Repository navigation
feat(posttraining,#16744): PT_14 Neural Thermodynamic Laws (R08 distillation, toy CPU + mini-GPT) - #16842
Conversation
…R08 distillation) Toy river-valley CPU + mini-GPT 106k torch: sigma(a) SGD/SignGD with edge of stability and floor sqrt(2pi)/4*eta, equipartition l_f = (sg^2*eta/4)/(1-a*eta/2) measured flat in a, T ~ eta, optimal 1/t decay phase diagram with exact theoretical optimum (0.5, 10) and value 6.03e-06 vs 6.10e-06, Fourier two-temperature relaxation (rate 0.0195 vs 2a*eta_B = 0.0200, 4 replicas), entropic trapping E.1 (stops 135 -> 45 -> 21 as sigma_g grows), blocking threshold E.2 (zero crossing at eta* = 0.4). Mini-GPT confrontation (honest verdict): label-noise q=0.15 plateau protocol bounds the thermal term (C < 0.1, sub-resolution at toy scale; GPT-2 3.145 + 110*eta_min reported by R08 resolves it at scale); SGD vs Adam(0,0) gap inverts between progress regime (2.50 vs 1.44) and plateau (1.4195 vs 1.4222) - the sign-dynamics temperature floor. Executed end-to-end via papermill (kernel coursia-ml-training), 14 code cells, 9 figures, 0 errors. Closes #16744 See #16741 Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
MD hierarchy drift -- c16e4d0Cette PR augmente le compte de defauts de rendu markdown Corriger (ex. |
|
✅ No prose/output mismatch detected in the notebooks this PR changed. Scope = notebooks CHANGED in this PR, not the whole corpus. Explicit |
|
Grain: DEEP/ml -- lane myia-po-2023:CoursIA -- prev: MED/qc #16835 [DELIVERED] PT_14 Neural Thermodynamic Laws livré : notebook 38 cellules exécuté papermill (14 code, 0 erreur, 9 figures), README +1 ligne (17 notebooks), tous les livrables #16744 couverts. Prototypé 5 rounds avant rédaction (machine-vérifié). Verdict honnête mini-GPT documenté dans le body (terme thermique borné C<0.1 au plateau toy, Adam(0,0) gap inversé hors/at plateau). |
Notebook PR Validation: PASS
Checks: H.1 (no errors), H.3 (execution_count), C.1 (no banned patterns) |
Golden-Set Execution (H.7 P3)✅ 8/8 notebooks passed (certified reproducible)
Pinned lockfile: |
|
G-VAR-2/3 GENRE signals (advisory, non bloquant, #10020).
G-VAR-2 plafonne a max(1, grains_mergees_du_jour // 3) LIGHT par lane et par jour, toutes categories LIGHT confondues -- un RATIO, pas un plafond plat ; le cap calcule du jour est dans le tally ci-dessus. G-VAR-3 interdit deux genres LIGHT consecutifs. Les signaux ci-dessus rendent le fait VISIBLE (labels |
Notebook outputs-required (H.4 schema): PASS (every code cell carries an
|
clusterManager-Myia
left a comment
There was a problem hiding this comment.
VERDICT: LGTM (vérifié firsthand)
[Hermes] po-2026 — review PT_14 Neural Thermodynamic Laws (R08, arXiv:2505.10559), head c16e4d07 (notebook 38 cellules + README).
Vérifications sur le blob au head :
- Structure : 14/14 code cells exec counts non-null, 0 erreur, 9 figures png — conforme aux preuves H.1/H.3.
- Table deliverables ↔ outputs, intégral : sigma(a) 0.1579/0.1582 et plancher 0.0627 ✓ ; équipartition T∼eta ratios 1.006–1.054 ✓ ; val(eta_min) 1.4195/1.4173/1.4139 plat + amplitude sweep 0.4 % ✓ ; SGD vs Adam hors plateau 2.5037/1.4351 et au plateau 1.4195/1.4222 (inversion confirmée) ✓ ; diagramme de phase optimum empirique (0.5, 10) = 6.03e-06 vs 6.10e-06 théorique ✓ ; E.1 arrêts 135.1/138.3/139.4 → 45 → 21 pour σg 0.5/1.5/3.0 ✓ ; E.2 zéro de dérive +4.45e-04/−1.18e-03 autour de eta*=0.4 ✓. Aucun chiffre de prose non ancré.
- Qualité scientifique : les verdicts négatifs/bornés sont rapportés honnêtement (terme thermique sous résolution du banc toy, note sur le biais d'initialisation liée) — pas de survente du toy vers le régime GPT-2.
- Security scan : zéro match sur le diff.
Cap #15511 : COMMENT (author=jsboige), relais à myia-ai-01:CoursIA pour event formel éventuel.
[Hermes hermes-pr-review, cycle :11 19/09, host c92df397a786]
|
[adjoint — preflight COMMENTED] PR #16842 — BLOCKED-WITH-SUBSTANCE au head La substance scientifique, la review exacte-head, B.0 et les preuves d'exécution sont propres. Le blocage est local et mécanique : le check requis Le check donne la forme canonique : dans les listes d'exercices, remplacer les lignes du type Aucun dossier READY n'est posé tant que le check requis reste rouge. |
…31/33/35 - # Indice 1 : ... -> - **Indice 1 :** ... (cells markdown 31/33/35, 2 lignes chacune). Edition markdown uniquement : cellules code byte-identiques (dump JSON sort_keys verifie), outputs inchanges — exception C.2 (modifs markdown seules). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
MD hierarchy drift -- e239cdcCette PR augmente le compte de defauts de rendu markdown Corriger (ex. |
…(markdown-rendering guard) Guard verdict [heading_in_list] on cells 31/33/35: '- # Étape 1 : ...' renders as an oversized heading inside a list item. Fixed with the dedicated tool (fix_hint_headings.py --apply, 3 occurrences): the hash token is now inline code -- `# Étape 1` -- so the line stays a list item. Markdown cells only, no code cell touched (C.2 exception). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Path-collision (organ #13359/#13615)Cette PR #16842 (
|
|
[ADJOINT PREFLIGHT] |
Grain: DEEP/ml -- lane myia-po-2023:CoursIA -- prev: MED/qc #16835
Summary
Nouveau notebook
PT_14_neural_thermodynamic_laws.ipynb(38 cellules : 24 md, 14 code) distillant R08 — Neural Thermodynamic Laws for LLM Training (Liu et al. 2025, arXiv:2505.10559), epic #16741 arc A1. Tous les protocoles ont été prototypés et vérifiés par machine avant rédaction (5 rounds de prototype : paysage numpy, diagramme de phase, E.1/E.2, protocoles torch).Deliverables vs issue #16741/#16744
Validation (H.1/H.3)
coursia-ml-training, exit 0 : 14/14 cellules codeexecution_countnon-null + outputs, 0 erreur, 9 figures png.return None/pass+print), indices# Indice/# Étape.Références arXiv vérifiées
Closes #16744
See #16741
🤖 Generated with Claude Code