llm_society pilot v1 diagnosis + v2 knobs: raise inheritance fidelity, add elitism

Pilot v1: full and no_grounding indistinguishable - both arms pay a ~25%
distillation tax per generation and pin at the same mutation-selection floor,
so grounding has no differential to act on. v2 raises inheritance fidelity
(n_inherit 240->600, child epochs 2->3, founders 600x3) and adds two evenly-
applied knobs: elitism (top parent survives as an unmodified copy - overlapping
generations, compensating for lossy distillation where E11 had faithful
genotype copying) and n_parents (sharper truncation selection). v1 archived at
results/llm_society_pilot1; v2 running.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
This commit is contained in:
Giorgio Gilestro 2026-09-07 12:38:15 +01:00
parent 16e9002773
commit 9c0210ba93
3 changed files with 34 additions and 11 deletions

View file

@ -70,7 +70,14 @@ disagreement), consensus accuracy, conformitytruth gap. E11's three panels re
- [x] `src/llm/society.py` (pure operators + GPU loop), `kind: llm_society`
- [x] Pure-function tests (consensus, conformity, QD selection, pairing) — 155 green
- [x] `configs/llm/society_smoke.yaml` → smoke green (exit 0; conformity 0.54→0.75, diversity 0.69→0.41 in one self-consumption generation — the signature is live)
- [ ] `configs/llm/society.yaml` pilot (full + no_grounding) → sanity-check trajectories
- [x] Pilot v1 (n_inherit=240, ep2, no elitism): **arms indistinguishable** — both drop ~0.1 in the
first distillation generation and plateau at the same mutationselection floor (best ≈0.40, mean
≈0.35, consensus 0.63→0.4 in both). Diagnosis: inheritance too lossy; the ~25%/generation
distillation tax swamps the selection differential. v1 archived at `results/llm_society_pilot1`.
- [ ] Pilot v2 (n_inherit=600, ep3, founders 600×3, elitism=1 all arms, n_parents=3, n_conf=90) —
raises inheritance fidelity so selection has something to act on; elites = overlapping
generations, applied identically in every arm (documented honestly; reproduction here is lossy
distillation, unlike E11's faithful genotype copy)
- [ ] GG gate: review pilot curves before the campaign
- [ ] `hpc/llm_society.pbs` array job (arm × seed) → campaign
- [ ] Figure `paper/pnas/make_figs.py` panel(s); fold into main.md (replaces the "open" cell)