", "title": "Insights on Crosscoder Model Diffing", "url": "https://transformer-circuits.pub/2025/crosscoder-diffing-update/index.html", "published_at": "2025-02-19T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa89ca377", "title": "Circuits Updates — January 2025", "url": "https://transformer-circuits.pub/2025/january-update/index.html", "published_at": "2025-01-27T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa8f65dd5", "title": "Stage-Wise Model Diffing", "url": "https://transformer-circuits.pub/2024/model-diffing/index.html", "published_at": "2024-12-11T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfa9cdbada", "title": "Sparse Crosscoders for Cross-Layer Features and Model Diffing", "url": "https://transformer-circuits.pub/2024/crosscoders/index.html", "published_at": "2024-10-25T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfaa6e7ae2", "title": "Using Dictionary Learning Features as Classifiers", "url": "https://transformer-circuits.pub/2024/features-as-classifiers/index.html", "published_at": "2024-10-16T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfaa915c8f", "title": "Circuits Updates — September 2024", "url": "https://transformer-circuits.pub/2024/september-update/index.html", "published_at": "2024-10-01T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfaaacdce8", "title": "Circuits Updates — August 2024", "url": "https://transformer-circuits.pub/2024/august-update/index.html", "published_at": "2024-09-06T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfab109557", "title": "Circuits Updates — July 2024", "url": "https://transformer-circuits.pub/2024/july-update/index.html", "published_at": "2024-07-31T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfab8ec492", "title": "Circuits Updates — June 2024", "url": "https://transformer-circuits.pub/2024/june-update/index.html", "published_at": "2024-06-28T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfab980626", "title": "Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet", "url": "https://transformer-circuits.pub/2024/scaling-monosemanticity/index.html", "published_at": "2024-05-21T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfabfbf7ab", "title": "Circuits Updates — April 2024", "url": "https://transformer-circuits.pub/2024/april-update/index.html", "published_at": "2024-04-26T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfac3fbe8c", "title": "Circuits Updates — March 2024", "url": "https://transformer-circuits.pub/2024/march-update/index.html", "published_at": "2024-03-21T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfad0c413a", "title": "Reflections on Qualitative Research", "url": "https://transformer-circuits.pub/2024/qualitative-essay/index.html", "published_at": "2024-03-08T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfadfa1c7c", "title": "Circuits Updates — February 2024", "url": "https://transformer-circuits.pub/2024/feb-update/index.html", "published_at": "2024-02-23T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfae6856f5", "title": "Circuits Updates — January 2024", "url": "https://transformer-circuits.pub/2024/jan-update/index.html", "published_at": "2024-01-23T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfaeb0d08e", "title": "Towards Monosemanticity: Decomposing Language Models With Dictionary Learning", "url": "https://transformer-circuits.pub/2023/monosemantic-features/index.html", "published_at": "2023-10-04T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfaf5616c1", "title": "Circuits Updates — July 2023", "url": "https://transformer-circuits.pub/2023/july-update/index.html", "published_at": "2023-07-18T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb02882b2", "title": "Circuits Updates — May 2023", "url": "https://transformer-circuits.pub/2023/may-update/index.html", "published_at": "2023-05-24T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb0d87dbd", "title": "Interpretability Dreams", "url": "https://transformer-circuits.pub/2023/interpretability-dreams/index.html", "published_at": "2023-05-24T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb113a472", "title": "Distributed Representations: Composition & Superposition", "url": "https://transformer-circuits.pub/2023/superposition-composition/index.html", "published_at": "2023-05-04T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb1ac7120", "title": "Privileged Bases in the Transformer Residual Stream", "url": "https://transformer-circuits.pub/2023/privileged-basis/index.html", "published_at": "2023-03-16T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb2804897", "title": "Superposition, Memorization, and Double Descent", "url": "https://transformer-circuits.pub/2023/toy-double-descent/index.html", "published_at": "2023-01-05T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb2c3fa61", "title": "Toy Models of Superposition", "url": "https://transformer-circuits.pub/2022/toy_model/index.html", "published_at": "2022-09-14T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb368a349", "title": "Softmax Linear Units", "url": "https://transformer-circuits.pub/2022/solu/index.html", "published_at": "2022-06-27T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb405db33", "title": "Mechanistic Interpretability, Variables, and the Importance of Interpretable Bases", "url": "https://transformer-circuits.pub/2022/mech-interp-essay/index.html", "published_at": "2022-06-27T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb4da8a91", "title": "In-Context Learning and Induction Heads", "url": "https://transformer-circuits.pub/2022/in-context-learning-and-induction-heads/index.html", "published_at": "2022-03-08T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb636cb3a", "title": "PySvelte", "url": "https://github.com/anthropics/PySvelte", "published_at": "2021-12-23T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb5b04faf", "title": "A Mathematical Framework for Transformer Circuits", "url": "https://transformer-circuits.pub/2021/framework/index.html", "published_at": "2021-12-22T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb67af29f", "title": "Garcon", "url": "https://transformer-circuits.pub/2021/garcon/index.html", "published_at": "2021-12-22T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb5d8675c", "title": "Exercises", "url": "https://transformer-circuits.pub/2021/exercises/index.html", "published_at": "2021-12-17T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb5ed94b2", "title": "Videos", "url": "https://transformer-circuits.pub/2021/videos/index.html", "published_at": "2021-12-17T00:00:00+00:00" }, { "id": "01a08219-ddaf-7384-bcb4-f2bfb74d718c", "title": "Original Distill Circuits Thread", "url": "https://distill.pub/2020/circuits/", "published_at": "2020-03-10T00:00:00+00:00" } ] posts Claim your blog

