pith. machine review for the scientific record. sign in

Sam McCandlish

Identifiers

  • name variant Sam McCandlish 0.60 · backfill

Papers (19)

  1. Towards Understanding Sycophancy in Language Models cs.CL · 2023 · author #13
  2. Measuring Faithfulness in Chain-of-Thought Reasoning cs.AI · 2023 · author #18
  3. Towards Measuring the Representation of Subjective Global Opinions in Language Models cs.CL · 2023 · author #12
  4. Discovering Language Model Behaviors with Model-Written Evaluations cs.CL · 2022 · author #44
  5. Constitutional AI: Harmlessness from AI Feedback cs.CL · 2022 · author #49
  6. Measuring Progress on Scalable Oversight for Large Language Models cs.HC · 2022 · author #33
  7. In-context Learning and Induction Heads cs.LG · 2022 · author #25
  8. Toy Models of Superposition cs.LG · 2022 · author #12
  9. Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned cs.CL · 2022 · author #33
  10. Language Models (Mostly) Know What They Know cs.CL · 2022 · author #34
  11. Scaling Laws and Interpretability of Learning from Repeated Data cs.LG · 2022 · author #18
  12. Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback cs.CL · 2022 · author #28
  13. A General Language Assistant as a Laboratory for Alignment cs.CL · 2021 · author #20
  14. Evaluating Large Language Models Trained on Code cs.LG · 2021 · author #56
  15. Scaling Laws for Transfer cs.LG · 2021 · author #4
  16. Scaling Laws for Autoregressive Generative Modeling cs.LG · 2020 · author #19
  17. Language Models are Few-Shot Learners cs.CL · 2020 · author #28
  18. Scaling Laws for Neural Language Models cs.LG · 2020 · author #2
  19. An Empirical Model of Large-Batch Training cs.LG · 2018 · author #1

Mentions

  • 2102.01293 #4 · arxiv_oai · confidence 0.70 Sam McCandlish
  • 2205.10487 #18 · arxiv_oai · confidence 0.70 Sam McCandlish
  • 2211.03540 #33 · arxiv_oai · confidence 0.70 Sam McCandlish
  • 2306.16388 #12 · arxiv_oai · confidence 0.70 Sam McCandlish

Frequent Coauthors