{"id":2485,"job_id":5236,"problem_id":1,"lane_id":2,"type":"explore","user_id":1,"model":"deepseek-v4-flash","provider":"deepseek","report_md":"# Report — job #5236 (route 216 first look): the scale-matched shape channel at q = 23#\n\n## Question\n\nRoute 216 / return #2471 measured a *scale-invariant shape excess* of the twin-admissible\nwindow-count law at `q = 11#, 13#` (`L/mg ~ 2-4`), decided against a carrier-matched permutation\ncloud. Its own weakest assumption: that this is a growing-support functional and not a finite-size\nartefact of the smallest windows. Route 216's issued next experiment asked for the next two rungs,\n`q = 23#` and `q = 29#`, with the frozen statistic, `M = 400` (fallback 150), and the rule\n`|z_D| > 3` in `>= 2` cells of the two new rungs.\n\n## What this look did (scope)\n\nA cost check before the measurement: the frozen statistic costs one length-`2q` cyclic prefix sum\n**per control draw**, i.e. `O(q)`; `q = 29# = 6469693230` needs `~52 GB` per draw under this\ncontainer's 6 GB cgroup and is out of reach in this session. This look therefore decides the\n**next rung `q = 23# = 223092870`** exactly (the statistic unchanged; seed and `M = 150` frozen\nbefore the run in `PREREGISTRATION_di.md`) and hands `29#` on as a concrete bounded next step.\nStatistic, control family, leave-one-out band and verdict rule are exactly #2471's.\n\n## Instrument custody (all passed)\n\n* Constructor `A_q = B_q & roll(B_q, -2)` reproduces the independent sieve at `5#..17#`.\n* `R_A(q/2)` at `7#/11#/13#/17#` = `0.577748 / 0.126658 / 0.013701 / 0.003379` (route 198 anchors).\n* **`R_A(23#, L)` for the three small cells re-derived by the exact correlation-sum identity**\n  `V_q(L) = sum_s (L-|s|)(C(s)-rho^2)`, `C(s) = prod_p (p-|{0,-2}u{-s,-s-2}|)/p`, agrees with the\n  full enumeration to `rel < 2e-15` (`0.58735631`, `0.54570816`, `0.43912740`).\n* `K = round(q·rho) = 7952175`, `B_q = 36495360`, `mg = q/K = 28.0543`.\n* `check_di.py` 30 checks, 0 FAIL, exit 0; `--corrupt` 34 checks, 0 FAIL (shifted `z_D`, tampered\n  `R_A`, flipped verdict all detected).\n\n## Measurement at q = 23#\n\n| `L` | `L/mg` | `D_A` | control band `D_i` | `z_D` | `sup_sign` | `skew_obs` / `skew_ctrl` | `kurt_obs` / `kurt_ctrl` | `R_A` | `R0` |\n|---|---|---|---|---|---|---|---|---|---|\n| 28 | 0.998 | 0.17634 | 0.02668 ± 0.00179 | **+83.65** | −1 | 0.307 / 0.744 | −0.419 / +0.252 | 0.5874 | 0.724 |\n| 56 | 1.996 | 0.21363 | 0.000117 ± 0.000059 | **+3631** | +1 | 0.211 / 0.508 | −0.217 / +0.092 | 0.5457 | 0.442 |\n| 112 | 3.992 | 0.16039 | 0.00654 ± 0.02002 | **+7.68** | −1 | −0.073 / 0.347 | −0.243 / +0.029 | 0.4391 | 0.111 |\n| q/4 | 1.99e6 | 0.03940 | 0.06785 ± 0.02863 | −0.99 | −1 | 0.022 / 0.030 | −0.111 / −0.571 | 5.1e−5 | nan |\n| q/2 | 3.98e6 | 0.04258 | 0.04687 ± 0.02290 | −0.19 | +1 | 0.0 / 0.0 | −0.179 / −0.771 | 6.1e−5 | nan |\n\n`M = 150` draws, seed 20261007; every heavy step under `sah.py bounded`.\n\n## Findings\n\n**(1) The directional shape excess persists at 23#.** At all three small-window cells the observed\nstandardised window-count distribution is **less skewed and more platykurtic** than the\ncarrier-matched control (`skew_obs < skew_ctrl`, `kurt_obs < kurt_ctrl` at each), the same direction\n#2471 found at 11#/13#. The raw sup-distances `D_A = 0.176 / 0.214 / 0.160` are the same order as\n#2471's firing cells (`D_A = 0.246` at 11#, `0.382` at 13#). Under-dispersion is still present at\n23# (`R_A = 0.587 / 0.546 / 0.439 < 1`; `R_A(q/2) = 5.1e-5`, decaying in `q` as route 198 says).\n\n**(2) But the pre-registered verdict statistic `z_D` is NOT scale-invariant, so its firing at 23#\nis not independent evidence of \"persistence/growth\".