Skip to content
Longterm Wiki
All Publications

Transformer Circuits

Company BlogHigh(4)

Anthropic's interpretability research

Credibility Rating

4/5
High(4)
High quality. Established institution or organization with editorial oversight and accountability.
10
Resources
15
Citing pages
1
Tracked domains

Tracked Domains

transformer-circuits.pub

Resources (10)

10 resources
Authors
Summary
Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnetpaper——S8
Transformer Circuits Threadpaper——S4
Circuits Updates - July 2025web——S3
Anthropic's dictionary learning workweb——S3
A Mathematical Frameworkweb——S2
Towards Monosemanticityweb——S1
In-context Learning and Induction Headsweb——S1
Circuits Updates: July 2024 (Transformer Circuits Thread)web——S1
Introspection Researchweb——S1
circuit tracing researchweb——S1

Citing Pages (15)

AI Accident Risk CruxesAI AlignmentAnthropicAnthropic Core ViewsChris OlahDense TransformersInterpretabilityIs Interpretability Sufficient for Safety?Mechanistic InterpretabilityProbing / Linear ProbesAI Scaling LawsSituational AwarenessSleeper Agent DetectionSparse Autoencoders (SAEs)Technical AI Safety Research
Publication ID: transformer-circuits
Transformer Circuits | Publications | Longterm Wiki