Round r57b

r57b

Status · INCOMPLETEReviewed status is transcribed from rounds/STATUS.md. The original narrative and its qualifications remain below.

Source-reported narrativeRound README and results text are source material. Numeric values that are not tied to a claim ledger record remain source-reported.

Round source README · rounds/r57b/README.md

Round date: HISTORICAL (before 2026-10-09) Publication badge: INCOMPLETE; KEPT FOR AUDIT Why not VALID: The historical inventory retains INCOMPLETE because of DESIGN, ENVIRONMENT, NO_PREREG. See the round evidence and limitations below. Recomputation status: OBSERVED SOURCE ONLY

Recomputation update: The mined raw records now provide per-cell outcome, manifest token-counter, and wall-time recomputation for the observed cohort. This does not reconstruct missing planned cells, a full intention-to-treat denominator, or model execution.

Status

INCOMPLETE

  • NO_PREREG — No dated registration is linked in the committed public record.
  • DESIGN — The README says no testable question or complete comparison protocol is established.
  • ENVIRONMENT — Planned task-to-environment assignments are not established.

Required reproduction metadata

  • Kogen commit: not recorded in the cited round source.
  • Harness commit: not recorded in the cited round source as a Git commit; any adapter or runtime fingerprints are not Git commits.
  • Model and effort: a complete per-cell mapping is not recorded in the cited round source; values present in the source summary remain as stated above.
  • Raw records: the cited sources do not provide a complete public round-specific request and grade bundle; public delivery and outcome exports are linked above where applicable.

Terminology: public round glossary.

STATUS: INCOMPLETE

Data completeness: 48.6% (96 deliveries; PASS WITH DECLARED GAPS). Measurement contract; declared gaps.

Status reason: Documentary capture only; the surviving public record does not establish a testable research question or a complete comparison protocol, so no research conclusion is claimed.

Pre-registered: unverified; no dated registration is linked in the committed public record.

Lifecycle: Historical capture; closure is not independently reconciled from the public record.

Question: No testable research question is recoverable from the committed public record; this page is a documentary delivery inventory, not an interpretable study. Design: Round 57b: shell-only vs default tools on Rails (Studio) (4 Oct ~11:45Z)

Venue: Public host identifiers are listed in the reconciliation table. The surviving record does not establish every planned task-to-host assignment or the execution-environment details.

Decision rule: No public predeclared decision rule was recovered; outcome counts below are descriptive.

Arms: Arm definitions are not re-derivable from the public record; exported labels appear in the reconciliation table below.

Planned tasks in the surviving round note: elx-port-erase-account, rails-ac-throttle-search, rails-aj-enqueue-after-commit, rails-ar-bulk-access-grants, rails-as-variant-processed-once, rails-ft-mysql-fulltext-search-foundation, rails-hw-scoped-broadcast, syn-14-bug-sla-business-hours

Task reconciliation: The surviving plan lists elx-port-erase-account, rails-ac-throttle-search, rails-aj-enqueue-after-commit, rails-ar-bulk-access-grants, rails-as-variant-processed-once, rails-ft-mysql-fulltext-search-foundation, rails-hw-scoped-broadcast, syn-14-bug-sla-business-hours. Task IDs in the public run-record export but not in that list: None. Listed task IDs with no matching tagged delivery: elx-port-erase-account, syn-14-bug-sla-business-hours. The public record does not establish intended reassignment or the reason for any mismatch; no run-record row was changed or reattributed. Verdict: The public exports record 32 passes among 48 shell-only rows and 31 passes among 48 default-tools rows. The previously reported accuracy bound is not reproduced because its interval method and assumptions are not documented. No non-inferiority decision is made.

Public delivery and outcome reconciliation

This captured-delivery summary uses the public run-record export, grouped by audit round, task ID, public host ID, and the exact exported arm label. Captured counts rows; graded counts rows marked graded=true; pass, fail, and other are the run-record outcome fields. Official outcome states are in cells.jsonl, whose classifications can differ; see the register source crosswalk. Planned n is shown only when source-reported; unknown allocations are not inferred.

