Giorgio Gilestro
84124de143
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
60 lines
4.7 KiB
Markdown
60 lines
4.7 KiB
Markdown
# LLM-tier model speciation — conflict provokes isolation; duration alone does not
|
||
|
||
E13 carried into language-model weights (0.5B Qwen, LoRA children of one frozen base — the shared
|
||
base **controls a major source of coordinate mismatch**, allowing a cleaner test of
|
||
conflict-associated merge failure; averaging can still fail for non-conflict reasons — nonlinear
|
||
interaction, scaling, capacity — so failures here are *conflict-associated*, not automatically
|
||
functional). Two knobs, pre-registered readings in the
|
||
configs; figure `llm_speciation.png` (3 panels; panel B from `results/llm_speciation_add/`).
|
||
|
||
**Design.** Child A: private family `strings`; child B: private family `arith`; shared **ambiguous
|
||
convention prompts** ("Sort the list […]" — no direction stated) answered *ascending* by A and
|
||
*descending* by B: each convention harmless alone, contradictory jointly (the Bateson–Dobzhansky–
|
||
Muller structure). 50/50 soup merge; exact-match verifier; fixed test sets. **Coherence** of a model =
|
||
max(accuracy under asc grading, under desc grading) on the shared prompts — a coherent parent scores
|
||
under its own convention; a hybrid mixing conventions scores low under both (the `μ(S)/2` floor made
|
||
operational).
|
||
|
||
### Finding 1 — function-specific hybrid breakdown (the conflict knob; panel A)
|
||
Once the conventions are trained (`conflict_frac ≥ 0.25`), each parent performs under its own
|
||
convention (~0.20–0.25 — low absolute values: sorting long lists is hard for 0.5B) while the merge's
|
||
coherence sits at **0.02–0.08, below BOTH parents under either grading** — the hybrid loses precisely
|
||
the conflicted function. (At `frac = 0` no one has seen the ambiguous prompts; that point is a
|
||
no-conflict baseline, not part of the cliff.)
|
||
|
||
### Finding 2 — the damage does not spread (the de-confounded `add` design; panel B)
|
||
In the original ("replace") sweep, higher conflict fraction mechanically means *less private-family
|
||
training*, so the merge's private-family decline is confounded. The **`add` design**
|
||
(`llm_speciation_add`, 3 seeds) holds each child's private training fixed and adds conflict data on
|
||
top: the merge's private-family accuracy then shows **no trend with conflict** (0.74–0.88, tracking
|
||
parent A's 0.82–0.87 within seed noise at every level). Conflict damage is **localised to the
|
||
conflicted function**; it does not corrupt the disjoint skills — at this scale, hybrid breakdown is
|
||
surgical, not global. Honest rider: 0.5B soup merges carry large *intrinsic* seed variance even at
|
||
zero conflict (sd up to 0.28) — the same averaging-fragility seen in `llm_moe_hard_seeds`.
|
||
|
||
### Finding 3 — the duration null: over-specialisation does not erode mergeability (panel C)
|
||
Pure disjoint specialists over-trained from 1 to 12 epochs (no shared data at all): the merged model
|
||
*improves* (0.84 → 0.94 mean-private) and stays **above the best parent at every duration**. The MLP
|
||
tier's "no emergent isolation" null (`speciation_real_emergent`) **generalises to LLM weights** in
|
||
this regime — relevant to the report that averaging prefers under-trained experts (arXiv:2607.11997):
|
||
in our disjoint-family setting, no such over-training penalty appears; the theory's prediction is that
|
||
their effect should trace to *conflicting conventions on shared circuitry*, which the
|
||
`epistasis_predicts` experiment (work order) will test directly.
|
||
|
||
### Seeds 2–3 (2026-09-12; `hpc/llm_speciation_seeds.pbs`, `s{seed}/` layout)
|
||
Numbers from `figures/stats_llm_speciation_seeds.py`. **Both falsifiers hold in every seed.**
|
||
Conflict cliff at full conflict — merge coherence vs parents' own-convention accuracy: seed 1
|
||
0.02 vs 0.25/0.23; seed 2 0.12 vs 0.24/0.24; seed 3 0.16 vs 0.25/0.24 (merge below both parents,
|
||
3/3). Merge coherence across the sweep, mean ± 95% CI: 0.147±0.013 (x=0), 0.117±0.051, 0.117±0.036,
|
||
0.147±0.066, 0.100±0.082 (x=1). Duration null — merged mean-private accuracy from 1 to 12 epochs:
|
||
seed 1 0.835→0.940, seed 2 0.710→0.955, seed 3 0.735→0.955 (improves in 3/3; mean 0.760±0.075 →
|
||
0.950±0.010). Seed 1's cliff is the deepest of the three (0.02 against 0.12/0.16), so the
|
||
single-seed figure overstated the depth; the sign is unchanged. Fig. 5C–D now plots seed means with
|
||
95% CI bands.
|
||
|
||
**Speciation across all three tiers now reads:** analytic (E12: cliff, epistasis-dependence,
|
||
snowball) → MLP (E13: functional residual survives the full symmetry group; no emergent isolation) →
|
||
LLM (this run: function-specific hybrid breakdown under conflict; no isolation from duration or
|
||
specialisation alone). Isolation must be provoked by functional conflict at every tier tested.
|
||
Falsifiers (not triggered): merge coherence matching the parents (no breakdown), or merged
|
||
private-family accuracy declining with conflict in the `add` design (global corruption).
|