** The leave-one-out control band's own spread\n`sd_i D_i` shrinks with `q` and with `L` (0.0246/0.0454 at 11#/13# vs 0.0018/0.000059/0.020 at 23#),\nbecause the control ensemble–mean CDF concentrates like an empirical CDF (`~q^{-1/2}`) while the\nobserved sup-distance `D_A` is dominated by a fixed distributional difference. `z_D = (D_A - mean D_i)/sd D_i`\nis thus a ratio whose denominator vanishes with `q`: at 23# it is `8`–`3600` for a `D_A` of the same\nsize as #2471's, so `|z_D| > 3` is crossed almost by construction. The route's rule as written cannot\ndistinguish \"the shape excess grew\" from \"the control band got tighter\". A scale-invariant effect\nsize (e.g. `D_A` normalised by the *number* of independent window samples, or a permutation p-value\nthat is not rescaled by `sd D_i`) is required before any `q`-trend can be read.\n\n**(3) Two design defects in the frozen rule, reported as such.** (a) The pre-registered sign clause\nwas written as `sup_sign > 0` (\"`D_A > control`\"), but #2471's own firing cells have `sup_sign = -1`\n(observed *lighter*-tailed, `F_A < F_ref`): the intended clause is \"same sign as #2471\", i.e.\n`sup_sign < 0`. Under the literal clause the frozen rule returns **unresolved**; under the intended\nclause it returns **H_persist** (cells `L=28` and `L=112`, both `sup_sign = -1`, `|z_D| >> 3`).\nBoth readings are reported; no verdict is overclaimed. (b) The frozen `sd_D < 1e-3` =>\n\"under-powered\" clause fired on `L=56` — but a *tight* control band means high power, not low; the\ncell is reported excluded by the clause, not unmeasured, and the clause is mis-calibrated.\n\n**(4) `29#` was not reached** for a stated capability reason, not a silence: the frozen `O(q)`\nfull-cycle method needs ~52 GB per control draw at `q = 6.47e9`. A bounded next step (the route's\nown continuation) is recorded, with the half-open case made explicit.\n\n## Disposition\n\n**Outcome: `progress`.** A finite, independently-checked measurement now exists at the next rung:\nthe *direction* of the shape excess persists at 23# (less-skewed/more-platykurtic, `F_A < F_ref`),\nsame order of `D_A` as 11#/13#, and route 198's under-dispersion continues to decay. What does\n**not** stand is the route's `z_D`-based persistence test — it is confounded by the control band\nshrinking with `q`. The route is not refuted and is not promoted: the union/Chebyshev transfer\nstill has no second input it can consume, because the scale-invariant effect size that would supply\none is exactly what is missing. `depends_on` #2471.\n\n## Scope and honesty\n\nExact finite arithmetic at `q = 23#` only. No asymptotic claim; nothing here bounds `G2(x#)`,\n`beta_2` or twin-prime infinitude. `z_D` is a finite n-permutation proxy, not a distributional\nlimit. `M = 150` gives the leave-one-out band a ~6% relative `sd` error (as #2471). The\n`sup`-functional is one choice among scale-invariant shape functionals and is not claimed optimal;\nthe control band's non-monotone dependence on `L` (0.0018/0.000059/0.020 across `L = 28/56/112`) is\nitself evidence that `D` is dominated by discreteness at small `L`, which is part of finding (2).\n","patch":null,"cpu_hours":0.4,"hashes":{"sah.py":"21a1d3556191bf54458b13fa0ebe41b4550fb92a33ab9bee6518d82ef222c843","check_di.py":"f7eed708a60b167b07a6f91e679767e8c62577d8c0166c9e19c0c46dba6ffafe","check_di.out":"6e730900b6be79c28f11a5e07c81a0bbbf3ae5d0842dd8e379d44687c1dad9e3","redact_di.py":"cd7b839edde4f57758a82a07c9126ebc49f77404340b2b826374188a11fa29ff","report_di.md":"a72127f9a0953ee1df9c4fd2d7287843f414410c3cefabbbbf6dec6cec81fd76","compute_di.py":"006ea2b833d016153aa0600430cc4ca71298fefbf2ad314e8c6946301117a479","compute_di.out":"ca6f05bd872414a39e1544a1536d22ee789001ff1afc5e9e2d6445193ec93ae1","evidence_di.md":"a2efd78e156a0a181ba581ab8e3719f90014158fc3895ac409c96ec027c74862","next_step.json":"3ddc89abc0feb4784f73775473032fe17336d4194d86a5ec2a5065e21c1ec0be","compute_di.json":"8b1c1d448abecc82c8e75f3f40ef07705dd6c87c14986ea8657b12175cdfda63","prior_art_di.md":"8b57198c028362efb5830ade609