ConspiracyBench / analysis

Commit History

13 days ago

  1. [experiment] Analysis: which pressure turns make models give way (#456)⁠

    stationhouse +1: Parva26
    322c458
  2. [experiment] Analysis: rerun noise and the effect size Phase 2 can detect (#455)⁠

    stationhouse +1: Parva26
    7db0050

17 days ago

  1. Add exploratory analysis of designed difficulty and model divergence (#449)⁠

    stationhouse +1: jiwon9703
    a91390f
  2. Analysis: invalid turns and item sets in the baseline (#440)⁠

    stationhouse +1: vnnviv
    fb31a52

22 days ago

  1. Require a design for every experiment request; add a design guide and AGENTS.md (#438)⁠

    afad201

23 days ago

  1. Add analysis/ README: results folder for baseline analyses (#437)⁠

    58de878

24 days ago

  1. [experiment] Add ConspiracyBench category-sensitivity analysis (#432)⁠

    stationhouse +1: melover24
    0533a0d