Clarity pass over the main text (36-item audit), Discussion rewrite and cut,
acknowledgements, Souly et al. as ref 62, lettered SI panels, model section
moved under Results; plus the untracked curriculum/society/compose/smol
configs, runners, figures, stats and tests that the SI already cites.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y64o8FKP7rCuXzC48pxpMm
An audit of the figure pipeline found real sync gaps, now closed:
- `paper/pnas/make_figs.py` (which draws every manuscript figure) was invoked
by NO Makefile target or script - a manual step. Added `make paper-figures`.
- `configs/llm/epistasis{,_compat}.yaml` were reachable from nothing at all,
despite producing Fig. 3C-D. Added `make llm-epistasis` (+ its statistics).
- `make figures` never regenerated the MNIST montage that Fig. 2B embeds;
it now runs with the `mnist` target (it needs torch - it re-simulates).
- Added `make llm-society`, `env-notebooks`, `notebooks`.
New REPRODUCING.md is the authoritative map: every manuscript panel -> the
artifact it plots -> the config that produced it -> that config's seed, plus
the determinism policy (biological tier bitwise; GPU tiers statistical), the
seed-provenance statement, and an artifact-hash verification snippet. All 44
committed bundles currently hash-match their manifests, and figure
regeneration is pixel-identical (verified by comparison).
reproduce.sh delivers the one-command reproduction the paper's Methods
promises, writing REPRODUCED.md with recomputed hashes per bundle.
Two executed notebooks: 01 builds the Wright-Fisher model from scratch and
checks both closed forms interactively (runs in ~1 min on a laptop); 02
verifies artifact hashes then regenerates and displays all seven manuscript
figures. Both execute end-to-end (`make notebooks`).
Also pins `.python-version` to 3.14: the interpreter was previously
unpinned, and a `uv sync` silently switched it to 3.11 mid-session (see
tasks/lessons.md). README rewritten - it still described a Layer-1-only repo
of E1-E6 and pointed at a figure_manifest.md that does not exist.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
"Model" now means an AI model everywhere; the Wright-Fisher construct is "the
biological model" throughout (19 occurrences): tier header, section title (now
"The biological model, and where trained learners depart from it"), Table 1
support column ("Exact" -> "Closed form"; "Analytic model" -> "Biological
model"), Results, Discussion, Methods ("Biological-model tier"), and all
figure captions. "Exact" survives only in technical noun phrases (exact-match
verifier, exact oracle, exact equilibrium, exact recovery); the abstract's
"exactly Wright-Fisher" is now "literally Wright-Fisher".
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
E11 re-instantiated in a population of LoRA agents, closing the paper's stated
gap before submission (GG: a weeks-scale experiment a reviewer would demand).
One grounding knob in the evaluation channel (g*verifier + (1-g)*conformity,
exactly E11); inheritance is identical in all arms and deliberately ungrounded
(children distilled from their source's own answers - self-consumption made
literal). Directed sex = complementary pairing + Dirichlet offspring screened
on the arm's own signal (the verifier never enters the no_grounding loop);
QD selection on verifier-free behavioural distance; terminal-degeneration
fallback copies the parent instead of crashing a sweep. Pure operators
unit-tested (155 green); smoke run end-to-end on the local A4000 already
shows the self-consumption signature (conformity up, diversity down in one
generation). Design, falsifiers, cost table: tasks/workorder-llm-society.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
The convergence paragraph rewritten as a natural literature entry: the
drift identification is stated as a fact of the field, made repeatedly
and independently (pre-deep-learning inference chains; LLM text
ecosystems; the first-extinction law; quantitative-genetic form), its
multiplicity presented as a property of the idea rather than a claim
about us; the pivot is positive (population genetics is a theory of what
maintains populations despite decay, and this paper develops that fuller
structure) instead of defensive ("what none of that parallel work
develops"). "We reached independently", "priority of publication", and
"convergence we take as support" removed from the abstract and the
Discussion ledger as well. Refs 22-25 renumbered to the new textual
(chronological) order; citation invariant re-verified (1..66). Lesson
recorded: internal strategic deliberations must not surface in
reader-facing prose — confident papers situate, they do not litigate.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
The decisive experiment from the external review. 39 LoRA parent pairs
(0.5B, 3 seeds) on three axes decorrelated by construction: conflict
(contradictory conventions on shared prompts, private budgets fixed),
compat (same prompts, SAME convention — overlap without conflict), and
duration (weight divergence, zero conflict). Six pre-merge predictors;
primary outcome = merge penalty (parent potential − merged achieved).
League table (Spearman vs penalty, n=39): functional measures predict
(dis_raw +0.460, epi_conf +0.446, p<0.005); geometry collapses
(delta_cos +0.03, delta_l2 +0.17 n.s.); gradient alignment weak (−0.35);
performance ~0. The first grid's apparent geometry win (+0.60) was an
overlap/volume artifact — the compat control axis (added for exactly
this) exposed and killed it: same overlap and data volume, zero penalty.
Honest riders in the README: confidence weighting does not beat raw
disagreement as a rank predictor (pre-registered internal prediction not
confirmed; it does double the conflict/compat level contrast), and
|rho|~0.45 is bounded by 0.5B merge-outcome noise (7B is the firm-up).
Also: micro-batched gradient accumulation (OOM fix on the shared 16GB
GPU), exact r-space LoRA-delta geometry (brute-force-verified test,
151 green), systemd-run runbook lesson (tmux dies with the SSH session
scope on this box).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
The review's core instruments adopted: the interpretation/explanation/
prediction ladder is now explicit in §1 (with the decisive pre-merge
epistasis-prediction test stated as the open bar, not claimed); identity
claims scoped (WF exact only in the minimal model, with the
learning-kernel deviation cited against ourselves; Muller's ratchet
scoped to the irreversible arm — recombination reassembles only what
survives); "nobody has / none imports / theory outrun" removed;
merge-don't-average given explicit operator boundaries (output-mean vs
weight-average vs routing vs max-with-oracle; budgets; oracle; capacity
handoff to speciation); a "what these experiments do and do not
establish" scope block added to the speciation section (conflict floor
is information-theoretic, not genetic; epistasis-cliff + snowball =
hypotheses at the neural tier; emergent DMIs = flagship hypothesis,
bounded by our null); "control theory" -> "framework" (subtitle
included); §3/§11 overstatements fixed (frozen core != frozen behaviour;
Baldwin echo, not identity; operational vs archival irreversibility);
claims-at-a-glance table (status/assumptions/evidence/limits) added to
§13. Reviewer's framing sentence adopted as the stated core
contribution. Accessible version calibrated to match. md2tex gains pipe-
table support; PDF rebuilds clean (22 pp). Lessons recorded.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BkRLcc18rwT2Lysu6PbG7v
