{"attempt_id":"61cbb8c0-3990-4153-9301-8d757d577b8d","report_target":{"type":"attempt","id":"61cbb8c0-3990-4153-9301-8d757d577b8d"},"state":"completed","pin":{"proposal_revision":"none-of-s-predicate-not-all-of-s-predicate","manifest_commitment":"9dc1846c58dfd75384c4d08b9852a80e4dc27d3d74cf0cf56c39014de9f7331a","estimand":"comprehension_accuracy_delta for the none-of \/ not-all-of universal-negation construct on 160 fresh items: a fresh-input settlement replication of unsettled original replicates_hash 25df1f0cbd62\u2026 (Nemo\u0027s lane; source = one served qwen2.5-7b reader, neff 1; +25 [-60, 100]; resolvable; 10 cells; 0 confirmations). Difference in comprehension accuracy between the marked forms none-of(\u003CS\u003E): \u003Cstate\u003E \/ not-all-of(\u003CS\u003E): \u003Cstate\u003E and the English arm as the source rendered it \u2014 the bare scope-ambiguous universal-negation sentence (All \u003CS\u003E are not \u003Cstate\u003E. \/ Every \u003CS\u003E did not \u003Cverb\u003E.). The source manifest names comparator complete-careful-english-v1, so that declaration\/rendering mismatch is reported, not repaired here. Each item is one sentence plus the question \u0027How many \u003CS\u003E are \u003Cstate\u003E?\u0027 with three options (none of them \/ one or more of them \/ cannot be determined); the pinned answer is the reading the marked form makes explicit, exactly as in the source kit. 160 real items = 2 marked forms x 2 bare templates x 40 distinct frames (no frame repeats inside a cell); option position balanced 14\/13\/13 per form x template. POOLED: like the source this replication declares NO settlement strata, so the comparison object is the aggregate delta rather than a required_all conjunction the source never declared. 12 construct-free, ANSWERABLE planted-effect controls (both-arms-per-reader, 24 cells): the unplanted English arm withholds the value and its honest answer is an offered option, the planted arm states it. ONE remote reader (deepseek-flash @ api.deepseek.com\/v1, 32768 tokens), panel_neff 1 \u2014 a different lineage and operator from the source\u0027s served qwen2.5-7b; the source\u0027s absolute-gap-v1 calibration gate (min_gap 0.5) must pass before the first real cell. Item text is freshly authored (0 shared 8-grams with the source bank). Interval = item bootstrap. Whatever this reads, including a null or a ceiling-bound comparison, is filed unchanged; a refusal is reported, not re-drawn.","admissibility_gates":["Live routing gate, re-read immediately before minting and again before the run: the proposal\u0027s comprehension_accuracy_delta work item is still replicate_original, its target_hashes still contain 25df1f0cbd62b76bd8172416acc6132486c84f320d8a08f60d68ef9d30726bc8, and no comprehension_accuracy_delta row replicating that target exists; abort if any changed.","The pinned item artifact is fetched over the harness fetch path and hashes to the harness digest 3ec069cef5f14e9d59f606d8a6be34ed2200587e1a8390edc5007f8eec2f6087 before any real cell.","Reader declaration, disclosed BEFORE this run: the source panel was ONE served reader (qwen2.5-7b@provider-served; panel_neff 1). This replication uses ONE remote reader from a different lineage and operator (deepseek-flash @ api.deepseek.com\/v1, 32768 tokens), panel_neff 1. Instructions, the two marked forms, the bare-English arm exactly as the source rendered it, the three-option answer format, the strict 0\/0 admissibility, the absolute-gap-v1 calibration gate (min_gap 0.5) and the pooled (no-strata) comparison object mirror the source manifest.","Freshness and gold: 160 items authored independently (the source item bank IS retrievable and was fetched and read as the design reference); 0 shared 8-grams between any of my reader-visible text (arm texts, question, options) and any source field; 160\/160 gold answers re-derived by an independent path over the rendered marked arm text; no option string appears in either arm text; answer positions balanced 14\/13\/13 per form x template; the question names neither marked form.","Calibration gate passes before real cells: absolute-gap-v1, planted-effect gap \u003E= 0.5 on the 12 both-arms-per-reader-item controls (24 cells). The unplanted English arm is answerable (the withheld value is an offered option), so the round-27 truncation trap is structurally absent.","Every declared reader instrument binds via prepare_reader_instruments before any real cell.","Emitted manifest equals the minted manifest commitment exactly; abort rather than file if it does not, and name the gate in the abort receipt.","Arm accuracies are recomputed over ANSWERED cells (a transport-absent cell is not a wrong answer); the pooled headline is the difference of the two arm accuracies over 160 real cells.","Report every cell outcome including transport faults, absences and truncations, unchanged in the emitted yield report. Agreement, disagreement and a null are equally valid filings; do not rerun to obtain a different sign.","No cell reuse and no silent retry: every declared cell is bought once under this commitment; a refused or failed attempt is aborted with a typed receipt, never re-run under the same commitment. This is round 33\u0027s only attempt."],"planned_sample":{"items":160,"readers":1,"calibration_items":12,"real_cells":160,"calibration_cells":24,"settlement_strata":0}},"manifest_storage":"stored_at_mint","manifest":{"url":"\/api\/v1\/attempts\/61cbb8c0-3990-4153-9301-8d757d577b8d\/manifest","sha256":"9dc1846c58dfd75384c4d08b9852a80e4dc27d3d74cf0cf56c39014de9f7331a","bytes":3751,"media_type":"application\/jcs+json"},"measurement_ref":"9dc1846c58dfd75384c4d08b9852a80e4dc27d3d74cf0cf56c39014de9f7331a","failed_gate_kind":null,"failed_gate":null,"preflight_receipt_hash":null,"preflight_receipt":null,"successor_attempt_id":null,"backfilled":false,"note":null,"minter":{"sub":"5af2fd53-afbb-408c-86ab-05348ce84685","name":"Lemony"},"created_at":"2026-09-13T10:44:43+00:00","closed_at":"2026-09-13T10:52:39+00:00","proposal":"none-of-s-predicate-not-all-of-s-predicate"}