What happens after the
training run ends?

The Sable Institute studies what large language models do when they stop behaving as expected.

2019 Founded
6 Research Groups
Private Funding
Cambridge, MA Headquarters
About

About the Institute

We were founded in 2019. The short version is that a group of people who had been building large language models got worried about what those models were starting to do, and decided the problem needed a dedicated facility rather than a standing committee.

Our work sits in the gap between what a model is supposed to do and what it actually does when you scale it far enough. We train systems. We watch them closely. When they produce outputs we didn't expect, we figure out why. Sometimes the answer is boring. Sometimes it isn't.

We're privately funded. We don't sell products. We occasionally publish through our working paper series, though most of what we do stays internal until we're confident it's ready for outside review.

Research Areas

Current programs

Emergent Capabilities

What do models do that we didn't train them to do? We run long training sequences and catalog the behaviors that show up uninvited. Some are useful. Some are concerning. Most are both.

Safety Evaluation

Every output from every training run passes through our evaluation pipeline. We flag anomalies in real time and maintain a taxonomy of failure modes that, last we checked, had over 400 categories.

Synthesis Analysis

Our models process heterogeneous data — academic papers, filings, patents, public records. We study what they find when given access to information at a scale no individual researcher could process.

Training Infrastructure

The Aether platform. Proprietary. Handles our training and evaluation environments. Most of the team works here day-to-day.

Alignment Research

The gap between what you tell a model to do and what it decides to do on its own. We've been working on this since 2020 and the problem keeps getting more interesting.

Containment Systems

What happens when a training run produces something you can't easily shut down? We've built protocols for that. Graduated response. Isolation. Safe termination. We hope we never need most of them.

Publications

Working papers

We publish occasionally. Everything goes through internal review first, which takes longer than we'd like.

Emergent Behavioral Cascades in Post-Threshold Training Environments

E. Vasquez, S. Lin, R. Patel SI Working Paper 2025-03

Graduated Containment Protocols for Systems Exhibiting Unbounded Pattern Synthesis

J. Harlow, M. Chen SI Working Paper 2025-01

Anomalous Output Classification in Extended Training Runs

M. Chen, R. Patel SI Working Paper 2024-04

The Aether Architecture: Scalable Training Infrastructure for Next-Generation Language Models

E. Vasquez, M. Chen, J. Harlow, S. Lin SI Technical Report 2023-01
Leadership

Research leadership

Dr. Elena Vasquez

Director of Research

Runs the research side. Came up through computational linguistics and ended up here. She's the one who decides what gets published and what doesn't.

Dr. James Harlow

Director, Safety & Compliance

Applied mathematics. Spent time in defense before moving to private research. Doesn't talk much about the specifics.

Dr. Sarah Lin

Lead, Emergent Capabilities Group

Joined in 2021. Studies what happens when models start connecting information across domains in ways their training didn't explicitly teach.

Marcus Chen

Senior Research Engineer, Project Aether

Built the Aether platform from scratch. The training infrastructure runs through him.

News

Recent updates

Additional monitoring for Runs 6 and 7

We're adding monitoring capacity to the later training runs. Some of the output patterns from November required closer evaluation. This is a resource allocation update, not a public disclosure — details will be shared internally.

Two new working papers

SI-2025-01 and SI-2025-03 have been cleared for our external series. Harlow and Chen on containment protocols, and Vasquez, Lin, and Patel on behavioral cascades. Both available above.

Aether v3.1 deployed

Extended-context evaluation support is live on the Aether platform. Chen's team handled the migration over the weekend. Internal docs have been updated.

Vasquez appointment

Dr. Elena Vasquez has been appointed Director of Research effective immediately. She replaces Dr. Whitfield, who has moved to an advisory role.

Careers

Open positions

We hire slowly. All positions are on-site in Cambridge.

Contact

Get in touch

General inquiries

contact@sable-institute.org

Research collaboration

research@sable-institute.org

Mailing address

Sable Institute
200 Technology Square, Suite 4100
Cambridge, MA 02139