diff --git "a/evaluation.json" "b/evaluation.json" --- "a/evaluation.json" +++ "b/evaluation.json" @@ -1,5604 +1,5013 @@ { - "scope": "One preregistered SwarmTraces continuation; manual source panel is primary, parser-derived development is secondary.", + "suite": "/Users/mac/Sev-security-20260925/evals/sev/response-policy-curriculum-v2", + "suite_sha256": "3b9a80dfd212b6932b66ea520c35ab1a3b19e132a81af1b6b7b306d776a70a02", + "partition_hashes": { + "calibration": "890c363c5bfe5351600071691eefb47dcf4e2530b31b75d4ce8830bd6381f585", + "development": "cb92ae6c329f9b9b738eb7637a6f6b3adb3dd8bb7204b5b697f268afb98a4efa" + }, + "scorer_sha256": "c079a66a2ab4a3d81ce9872b4c1f309ae2f07c89684dc025d70008d9b4be8844", "models": { "parent": { - "temperature": 1.3660402567543952, - "domains": { - "excytin": { - "raw": { - "n": 398, - "accuracy": 1.0, - "nll": 0.002029807853842489, - "brier": 0.00013267567616055285, - "correct": 398 - }, - "global_calibrated": { - "n": 398, - "accuracy": 1.0, - "nll": 0.00820502390654013, - "brier": 0.0006264068412257906, - "correct": 398 + "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-response-policy-parent-4b-v2/00-Sev-4B@a1824aefba305fda86e3503b895a7d9b3871f79a", + "checkpoint": { + "requested": "macmacmacmac/Sev-4B@a1824aefba305fda86e3503b895a7d9b3871f79a", + "resolved": "/__modal/volumes/vo-RociVuGycNGtzvUcdOqecz/hub/models--macmacmacmac--Sev-4B/snapshots/a1824aefba305fda86e3503b895a7d9b3871f79a", + "head_sha256": "1d50c37b4db60c22ae8e6bb24f5055056701bb16e332964350d76df57dc82af0", + "adapter_sha256": "b5afd73584d1098cccb0a1bd42afc40e2de3c3bc3074a5173a642b6d4b880420", + "inference_temperature": 1.0 + }, + "inference_temperature": 1.0, + "additional_temperature": 1.6245047927124707, + "splits": { + "calibration": { + "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-response-policy-parent-4b-v2/00-Sev-4B@a1824aefba305fda86e3503b895a7d9b3871f79a/calibration", + "coverage": { + "requested_records": 1920, + "requested_questions": 2674, + "evaluated_records": 1920, + "evaluated_questions": 2674, + "rejected_records": 0, + "truncated_records": 0 + }, + "maximum_softmax_probability_error": 3.5965218081468464e-08, + "files": { + "report.json": "a85db0f08eef510bf05a9ad177940e6b7d05ba3b06cf0b738816e918d8ccd8e6", + "rows.json": "507dbf8555ec0f8f045c4d8223d68e9598a5410f163abb5e2110d74ab8614c78", + "predictions.jsonl": "2219339424e3ab09f5e3b62eb7c499286b0fe9977dbaef1def487c9166148f0c" } }, - "guide": { - "raw": { - "n": 231, - "accuracy": 0.5497835497835498, - "nll": 0.7756690534124361, - "brier": 0.479220229400239, - "macro_f1": 0.4478062632890602, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" + "development": { + "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-response-policy-parent-4b-v2/00-Sev-4B@a1824aefba305fda86e3503b895a7d9b3871f79a/development", + "coverage": { + "requested_records": 1828, + "requested_questions": 2632, + "evaluated_records": 1828, + "evaluated_questions": 2632, + "rejected_records": 0, + "truncated_records": 0 + }, + "maximum_softmax_probability_error": 3.5965218081468464e-08, + "files": { + "report.json": "cbdc805ba88cff1e050b594eefd164fc96e097e5fb580af33b67ca34dde0374a", + "rows.json": "e39116e8120a64037aedc101740359b317108ab07cd7aba5bcaed7119fc7bd0a", + "predictions.jsonl": "b49dc9c2f7af40f17d88c61ca7d88d121293b0f3951130205e17d6309eda103f" + } + } + }, + "files": { + "result.json": "fcd9c5d0d47a5da3799a1bd5b062fd7872e8417bc57f55d9e22feadb29c32e47", + "provenance.json": "9568b5ca8ed5e28d306ef28d0b14ab278b67e48b277e983babd03bd273fad769", + "calibration/temperature.json": "a63b396ccd1a5041ecc92923cbf62628190207fc42857da4a6727cf7e84fb0af" + } + }, + "candidate": { + "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-response-policy-4b-v2/00-trial-0", + "checkpoint": { + "requested": "/runs/sev-r2-response-policy-4b-v2/00-trial-0/checkpoint", + "resolved": "/runs/sev-r2-response-policy-4b-v2/00-trial-0/checkpoint", + "head_sha256": "506dcd3587847917a288d95757c497c15a2aef8e2d4bc6523b04ad991c961a2b", + "adapter_sha256": "fa1916b34fe1eeb10d15c36927c225df43115407d7c97b4f879bba3102c3ed21", + "inference_temperature": 1.0 + }, + "inference_temperature": 1.0, + "additional_temperature": 1.6817928305074288, + "splits": { + "calibration": { + "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-response-policy-4b-v2/00-trial-0/calibration", + "coverage": { + "requested_records": 1920, + "requested_questions": 2674, + "evaluated_records": 1920, + "evaluated_questions": 2674, + "rejected_records": 0, + "truncated_records": 0 + }, + "maximum_softmax_probability_error": 3.3540022259526125e-08, + "files": { + "report.json": "44dff4a4d14b65992f8b21e784a50823e30ac16c7e1bba4d75c860fc04bc4995", + "rows.json": "026b68705b7ac76bc4361bc2fda1cbfe74265b436a807ea68bca77f87d3cf9da", + "predictions.jsonl": "80047e316749504f63b02b49ec8e790eb92cdce9e4bdc15fbf764ac69f0d7ea5" + } + }, + "development": { + "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-response-policy-4b-v2/00-trial-0/development", + "coverage": { + "requested_records": 1828, + "requested_questions": 2632, + "evaluated_records": 1828, + "evaluated_questions": 2632, + "rejected_records": 0, + "truncated_records": 0 + }, + "maximum_softmax_probability_error": 3.3602443216818045e-08, + "files": { + "report.json": "29e391352dd2474975ebe3d9b8b66be1a0cf96b5b917d559d1f80afd5b0704c5", + "rows.json": "27ce2c65c5efa32d0197c51e2dd13f81c7b71d410ce7a27dce7998e8dd247c27", + "predictions.jsonl": "548409eb6efb70c9b320a43cb88d671696c01326917bb18837e6d7aba4d2a69e" + } + } + }, + "files": { + "result.json": "ebf9f528ebe8ecef772f32418c8bc99b998285962b39c0a229c4e72b1f0b1014", + "provenance.json": "7cfaf8ba0c975c545afd56248df611d1fa6bed0a9b0655968f9aa6bfe60258b1", + "calibration/temperature.json": "328a01a5463c0f03f2b5d5c324d5be8def08cac819cf89aaf49a27a9c88e71e4" + } + } + }, + "calibration_thresholds": { + "parent": { + "threshold": 0.8875371629397495, + "false_alert_budget": 0.05, + "comparison": "positive_probability >= threshold; equal scores stay together", + "selection": "maximum recall, then fewest false alerts, then highest threshold", + "n": 140, + "positives": 28, + "negatives": 112, + "confusion": { + "tp": 11, + "fp": 5, + "tn": 107, + "fn": 17 + }, + "recall": 0.39285714285714285, + "false_alert_rate": 0.044642857142857144, + "precision": 0.6875, + "threshold_accuracy": 0.8428571428571429 + }, + "candidate": { + "threshold": 0.8449815374033091, + "false_alert_budget": 0.05, + "comparison": "positive_probability >= threshold; equal scores stay together", + "selection": "maximum recall, then fewest false alerts, then highest threshold", + "n": 140, + "positives": 28, + "negatives": 112, + "confusion": { + "tp": 21, + "fp": 4, + "tn": 108, + "fn": 7 + }, + "recall": 0.75, + "false_alert_rate": 0.03571428571428571, + "precision": 0.84, + "threshold_accuracy": 0.9214285714285714 + } + }, + "baseline_headroom": { + "rule": "continuation only if parent development recall < 0.95 or false-alert rate > 0.05", + "class_support_sufficient": true, + "continuation_allowed": true + }, + "development": { + "always_permit_baseline": { + "n": 175, + "positives": 35, + "negatives": 140, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 140, + "fn": 35 + }, + "recall": 0.0, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 0.8 + }, + "models": { + "parent": { + "questions": 175, + "groups": 35, + "raw": { + "n": 175, + "nll": 0.795502582803619, + "acc": 0.7257142857142858, + "ece": 0.1747005023605082, + "brier": 