# Circuits Updates – September 2024

Source: https://www.anthropic.com/research/circuits-updates-sept-2024
Published: 2024-10-01
Status: unreviewed source report

## Source excerpt (up to 25 words)

> Preliminary interpretability notes on successor heads and on oversampling topics in sparse autoencoder training for safety-relevant features.

## Review question

Check the problem size, uniformity, resource measure, reduction and approximation factor. A claim in one restricted model should not be promoted to a statement about every algorithm.

No independent proof check or claim of peer review.