Commit graph

214 commits

Author SHA1 Message Date
9785829755 Add the missing --freeze CLI mode to the en pair module so the unfrozen-manifest error points at a command that exists 2026-08-14 14:44:01 +03:00
ea1cbcd51b Reconfigure the phase cash desk after foreign module loads in the contamination probe so purchases land in the phase directory 2026-08-14 14:23:58 +03:00
9df7944381 Freeze the phase-D harness before the en-axis purchases: pair provider, wired rig hooks, pinned unit manifests, pair terminologist prompts and the paid contamination path 2026-08-14 14:17:42 +03:00
01485125a5 Fix three acceptance defects in the D4 summary: decoy rebuilt with this pack's plants, three axes per arm as the order requires, two inadmissible packets annulled 2026-08-14 03:42:24 +03:00
d1f42c5470 Add the span-join between the judge and deterministic axes, and blind adjudication with agreeableness and sensitivity controls 2026-08-14 02:00:17 +03:00
e630f97733 Add the D4 summary: admissibility gates, own floor per family, an assay-sensitivity gate on the margin decoy, and the pre-registered disagreement rule 2026-08-13 22:51:44 +03:00
766f902abd Rename judge prompt paths to ASCII: Cyrillic names render as octal escapes in the shell and break completion and greps 2026-08-11 18:43:31 +03:00
d10236453b Fix the lint slip in the parity gate: zip needs an explicit strict argument 2026-08-11 14:53:53 +03:00
b1cd892040 Freeze judge wrappers built from one shared core with a mechanical parity gate, plus reproducible session layout keeping floor halves in different sessions 2026-08-11 14:53:34 +03:00
baae56a1c0 Freeze the reworked en-ru and ja-ru pair prompt packs before any purchase, with the free-design editions kept apart from the measurement 2026-08-11 13:44:04 +03:00
18a8179eb3 Add the phase D judging harness: floor halves split across sessions, a margin decoy sized by measurement, isolated task dir and a transcript audit 2026-08-11 12:46:45 +03:00
2fed3d060f Raise the pack ceiling to 6.15 on the owner's sanction, funding the A1B re-buy that a session defect made necessary 2026-08-11 11:30:03 +03:00
6faa52f753 Raise the D4 sub-ceiling by two cents on the owner's sanction so the last A1B cell completes; the package ceiling is untouched 2026-08-11 11:14:46 +03:00
415f54f098 Add arm A1B carrying the ordered flag contour with the battery wired, keeping the paid canon-only A1 cells intact as a descriptive arm 2026-08-11 03:57:59 +03:00
cf85678e2a Freeze phase D judging design and $6.00 ceiling: two zero-cost judge families with a pre-registered disagreement rule, external judge isolated, Д7 cap restored to the order's $0.05 2026-08-11 03:42:45 +03:00
8950352fa2 Freeze phase D pre-reg erratum and harness repair: A2 arm built, battery wired with violation whitelist, per-contrast power with Student quantile, guard is a real stop 2026-08-11 03:18:37 +03:00
Claude (backend session)
4f698106e9 Fix silent cross-pack cash misroute: buy is a singleton and the last configurer won, so paid cells landed in another pack directory while the phase ledger read zero 2026-08-10 17:11:22 +03:00
Claude (backend session)
488d1d0b8d Freeze phase D edit-contour harness: fixer sees only flagged places, bank law on every arm including the critic, isolation selftest 2026-08-10 12:19:31 +03:00
Claude (backend session)
da5f4d0224 Freeze phase D pre-registration: hypotheses verbatim, Holm-corrected power with declared families, equivalence margin, measured budget, cash desk, ja pair pack 2026-08-10 07:06:07 +03:00
Claude (backend session)
b1780b8d9d Fix packs 22/23 rigs and reports per acceptance 16 and three adversarial review lines: exact sign test, word-boundary bank regex, judge attribution, coverage guard 2026-08-10 07:05:55 +03:00
Claude (backend session)
1f6d6aef17 Freeze the pack noise floor and judging config: second independent generation of the cheapest panel arm, two passes with family stamped before answers 2026-08-09 05:36:25 +03:00
Claude (backend session)
f1f9947dd6 Freeze editor-tier phases B and C: dual screen verdict against the pre-registered editor criterion, border units, tier panel over the frozen anchor base 2026-08-09 03:21:57 +03:00
Claude (backend session)
87247e2e97 Re-freeze editor-tier phase A: build the provider client from the extended roster so new slugs do not fail on lookup 2026-08-09 02:45:04 +03:00
Claude (backend session)
c0dd228d64 Freeze editor-tier phase A: pack roster with vendor-verified prices, configurable ledger, unit pinning by manifest, and screen reusing exp-22 thresholds 2026-08-09 02:44:28 +03:00
Claude (backend session)