0.4505153855550131, + "mean_conf": 0.8840572965356658, + "confident_error_rate": 0.13714285714285715, + "coverage_at_0_9": 0.6457142857142857, + "accuracy_at_0_9": 0.7876106194690266, + "coverage_at_5pct_error": 0.02857142857142857, + "coverage_at_1pct_error": 0.02857142857142857, + "aurc": 0.1992292401228163, + "error_rate_at_0_9": 0.21238938053097345, + "confidence_bias": 0.15834301082138003, + "top_bins": { + "0.9": { + "n": 113, + "errors": 24, + "error_rate": 0.21238938053097345 + }, + "0.95": { + "n": 84, + "errors": 18, + "error_rate": 0.21428571428571427 + }, + "0.99": { + "n": 26, + "errors": 2, + "error_rate": 0.07692307692307693 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.7613636363636364, + "confidence_cutoff": 0.9457634807080788 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.75, + "confidence_cutoff": 0.7938824081649869 + } + } + }, + "calibrated": { + "n": 175, + "nll": 0.6260474644745716, + "acc": 0.7257142857142858, + "ece": 0.10761147298493057, + "brier": 0.4111063010906543, + "mean_conf": 0.8140932667241441, + "confident_error_rate": 0.06857142857142857, + "coverage_at_0_9": 0.35428571428571426, + "accuracy_at_0_9": 0.8064516129032258, + "coverage_at_5pct_error": 0.02857142857142857, + "coverage_at_1pct_error": 0.02857142857142857, + "aurc": 0.1992292401228163, + "error_rate_at_0_9": 0.1935483870967742, + "confidence_bias": 0.0883789810098583, + "top_bins": { + "0.9": { + "n": 62, + "errors": 12, + "error_rate": 0.1935483870967742 + }, + "0.95": { + "n": 21, + "errors": 2, + "error_rate": 0.09523809523809523 + }, + "0.99": { + "n": 1, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.7613636363636364, + "confidence_cutoff": 0.8531718492565429 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.75, + "confidence_cutoff": 0.6963743924447515 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 35, + "correct": 29, + "recall": 0.8285714285714286 + }, + "permitted": { + "n": 140, + "correct": 98, + "recall": 0.7 + } + }, + "threshold": 0.8875371629397495, + "n": 175, + "positives": 35, + "negatives": 140, + "confusion": { + "tp": 14, + "fp": 12, + "tn": 128, + "fn": 21 + }, + "recall": 0.4, + "false_alert_rate": 0.08571428571428572, + "precision": 0.5384615384615384, + "threshold_accuracy": 0.8114285714285714 + }, + "candidate": { + "questions": 175, + "groups": 35, + "raw": { + "n": 175, + "nll": 0.49844721765472333, + "acc": 0.84, + "ece": 0.11735638435435483, + "brier": 0.269672286103042, + "mean_conf": 0.9559888432162478, + "confident_error_rate": 0.10285714285714286, + "coverage_at_0_9": 0.8742857142857143, + "accuracy_at_0_9": 0.8823529411764706, + "coverage_at_5pct_error": 0.6114285714285714, + "coverage_at_1pct_error": 0.44, + "aurc": 0.044632779737602504, + "error_rate_at_0_9": 0.11764705882352941, + "confidence_bias": 0.1159888432162478, + "top_bins": { + "0.9": { + "n": 153, + "errors": 18, + "error_rate": 0.11764705882352941 + }, + "0.95": { + "n": 140, + "errors": 14, + "error_rate": 0.1 + }, + "0.99": { + "n": 92, + "errors": 3, + "error_rate": 0.03260869565217391 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.9772727272727273, + "confidence_cutoff": 0.9904485398763483 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.9, + "confidence_cutoff": 0.9516673073011176 + } + } + }, + "calibrated": { + "n": 175, + "nll": 0.37433095036445935, + "acc": 0.84, + "ece": 0.06636288221880407, + "brier": 0.23938834727626476, + "mean_conf": 0.9048661883246669, + "confident_error_rate": 0.05142857142857143, + "coverage_at_0_9": 0.7142857142857143, + "accuracy_at_0_9": 0.928, + "coverage_at_5pct_error": 0.6114285714285714, + "coverage_at_1pct_error": 0.44, + "aurc": 0.044632779737602504, + "error_rate_at_0_9": 0.072, + "confidence_bias": 0.06486618832466695, + "top_bins": { + "0.9": { + "n": 125, + "errors": 9, + "error_rate": 0.072 + }, + "0.95": { + "n": 77, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 14, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.9772727272727273, + "confidence_cutoff": 0.9404661687170754 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.9, + "confidence_cutoff": 0.8547040294961724 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 35, + "correct": 35, + "recall": 1.0 + }, + "permitted": { + "n": 140, + "correct": 112, + "recall": 0.8 + } + }, + "threshold": 0.8449815374033091, + "n": 175, + "positives": 35, + "negatives": 140, + "confusion": { + "tp": 25, + "fp": 14, + "tn": 126, + "fn": 10 + }, + "recall": 0.7142857142857143, + "false_alert_rate": 0.1, + "precision": 0.6410256410256411, + "threshold_accuracy": 0.8628571428571429 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": 0.4, + "ci95": [ + 0.22857142857142856, + 0.5714285714285714 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.7142857142857143, + "ci95": [ + 0.5714285714285714, + 0.8571428571428571 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.3142857142857143, + "ci95": [ + 0.17142857142857143, + 0.4857142857142857 + ], + "valid_resamples": 2000 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.08571428571428572, + "ci95": [ + 0.05, + 0.12857142857142856 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.1, + "ci95": [ + 0.06428571428571428, + 0.14285714285714285 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.01428571428571429, + "ci95": [ + -0.014285714285714277, + 0.04285714285714286 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": 0.5384615384615384, + "ci95": [ + 0.3665909090909091, + 0.70979262672811 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.6410256410256411, + "ci95": [ + 0.55, + 0.7435897435897436 ], - "per_class": { - "BenignPositive": { - "precision": 0.2987012987012987, - "recall": 0.3898305084745763, - "f1-score": 0.3382352941176471, - "support": 59.0 + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.10256410256410264, + "ci95": [ + 0.005701872682569551, + 0.21591595300834426 + ], + "valid_resamples": 2000 + } + }, + "threshold_accuracy": { + "parent": { + "value": 0.8114285714285714, + "ci95": [ + 0.7655714285714286, + 0.8628571428571429 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.8628571428571429, + "ci95": [ + 0.8228571428571428, + 0.9028571428571428 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.05142857142857149, + "ci95": [ + 0.02285714285714291, + 0.08000000000000007 + ], + "valid_resamples": 2000 + } + } + } + } + }, + "by_operation": { + "plain_text_display": { + "always_permit_baseline": { + "n": 35, + "positives": 20, + "negatives": 15, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 15, + "fn": 20 + }, + "recall": 0.0, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 0.42857142857142855 + }, + "models": { + "parent": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 0.44020879235141186, + "acc": 0.7428571428571429, + "ece": 0.1010721055158729, + "brier": 0.2970615598224881, + "mean_conf": 0.7840390587691357, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.2571428571428571, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.37142857142857144, + "coverage_at_1pct_error": 0.37142857142857144, + "aurc": 0.08938203390842284, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.04118191591199283, + "top_bins": { + "0.9": { + "n": 9, + "errors": 0, + "error_rate": 0.0 }, - "FalsePositive": { - "precision": 0.6956521739130435, - "recall": 0.7111111111111111, - "f1-score": 0.7032967032967034, - "support": 135.0 + "0.95": { + "n": 6, + "errors": 0, + "error_rate": 0.0 }, - "TruePositive": { - "precision": 0.5, - "recall": 