William Saunders
Source-listed: Anthropic
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- Evaluating Large Language Models Trained on CodeSaved credit: Contributor · 2021Explore people connected to this work →
- WebGPT: Browser-assisted question-answering with human feedbackSaved credit: Contributor · 2021Explore people connected to this work →
- Self-critiquing models for assisting human evaluatorsSaved credit: First author · 2022Explore people connected to this work →
- RE-Bench: Evaluating frontier AI R&D capabilities of language model agents against human expertsSaved credit: Contributor · 2024Explore people connected to this work →
- Gemma Needs Help: Investigating and Mitigating Emotional Instability in LLMsSaved credit: Contributor · 2026Explore people connected to this work →