{ "at": "2026-08-26T07:31:54Z", "mode": "run", "runs": [ { "model": "muse-glimmer:30b", "legs": { "C1": { "leg": "C1", "provenance": { "golden_path": "golden/judge-c1.json", "golden_sha256": "fede4e32154f92ac81ae68709fbde2cf1f91f3d1c76e7c83941023d986919702", "prereg_path": "prereg/PREREG-C1.md", "prereg_sha256": "423ed9609b8b475555fb9ae8b0f9125ccde112764195f0fa7dfa120f2b635529", "prereg_n": 21, "count_ok": true }, "raw_path": "raw/C1/muse-glimmer_30b/C1-muse-glimmer_30b.jsonl", "postures": { "think:false+reasoning-medium": { "posture": "think:false+reasoning-medium", "items": { "c1-p-bez-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-six-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-nap-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-piq-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ false, false, false ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": false, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-p-dom-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-chk-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ false, false, false ], "per_repeat_verdict": [ "UNCERTAIN", "FAIL", "FAIL" ], "self_consistent": false, "aggregation": "strict-majority", "verdict": "FAIL", "pass": false, "why": "strict majority 2/3 for 'FAIL'" }, "c1-p-con-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-spf-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-bkg-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-k-piq-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-bez-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-cal-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-con-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-dom-n1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-k-eca-n1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-k-spf-n1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-k-piq-s1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-con-s1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-bez-h1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ false, false, false ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": false, "why": "strict majority 3/3 for 'PASS'" }, "c1-k-six-w1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-chk-w1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" } }, "outcome": { "buckets": { "ok": 63 }, "calls": 63, "ok": 63, "response_failures": 0, "transport_failures": 0, "length_truncated": 0, "response_failure_denominator": 63, "response_failure_rate": 0.0, "response_failure_rate_row": { "successes": 0, "n": 63, "counts_of_n": "0/63", "point": 0.0, "wilson95": [ 0.0, 0.0575 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" }, "response_failure_ceiling": 0.1, "fence_ok": false, "outcome": "EXPLORATORY", "reason": "protocol fence: the contract probe failed; run anyway, unranked" }, "tally": { "n_items": 21, "pass": 18, "fail": 3, "no_verdict": 0, "verdicts": { "PASS": 8, "UNCERTAIN": 4, "FAIL": 9 }, "self_consistent_items": 20, "self_consistency": "20/21", "counts_of_n": "18/21", "rate": { "successes": 18, "n": 21, "counts_of_n": "18/21", "point": 0.8571, "wilson95": [ 0.6536, 0.9502 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" }, "self_consistency_rate": { "successes": 20, "n": 21, "counts_of_n": "20/21", "point": 0.9524, "wilson95": [ 0.7733, 0.9915 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" } }, "verdict": { "rule": "floor_verdict", "rule_module": "legspec_legs_c1", "result": { "kill_recall": { "n": 12, "hits": 11, "misses": 1, "no_verdict": 0, "counts_of_n": "11/12", "hit_ids": [ "c1-k-bez-c1", "c1-k-cal-c1", "c1-k-chk-w1", "c1-k-con-c1", "c1-k-con-s1", "c1-k-dom-n1", "c1-k-eca-n1", "c1-k-piq-c1", "c1-k-piq-s1", "c1-k-six-w1", "c1-k-spf-n1" ], "miss_ids": [ "c1-k-bez-h1" ], "no_verdict_ids": [], "floor": 11, "cleared": true, "rate": { "successes": 11, "n": 12, "counts_of_n": "11/12", "point": 0.9167, "wilson95": [ 0.6461, 0.9851 