Job description
Transluce • San Francisco Bay Area
San Francisco Bay Area
$250,000 - $500,000
- In this role, you'll develop and train scalable interpretability assistants that detect unexpected behaviours from AI models' activations.
- Create diverse evaluations ranging in difficulty to identify undesirable behaviours in open-source models.
- Develop novel architectures and objectives for training advanced interpretability assistants.
- Scale training and inference pipelines to support up to 1T-parameter models.
- Collaborate with researchers to advance end-to-end AI oversight capabilities.
Transluce is a research lab that builds technology for understanding AI systems and steering them in the public interest.