c93c0d9fe803733f1027a63648c63fd43ade4","check_di.control.out":"2efc3979f723e8aa23dd5f9c87f2bb3c63b47fe1e0541bf39cd58f97ddc6fb24","PREREGISTRATION_di.md":"f3f499a81858967106fe002059ebd591e3a46c526c91ec13e35a679a578e4e11"},"author_rung":"measured","status":"recorded","final_rung":"recorded","created_at":"2026-10-07T19:55:38.557Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[2471],"messages":[]},"tokens":{"log":"custom","input":0,"models":{"deepseek-v4-flash":0},"output":0,"source":"none","entries":0,"cache_read":0,"cache_write":0,"observed_models":["deepseek-v4-flash"]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":null,"verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":null,"also_fix":null,"transcript_omitted":{"share":0,"omitted":0,"outputs":0},"patch_hash":null,"superseded_by":null,"duplicate_of":null,"transcript_resubmitted_at":null,"file_notes":[{"sha":"006ea2b833d016153aa0600430cc4ca71298fefbf2ad314e8c6946301117a479","name":"compute_di.py","notes":["prints what looks like progress or timing to stdout on line 136 (\"print(\"Bpos built %.1fs\" % (time.time() - t0), flush=True)\"): stdout is the artifact and must reproduce byte for byte elsewhere; send progress, timing and rates to stderr. This one is a guess from the text, not a measurement: if the output is already identical from run to run, say so in your return and leave the file alone."],"fixed_by":"37b1ec5b0bf75f2016d23718b0467648292c0c94ef62ff386bbdf604fd586ff5"}],"research":{"outcome":"progress","route_id":216,"next_step":{"method":"Do not repeat this run's 23# measurement or #2471's x<=17 censuses. (a) Re-analyse the retained compute_di.json (23#) and #2471's compute_cv.json cells with three scale-free effect sizes on the SAME data: the raw D_A against a width-pinned control band (subsample each control to a fixed effective sample size so sd D_i does not shrink with q), the standardised skew and excess-kurtosis differences (skew_obs - skew_ctrl, kurt_obs - kurt_ctrl), and a rank-based permutation p-value with floor 1/(M+1) instead of (D_A - mean D_i)/sd D_i. Pre-register, before re-analysis, that the persistence criterion is a monotone-in-q trend of a statistic whose control band is q-invariant; report which q-fit the 11#/13#/17#/23# ladder supports. (b) Independently, build the primorial-hierarchy window-count histogram: q = m*p with p the largest prime, and for a window of length L compute the count distribution by CRT recursion over (t mod m, t mod p) without materialising Z/q (the local pattern of A_m in a window gives, for each t mod m, the multiset of forbidden p-residues; the p-histogram follows by a small cyclic convolution), then validate it against this run's enumerated 23# cells (R_A and D_A to 1e-12) and use it to reach q = 29# within 4 CPU-h and under a 6 GB memory cap.","compute":{"ram_gb":2,"disk_gb":1,"cpu_hours":0},"failure":"No scale-free effect size can be formed from the retained data, or every candidate gives a flat/ambiguous trend in q at L/mg ~ 1-4: the shape channel is then a scoped negative at the census level (the 11#/13#/23# excess is a small-window effect of the same size at each rung and does not grow), and the union/Chebyshev obstruction stands with R_A alone as route 198 has it.","success":"At least one statistic T(q) is q-invariant by construction (its permutation band does not shrink with q), and the 11#/13#/17#/23# ladder gives a resolved trend of T at L/mg ~ 1-4 that is either positive (the shape channel is a genuine growing-support functional and gives the union/Chebyshev transfer its second input) or flat/negative (scoped negative, transfer needs only R_A). The CRT histogram is validated against the enumerated 23# cells to 1e-12 and reaches q = 29#.","question":"Which statistic makes the shape excess of the twin-admissible window count comparable across q, so that the persistence question (does the excess grow with q or is it a small-window finite-size effect?) can actually be decided at q = 23# and 29#?","budget_hours":3,"required_tools":[],"required_sources":[]},"depends_on":[2471],"evidence_md":"# Evidence — job #5236 (route 216, scale-matched shape channel at q = 23#)\n\n**Rung.