Round r20

r20

Status · DESCRIPTIVEReviewed 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/r20/README.md

Round date: HISTORICAL (before 2026-10-09) Publication badge: DESCRIPTIVE; KEPT FOR AUDIT Why not VALID: The historical inventory retains DESCRIPTIVE because of ARM_RECIPE, DECISION_RULE, ENVIRONMENT, PRE_REGISTRATION. 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.

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.
  • Historical execution command: not recorded in the cited round source.
  • 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

STATUS: DESCRIPTIVE

Why not VALID

  • PRE_REGISTRATION: The README says the round was not pre-registered; no rule timestamp predating the first result is shown.
  • DECISION_RULE: The README says no public predeclared decision rule was recovered.
  • ARM_RECIPE: The README says exact arm definitions or recipes are not re-derivable from the public record.
  • ENVIRONMENT: The round measurement contract and missing-field record show incomplete per-cell environment metadata.

Data completeness: 45.3% (6 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: no

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 20 diagnostic + rounds 21/22 interim (11:30Z)

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: syn-14-bug-sla-business-hours

Task reconciliation: The surviving plan lists syn-14-bug-sla-business-hours. Task IDs in the public run-record export but not in that list: rails-ar-bulk-access-grants, rails-as-variant-processed-once. Listed task IDs with no matching tagged delivery: 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: Public run-record outcomes are summarized below. No comparison claim is made where the arm definition, decision rule or source report is not re-derivable.

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-ar-bulk-access-grants studio solhigh-direct not re-derivable 3 3 1 2 0
rails-as-variant-processed-once studio solhigh-direct not re-derivable 3 3 3 0 0

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 6 rows (fail 2, pass 4); overall pass rate is 66.7% (4/6) among pass/fail grades. Per-arm detail and coverage counts are in the JSON.

Published-export outcome cross-check: 6/6 shared cell IDs match; 0/0 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: 6/6 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.

6 records · 38,838 bytesDownload r20.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