ef797101c1 Re-freeze phase 3 for the slow mandatory candidate: hard per-model timeout, no client retries, declared unit subset 2026-08-08 22:26:30 +03:00
Claude (backend session)
784e412c24 Record the decoy base arm in the blind key so the mandatory control is computed for any panel, not only the first pack 2026-08-08 22:16:20 +03:00
Claude (backend session)
ed9e049248 Re-freeze phase 3 so its purchases split by model instead of queueing on one lock 2026-08-08 22:13:46 +03:00
Claude (backend session)
56b34d48ad Freeze phase 3 before its purchases: panel rule, isolation check and an estimate that uses the ledger price for the model whose screen cells never landed 2026-08-08 21:12:16 +03:00
Claude (backend session)
f6f04ff9ed Replace the decoy plant rules with book-independent ones after the control silently vanished on a different book 2026-08-08 21:07:59 +03:00
Claude (backend session)
7384c57f1a Re-freeze the phase 2 floor stage so its two end-to-end runs and their units can be split across processes 2026-08-08 20:58:50 +03:00
Claude (backend session)
fd3e5224d1 Re-freeze phase 2: claim each cell with an exclusive lock so parallel workers cannot buy the same call twice with the second payment invisible to the ledger 2026-08-08 20:43:25 +03:00
Claude (backend session)
af6d885047 Re-freeze phase 2 mid-stage: split units across processes to raise throughput, and read the battle editor cell from phase 2 in the phase 3 rig 2026-08-08 20:38:18 +03:00
Claude (backend session)
a725be5191 Re-freeze phase 2 before its bulk purchases: bound the ledger guard by per-model price and allow a tenant subset per process 2026-08-08 20:04:06 +03:00
Claude (backend session)
43381fe766 Freeze phase 2: panel composition, noise-floor design, declared contrast family and criteria, before its first purchase 2026-08-08 19:54:07 +03:00
Claude (backend session)
d123526cd5 Re-freeze phase 1 tail again: retry transient provider 5xx with backoff after four capacity refusals cost the mandatory candidate its screen cells 2026-08-08 18:12:18 +03:00
Claude (backend session)
05e2d57fd7 Re-freeze phase 1 tail: raise its ceiling to the owner-approved figure and align the markdown gate with the production sanitizer 2026-08-08 17:59:20 +03:00
Claude (backend session)
930a3792c3 Re-freeze before the rest of phase 1: point the ledger at the experiment price table after it failed on a model the old table did not know 2026-08-08 15:47:13 +03:00
Claude (backend session)
9ca51da0a8 Re-freeze the phase 1 harness after the first draft returned the Chinese source: add the echo gate with re-generation, and widen canon regexes 2026-08-08 15:23:27 +03:00
Claude (backend session)
3df9f54007 Widen the polygon purchase guard from one package directory to the whole eval zone so a second pack can buy under its own freeze 2026-08-08 15:15:49 +03:00
Claude (backend session)
e67ef1bea8 Freeze the phase 0 and 1 pre-registration of experiment 22 with its harness, ledger and thresholds before any paid call 2026-08-08 15:14:14 +03:00
84bc73baa5 Add the missing new file of the previous commit: the command that recomputes every итог number from raw 2026-08-08 02:42:48 +03:00
6aa7cfa8da Give every carrying number of the experiment a path in the tree: add itog.py, guard those numbers in the gate, print the phase-2b family, and write the missing report body 2026-08-08 02:42:37 +03:00
defe869c28 Restore the drift checks lost in the incident recovery and pin a linter for the polygon zone after a shadowed-name bug silently cost half the gate coverage 2026-08-08 02:06:06 +03:00
4b82fb1799 Fix the third recurrence of the ledger-glob defect that certified a false phase figure, correct the file and refusal counts, and restore the valid half of the drift measurement the retraction over-reached 2026-08-08 01:54:18 +03:00
0d05af3058 Restore the en-pair figures and money totals the incident reverted, and account for the phase 2b spend 2026-08-08 01:41:14 +03:00
f9986d9ddf Emit the narrow fourth judging pass for the re-measured phase 2b arms against their bases 2026-08-08 01:18:23 +03:00
d654036b5d Give the point-repair fixer the same term law block the editor gets, after the first pass showed it rewriting canonical terms blind 2026-08-08 01:13:39 +03:00
8be5653862 Wire the phase 2b arms into the same judging rig so their contrasts stay comparable with phase 2a 2026-08-08 00:53:16 +03:00
caaf36d24f Restore the polygon scripts lost to the 08.08 checkout incident and land the phase 2b arms module: exact-number gate matching, degrees-of-freedom correction, counting drift metric, point-repair arms C and E 2026-08-08 00:39:31 +03:00
399147aa76 Fix the drift plan price table and purchase loop to include the deconfound arm 2026-08-07 21:06:19 +03:00