Block a user
bbf8a02c6b
item 28 pre-registered (Nils's variant): teacher-state distillation — frozen full-cot teacher's band-exit state at step end, cosine into burst s^10; λ amended 1.0→5.0 pre-run (smoke: baseline cos-dist 0.113)
62ce30e557
item 25 closed (truncated): writing is not computing — lce 1.9 with 2:12 flat at 31.2 (log-only; eval json never written due to kill); λ=1.0 arm cancelled; item 27 promoted
44e814e78c
item 27 pre-registered: zero-pause internal band looping — single M=10 in-place burst at prompt end, iteration-aligned lens targets, ii0 inference ablation; carry_steps gains inplace updates, trainer gains --inner-iters/--base-pauses
0cffce876b
item 26 pre-registered: result-staging lens supervision at pre-'=' positions (no causal leakage — low loss requires computation); arms lg03 and lt03+lg03; item-25 in-flight note (lce 10.3→2.3)
64abfe66f3
item 25 pre-registered: latent process supervision via differentiable lens readout — pause j trained to lens-encode deleted-step token j (λ=0.3/1.0 arms); carry_logits gains return_states
f31e6e7e30
item 24 closed (stopped by Nils): trainable band doesn't improve fit either; E2-L internalization line closed on all four axes; LoopLoRA k=0 bit-exactness validated in passing
fe89820e64
item 24 pre-registered: d=1 retry with loop-only band-LoRA r=16 (4.8M params, whole band, k=0 bit-exact); trainer/eval gain --bandlora; smoke-tested train+eval
7a141948fa
item 23 scored: both control axes confirm — pp30 29.3 (within ±5 of 31.6, 3x pauses bought nothing), x600 no recovery; d=1 ceiling is structural, ladder chapter closed
46b5e1b802
whiteboard microscopy on hard(795)/drop(430) specimens: FF slips at the 3-term sum (155), carry re-expands and defers (165 ✓); boards show live continuation-plan vs premature digit-commit at identical prefixes
f143c67b3d
probe_discount: single-prompt whiteboard microscopy (lens table + digit probs, carry vs FF control) on the 80/15% problem
001a4ca5f3
item 23 pre-registered: d=1 ceiling controls (x600 undertrain arm, pp30 pause-bandwidth arm, +-5pt decision rule); trainer gains --tag-suffix
7d60f37cf1
item 22 scored: ladder breaks at d=1 (57.4->31.6), plateaus ~19 at d=2/3 — 2x cold floor (d=3 vs base McNemar p=0.0025); latent loop trades easy reliability for hard/drop reach
4e744e7d94
26B-A4B lens mapped: 30 layers (not 48), same early-persist/terminal-motor family as E4B+31B; patched save delivered full regimes.pt incl. per-layer ignition; 5-scale figure complete
20f7014f5b
item 22 pre-registered: E2-L rung B internalization ladder (front-first deletion, 10 pauses/step, warm-started d=1..3); trainer gains --drop-steps/--warm-start/--steps/--lr; smoke-tested 2 steps on Spark
272a7b1d1f
31B regimes parsed (early persist bump L7-16, motor only terminal — E2B signature absent, matches E4B); exp4_regimes.py takes explicit out path / defaults next to input jbar (fix pushed to node before 26B stage)
34497b8835
item 21 McNemar scored: drop-bucket edge significant (2:6 p=0.0094, survives Bonferroni x4); overall A-vs-B ns; + session handoff, fig cleanup
051e90e805
item 21: GSM carry-cot 57.4% (5x prior best); control isolates whiteboard to drop-bucket; E2-N/A2 planned
4b9e168838
item 20: gate threshold curve + oracle bound (probs pass, merge per-item LUT, offline sweep)