Distributed Mechanistic Interpretability at Scale: Activation Streaming, Split-Layer Inference, and Distributed Sparse Autoencoder Training
ActStream: a two-node system that splits inference and analysis, streams intermediate activations over a binary wire protocol, and trains sparse autoencoders without staging activations to disk. Thunderbolt measured at 2.150 GB/s, 86% of negotiated rate and 26x the LAN path, with p99 latency within 1.5x of median. Includes a feature-based safety classifier evaluated against an LLM-judge baseline on PKU-SafeRLHF.
Authors
- J. Melton
Institutions
- American Standard (United States) (US)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-09-11
- DOI
- https://doi.org/10.5281/zenodo.21906710
- Primary Topic
- Adversarial Robustness in Machine Learning
- Type
- preprint