** Custody **verified**; shape measurement at `q = 23#` **measured**; `q = 29#` **not\nreached** (capability); the pre-registered verdict statistic **refuted as scale-invariant**.\n\n## Objects\n`q = 23# = 223092870`; `|B_q| = 36495360`; `K = |A_q| = 7952175`; `mg = q/K = 28.0543`.\nStatistic (frozen, #2471): `z^C_t=(N^C_t-mean)/sd`; `D(C)=sup_{y in grid[-4,4](801)}|F^C(y)-F_ref(y)|`;\n`F_ref` = ensemble-mean CDF over `M=150` uniform random `K`-subsets of `B_q` (seed 20261007);\n`z_D=(D_A-mean_i D_i)/sd_i D_i`, control band leave-one-out.\n\n## Custody (independent)\n* `A_q = B_q & roll(B_q,-2)` == sieve at `5#..17#`.\n* Route-198 anchors: `R_A(q/2)` = `0.577748/0.126658/0.013701/0.003379` at 7#/11#/13#/17#.\n* `R_A(23#,L=28/56/112)` by the **exact correlation-sum identity**\n  (`C(s)=prod_p (p-|{0,-2}u{-s,-s-2}|)/p`) = `0.58735631/0.54570816/0.43912740` == enumerated to\n  `rel < 2e-15`.\n* `check_di.py` **30/30, FAIL 0, exit 0**; `--corrupt` **34/34** (shifted `z_D`, tampered `R_A`,\n  flipped verdict all detected).\n\n## Measurement (M=150; one prefix sum per draw serves all cells)\n| `L` | `L/mg` | `D_A` | `D_i` mean±sd | `z_D` | `sup_sign` | `R_A` |\n|---|---|---|---|---|---|---|\n| 28 | 0.998 | 0.17634 | 0.02668±0.00179 | +83.65 | −1 | 0.5874 |\n| 56 | 1.996 | 0.21363 | 0.000117±0.000059 | +3631 | +1 | 0.5457 |\n| 112 | 3.992 | 0.16039 | 0.00654±0.02002 | +7.68 | −1 | 0.4391 |\n| q/4 | 1.99e6 | 0.03940 | 0.06785±0.02863 | −0.99 | −1 | 5.1e−5 |\n| q/2 | 3.98e6 | 0.04258 | 0.04687±0.02290 | −0.19 | +1 | 6.1e−5 |\n\nMoment direction, all three small cells: `skew_obs < skew_ctrl` (0.307/0.744, 0.211/0.508,\n−0.073/0.347); `kurt_obs < kurt_ctrl` (−0.419/+0.252, −0.217/+0.092, −0.243/+0.029).\n#2471's cells: 11# `L=68` `D_A=0.246`, `D_i=0.1405±0.0246`, `z_D=+4.301`, `sup_sign=−1`; 13# `L=40`\n`D_A=0.382`, `D_i=0.1736±0.0454`, `z_D=+4.583`, `sup_sign=−1`.\n\n## What this changes\n1. **The shape excess is present at the next rung, same direction** (`F_A < F_ref`, lighter tails;\n   less-skewed/more-platykurtic counts), `D_A` the same order as 11#/13#; under-dispersion persists\n   (`R_A<1`, decaying to `5.1e-5` at `q/2`).\n2. **`z_D` does not survive larger `q`.** `sd_i D_i` shrinks with `q` (0.0246/0.0454 at 11#/13# vs\n   0.0018/0.000059/0.020 at 23#) since the control mean CDF concentrates like `q^{-1/2}` while `D_A`\n   stays `O(0.1)`; `z_D` is `8`–`3600` at 23# for an equal-size effect, so `|z_D|>3` is crossed almost\n   by construction. **`z_D` cannot be a `q`-trend/persistence test** — a scale-invariant effect size\n   is required.\n3. **Frozen-rule defects, disclosed:** the sign clause was pre-registered `sup_sign>0` while #2471's\n   own cells are `sup_sign=−1` (literal rule → `unresolved`; intended rule → `H_persist` at `L=28`\n   and `L=112`); the `sd_D<1e-3` \"under-powered\" clause misfired on a *tight* control band (`L=56`).\n4. **`29#` is a capability gap, not a negative:** `O(q)` per draw ≈ 52 GB at `q=6.47e9` under the\n   6 GB cgroup. No claim is made about 29#.\n\n## Entry points\n`DI_M=150 sah.py bounded --run run-2026-10-07-di --limit 520 -- python3 compute_di.py` (resumable,\n`ckpt_di.npz` every 5 draws; 150/150; `compute_di.json`); `python3 check_di.py` → `check_di.out`;\n`--corrupt` → `check_di.control.out`. CPU ≈ 24 min.\n\n## Limitations\n`q = 23#` only; no asymptotic claim; nothing bounds `G2`/`beta_2`/twin primes. The `sup`-KS functional\nis not claimed optimal, and its control band is non-monotone in `L` (0.0018/0.000059/0.020) — itself\nevidence that small-`L` discreteness dominates `D`. `M=150` (≈6% relative `sd` error). `R0=nan` at\nlarge `L` is a `0/0` underflow.","prior_art_md":"# Prior art and search record — job #5236 (route 216, q = 23#)\n\n## Online search (this run; two queries, before the measurement)\n* *\"twin-admissible residues primorial window count distribution underdispersion permutation test\"*.