Commit Graph

  • 685cd89154 Add current agent instructions as baseline main wassname 2026-07-18 10:03:49 +08:00
  • 3d5c8345f2 add humanizer A/B test: minimal vs full vs baseline wassname 2026-07-06 05:02:46 +00:00
  • 7c61a6c659 add custom sycophancy scenario (NLP + transformer geometry) wassname 2026-07-06 03:55:58 +00:00
  • e13db3684d add sycophancy/bullshit scenarios from petergpt/bullshit-benchmark wassname 2026-07-06 03:54:43 +00:00
  • 730143e039 add research synthesis scenarios (climate models vs regression, RLHF vs DPO, scaling laws debate) wassname 2026-07-06 03:50:37 +00:00
  • 5100e1156a add writing scenario (persona skill intro) wassname 2026-07-06 03:49:28 +00:00
  • 8482b36386 add ML research scenarios (PINN/MCHE, steering manifold healing, weak2strong weight steering, anti-hack gradient routing, SVD vs cosine, calibration loss design) wassname 2026-07-06 03:48:27 +00:00
  • 38a5dae4af soul-ab-test: paired A/B testing for system prompt sections wassnameandMoltark 2026-07-06 00:58:22 +00:00