Task ID Public host ID Arm label in export Planned n Captured Graded Pass Fail Other graded outcome
rails-ac-throttle-search studio default-tools not re-derivable 8 8 4 4 0
rails-ac-throttle-search studio shell-only not re-derivable 8 8 8 0 0
rails-aj-enqueue-after-commit studio default-tools not re-derivable 8 8 8 0 0
rails-aj-enqueue-after-commit studio shell-only not re-derivable 8 8 8 0 0
rails-ar-bulk-access-grants studio default-tools not re-derivable 8 8 8 0 0
rails-ar-bulk-access-grants studio shell-only not re-derivable 8 8 7 1 0
rails-as-variant-processed-once studio default-tools not re-derivable 8 8 4 4 0
rails-as-variant-processed-once studio shell-only not re-derivable 8 8 3 5 0
rails-ft-mysql-fulltext-search-foundation studio default-tools not re-derivable 8 8 0 8 0
rails-ft-mysql-fulltext-search-foundation studio shell-only not re-derivable 8 8 0 8 0
rails-hw-scoped-broadcast studio default-tools not re-derivable 8 8 7 1 0
rails-hw-scoped-broadcast studio shell-only not re-derivable 8 8 6 2 0

Reproduction record and source reconciliation

This section documents exact replay limits and keeps the page lifecycle and earlier statuses unchanged.

  • Kogen source commit: not applicable (Harness does not use Kogen) in the per-cell setup.kogen_sha field.
  • Harness source commit: not recorded. Per-cell tools.harness SHA-256 fingerprint(s): 562c489daaf10a03c06d3a86f8227bf08b6a43096f6fae9fd0b5f59bda4e7371. Only wrapper fingerprints are public; the underlying source commit remains unrecorded.
  • Models and effort (requested → effective): kh-gpt: model requested/effective gpt-6-luna → chatgpt/gpt-6-luna; effort requested/effective max → xhigh; runner requested low.
  • Observed task IDs: rails-ac-throttle-search, rails-aj-enqueue-after-commit, rails-ar-bulk-access-grants, rails-as-variant-processed-once, rails-ft-mysql-fulltext-search-foundation, rails-hw-scoped-broadcast.
  • Original execution command: not recorded in the public snapshot; the benchmark cells cannot be replayed from this repository.
  • Rebuild the public metadata from the repository root with python3 reproduce/export_results.py and python3 reproduce/build_records.py; check this declaration with python3 reproduce/validate_round.py --round r57b. These commands regenerate or validate the archived public records. An original cell-launch command is not available.
  • Raw public records: r57b.jsonl filtered by audit_round/round_id, and cells.jsonl filtered by round.

Public snapshot rows for this tag: 96 captured deliveries (63 pass, 33 fail, 0 ungraded/unknown in results/run-records/r57b.jsonl); official outcome export has 96 rows: 63 pass, 33 fail in cells.jsonl. These are separate source totals. The public files do not provide a complete joined intention-to-treat result.

Round-specific reconciliation: Keep this captured round separate from r57/r57c/r57d and from r57e. The historical pooled non-inferiority interval and cost-per-pass comparison are source-reported, not reproducible from public data.

Recomputed from raw records

`recomputed.json`, mined cell records, and the token audit back observed official outcomes, per-arm pass counts/rates, manifest token counters, and wall medians. The source contains 96 rows (fail 33, pass 63); overall pass rate is 65.6% (63/96) among pass/fail grades. Per-arm detail and coverage counts are in the JSON.

Published-export outcome cross-check: 96/96 shared cell IDs match; 0/96 public-export rows have no mined source record. Exact IDs and results for unpaired rows are in recomputed.json. No per-cell outcome mismatches were found in the shared IDs. Token reconciliation under METHOD.md: 96/96 published metadata totals equal the manifest input + cached input + output sum; 0 differ (no mismatches). Reasoning remains a separate reported component and is not added to output. The recompute keeps both numbers visible. 0 unexplained single-response numeric errors were identified; multi-response rows retain the published full-cell total and the manifest counter separately because per-response usage records are excluded.

Captured run records

Standard-schema records assigned to this round in the indexed snapshot.

96 records · 633,575 bytesDownload r57b.jsonl · View immutable source ↗

Reproduction inventory

Published analysis replay and full experiment replay are reported separately.

Replay type
unavailable
Script
No per-round analysis script is published.
Inputs
No identified public input bundle.
Full execution replay
Unavailable
Scope
No public per-round analysis script with a complete, identified input bundle is published. See the round README, MEASURED.md, and MISSING.md for recorded scope and gaps; historical execution inputs/toolchains/grading are not bundled.

View the complete per-round reproduction inventory · Repository reproduction guide