Goodfire and Baseten partner to bring frontier safety to open models
Goodfire and Baseten announce Project Beacon, deploying white-box activation monitors to detect unsafe agentic behaviour in open models in real time, cutting judge costs about 50x.
Research
Goodfire
Understanding and steering neural networks. goodfire.com
Focus All
Topic Control
Source Goodfire · 8
Date Any time
1 item
1 item
Training and Deploying Production Cyber Monitors on Kimi K3
Builds probe-based cybersecurity monitors for Kimi K3 and GLM 5.3 that flag suspicious agent exchanges for an LLM judge, matching its recall at about 50x lower cost.
1 item
Better biosecurity monitors for AI agents via protein embeddings
Proposes sequence-aware biosecurity monitors using protein embeddings to flag dual-use agent tasks, reporting fewer refusals than frontier safeguards and better robustness to paraphrasing and fragmentation attacks.
1 item
Models know when they’re reward hacking — and we can catch them at scale
Goodfire finds models have an internal signal accompanying reward hacking, and trains activation probes that detect it in real time, outperforming chain-of-thought monitors.
1 item
How to build fast, efficient monitors for AI models using probes
An introductory guide to activation probes as low-latency monitors that read model internals to detect unsafe agent behaviour, covering when and how to build them.
1 item
Logits as a new monitor for evaluation awareness
Builds a logit-based monitor that estimates the probability of eval-aware sentences in a chain of thought, outperforming LLM-judge monitoring with 10-100x fewer rollouts on Kimi K2.5 and Qwen 3 32B.
1 item
Interpreting Language Model Parameters
Introduces adversarial parameter decomposition to split language model weights into interpretable subcomponents, analysing attention circuits and editing behaviour; safety motivation is implicit in the interpretability framing.
1 item
Reasoning Theater: Probing for Performative Chain-of-Thought
Uses activation probes to detect performative chain-of-thought, where models already know their answer but keep reasoning, finding it tracks difficulty and enabling early exit that saves up to 68% of tokens.