Alphabell.
Oversight

Interpretability and oversight

AI used to explain, audit and supervise other AI, with the findings fed back into training.

Recently on the radar

More →
Forum postLessWrong·11 Nov 2026·Interpretability and oversight

The Chain Lens [interim]

The loop: The chain lens extracts multi-token concepts from a model's latent representations using contrastive decoding. This surfaces specific computational intermediates that can be used to oversee and refine the model's reasoning.

loop fit 6/10vgelvia LessWrong
The Chain Lens [interim]
PaperarXiv·6 Nov 2026·Interpretability and oversight

J-lens preferentially reads from coordinate-aligned subspaces

The loop: The J-lens tool identifies coordinate-aligned subspaces in Gemma-4, which improves model interpretability and editing, making future versions of Gemma easier to verify and steer.

loop fit 6/10Yubo Wang, Rui Meng, Jaehyun Nam, Tomas Pfister, Jinsung Yoonvia arXiv
J-lens preferentially reads from coordinate-aligned subspaces
Forum postLessWrong·5 Nov 2026·Interpretability and oversight

J-lens preferentially reads from coordinate-aligned subspaces

The loop: The J-lens tool identifies coordinate-aligned subspaces in Gemma-4, which improves model interpretability and editing, making future versions of Gemma easier to verify and steer.

loop fit 6/10Caleb Briggsvia LessWrong
J-lens preferentially reads from coordinate-aligned subspaces
NewsAlignment Forum·3 Nov 2026·Interpretability and oversight

Automating subtle emergent misalignment research

The loop: A Claude model is instructed to write training data that induces misalignment while bypassing a production safety monitor. The successful attacks it discovers are then used to test and improve the robustness of future monitors.

loop fit 9/10evhubvia Alignment Forum
Automating subtle emergent misalignment research
ReleaseGitHub·2 Nov 2026·Interpretability and oversight

Training activation classifiers for safety and alignment

The loop: Activation classifiers monitor a model's internal activity to detect policy violations. This improves the safety and alignment of the model, making it safer to deploy and use for further development.

loop fit 7/10codelionvia GitHub releases
Training activation classifiers for safety and alignment
Blog postOpenAI·2 Nov 2026·Interpretability and oversight

Training activation classifiers for safety and alignment

The loop: Activation classifiers monitor a model's internal activity to detect policy violations, which improves the safety and alignment of the model, which makes the model safer to deploy and use for further development.

loop fit 7/10Alex Makelov, Xiaojun Xu, Sundaram Ananthanarayanan, Ethan Mcdonald, Tom Dupre...via OpenAI Alignment
Training activation classifiers for safety and alignment

Other loop types