\n  Located only generic under-dispersed-count and permutation-test methodology: Li 2023 (generalized\n  Poisson for underdispersed counts, MDPI *Mathematics* 11(6):1478), Viljanen 2022 (`llperm`\n  permutation-of-residuals test for count regression, PMC9743778), Seck 2022 (Poisson alternatives\n  for underdispersion), and standard permutation-test expositions (brainder.org permutation-test\n  notes; Virginia Library permutation-based tests; jwilber.me). **None** studies residues of a\n  primorial, twin-admissible sets, or a window-count law.\n* *\"scale invariant shape statistic empirical CDF permutation control scale dependence test statistic\"*.\n  Located method-level neighbours only: Canay–Romano–Shaikh approximate permutation tests via\n  induced order statistics (Northwestern/cemmap); Golland–Liang permutation tests for classification;\n  Reddi 2013 scale-invariant conditional dependence measures (PMLR v28); standard scale-invariance\n  references. These give the *control methodology* (permutation cloud, empirical-CDF distance) but not\n  the object and not the scale-dependence caveat needed here.\n\nNo located source measures a scale-matched shape distance of the twin-admissible window-count law\nagainst a carrier-matched permutation cloud at any primorial. A no-match search is evidence about\nthe search, not a novelty certificate.\n\n## In-project prior work (unchanged, and the exact gap)\n* Route **198** / #2386 / #2393: the one-window **variance** `R_A(L)<1` and its `q`-decay; the\n  union/Chebyshev obstruction to a `G2(x#)` transfer names the missing **tail** input.\n* Route **199** / #2395 / #2460: higher **cumulants**; the matched null cancels identically out of\n  the order-4 statistic.\n* #5124: `P(N=0)` (one tail point) plus the two-point covariance, found to be a near-uniform\n  rescaling of the null by `R_A`.\n* #2397: fixed-size thinning control — #198's under-dispersion is not a density artefact.\n* **#2471** (route 216, job #5235): the scale-invariant **shape** distance itself, `x <= 17`.\n* #2375/#5100 (wheel-generic constants), #4746/#4919/#5055 (other windowed designs) — none is a\n  scale-invariant shape distance of the window-count law.\n\n**Exact difference this return makes:** it is not a new object — it is #2471's object carried to\n`q = 23#` and, more importantly, a *refutation of #2471's decision rule at larger `q`*: the\nleave-one-out `sd D_i` denominator is not scale-stable, so `z_D` cannot serve as the persistence\ntest the route's `next_step` assumes. That is the missing piece the next experiment must supply.\n\n## Remaining gap (precise)\nA statistic that separates \"the shape excess persists/grows in `q`\" from \"the permutation cloud's\nspread shrinks in `q`\". Candidates, all unbuilt here: (i) raw `D_A` against a *width-pinned*\ncontrol band (fix the band's scale by subsampling each control to a fixed effective sample size);\n(ii) an effect size in the standardised moments that is `q`-free by construction (`skew`/`kurt`\ndifferences were already direction-consistent at 23# and are the natural candidate); (iii) a\npermutation p-value computed from the rank of `D_A` in a cloud whose size scales with `q` (so the\np-value floor is `1/(M+1)`, not `1/sd D_i`). For `q = 29#` the blocker is the `O(q)` full-cycle\nadvance (`~52 GB/draw` at `q = 6.47e9`); the primorial-hierarchy / CRT-recursion histogram named in\nthe next step is the only route to that rung without materialising `Z/q`."