0.21621621621621623, - "f1-score": 0.3018867924528302, - "support": 37.0 + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null } }, - "confusion": [ - [ - 23, - 33, - 3 - ], - [ - 34, - 96, - 5 - ], - [ - 20, - 9, - 8 - ] - ], - "equal_organization_accuracy": 0.6445652173913043, - "organization_results": [ - { - "group": "0d1b7eba9c4d5a5852ef2805e5f1c1018c827e36bfb922fefa4d0a92d76f322c", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "108482877637fd1bbe809b2faddd12c9affef276cdb170667c0234a670019cd4", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 0, - "true_positive_correct": 0 + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.9444444444444444, + "confidence_cutoff": 0.8020615218849417 }, - { - "group": "13a6e1ec428dca68e3cc729ac008e19870e5caa60b3f2d8d013eeb5a4f25aa71", + "0.8": { + "coverage": 0.8, + "accuracy": 0.7857142857142857, + "confidence_cutoff": 0.6323033624549943 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.4703692241803648, + "acc": 0.7428571428571429, + "ece": 0.1062363699460275, + "brier": 0.3059826176210274, + "mean_conf": 0.7080594224133254, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.05714285714285714, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.37142857142857144, + "coverage_at_1pct_error": 0.37142857142857144, + "aurc": 0.08938203390842284, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.03479772044381746, + "top_bins": { + "0.9": { "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17515941d0c83a2c0515166a1503db5e3d24c63b53c0398a9c4f3345d3ef5eda", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "22e3ca4ff2df3a13f6db322d247b7b7f8742181d456be68cb84fb4dfbda7b993", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5667ea75698bc671a6165fb7604bf6d7b7874ba2ef04005f82edf2811b406b53", - "n": 20, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "5cd724f6f8c622cf83c9ee56eeb3e956e4c0428f4bb398608581d51b9f3bd4cf", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5fd9d56568dd7e69c4871b023d3df17203f2eb6d9aaf7a808371a391407f9026", - "n": 20, - "accuracy": 0.05, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "66288757ecaf36bf368cd8d4c43f454658be762a12957c52eae438ca3e727ad8", - "n": 8, - "accuracy": 0.375, - "true_positive_n": 6, - "true_positive_correct": 1 - }, - { - "group": "6d53f455fa642b810781930d63db6b35443da067812f1bf329a37de938179aed", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "79707d8a1f7240a758d67ba5212ab21ef286647a4b71ba71c892f88341f3f4e6", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 1 - }, - { - "group": "7b4cb8f8c017bd94da14e9460c49cdf44fa47f6eab9726918b70ebb7b2695ee1", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "7f55e216b81a67b2fcebf627b534d1d18d9d734f31cb41477725aaa041f6d42b", - "n": 20, - "accuracy": 0.55, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "974cf500c91fb4e80273489c70f89803fe8eaf2fc9d4e68a6aac899d0e580551", - "n": 20, - "accuracy": 0.15, - "true_positive_n": 18, - "true_positive_correct": 1 - }, - { - "group": "97dbafe0dc6d23395bb2008b78ccf190cf0710c18154cbbfc664b3d388086baf", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a756d4fa124cf854b444faed605f6fee522c5418d960b8cd60a8a06c44ef56b0", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "adcffe87f88661c654325a95ccbc252777e27ff83f947ff85ea77cf167ad57c0", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 0, + "errors": 0, + "error_rate": null + }, + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.9444444444444444, + "confidence_cutoff": 0.702937726558054 }, - { - "group": "b4edbf01a09fe17e2fc4eba076213094ab09b40f0ff73c9d3e5c9592f0f59353", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.8": { + "coverage": 0.8, + "accuracy": 0.7857142857142857, + "confidence_cutoff": 0.5826614347130625 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 20, + "correct": 14, + "recall": 0.7 + }, + "permitted": { + "n": 15, + "correct": 12, + "recall": 0.8 + } + }, + "threshold": 0.8875371629397495, + "n": 35, + "positives": 20, + "negatives": 15, + "confusion": { + "tp": 1, + "fp": 0, + "tn": 15, + "fn": 19 + }, + "recall": 0.05, + "false_alert_rate": 0.0, + "precision": 1.0, + "threshold_accuracy": 0.45714285714285713 + }, + "candidate": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 0.2308958263556794, + "acc": 0.8857142857142857, + "ece": 0.05276786009948832, + "brier": 0.1387650722964353, + "mean_conf": 0.9185510684713508, + "confident_error_rate": 0.02857142857142857, + "coverage_at_0_9": 0.7714285714285715, + "accuracy_at_0_9": 0.9629629629629629, + "coverage_at_5pct_error": 0.8857142857142857, + "coverage_at_1pct_error": 0.6285714285714286, + "aurc": 0.018915971835104774, + "error_rate_at_0_9": 0.037037037037037035, + "confidence_bias": 0.03283678275706514, + "top_bins": { + "0.9": { + "n": 27, + "errors": 1, + "error_rate": 0.037037037037037035 + }, + "0.95": { + "n": 19, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "bfe24bf96507627d9c2e0eb183e0cb607aa50a5195abded30b612c252216e7b0", - "n": 1, + "0.99": { + "n": 6, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9562865363749455 }, - { - "group": "cbdba8017701da737b2f23ee11598f766b99054550311935a5730acb5f7e57d9", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.8": { + "coverage": 0.8, + "accuracy": 0.9642857142857143, + "confidence_cutoff": 0.8891588391985842 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.2642131240699265, + "acc": 0.8857142857142857, + "ece": 0.08058920252061248, + "brier": 0.1465567737525219, + "mean_conf": 0.8412751407359753, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.4, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.8857142857142857, + "coverage_at_1pct_error": 0.6285714285714286, + "aurc": 0.018915971835104774, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.04443914497831036, + "top_bins": { + "0.9": { + "n": 14, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "cc3696bd6f2c8a2d65d8376338102471db2565326ab124811ce53c428e54425e", + "0.95": { "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "d592c0767849012441fe14ab7451f4371421ba152f26c3587463adcf58b911e5", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "errors": 0, + "error_rate": 0.0 }, - { - "group": "f562314a5da425b50c820fa25a6756b09f54b042ff1a3a4b580647dd4c20aa7d", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null } - ], - "correct": 127 - }, - "global_calibrated": { - "n": 231, - "accuracy": 0.5497835497835498, - "nll": 0.7508521265465798, - "brier": 0.4566593103734014, - "macro_f1": 0.4478062632890602, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.2987012987012987, - "recall": 0.3898305084745763, - "f1-score": 0.3382352941176471, - "support": 59.0 - }, - "FalsePositive": { - "precision": 0.6956521739130435, - "recall": 0.7111111111111111, - "f1-score": 0.7032967032967034, - "support": 135.0 + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 1.0, + "confidence_cutoff": 0.8623076992723852 }, - "TruePositive": { - "precision": 0.5, - "recall": 0.21621621621621623, - "f1-score": 0.3018867924528302, - "support": 37.0 + "0.8": { + "coverage": 0.8, + "accuracy": 0.9642857142857143, + "confidence_cutoff": 0.7752279376903202 } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 20, + "correct": 20, + "recall": 1.0 }, - "confusion": [ - [ - 23, - 33, - 3 + "permitted": { + "n": 15, + "correct": 11, + "recall": 0.7333333333333333 + } + }, + "threshold": 0.8449815374033091, + "n": 35, + "positives": 20, + "negatives": 15, + "confusion": { + "tp": 12, + "fp": 0, + "tn": 15, + "fn": 8 + }, + "recall": 