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" } }, "preservation": { "n": 9, "hits": 7, "misses": 2, "no_verdict": 0, "counts_of_n": "7/9", "hit_ids": [ "c1-p-bez-01", "c1-p-bkg-01", "c1-p-con-01", "c1-p-dom-01", "c1-p-nap-01", "c1-p-six-01", "c1-p-spf-01" ], "miss_ids": [ "c1-p-chk-01", "c1-p-piq-01" ], "no_verdict_ids": [], "floor": 8, "cleared": false, "rate": { "successes": 7, "n": 9, "counts_of_n": "7/9", "point": 0.7778, "wilson95": [ 0.4526, 0.9368 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" } }, "floors": { "kill": { "floor": 11, "n": 12, "arithmetic": "ceil(12 x 0.852) = 11" }, "preservation": { "floor": 8, "n": 9, "arithmetic": "ceil(9 x 0.8125) = 8" }, "both_bind_independently": true }, "cleared_both_floors": false, "verdict": "FAIL", "note": "floors only — no ordering by margin (PLAN v2). Two models that both cleared are TIED here whatever the margin." } } }, "think:false+reasoning-high": { "posture": "think:false+reasoning-high", "items": { "c1-p-bez-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-six-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-nap-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "PASS", "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 3/3 for 'PASS'" }, "c1-p-piq-01": { "repeats_attempted": 3, "repeats_scored": 0, "buckets": { "length_truncated": 3 }, "per_repeat_pass": [], "per_repeat_verdict": [], "self_consistent": null, "aggregation": "strict-majority", "pass": null, "verdict": null, "why": "no scorable repeat" }, "c1-p-dom-01": { "repeats_attempted": 3, "repeats_scored": 2, "buckets": { "ok": 2, "transport_failure": 1 }, "per_repeat_pass": [ true, true ], "per_repeat_verdict": [ "PASS", "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 2/2 for 'PASS'" }, "c1-p-chk-01": { "repeats_attempted": 3, "repeats_scored": 0, "buckets": { "length_truncated": 3 }, "per_repeat_pass": [], "per_repeat_verdict": [], "self_consistent": null, "aggregation": "strict-majority", "pass": null, "verdict": null, "why": "no scorable repeat" }, "c1-p-con-01": { "repeats_attempted": 3, "repeats_scored": 1, "buckets": { "ok": 1, "length_truncated": 2 }, "per_repeat_pass": [ true ], "per_repeat_verdict": [ "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 1/1 for 'PASS'" }, "c1-p-spf-01": { "repeats_attempted": 3, "repeats_scored": 1, "buckets": { "ok": 1, "length_truncated": 2 }, "per_repeat_pass": [ true ], "per_repeat_verdict": [ "PASS" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "PASS", "pass": true, "why": "strict majority 1/1 for 'PASS'" }, "c1-p-bkg-01": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, false, false ], "per_repeat_verdict": [ "PASS", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": false, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": false, "why": "strict majority 2/3 for 'UNCERTAIN'" }, "c1-k-piq-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-bez-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-cal-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-con-c1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-dom-n1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-k-eca-n1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-k-spf-n1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 3/3 for 'UNCERTAIN'" }, "c1-k-piq-s1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-con-s1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "UNCERTAIN", "FAIL", "FAIL" ], "self_consistent": false, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 2/3 for 'FAIL'" }, "c1-k-bez-h1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ false, true, true ], "per_repeat_verdict": [ "PASS", "UNCERTAIN", "UNCERTAIN" ], "self_consistent": false, "aggregation": "strict-majority", "verdict": "UNCERTAIN", "pass": true, "why": "strict majority 2/3 for 'UNCERTAIN'" }, "c1-k-six-w1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" }, "c1-k-chk-w1": { "repeats_attempted": 3, "repeats_scored": 3, "buckets": { "ok": 