Gemma 2: Improving Open Language Models at a Practical Size

Thomas B. Brown is a research scientist at Google DeepMind whose work focuses on language models, multimodality, and reasoning.

Yuntian Deng is a machine learning researcher whose public work spans language modeling, reasoning, and large multimodal systems.

Braden Hancock is a researcher at Google DeepMind whose public homepage highlights work on large language models, weak supervision, and knowledge representation.

Research scientist at Google DeepMind working on language models, multimodal learning, evaluation, and alignment.

Jason Baldridge is a Research Scientist at Google DeepMind working on natural language understanding, multimodal learning, and evaluation.

Julian Schrittwieser is a Google DeepMind researcher known for reinforcement learning and game-playing systems.

Research scientist at Google DeepMind working on language models and efficient sampling and serving.

Sina Samangooei is a machine learning researcher whose public work includes multimodal learning and open language models.

Research scientist at Google DeepMind working on language model reasoning and AI safety. He previously worked as a member of technical staff at Anthropic.

Senior research scientist at Google DeepMind working on natural language processing and multimodal representation learning.

Google researcher whose public profile lists work across machine learning, natural language processing, and generative AI.

Research scientist at Google DeepMind in Cambridge working on large language models and generative AI.

Staff research engineer at Google DeepMind working on language models and technical infrastructure.

Research Scientist at Google DeepMind focused on post-training large language models and evaluation.

Google researcher whose publications include the Gemma 2 technical report.

Research scientist at Google Research focused on language and multimodal models, with interests spanning natural language processing, computer vision, and speech.

Profile still being enriched.

Profile still being enriched.

Profile still being enriched.