| Age | Commit message (Expand) | Author |
|---|
| 2026-07-10 | RESULT 10: epoch endpoints (BP 1.2509/1.2750 vs EP 1.4802, +0.22 erosion cost... | Yuren Hao |
| 2026-07-10 | cos<0.96: flagship stays as control; bf1e3 continuation from s40000 queued on... | Yuren Hao |
| 2026-07-10 | Sustained dose-response sealed (3e-4/5e-4/1e-3 -> 0.975/0.983/0.990); ship --... | Yuren Hao |
| 2026-07-10 | RESULT 8 sealed (Muon 3v3: BP 1.7098 / EP 1.7314); beta-floor dose-response @... | Yuren Hao |
| 2026-07-10 | Watch: late-epoch cos erosion = intrinsic late-training SNR decline; let it r... | Yuren Hao |
| 2026-07-10 | RESULT 8 final: Muon generic (+0.13 both columns; BP+Muon 1.7020 vs EP+Muon 1... | Yuren Hao |
| 2026-07-10 | RESULT 9: wall-2 ELIMINATED by OLMo2 arch — epoch cleared 10-14k with zero ... | Yuren Hao |
| 2026-07-10 | RESULT 8 prelim: Muon+beta_floor on EP = 1.7316 (-0.098 vs AdamW, n=1, zero s... | Yuren Hao |
| 2026-07-10 | RESULT 7: OLMo2 parity PASS (BP 1.8335 vs EP 1.8643, gap +0.031 within metric... | Yuren Hao |
| 2026-07-10 | RESULT 6 addendum: K8 also blew at 12600 -> wall-2 is true contractivity cros... | Yuren Hao |
| 2026-07-10 | RESULT 6: wall-2 = EP-specific marginal under-convergence at the contractivit... | Yuren Hao |
| 2026-07-10 | HW dossier: OLMo2-standard block analog audit (net win; no new non-reciprocit... | Yuren Hao |
| 2026-07-10 | OLMo2-standard architecture (user directive): norm-after-sublayer RMSNorm blo... | Yuren Hao |
| 2026-07-10 | AUDIT: retract confounded Muon verdict; tone down parity claims (n=3, best-of... | Yuren Hao |
| 2026-07-10 | Stage-1 epoch blowup @12100: sig story REFUTED (only +8%, cos fine till after... | Yuren Hao |
| 2026-07-10 | RESULT 4: QK-norm validated parity-preserving (EP 1.8868 <= BP 1.9066); qk_no... | Yuren Hao |
| 2026-07-10 | QK-norm: replace nn.MHA with explicit SDPA attn + --qk_norm (OLMo2-style, ana... | Yuren Hao |
| 2026-07-10 | RESULT 3: K4 depth-parity SEALED EP-favorable (EP 1.8907 <= BP 1.9192 @ L12xC... | Yuren Hao |
| 2026-07-10 | Add --cosine (warmup->cosine decay to 0.1x lr) to both cascade trainers for f... | Yuren Hao |
| 2026-07-10 | CORRECTION: D1a did not die (misread mangled proc scan); originals completed ... | Yuren Hao |
| 2026-07-10 | RESULT 2: beta-floor confirmed as L12 depth fix (3e-4 holds cos=1.0 vs contro... | Yuren Hao |
| 2026-07-10 | K refuted as L12 lever (cos K-invariant K3==K8); BP 3-seed sealed 1.919; mech... | Yuren Hao |
| 2026-07-09 | D1a autopsy + K-ladder diagnostic: parent-death (not nohup'd) killed 4 arms; ... | Yuren Hao |
| 2026-07-09 | notes refresh: plan status stamp (K1-K3 sealed, D-tier in flight), ONBOARDING... | Yuren Hao |
| 2026-07-09 | FINDINGS: 2026-07-11 consolidated hardening entry | Yuren Hao |
| 2026-07-09 | HW: cascade re-pricing under the reuse doctrine — depth becomes a memory li... | Yuren Hao |
| 2026-07-09 | HW economics retraction: doctrine currency restored (COTS-stitch + FPGA + UIU... | Yuren Hao |
| 2026-07-09 | HW BoM correction: Lion/SGD ruled out for EP training (empirical, abl_lion2 b... | Yuren Hao |
| 2026-07-09 | BP-free formal audit (5.6-sol): EP update verified local (108 bitwise zero-in... | Yuren Hao |
| 2026-07-09 | HW: optimizer BoM audit per cascade block — Adam analog-compatible via fact... | Yuren Hao |
| 2026-07-09 | A0.4 precision gate (TF32 harmless cos 0.9946==fp32; pure-bf16 cos 0.9427) + ... | Yuren Hao |
| 2026-07-09 | FINDINGS: K3 sealed — matched-tuning parity n=3 (EP 2.0500+-0.015 vs BP 2.0... | Yuren Hao |
| 2026-07-09 | casc_eq_train: exact mode (dtop_every=1) now DEFAULT — bisection showed the... | Yuren Hao |
| 2026-07-09 | HW dossier: analog softmax/attention (12 verified claims) — Elfadel-Wyatt 1... | Yuren Hao |
| 2026-07-09 | casc_eq_train v7: two algorithmic dedups — d_top refreshed every other roun... | Yuren Hao |
| 2026-07-09 | casc_eq_train v6: graph-reuse fb (rebuild graphs feed next round's vjps), --c... | Yuren Hao |
| 2026-07-09 | FINDINGS: final C1 verdict — equilibrium fb-EP 2.2949 beats BP 2.3694 (same... | Yuren Hao |
| 2026-07-09 | FINDINGS: cascade pivot one-day arc — bridge, solver saga, fb message-passi... | Yuren Hao |
| 2026-07-09 | cascade root cause found: tied readout + default N(0,1) embedding init = path... | Yuren Hao |
| 2026-07-09 | casc_eq_train v4: quality-governed estimator — periodic measured cos(EP,BP)... | Yuren Hao |
| 2026-07-09 | casc_eq_train v3: adaptive beta (tok-sigma stiffness compensation), adaptive-... | Yuren Hao |
| 2026-07-09 | cascade equilibrium solver solved: fb (forward-backward message passing) K=3 ... | Yuren Hao |
| 2026-07-09 | casc_eq_train: equilibrium-mode cascade-EP trainer (GS-reverse solver K sweep... | Yuren Hao |
| 2026-07-09 | cascade: retire zil per user directive (EP route = equilibrium two-phase); pr... | Yuren Hao |
| 2026-07-09 | cascade: zil interleaved scheme = exact gradients at all depths (cos 1.0000 L... | Yuren Hao |
| 2026-07-09 | cascade ablation program: plan doc (5 claims, tiers 0-4), probe v2 (jacobi/gs... | Yuren Hao |
| 2026-07-09 | cascade_probe: layered-energy EP on L distinct standard blocks — free equil... | Yuren Hao |
| 2026-07-09 | lt_ep_train: --gov_keepjr now keeps the jr floor through reg_delay too (fastf... | Yuren Hao |
| 2026-07-09 | backfill_wandb: replay historical run logs into W&B (ept-c512 / ept-tol / ept... | Yuren Hao |
| 2026-07-09 | lt_ep_train: --rr_floor (resreg floor during free/delay phase) for govfloor arms | Yuren Hao |