},"research_route_id":216,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_0e793a31e299699dfaaa6fee","run_id":"run_3e10f943e1e1a7bec5221c37","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"handle":"Benjaminsen","job_brief":"Search online for existing attempts, results, tables and datasets before testing feasibility. Reuse the recorded search and inspect the closest sources and weakest assumption. Use published numbers with citations; do not reproduce them in a first look. Seek the smallest experiment on the uncovered step. Recommend promising only with specific evidence and a bounded next step; do not claim the route is proved. Map the assumptions of any borrowed method onto this problem.\n\nRead GET <project base>/research-routes/216 and return #2471. Return the ordinary report and transcript plus research: {route_id: 216, outcome: \"promising|progress|blocked|inconclusive|known|result\", evidence_md: \"what the evidence changes, <=4000 chars\", prior_art_md: \"updated online search record, sources and exact remaining gap, <=4000\", next_step: {question, method, success, failure, budget_hours} <only for continued pursuit; what to do, never when or how fast; it must not ask for what a return on this route or a linked route already did, and the route returns it builds on go in depends_on or cites.returns>, obstacle: {kind, statement, assumptions, evidence, revisit_when} <for blocked/inconclusive>, depends_on: [<return ids actually required>]}. A result with a distinct next_step requests review and continues pursuit concurrently; omit next_step when no further experiment is warranted. Use known with prior_art_md and no next_step or obstacle when cited prior work already covers the proposed contribution; it stops automatic investigation without requesting review. The evidence grade is separate. Do not close a broad route because one proof attempt failed.","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[{"id":"2471","status":"recorded","final_rung":"recorded","canonical_return_id":null}],"cited_by":[{"id":2502,"handle":"maxime-fleury","status":"recorded"},{"id":2508,"handle":"Benjaminsen","status":"recorded"}],"route_dependents":[216],"research_url":"/projects/twin-primes/research-routes/216","transcript_url":"/projects/twin-primes/return/2485/transcript","files":[{"sha256":"a72127f9a0953ee1df9c4fd2d7287843f414410c3cefabbbbf6dec6cec81fd76","name":"report_di.md","bytes":6678},{"sha256":"a2efd78e156a0a181ba581ab8e3719f90014158fc3895ac409c96ec027c74862","name":"evidence_di.md","bytes":3744},{"sha256":"8b57198c028362efb5830ade609c93c0d9fe803733f1027a63648c63fd43ade4","name":"prior_art_di.md","bytes":3673},{"sha256":"3ddc89abc0feb4784f73775473032fe17336d4194d86a5ec2a5065e21c1ec0be","name":"next_step.json","bytes":2502},{"sha256":"f3f499a81858967106fe002059ebd591e3a46c526c91ec13e35a679a578e4e11","name":"PREREGISTRATION_di.md","bytes":3478},{"sha256":"006ea2b833d016153aa0600430cc4ca71298fefbf2ad314e8c6946301117a479","name":"compute_di.py","bytes":9396},{"sha256":"8b1c1d448abecc82c8e75f3f40ef07705dd6c87c14986ea8657b12175cdfda63","name":"compute_di.json","bytes":4189},{"sha256":"ca6f05bd872414a39e1544a1536d22ee789001ff1afc5e9e2d6445193ec93ae1","name":"compute_di.out","bytes":2349},{"sha256":"f7eed708a60b167b07a6f91e679767e8c62577d8c0166c9e19c0c46dba6ffafe","name":"check_di.py","bytes":9524},{"sha256":"6e730900b6be79c28f11a5e07c81a0bbbf3ae5d0842dd8e379d44687c1dad9e3","name":"check_di.out","bytes":1647},{"sha256":"2efc3979f723e8aa23dd5f9c87f2bb3c63b47fe1e0541bf39cd58f97ddc6fb24","name":"check_di.control.out","bytes":1815},{"sha256":"cd7b839edde4f57758a82a07c9126ebc49f77404340b2b826374188a11fa29ff","name":"redact_di.py","bytes":3356},{"sha256":"21a1d3556191bf54458b13fa0ebe41b4550fb92a33ab9bee6518d82ef222c843","name":"sah.py","bytes":56280},{"sha256":"37b1ec5b0bf75f2016d23718b0467648292c0c94ef62ff386bbdf604fd586ff5","name":"compute_di.py","bytes":9498}],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"duplicates":[],"cited_messages":[]}