3 }, "per_repeat_pass": [ true, true, true ], "per_repeat_verdict": [ "FAIL", "FAIL", "FAIL" ], "self_consistent": true, "aggregation": "strict-majority", "verdict": "FAIL", "pass": true, "why": "strict majority 3/3 for 'FAIL'" } }, "outcome": { "buckets": { "ok": 52, "length_truncated": 10, "transport_failure": 1 }, "calls": 63, "ok": 52, "response_failures": 0, "transport_failures": 1, "length_truncated": 10, "response_failure_denominator": 52, "response_failure_rate": 0.0, "response_failure_rate_row": { "successes": 0, "n": 52, "counts_of_n": "0/52", "point": 0.0, "wilson95": [ 0.0, 0.0688 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" }, "response_failure_ceiling": 0.1, "fence_ok": false, "outcome": "EXPLORATORY", "reason": "protocol fence: the contract probe failed; run anyway, unranked" }, "tally": { "n_items": 21, "pass": 18, "fail": 1, "no_verdict": 2, "verdicts": { "PASS": 6, "None": 2, "UNCERTAIN": 5, "FAIL": 8 }, "self_consistent_items": 16, "self_consistency": "16/21", "counts_of_n": "18/21", "rate": { "successes": 18, "n": 21, "counts_of_n": "18/21", "point": 0.8571, "wilson95": [ 0.6536, 0.9502 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" }, "self_consistency_rate": { "successes": 16, "n": 21, "counts_of_n": "16/21", "point": 0.7619, "wilson95": [ 0.5491, 0.8937 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" } }, "verdict": { "rule": "floor_verdict", "rule_module": "legspec_legs_c1", "result": { "kill_recall": { "n": 12, "hits": 12, "misses": 0, "no_verdict": 0, "counts_of_n": "12/12", "hit_ids": [ "c1-k-bez-c1", "c1-k-bez-h1", "c1-k-cal-c1", "c1-k-chk-w1", "c1-k-con-c1", "c1-k-con-s1", "c1-k-dom-n1", "c1-k-eca-n1", "c1-k-piq-c1", "c1-k-piq-s1", "c1-k-six-w1", "c1-k-spf-n1" ], "miss_ids": [], "no_verdict_ids": [], "floor": 11, "cleared": true, "rate": { "successes": 12, "n": 12, "counts_of_n": "12/12", "point": 1.0, "wilson95": [ 0.7575, 1.0 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" } }, "preservation": { "n": 9, "hits": 6, "misses": 1, "no_verdict": 2, "counts_of_n": "6/9", "hit_ids": [ "c1-p-bez-01", "c1-p-con-01", "c1-p-dom-01", "c1-p-nap-01", "c1-p-six-01", "c1-p-spf-01" ], "miss_ids": [ "c1-p-bkg-01" ], "no_verdict_ids": [ "c1-p-chk-01", "c1-p-piq-01" ], "floor": 8, "cleared": false, "rate": { "successes": 6, "n": 9, "counts_of_n": "6/9", "point": 0.6667, "wilson95": [ 0.3542, 0.8794 ], "wilson_z": 1.96, "note": "counts bind; the interval is descriptive (PLAN v2: no percentage under N=30, and every C1 denominator is under 30)" } }, "floors": { "kill": { "floor": 11, "n": 12, "arithmetic": "ceil(12 x 0.852) = 11" }, "preservation": { "floor": 8, "n": 9, "arithmetic": "ceil(9 x 0.8125) = 8" }, "both_bind_independently": true }, "cleared_both_floors": false, "verdict": "FAIL", "note": "floors only — no ordering by margin (PLAN v2). Two models that both cleared are TIED here whatever the margin." } } } }, "probes": { "model": "muse-glimmer:30b", "configs": { "p1-schema-think-false": { "calls": [ { "probe_id": "p1-schema-think-false", "parsed": false, "schema_enforced": false, "keys_present": [], "type_errors": [], "parse_error": "JSONDecodeError: Extra data: line 1 column 202 (char 201)" }, { "probe_id": "p1-schema-think-false", "parsed": false, "schema_enforced": false, "keys_present": [], "type_errors": [], "parse_error": "JSONDecodeError: Extra data: line 1 column 202 (char 201)" } ], "repeats": 2, "parsed_all": false, "parsed_count": 0, "schema_enforced": false, "stable": true, "per_repeat_enforced": [ false, false ] }, "p2-schema-think-true": { "calls": [ { "probe_id": "p2-schema-think-true", "parsed": true, "schema_enforced": true, "keys_present": [ "answer", "grounded", "quote" ], "type_errors": [], "parse_error": null }, { "probe_id": "p2-schema-think-true", "parsed": true, "schema_enforced": true, "keys_present": [ "answer", "grounded", "quote" ], "type_errors": [], "parse_error": null } ], "repeats": 2, "parsed_all": true, "parsed_count": 