← People

Ben Mann

Source-listed: Anthropic

Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.

Selected work

5
  1. Constitutional AI: Harmlessness from AI FeedbackSaved credit: Contributor · 2022Explore people connected to this work →
  2. Training a Helpful and Harmless Assistant with Reinforcement Learning from Human FeedbackSaved credit: Contributor · 2022Explore people connected to this work →
  3. Discovering Language Model Behaviors with Model-Written EvaluationsSaved credit: Contributor · 2023Explore people connected to this work →
  4. Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons LearnedSaved credit: Contributor · 2022Explore people connected to this work →
  5. Scaling Laws and Interpretability of Learning from Repeated DataSaved credit: Contributor · 2022Explore people connected to this work →