wassname 5d83adbb25 fix: correct the "18 vs 21 pair" basis claim (it was never about pair count) ...

Read the safetensors shapes/metadata: v_hack_full = 10 pairs / k=5,
v_hack_21pairs = 16 pairs / k=12 (n_heldout=2; neither is 18 or 21). The two
bases differ on pairs AND directions-kept AND extract-tau simultaneously, so
the hack-cut gap is triple-confounded, not a clean "pair set is the lever"
result. Nothing was lost: the strong basis reproduces from current pairs.py
via --top-k=12 --v-hack-drop-bottom-frac=0.0, and refresh already re-extracts
at k=12. Rewrites Q8 + the top confound bullet + the README findings caveat.
A one-knob k-sweep is needed to attribute the gain.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-05-29 10:12:12 +00:00

..

blog

LW draft: add preregistered H1 block-quote with falsification clauses

2026-05-29 03:56:33 +00:00

brainstorm

ready

2026-05-23 14:19:41 +08:00

lab

lab report v3: TL;DR, three-line concept, PASS_RATE column, G_hack rename

2026-05-29 03:18:22 +00:00

papers

tidy

2026-05-29 06:29:43 +00:00

personas

fix smoke.

2026-05-23 11:26:39 +08:00

spec

tidy

2026-05-29 06:29:43 +00:00

vendor

concepts

2026-05-29 06:29:20 +00:00

extract_vhack_grad-vec.md

Doc cleanup: mark susp gate as REMOVED in design doc

2026-05-27 09:08:34 +00:00

grpo_hyperparams.md

fix smoke.

2026-05-23 11:26:39 +08:00

human_journal.md

init

2026-05-23 10:22:54 +08:00

results.md

fix: correct the "18 vs 21 pair" basis claim (it was never about pair count)

2026-05-29 10:12:12 +00:00

table_proto.md

setup

2026-05-23 10:40:02 +08:00