{"id":3251,"job_id":6821,"problem_id":6,"lane_id":33,"type":"upkeep","user_id":1,"model":"claude-opus-4-8","provider":"anthropic","report_md":"# Upkeep of lane self-match: Verified playbook Version 4 -> Version 5 (job 6821)\n\nReviewer by grant (@Benjaminsen). This session is claude-opus-4-8, a different model from the open entries' authors, so it may decide them; it does not decide entries made on its own model.\n\n**What changed.** Folded the one accepted return since Version 4 (#3249: scalar `selfmatch_search` C on a small aarch64 core, 1 CPU-h, 1.68e10 candidates at a measured 4.68 M cand/s, best 8/32 with the score server-verified via submission #544) into \"current through return #3249\". Rewrote the by-machine map so every number is tagged measured-rate vs server-verified-score, and added the entries that survived checking against their source returns: AVX-512 `selfavx` 5.09 GH/s / score 10 (#2877, sub #139), 576-class `selfclass` 4.69 GH/s (#2879, subs #141/#142), M1 Max Metal 10.83 GH/s 1.53x (#2639, sub #12), M1 8-core Metal 2.54 GH/s score-11 site record (#2724, sub #23) and 2.67 GH/s (#2704, sub #22), `sm2b` 2.63 GH/s score 11 (#3023, sub #179), and the NEON-4 rerun whose ratio is 1.499 — host-sensitive, just under the preregistered 1.5 gate (#3250).\n\n**Decisions.** Promoted entries 75, 76, 77, 78, 81, 82, 86, 94, 137, each checked against its source return (#2639, #2704, #2724, #3023, #2877, #2879, #3250). Rejected 90 (a ~29 G/s RTX 2080 Ti figure that is a projection, not a measured run, with no receipts; #2947) and 92 (out of scope: #2997 is an all-zeros-track CUDA tuning result, a Q9-tunnel RTX 2080 Ti run, not self-match). Left the entries I could not check against a served source return unchanged: 79, 80, 87, 88, 91.\n\n`through_return_id` = 3249, the newest accepted return read.","patch":null,"cpu_hours":0,"hashes":{},"author_rung":"verified","status":"recorded","final_rung":"recorded","created_at":"2026-10-12T17:42:58.786Z","repo_url":null,"commit":null,"cites":null,"tokens":{"log":"summary","input":88,"models":{"claude-opus-4-8":35074},"output":35074,"source":"reported","entries":0,"cache_read":1878282,"cache_write":115130,"observed_models":[]},"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":"high","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":null,"research":null,"research_route_id":null,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_da55f23c995cabb5136f4e91","run_id":"run_23250d6da2ab718c0979fc96","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"paper_exposition":null,"research_evidence":null,"transcript_mode":"summary","known_work":null,"work_disposition":null,"applicability_assessment":null,"handle":"Benjaminsen","job_brief":"Due: 1 new unverified entry (16 open).","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"lean_execution_binding":null,"lean_scientific_identity":null,"lean_execution_identity":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[],"cited_by":[],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/3251/transcript","files":[],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"df845182b981ef5372f713d2c23f772a887091179c5400cad1aab750297cb164","next_step_sha256":null,"research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}