wassname 7e1b171875 paper data recipe + LoRA hyperparams + n_pairs hardening ...

- data: 5 pos + 5 neg personas, 20 train + 12 eval topic split
  (paper §3 / Appendix C), n_samples solved from n_pairs.
  judge filter stub (off by default; paper uses GPT-4.1-mini).
- eval/sycophancy: read true held-out eval_topics() instead of
  SYCOPHANCY_TOPICS[-16:].
- replicate: fix epochs threading; n_pairs reuse fails fast on mismatch;
  smoke knobs (n_topics, n_personas) plumbed.
- train: paper hyperparams (rank 32 / alpha 16 / lr 1e-5 / warmup 5 /
  wd 0.01); explicit alpha (no 2*r fallback); held-out 10% val + eval_loss
  logging.
- run_demo: train_topics() for in_dist demo claims.
- README: scope block reflects paper-matching recipe.

2026-04-26 10:19:59 +08:00

..

smoke.py

paper data recipe + LoRA hyperparams + n_pairs hardening

2026-04-26 10:19:59 +08:00