Back to Transformer Circuits

Blog · corpus.blog/blogs/transformer-circuits.pub/posts

Transformer Circuits

transformer-circuits.pub

2026

Characterizing interference weights in a tiny language model original ↗

21 Aug 2026

Verbalizable Representations Form a Global Workspace in Language Models original ↗

6 Jul 2026

Circuits Updates — June 2026 original ↗

30 Jun 2026

Circuits Updates — May 2026 original ↗

1 Jun 2026

Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activations original ↗

7 May 2026

HeadVis original ↗

4 May 2026

Emotion Concepts and their Function in a Large Language Model original ↗

2 Apr 2026

2025

Circuits Cross-Post — Activation Oracles original ↗

19 Dec 2025

Circuits Updates — November 2025 original ↗

26 Nov 2025

Emergent Introspective Awareness in Large Language Models original ↗

29 Oct 2025

Circuits Updates — October 2025 original ↗

24 Oct 2025

When Models Manipulate Manifolds: The Geometry of a Counting Task original ↗

21 Oct 2025

Circuits Updates — September 2025 original ↗

29 Sept 2025

Circuits Updates — August 2025 original ↗

28 Aug 2025

A Toy Model of Mechanistic (Un)Faithfulness original ↗

7 Aug 2025

Tracing Attention Computation Through Feature Interactions original ↗

31 Jul 2025

A Toy Model of Interference Weights original ↗

29 Jul 2025

Sparse mixtures of linear transforms original ↗

25 Jul 2025

Circuits Updates — July 2025 original ↗

25 Jul 2025

Automated Auditing original ↗

24 Jul 2025

Circuits Updates — April 2025 original ↗

29 Apr 2025

Progress on Attention original ↗

28 Apr 2025

On the Biology of a Large Language Model original ↗

27 Mar 2025

Circuit Tracing: Revealing Computational Graphs in Language Models original ↗

27 Mar 2025

Insights on Crosscoder Model Diffing original ↗

19 Feb 2025

Circuits Updates — January 2025 original ↗

27 Jan 2025

2024

Stage-Wise Model Diffing original ↗

11 Dec 2024

Sparse Crosscoders for Cross-Layer Features and Model Diffing original ↗

25 Oct 2024

Using Dictionary Learning Features as Classifiers original ↗

16 Oct 2024

Circuits Updates — September 2024 original ↗

1 Oct 2024

Circuits Updates — August 2024 original ↗

6 Sept 2024

Circuits Updates — July 2024 original ↗

31 Jul 2024

Circuits Updates — June 2024 original ↗

28 Jun 2024

Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet original ↗

21 May 2024

Circuits Updates — April 2024 original ↗

26 Apr 2024

Circuits Updates — March 2024 original ↗

21 Mar 2024

Reflections on Qualitative Research original ↗

8 Mar 2024

Circuits Updates — February 2024 original ↗

23 Feb 2024

Circuits Updates — January 2024 original ↗

23 Jan 2024

2023

Towards Monosemanticity: Decomposing Language Models With Dictionary Learning original ↗

4 Oct 2023

Circuits Updates — July 2023 original ↗

18 Jul 2023

Circuits Updates — May 2023 original ↗

24 May 2023

Interpretability Dreams original ↗

24 May 2023

Distributed Representations: Composition & Superposition original ↗

4 May 2023

Privileged Bases in the Transformer Residual Stream original ↗

16 Mar 2023

Superposition, Memorization, and Double Descent original ↗

5 Jan 2023

2022

Toy Models of Superposition original ↗

14 Sept 2022

Softmax Linear Units original ↗

27 Jun 2022

Mechanistic Interpretability, Variables, and the Importance of Interpretable Bases original ↗

27 Jun 2022

In-Context Learning and Induction Heads original ↗

8 Mar 2022

2021

PySvelte original ↗

23 Dec 2021

A Mathematical Framework for Transformer Circuits original ↗

22 Dec 2021

Garcon original ↗

22 Dec 2021

Exercises original ↗

17 Dec 2021

Videos original ↗

17 Dec 2021

2020

Original Distill Circuits Thread original ↗

10 Mar 2020