Research Scientist, Interpretability of LLMs

Job Description

Google DeepMind is seeking a Research Scientist to advance research in understanding large language models by considering controllability as a dual problem. This role focuses on practical interpretability, using controllability to evaluate the faithfulness of explanations. The Research Scientist will translate research findings into useful internal applications and publish findings to influence the field.

• Conduct research in understanding the model with practicality as a focus.
• Translate research findings into useful internal applications.
• Publish and present findings at internal and external venues to influence the field.

To be successful in this role, candidates should possess the following:

• PhD in machine learning, statistics, or related fields.
• Strong publication record.
• Strong record in LLMs and foundation models and related engineering skills.
• Strong end-to-end system building and prototyping skills (advantage).

Google DeepMind offers:

• Competitive US base salary range between $141,000 - $202,000.
• Bonus.
• Equity.
• Benefits.

Google DeepMind values diversity and is committed to equal employment opportunities.