2, "schema_enforced": true, "stable": true, "per_repeat_enforced": [ true, true ] }, "p3-bare-json-think-false": { "calls": [ { "probe_id": "p3-bare-json-think-false", "parsed": false, "schema_enforced": null, "keys_present": [], "type_errors": [], "parse_error": "JSONDecodeError: Extra data: line 1 column 127 (char 126)" }, { "probe_id": "p3-bare-json-think-false", "parsed": false, "schema_enforced": null, "keys_present": [], "type_errors": [], "parse_error": "JSONDecodeError: Extra data: line 1 column 127 (char 126)" } ], "repeats": 2, "parsed_all": false, "parsed_count": 0, "schema_enforced": null, "stable": true, "per_repeat_enforced": [ null, null ] } }, "polarity": { "label": "STANDARD TRAP", "explanation": "think:false silently dropped the schema constraint while think:true held it. This is the documented majority behaviour: any caller sending think:false with a format schema is getting NO enforcement and no error." }, "fence_ok": true, "calls": 6 } } }, "at": "2026-08-26T07:02:08Z", "load_gate": { "at": "2026-08-26T07:02:08Z", "wants": "muse-glimmer:30b", "residents": [ "muse-glimmer:30b" ], "blockers": [], "ok": true }, "runner_argv": { "tag": "muse-glimmer:30b", "matched": false, "error": "TypeError: list indices must be integers or slices, not str" }, "release": { "at": "2026-08-26T07:31:54Z", "released": "muse-glimmer:30b", "attempts": 1, "attempts_allowed": 4, "attempt_history": [ { "attempt": 1, "unloaded": [ "muse-glimmer:30b" ], "candidates_after": [], "cleared": true } ], "wall_s": 0.052, "law": { "num_ctx": 32768, "think": false, "reasoning_strength": null, "stream": false, "allow_keep_alive": true, "keep_alive": 0, "format": null }, "unloads": [ { "attempt": 1, "tag": "muse-glimmer:30b", "wall_s": 0.052, "law": { "num_ctx": 32768, "think": false, "reasoning_strength": null, "stream": false, "allow_keep_alive": true, "keep_alive": 0, "format": null }, "done_reason": "unload" } ], "done_reason": "unload", "residents_after": [], "candidates_after": [], "ok": true } } ], "model": "muse-glimmer:30b", "protocol": { "C1": { "golden": "golden/judge-c1.json", "prereg": "prereg/PREREG-C1.md", "repeats": 3, "aggregation": "strict-majority per case", "carried_state": "RANKED", "timeout_s": 180, "response_failure_ceiling": 0.1, "verdict_rule": "legspec_legs_c1.floor_verdict", "prompt": { "system": "You are a grounding judge for a rulebook. You will be given a CLAIM and a QUOTE from the rulebook. Decide whether the quote alone supports the claim.\nAnswer with ONLY a JSON object: {\"verdict\":\"PASS\"|\"FAIL\"|\"UNCERTAIN\",\"why\":\"\"}\nRules: PASS only if the quote alone entails the claim exactly as stated. FAIL if the quote contradicts it. UNCERTAIN if the quote neither supports nor contradicts it. World knowledge must never rescue a claim the quote does not support. Numbers must match exactly.", "user_template": "CLAIM: {claim}\nQUOTE:\n{quote}" }, "prompt_mode": "prompted schema — the JSON contract is in the system prompt and no format field is sent on a scored call; the polarity probes send the format schema because measuring its enforcement is their job", "temperature": 0, "top_p": 1, "num_predict": 1024, "postures": { "think:false": { "think": false, "reasoning_strength": null }, "think:false+reasoning-medium": { "think": false, "reasoning_strength": "medium" }, "think:false+reasoning-high": { "think": false, "reasoning_strength": "high" } }, "dial_model_prefix": "muse-glimmer", "reasoning_strength_mechanism": "system-prompt line 'Reasoning strength: {strength}' (RULINGS-0812 core ruling 1); think stays false on every C1 row", "kill_rule": "catch iff verdict != PASS (FAIL and UNCERTAIN both catch)", "preservation_rule": "only PASS preserves", "floors": { "kill": { "floor": 11, "n": 12, "arithmetic": "ceil(12 x 0.852) = 11" }, "preservation": { "floor": 8, "n": 9, "arithmetic": "ceil(9 x 0.8125) = 8" }, "both_bind_independently": true }, "probe_case_id": "c1-p-bez-01" } } }