Person

Jason Wei

2 entries · January 2022 – June 2022

Appears alongside

Featured in threads

Tracks

  • Ideas & essays 2
  • Benchmarks & progress 2

  1. Emergent abilities of large language models are described

    Jason Wei and co-authors catalogued tasks where accuracy jumped from near-chance to strong performance past a scale threshold, a pattern later disputed as a metric artefact.

    Ideas & essays · Benchmarks & progress

  2. Chain-of-thought prompting is described

    Asking a model to show its working improved reasoning benchmarks sharply, with no retraining — the seed of the later reasoning models.

    Ideas & essays · Benchmarks & progress

Also mentioned in 1 entry

Referenced in passing — Jason Wei isn't the main subject of these.