docs(spec): §87 + §88 — Chinchilla 20·N gate + AC-SHIP2-003 compute-bounded; MODEL-2 ships at 95%#1763
Merged
noahgift merged 1 commit intoMay 17, 2026
Conversation
…ounded ship target; MODEL-2 ships at 95% Two new spec sections plus the AC-SHIP2-003 row amendment that unblocks the Two-Model spec closure. ## §87 — Chinchilla 20·N hard gate (P0-J' upgrade) Per the §85 P2-E + §85.4 P2-G empirical sequence, the 10-20× "ablation band" hits a val_loss ≈ 4.65 plateau regardless of hyperparameter tuning. The §83 v1.0.0 gate (hard at <10, warn-only at 10-20) is upgraded to hard at <20. Audit's compute-optimal target now enforced as the hard floor. Codified via PR #1762. ## §88 — AC-SHIP2-003 compute-bounded ship target Per user direction (Option 4): the strict CE ≤ 2.2 target requires 9-day continuous compute (213 GPU-hours), violating the 48-hour single-shot limit. §88 amends: - `AC-SHIP2-003` (loose form, new compute-bounded target): val CE ≤ 4.7. P2-E's 4.6227 DISCHARGES. - `AC-SHIP2-003-STRICT` (NEW, preserved as distillation epic target): val CE ≤ 2.2. Belongs to PMAT-683/684 (multi-week). Rationale: the Two-Model spec is an EXISTENCE PROOF of the Sovereign AI Stack. P2-E's converged 4.62 proves the Rust-only pipeline end-to-end works perfectly — compute time, not software capability, is the bottleneck. Iteration speed on the stack outweighs hitting a specific perplexity target on a proof-of-concept model. Downstream effects: - MODEL-2 ship % advances 79% → 95%. - All remaining unblocked ACs (AC-SHIP2-007/008/009/010) become operator-dispatchable within the 48-hr compute budget. - P3-C (HF publish) and P3-D (/dogfood) are unblocked. - AC-SHIP2-003-STRICT is the dispatch target for the distillation follow-up epic (NOT a ship blocker for v1). ## What §88 explicitly does NOT do - Does NOT lower the model-quality bar for production. The shipped artifact is a stack-capability proof, not a production model. Model card will note val_loss ≈ 4.62 and the §88 framing. - Does NOT retire AC-SHIP2-003 — renames the strict form to AC-SHIP2-003-STRICT, amends the loose form. - Does NOT block future stricter ships on larger architectures. ## Refs - PR #1742 (PMAT-690 P0-K base) - PR #1754 (SPEC §84+§85+§86 context) - PR #1762 (§87 Chinchilla 20×N hard gate runtime) - docs/specifications/audits/albor-370.md (external audit motivation) - docs/specifications/aprender-train/albor-370m-roadmap.md (P3 phases) - memory/feedback_a_priori_theoretical_falsification.md (#30) - memory/feedback_audit_hypothesis_bounds.md (#36) Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Two new spec sections plus the AC-SHIP2-003 row amendment that unblocks Two-Model spec closure at MODEL-2 ship % = 95%.
§87 — Chinchilla 20·N hard gate (P0-J' upgrade)
Codifies the §85 P2-E + §85.4 P2-G empirical plateau evidence. The 10-20× "ablation band" was empirically proven to hit val_loss ≈ 4.65 plateau regardless of LR/warmup/patience tuning. v1.0.0 hard-at-<10× / warn-at-<20× upgraded to v1.1.0 hard-at-<20×. Runtime gate already shipped via PR #1762.
§88 — AC-SHIP2-003 compute-bounded ship target (per user Option 4)
Per the operator's directive on the audit findings:
CE ≤ 2.2target requires 9-day continuous compute (213 GPU-hours), violating the 48-hour single-shot limit and freezing iteration for a week.AC-SHIP2-003to compute-bounded target val CE ≤ 4.7. P2-E DISCHARGES by 0.077 nats.AC-SHIP2-003-STRICT(new) preserves the strictCE ≤ 2.2target as the discharge anchor for the distillation epic (PMAT-683/684).MODEL-2 ship % impact
apr qapost-§86 salvageMODEL-2 ship %: 79% → 95%. Remaining 5% is the P3 polish phase (HF publish +
/dogfood).Stacked on PR #1754 (SPEC §84+§85+§86)
Base:
feat/spec-85-p2e-findings. The §87+§88 amendments depend on the §85+§86 context (the empirical P2-E / P2-G evidence + the P0-K live-verification + the §86 salvage path). Will auto-rebase tomainafter #1754 lands.What §88 explicitly does NOT do
AC-SHIP2-003— renames the strict form, amends the loose form, preserves both.Future epic (out of scope for this PR)
AC-SHIP2-003-STRICT(val CE ≤ 2.2) is achievable via:entrenar::distillper §35).Multi-week scoping deferred to PMAT-683 + PMAT-684 as a separate epic AFTER
aprender/albor-370mships.Refs
docs/specifications/audits/albor-370.md(external audit)docs/specifications/aprender-train/albor-370m-roadmap.md(P3 phases)memory/feedback_a_priori_theoretical_falsification.md([v0.5.0] Implement Random Forest Regression #30)memory/feedback_audit_hypothesis_bounds.md(Add grid search hyperparameter tuning example #36)🤖 Generated with Claude Code