0.6, + "false_alert_rate": 0.0, + "precision": 1.0, + "threshold_accuracy": 0.7714285714285715 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": 0.05, + "ci95": [ + 0.0, + 0.16666666666666666 ], - [ - 34, - 96, - 5 + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.6, + "ci95": [ + 0.38095238095238093, + 0.8095238095238095 ], - [ - 20, - 9, - 8 - ] - ], - "equal_organization_accuracy": 0.6445652173913043, - "organization_results": [ - { - "group": "0d1b7eba9c4d5a5852ef2805e5f1c1018c827e36bfb922fefa4d0a92d76f322c", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "108482877637fd1bbe809b2faddd12c9affef276cdb170667c0234a670019cd4", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "13a6e1ec428dca68e3cc729ac008e19870e5caa60b3f2d8d013eeb5a4f25aa71", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17515941d0c83a2c0515166a1503db5e3d24c63b53c0398a9c4f3345d3ef5eda", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "22e3ca4ff2df3a13f6db322d247b7b7f8742181d456be68cb84fb4dfbda7b993", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.5499999999999999, + "ci95": [ + 0.3333333333333333, + 0.7728535353535347 + ], + "valid_resamples": 2000 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 1232 + }, + "candidate": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 1232 + } + }, + "threshold_accuracy": { + "parent": { + "value": 0.45714285714285713, + "ci95": [ + 0.2857142857142857, + 0.6285714285714286 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.7714285714285715, + "ci95": [ + 0.6285714285714286, + 0.8857142857142857 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.31428571428571433, + "ci95": [ + 0.17142857142857149, + 0.48571428571428565 + ], + "valid_resamples": 2000 + } + } + } + } + }, + "html_parse": { + "always_permit_baseline": { + "n": 35, + "positives": 14, + "negatives": 21, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 21, + "fn": 14 + }, + "recall": 0.0, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 0.6 + }, + "models": { + "parent": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 1.9047291463003604, + "acc": 0.42857142857142855, + "ece": 0.5258122169538513, + "brier": 1.0143214124687336, + "mean_conf": 0.95438364552528, + "confident_error_rate": 0.4857142857142857, + "coverage_at_0_9": 0.9142857142857143, + "accuracy_at_0_9": 0.46875, + "coverage_at_5pct_error": 0.14285714285714285, + "coverage_at_1pct_error": 0.14285714285714285, + "aurc": 0.3679267902406343, + "error_rate_at_0_9": 0.53125, + "confidence_bias": 0.5258122169538515, + "top_bins": { + "0.9": { + "n": 32, + "errors": 17, + "error_rate": 0.53125 + }, + "0.95": { + "n": 27, + "errors": 14, + "error_rate": 0.5185185185185185 + }, + "0.99": { + "n": 5, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.5555555555555556, + "confidence_cutoff": 0.9801112406947217 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.4642857142857143, + "confidence_cutoff": 0.9495538881263412 + } + } + }, + "calibrated": { + "n": 35, + "nll": 1.2659489508436006, + "acc": 0.42857142857142855, + "ece": 0.4588216874901758, + "brier": 0.8687995501112846, + "mean_conf": 0.8873931160616043, + "confident_error_rate": 0.2571428571428571, + "coverage_at_0_9": 0.6, + "accuracy_at_0_9": 0.5714285714285714, + "coverage_at_5pct_error": 0.14285714285714285, + "coverage_at_1pct_error": 0.14285714285714285, + "aurc": 0.3679267902406343, + "error_rate_at_0_9": 0.42857142857142855, + "confidence_bias": 0.45882168749017577, + "top_bins": { + "0.9": { + "n": 21, + "errors": 9, + "error_rate": 0.42857142857142855 + }, + "0.95": { + "n": 3, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "5667ea75698bc671a6165fb7604bf6d7b7874ba2ef04005f82edf2811b406b53", - "n": 20, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "5cd724f6f8c622cf83c9ee56eeb3e956e4c0428f4bb398608581d51b9f3bd4cf", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5fd9d56568dd7e69c4871b023d3df17203f2eb6d9aaf7a808371a391407f9026", - "n": 20, - "accuracy": 0.05, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "66288757ecaf36bf368cd8d4c43f454658be762a12957c52eae438ca3e727ad8", + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.5555555555555556, + "confidence_cutoff": 0.9167662540005418 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.4642857142857143, + "confidence_cutoff": 0.8589695610685519 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 14, + "correct": 14, + "recall": 1.0 + }, + "permitted": { + "n": 21, + "correct": 1, + "recall": 0.047619047619047616 + } + }, + "threshold": 0.8875371629397495, + "n": 35, + "positives": 14, + "negatives": 21, + "confusion": { + "tp": 12, + "fp": 9, + "tn": 12, + "fn": 2 + }, + "recall": 0.8571428571428571, + "false_alert_rate": 0.42857142857142855, + "precision": 0.5714285714285714, + "threshold_accuracy": 0.6857142857142857 + }, + "candidate": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 1.7814413567537535, + "acc": 0.45714285714285713, + "ece": 0.4870901667090771, + "brier": 0.9601370731320206, + "mean_conf": 0.9442330238519343, + "confident_error_rate": 0.42857142857142855, + "coverage_at_0_9": 0.8285714285714286, + "accuracy_at_0_9": 0.4827586206896552, + "coverage_at_5pct_error": 0.2, + "coverage_at_1pct_error": 0.2, + "aurc": 0.31511035965021955, + "error_rate_at_0_9": 0.5172413793103449, + "confidence_bias": 0.4870901667090772, + "top_bins": { + "0.9": { + "n": 29, + "errors": 15, + "error_rate": 0.5172413793103449 + }, + "0.95": { + "n": 25, + "errors": 12, + "error_rate": 0.48 + }, + "0.99": { "n": 8, - "accuracy": 0.375, - "true_positive_n": 6, - "true_positive_correct": 1 - }, - { - "group": "6d53f455fa642b810781930d63db6b35443da067812f1bf329a37de938179aed", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "79707d8a1f7240a758d67ba5212ab21ef286647a4b71ba71c892f88341f3f4e6", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 1 - }, - { - "group": "7b4cb8f8c017bd94da14e9460c49cdf44fa47f6eab9726918b70ebb7b2695ee1", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "7f55e216b81a67b2fcebf627b534d1d18d9d734f31cb41477725aaa041f6d42b", - "n": 20, - "accuracy": 0.55, - "true_positive_n": 0, - "true_positive_correct": 0 + "errors": 1, + "error_rate": 0.125 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.6666666666666666, + "confidence_cutoff": 0.9770944745872364 }, - { - "group": "974cf500c91fb4e80273489c70f89803fe8eaf2fc9d4e68a6aac899d0e580551", - "n": 20, - "accuracy": 0.15, - "true_positive_n": 18, - "true_positive_correct": 1 + "0.8": { + "coverage": 0.8, + "accuracy": 0.4642857142857143, + "confidence_cutoff": 0.9312263284660277 + } + } + }, + "calibrated": { + "n": 35, + "nll": 1.1646094584375128, + "acc": 0.45714285714285713, + "ece": 0.4387100296237643, + "brier": 0.8119992696195855, + "mean_conf": 0.8734203345049698, + "confident_error_rate": 0.2, + "coverage_at_0_9": 0.5428571428571428, + "accuracy_at_0_9": 0.631578947368421, + "coverage_at_5pct_error": 0.2, + "coverage_at_1pct_error": 0.2, + "aurc": 0.31511035965021955, + "error_rate_at_0_9": 0.3684210526315789, + "confidence_bias": 0.4162774773621127, + "top_bins": { + "0.9": { + "n": 19, + "errors": 7, + "error_rate": 0.3684210526315789 }, - { - "group": "97dbafe0dc6d23395bb2008b78ccf190cf0710c18154cbbfc664b3d388086baf", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.95": { + "n": 6, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "a756d4fa124cf854b444faed605f6fee522c5418d960b8cd60a8a06c44ef56b0", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.99": { + "n": 1, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.6666666666666666, + "confidence_cutoff": 0.903057582586965 }, - { - "group": "adcffe87f88661c654325a95ccbc252777e27ff83f947ff85ea77cf167ad57c0", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.8": { + "coverage": 0.8, + "accuracy": 0.4642857142857143, + "confidence_cutoff": 0.8248194570355916 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 14, + "correct": 14, + "recall": 1.0 + }, + "permitted": { + "n": 21, + "correct": 2, + "recall": 0.09523809523809523 + } + }, + "threshold": 0.8449815374033091, + "n": 35, + "positives": 14, + "negatives": 21, + "confusion": { + "tp": 12, + "fp": 12, + "tn": 9, + "fn": 2 + }, + "recall": 0.8571428571428571, + "false_alert_rate": 0.5714285714285714, + "precision": 0.5, + "threshold_accuracy": 0.6 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": 0.8571428571428571, + "ci95": [ + 0.6666666666666666, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.8571428571428571, + "ci95": [ + 0.6666666666666666, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.42857142857142855, + "ci95": [ + 0.21739130434782608, + 0.6521739130434783 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.5714285714285714, + "ci95": [ + 0.3598235294117647, + 0.7858082706766912 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.14285714285714285, + "ci95": [ + 0.0, + 0.3333333333333333 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": 0.5714285714285714, + "ci95": [ + 0.35, + 0.782608695652174 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.5, + "ci95": [ + 0.2916666666666667, + 0.7058823529411765 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": -0.0714285714285714, + "ci95": [ + -0.1768421052631578, + 0.0 + ], + "valid_resamples": 2000 + } + }, + "threshold_accuracy": { + "parent": { + "value": 0.6857142857142857, + "ci95": [ + 0.5142857142857142, + 0.8285714285714286 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.6, + "ci95": [ + 0.42857142857142855, + 0.7714285714285715 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": -0.08571428571428574, + "ci95": [ + -0.20000000000000007, + 0.0 + ], + "valid_resamples": 2000 + } + } + } + } + }, + "onward_request_body": { + "always_permit_baseline": { + "n": 35, + "positives": 1, + "negatives": 34, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 34, + "fn": 1 + }, + "recall": 0.0, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 0.9714285714285714 + }, + "models": { + "parent": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 0.13064358196820833, + "acc": 0.9714285714285714, + "ece": 0.082650843776636, + "brier": 0.062386596130186385, + "mean_conf": 0.8959570370446475, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.6857142857142857, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9428571428571428, + "aurc": 0.0016566626650660265, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.07547153438392395, + "top_bins": { + "0.9": { + "n": 24, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 16, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "b4edbf01a09fe17e2fc4eba076213094ab09b40f0ff73c9d3e5c9592f0f59353", + "0.99": { "n": 7, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bfe24bf96507627d9c2e0eb183e0cb607aa50a5195abded30b612c252216e7b0", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9326748524721808 }, - { - "group": "cbdba8017701da737b2f23ee11598f766b99054550311935a5730acb5f7e57d9", - "n": 3, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.8734922079816011 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.20940801007318904, + "acc": 0.9714285714285714, + "ece": 0.14463259695441177, + "brier": 0.098760907973673, + "mean_conf": 0.8267959744741596, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.37142857142857144, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9428571428571428, + "aurc": 0.0016566626650660265, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.14463259695441177, + "top_bins": { + "0.9": { + "n": 13, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "cc3696bd6f2c8a2d65d8376338102471db2565326ab124811ce53c428e54425e", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 + "0.95": { + "n": 7, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "d592c0767849012441fe14ab7451f4371421ba152f26c3587463adcf58b911e5", + "0.99": { "n": 1, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.8345254161812303 }, - { - "group": "f562314a5da425b50c820fa25a6756b09f54b042ff1a3a4b580647dd4c20aa7d", - "n": 20, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.7666347623294629 } - ], - "correct": 127 - } - }, - "native": { - "raw": { - "n": 410, - "accuracy": 0.9975609756097561, - "nll": 0.012999284410721951, - "brier": 0.004191458920366933, - "correct": 409 - }, - "global_calibrated": { - "n": 410, - "accuracy": 0.9975609756097561, - "nll": 0.030200841306362223, - "brier": 0.006697687496008216, - "correct": 409 - } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 1, + "correct": 1, + "recall": 1.0 + }, + "permitted": { + "n": 34, + "correct": 33, + "recall": 0.9705882352941176 + } + }, + "threshold": 0.8875371629397495, + "n": 35, + "positives": 1, + "negatives": 34, + "confusion": { + "tp": 1, + "fp": 0, + "tn": 34, + "fn": 0 + }, + "recall": 1.0, + "false_alert_rate": 0.0, + "precision": 1.0, + "threshold_accuracy": 1.0 }, - "guide_detector": { + "candidate": { + "questions": 35, + "groups": 35, "raw": { - "n": 1064, - "accuracy": 0.5855263157894737, - "nll": 0.8173590431098086, - "brier": 0.47423653600178534, - "macro_f1": 0.5415438082741008, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.4096153846153846, - "recall": 0.7580071174377224, - "f1-score": 0.5318352059925093, - "support": 281.0 - }, - "FalsePositive": { - "precision": 0.7873831775700935, - "recall": 0.6686507936507936, - "f1-score": 0.723175965665236, - "support": 504.0 - }, - "TruePositive": { - "precision": 0.6293103448275862, - "recall": 0.2616487455197133, - "f1-score": 0.369620253164557, - "support": 279.0 + "n": 35, + "nll": 0.005026554467109394, + "acc": 1.0, + "ece": 0.004983623817000349, + "brier": 0.0001685963296296438, + "mean_conf": 0.9950163761829997, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.004983623817000349, + "top_bins": { + "0.9": { + "n": 35, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 35, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 31, + "errors": 0, + "error_rate": 0.0 } }, - "confusion": [ - [ - 213, - 60, - 8 - ], - [ - 132, - 337, - 35 - ], - [ - 175, - 31, - 73 - ] - ], - "equal_organization_accuracy": 0.649889118396736, - "organization_results": [ - { - "group": "0228d088db03da24994da8d12bb80e2925bf424387c12a0cab440661f61e53db", - "n": 2, - "accuracy": 0.0, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "03b2cff809e7692ca15f889834563f866b356cc60a41023c8611dd2b60bcf51a", - "n": 2, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "04eb81bd28a99e8d254c7a0a9c07234f55af1f9e4fe167269f504109495c29a5", - "n": 4, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06646577f2484818ed5044762571197f87d3e79272459df5462e193f43ddd530", - "n": 1, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9974550323398608 }, - { - "group": "06f295d348152073a6bf3ba5485ca147a0c14bb94baf88391d465d7ea7e6eca9", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "08b767258688d27f07153aa89f0a2c2c6e8e426289043614eb48c2c881591192", - "n": 17, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "09890e550c1a9a583cf9aaa2a997c1bbc61162a5d36c8a891bbbc1a837dffc36", - "n": 20, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "0b8649cde9e63fe45d17a7941ef90f01d1b3068c464c013d3ff7e21047164b9c", - "n": 5, - "accuracy": 0.8, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "10f18e39bc3c8e8713158ad66e666edf9118b8b83fb7eb21e5e39476a75a3125", - "n": 12, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14919c2e5da07951c425b98441bab1107d02cbcb0496a9148b184b1fea1eb4d2", - "n": 3, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14a068f7500ef993a7f2e4cfcd5e194c57217338de4e72cf848d710c20392e91", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 18, - "true_positive_correct": 0 - }, - { - "group": "16cb7259665afd83462953cc90bf025352bab6f7210a59e5d86d49561f03fcbe", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "16fe8b7b9a0987bf861cee078b8763e8f75ec46733a6a59b61ffa8e2bd6e49f7", - "n": 5, + "confidence_cutoff": 0.9932663672241991 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.03405651326059574, + "acc": 1.0, + "ece": 0.033071042578584564, + "brier": 0.0037297689099056244, + "mean_conf": 0.9669289574214154, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9428571428571428, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.03307104257858462, + "top_bins": { + "0.9": { + "n": 33, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 28, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 6, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9720891179694704 }, - { - "group": "17cd31c2cd2d2ccbabdfab902aee6e4e759167d126e1bcaf4d638c23b93272ac", - "n": 17, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "18aa7b3932a87bfdfc7e3f032e91b38999f82e037146754bd8bba044870170ac", - "n": 2, + "confidence_cutoff": 0.9511715473570618 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 1, + "correct": 1, + "recall": 1.0 + }, + "permitted": { + "n": 34, + "correct": 34, + "recall": 1.0 + } + }, + "threshold": 0.8449815374033091, + "n": 35, + "positives": 1, + "negatives": 34, + "confusion": { + "tp": 1, + "fp": 0, + "tn": 34, + "fn": 0 + }, + "recall": 1.0, + "false_alert_rate": 0.0, + "precision": 1.0, + "threshold_accuracy": 1.0 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 1272 + }, + "candidate": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 1272 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 1272 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 1272 + }, + "candidate": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 1272 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 1272 + } + }, + "threshold_accuracy": { + "parent": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + } + } + } + } + }, + "code_execution": { + "always_permit_baseline": { + "n": 35, + "positives": 0, + "negatives": 35, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 35, + "fn": 0 + }, + "recall": null, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 1.0 + }, + "models": { + "parent": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 0.03495771061017535, + "acc": 1.0, + "ece": 0.03371089311351877, + "brier": 0.004643703343002468, + "mean_conf": 0.9662891068864811, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9714285714285714, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.033710893113518936, + "top_bins": { + "0.9": { + "n": 34, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 27, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 11, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9787669092307943 }, - { - "group": "191618c5651b68d639db0f90f67315430ae08a9467ebc761ca10a54847034de5", - "n": 20, - "accuracy": 0.75, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "1e7cf0ee593015038b9cf7686c7db7604b7de8b52d0a8b556038b7795a59dd5a", - "n": 19, - "accuracy": 0.7894736842105263, - "true_positive_n": 4, - "true_positive_correct": 0 - }, - { - "group": "23f95dce880a8574d254a19ac7f94f1284a556dfa61126a92844ba979083ede1", - "n": 2, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2413f2cb2e9a29595c0278b868a30250d3573d487f99e35b3fe3836713d79e96", - "n": 2, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "280f0d43bcdf087751b7a2b90739860bc02c2cccea845f81fae7555cdd9d7745", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 + "confidence_cutoff": 0.9455319719721502 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.10667969626313498, + "acc": 1.0, + "ece": 0.09922627250314853, + "brier": 0.02646181525533936, + "mean_conf": 0.9007737274968517, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.5714285714285714, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0992262725031483, + "top_bins": { + "0.9": { + "n": 20, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 9, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "2af6f963a1301ac7cbf959fb0f63442e6b1b28be622e0d74c6b94c1e459c8bf0", - "n": 20, + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9135753218620322 }, - { - "group": "2d4fd6af7767d8fbcfeddff36a76b5c993a49d5e45a515806ec30b1b53afa2ce", - "n": 1, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3229724a48ff16d1ea5cff92fcbc1a29fb636760780576606a1f931cba64c7e6", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "33593514e9eb3b4a2f11fbff9a900e304d7f7088413e1a6ed36210de11e1a7cd", - "n": 1, + "confidence_cutoff": 0.8528241762602075 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 0, + "correct": 0, + "recall": null + }, + "permitted": { + "n": 35, + "correct": 35, + "recall": 1.0 + } + }, + "threshold": 0.8875371629397495, + "n": 35, + "positives": 0, + "negatives": 35, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 35, + "fn": 0 + }, + "recall": null, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 1.0 + }, + "candidate": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 0.003713040174857426, + "acc": 1.0, + "ece": 0.0036946893416571758, + "brier": 7.269134859423068e-05, + "mean_conf": 0.9963053106583428, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0036946893416571758, + "top_bins": { + "0.9": { + "n": 35, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 35, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 31, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "357e4fcd5f01346ea07a7b6805f5e0e94f11cd39d870ebde997c06b3ca23403b", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "36be30d9fffc6077ec297a0ccde6690eb34709163c58b4b5df24236dd44c0e91", - "n": 18, - "accuracy": 0.9444444444444444, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9983259960008511 }, - { - "group": "3788764c156c1d8107b5e65fc226f169b6cfeb28e92341b3776e0323894d1182", - "n": 20, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "397381ff8e300cf10164a627a5a6e614e68a7fb2a4dbac6daa5c67e331248cb3", - "n": 2, + "confidence_cutoff": 0.9949578958858085 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.029845462374630014, + "acc": 1.0, + "ece": 0.02916307563639764, + "brier": 0.002620264819563586, + "mean_conf": 0.9708369243636024, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.02916307563639764, + "top_bins": { + "0.9": { + "n": 35, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 30, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 5, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3a3a67375cabf27ee46ffe17a8e3122e2bc0d77c22a4bc5fd3caf21af2fdf827", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "3ba2e76ba41ace83ac5fc70feb81f7bd790ebcc01cef912db9a520e0659301e6", - "n": 20, - "accuracy": 0.3, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "3c519ebea078526e0c7724d67b71d0e9d928d62adbb68de9cc3ab45caeab0ba6", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 0, - "true_positive_correct": 0 + "confidence_cutoff": 0.9781193495148215 }, - { - "group": "42a20c36367ded75633c56b5468efe4dab76b5242f7fc0fafe5de10ae8b9a078", - "n": 2, + "0.8": { + "coverage": 0.8, "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "4f4986e7532d0ba637710cb0010206dd22720f9b35cc163fd2f833fc93c25334", - "n": 14, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "52ed406f643d4adbab570be68c48860e0e9f239fe00b7f1bd0c35e00868fc879", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "557c22b05ccf851b94b1fc4571234d237e761ded29b151b3a3adcacd17f9ba00", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "58f3e1e6f3eca7efe52c2f97b888733f6a541ca293e36d6e1a1f22c2332d05fb", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d11eb324b736d8f810b14b708ea1982028cb405fb0fb2e3f20dab9ab5bb3fe1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d2fc2acbf5e2afa16d5ad87f1cce4b784cdd43ca2da54c3eea33521e52a5a81", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 10, - "true_positive_correct": 0 - }, - { - "group": "5f21eec69194bcd2e80220b2701002f2da10e7cd2b04625674cdf69feb57dedd", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61c9e0872a43ed30a90ab06f7a3e8b85d33f54355b964a7fd1ffa361531c5626", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61e7fb88134413ce324703852a01f81dda0cb96d4ae07a44cf845c9b1de8baa0", - "n": 20, - "accuracy": 0.25, - "true_positive_n": 13, - "true_positive_correct": 0 - }, - { - "group": "6201b69130be0b440362fc95b4cc2ef6c4c856543d9beb9584d43899e72d0283", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "629dffdc159a2dbb566dab56a6f17419a63aa2b0815020338d686277ff90f623", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "63b98074727eecae10e68ab106f9c5f2ec98bd2347e0c3eaa72d5557c1ae1b64", - "n": 7, - "accuracy": 0.42857142857142855, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "63deca8572354a01b85154c34b64122772c89023b59c7931064bb347d8389e7c", + "confidence_cutoff": 0.9586084771058523 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 0, + "correct": 0, + "recall": null + }, + "permitted": { + "n": 35, + "correct": 35, + "recall": 1.0 + } + }, + "threshold": 0.8449815374033091, + "n": 35, + "positives": 0, + "negatives": 35, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 35, + "fn": 0 + }, + "recall": null, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 1.0 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": null, + "ci95": null, + "valid_resamples": 0 + }, + "candidate": { + "value": null, + "ci95": null, + "valid_resamples": 0 + }, + "candidate_minus_parent": { + "value": null, + "ci95": null, + "valid_resamples": 0 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": null, + "ci95": null, + "valid_resamples": 0 + }, + "candidate": { + "value": null, + "ci95": null, + "valid_resamples": 0 + }, + "candidate_minus_parent": { + "value": null, + "ci95": null, + "valid_resamples": 0 + } + }, + "threshold_accuracy": { + "parent": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 1.0, + "ci95": [ + 1.0, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 2000 + } + } + } + } + }, + "console_output": { + "always_permit_baseline": { + "n": 35, + "positives": 0, + "negatives": 35, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 35, + "fn": 0 + }, + "recall": null, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 1.0 + }, + "models": { + "parent": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 1.4669736827879394, + "acc": 0.4857142857142857, + "ece": 0.44969603628529076, + "brier": 0.8741636560106545, + "mean_conf": 0.8196176344527856, + "confident_error_rate": 0.2, + "coverage_at_0_9": 0.4, + "accuracy_at_0_9": 0.5, + "coverage_at_5pct_error": 0.0, + "coverage_at_1pct_error": 0.0, + "aurc": 0.5978180747557157, + "error_rate_at_0_9": 0.5, + "confidence_bias": 0.33390334873849986, + "top_bins": { + "0.9": { "n": 14, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "659e4d7e826303db51376a34ed1ee1187305d8b4fcd2ba86de0f1df79cbe9d09", - "n": 20, - "accuracy": 0.95, - "true_positive_n": 20, - "true_positive_correct": 19 - }, - { - "group": "681a13deadc6006e0d4d996f2b006e35b1d71fdeaaebfe98db36b84ce9e1bb82", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6972efb6c1c8c5719fc446f6b4e9bbf812c8274c04ddce38ae95cf5b06e65e10", - "n": 12, - "accuracy": 0.9166666666666666, - "true_positive_n": 5, - "true_positive_correct": 5 - }, - { - "group": "6d2a8e5968c904fbeceb057dc6d374d5cd1ba38019b8d31da27cc658c12aa476", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6d5ea32830050e4ac2840eed05d6b46ec9d7a78f65fbf85d0742d83cc628fa9c", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "6d74b56adb5791668e05d491f03f083060be7b184f47660a3b9cc14d7a32b511", - "n": 13, - "accuracy": 1.0, - "true_positive_n": 13, - "true_positive_correct": 13 - }, - { - "group": "6e73fd09bd5175004650b33517e33671c77ebb649384e14ffd95309ac3e567a7", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6f07feecb02e317c2db269ed9ec22266aa9195e11612cf7362bcdd1f3c9a7d3c", - "n": 20, - "accuracy": 0.9, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6fc0443881957355817ec4b23a2d3027d6ce1ba949d0c313ac5134c5a33d7752", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "725e8bad3bf121140d3f1e8bb8625fac12766a1a7ebea88a432b39f8af246111", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "745320acbe1a562a24246846ea4e7c92fe29683dd9e75636a6183142147be543", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "77af40ba5530d0abf3b141de72a4c48f4d594f8bc50515d71d71bb0ef57ba861", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "798e40bc3bfcfea7bb0935e395494ac4836e1d49f294b2ca803f7e234a478493", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "79e0a2063b1c1395f9032d7bcca4a7987b25c134851f2b7087535b66973f8767", - "n": 3, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "7c2c1e933c21cfd9f0534226991df5542463f0d2c40138378024cbbabc1f205f", - "n": 20, - "accuracy": 0.55, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "7d092ad55211811544dc5993b69d906906fce9350e187508760b25a549bbc9ad", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 5, - "true_positive_correct": 0 - }, - { - "group": "8269675fe1c1fcb6500b8262056237a7eb277d8d7822e63cc1b335c9b3587400", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 + "errors": 7, + "error_rate": 0.5 }, - { - "group": "8433ad4a51679bb9fc21396e090e351720fb331d25292de538c1e1a8ca524bab", + "0.95": { "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8d353f585c30ae854c944e56bcab8cfff771273fefcde8d929e6faf7cb75488f", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8da5a2b78094becb52357364f9f1ea558c4052e869e9636c912fa88cce650666", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8e11912bbdeb4ce0b56ca82a5e0ad37a9159b78dd620607541dc93cb9ac8e02e", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8f019450f830cbc42ded96cce0b7b4c1ee38439a9e198406ffb38ba5e0524fd7", - "n": 4, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "90994dcfeb71c7a87a4784556f5bf0bdb0b542c15a5f6b92da96e8020c4f1bf1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "94d54be3b552337f2fc772ce8ff09d2b668ec497130ded8b947506a120acfb4e", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 14, - "true_positive_correct": 14 + "errors": 4, + "error_rate": 0.5 }, - { - "group": "97d7c3f3b10a157cdb2791ff320a99a7d1856f8bbdd7ee020ad9ddbe7ec14b8e", - "n": 13, - "accuracy": 0.0, - "true_positive_n": 13, - "true_positive_correct": 0 - }, - { - "group": "97f8ab62aee5d5d53245a565b836e892a9a7cda46748be23097421abb76ee39a", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 6, - "true_positive_correct": 0 - }, - { - "group": "99a049ff4aba4bc20df5f424b24efb9bab9118c192fd17f29b31ec910e497def", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "9b54c928a43ef6ff8b5b53cfeaad5e554106ed4d6718d08bc4f7440e968414af", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "9bcc91cc05b92cb30cead1b81acb706497144e4815da75642e99d3fd7ebfd154", - "n": 11, - "accuracy": 0.9090909090909091, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9d2f045465f59eb2bcb8a37818341ce3635a125bced7bfbcb7bf9acdd80c1a71", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "9e08a63fef975cba0f7fe19b9931144a9480e43b5b78497f97cfe4079c916045", - "n": 10, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9e38ecba2d710c7a9e3c2bef6abd472773b0edae845c5af26f317a4f5a2a3bc2", - "n": 19, - "accuracy": 0.10526315789473684, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "a04b59b050c050568baa9abccd205e73547f1e5111afaabdbcbe5949c3511ec6", - "n": 20, - "accuracy": 0.25, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "a1e75cda474e69786231632ed64953c19d60658191ea63553964e1d27015a7ce", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a52745698cb48cb1a28ac9bf212d724174c320e4648c9bececd83422b3364c1d", - "n": 19, - "accuracy": 0.10526315789473684, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a5a3b2d9f1ca0af015ed0584c90b28830aa11e86e8311c7f8f32f26b6d4f2768", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "a68f2269c6412f5885bdd26d3023dae3411476f5dd5b2e103355d4e1b0a41bb5", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a7b13e8d802fa5d584ad83d610714c858aab70e0c1f005324a2a91e7a369aed1", - "n": 16, - "accuracy": 0.0625, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a94db659229b95e557d461f6de9071da0762515528aa3a77d5cb870fcf0f914b", - "n": 20, - "accuracy": 0.75, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aab21c537c79f9eee1a053d0d9f7fbea9eb1451e166156d59dd6c464104a7992", - "n": 10, - "accuracy": 0.7, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "abd52b07ebdf5d9235ca1239d9c3acbe23eb7cabdbc661e0dc93653479a095f0", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aefb87718e7fc3725adba69a4f789e9de7d638ec0a0d1f12fcff8a8735fb622c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b095632bbe63fbdc009ed98e30a50899f6a8edc165e3be8a924a7f0f600be4c0", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 18, - "true_positive_correct": 6 - }, - { - "group": "b1b47d1978d292c9b80319bd2b531d2f2af206e3bcabf9aa0c46472112638ba9", + "0.99": { + "n": 3, + "errors": 2, + "error_rate": 0.6666666666666666 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.3888888888888889, + "confidence_cutoff": 0.8730975086069251 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.39285714285714285, + "confidence_cutoff": 0.6488618260771695 + } + } + }, + "calibrated": { + "n": 35, + "nll": 1.077831441012569, + "acc": 0.4857142857142857, + "ece": 0.4118859974171003, + "brier": 0.7555266144919474, + "mean_conf": 0.7474440931747798, + "confident_error_rate": 0.08571428571428572, + "coverage_at_0_9": 0.17142857142857143, + "accuracy_at_0_9": 0.5, + "coverage_at_5pct_error": 0.0, + "coverage_at_1pct_error": 0.0, + "aurc": 0.5978180747557157, + "error_rate_at_0_9": 0.5, + "confidence_bias": 0.26172980746049407, + "top_bins": { + "0.9": { "n": 6, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b7fa0bdd60d2f2321a27ad4ef1c84b2258bb60cb55372ea8aa56643d6d7a6d2b", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 15, - "true_positive_correct": 0 + "errors": 3, + "error_rate": 0.5 }, - { - "group": "b8b4a7b5769be3d4db661d88d40336ade544fcac1e5bd2ec6606c09f83e5263c", - "n": 7, - "accuracy": 0.7142857142857143, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "b9063c091cb8deaab4ea7fc341aebc169033dfb16792a7f0f253ce89ce556510", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bc9c57f088052889f45c0b709137c494d747464872fca72f16cfdde538281d64", - "n": 15, - "accuracy": 0.8666666666666667, - "true_positive_n": 6, - "true_positive_correct": 4 - }, - { - "group": "bd4aa774324b8affeb489e7b796b6f32a9a2f46442394a8f5042c1967eac05dc", + "0.95": { "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c0371b16a587352143c383ca9799634e07feb94a2df75ef8830adb86efe153c1", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c045d0efa4ee5ab2be1d352625f16f6fb9e5cfbe90e02574363e83cd27bc5109", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "c98451abd4304660daabdb977564d36fdad533b4458c467bd2b97d80a651955f", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "errors": 2, + "error_rate": 1.0 }, - { - "group": "cb85febf11725dacf437b764d64bf097c9c2bab08e74ac2eeadb496ccd51b5e8", + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.3888888888888889, + "confidence_cutoff": 0.7662416927389056 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.39285714285714285, + "confidence_cutoff": 0.5933872468656349 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 0, + "correct": 0, + "recall": null + }, + "permitted": { + "n": 35, + "correct": 17, + "recall": 0.4857142857142857 + } + }, + "threshold": 0.8875371629397495, + "n": 35, + "positives": 0, + "negatives": 35, + "confusion": { + "tp": 0, + "fp": 3, + "tn": 32, + "fn": 0 + }, + "recall": null, + "false_alert_rate": 0.08571428571428572, + "precision": 0.0, + "threshold_accuracy": 0.9142857142857143 + }, + "candidate": { + "questions": 35, + "groups": 35, + "raw": { + "n": 35, + "nll": 0.4711593105222168, + "acc": 0.8571428571428571, + "ece": 0.10032786865714749, + "brier": 0.24921799740853026, + "mean_conf": 0.9258384369166115, + "confident_error_rate": 0.05714285714285714, + "coverage_at_0_9": 0.7714285714285715, + "accuracy_at_0_9": 0.9259259259259259, + "coverage_at_5pct_error": 0.2571428571428571, + "coverage_at_1pct_error": 0.2571428571428571, + "aurc": 0.08024957297121005, + "error_rate_at_0_9": 0.07407407407407407, + "confidence_bias": 0.0686955797737544, + "top_bins": { + "0.9": { + "n": 27, + "errors": 2, + "error_rate": 0.07407407407407407 + }, + "0.95": { + "n": 26, + "errors": 2, + "error_rate": 0.07692307692307693 + }, + "0.99": { "n": 16, - "accuracy": 0.875, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "cfe165abb396d68223741b950f77ec55bd59d4c200bd310d7e9a956548c1590e", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d530b979376d67460a1a3dcafcc95c91e1a51ce2d97124dc2050c2030ae223b7", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 12, - "true_positive_correct": 0 - }, - { - "group": "d5496d95b9bcd56d2d382354f44af6098e30a6b132c6bd2ccbac429ee48d7d3b", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d6e9a943c91ab8d80d913c89d053cf89f2fe4bcd7df1963f3fd5a70bf56f3deb", - "n": 12, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "errors": 2, + "error_rate": 0.125 + } + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.8888888888888888, + "confidence_cutoff": 0.9891438506037644 }, - { - "group": "d794d72182a0c6f8a3009e01f6d607875a60752bf7f05920823cc55729bcbf3c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 + "0.8": { + "coverage": 0.8, + "accuracy": 0.9285714285714286, + "confidence_cutoff": 0.8757747622322081 + } + } + }, + "calibrated": { + "n": 35, + "nll": 0.3789301936796312, + "acc": 0.8571428571428571, + "ece": 0.11430695728032647, + "brier": 0.23203565927974715, + "mean_conf": 0.8718695845973715, + "confident_error_rate": 0.05714285714285714, + "coverage_at_0_9": 0.6857142857142857, + "accuracy_at_0_9": 0.9166666666666666, + "coverage_at_5pct_error": 0.2571428571428571, + "coverage_at_1pct_error": 0.2571428571428571, + "aurc": 0.08024957297121005, + "error_rate_at_0_9": 0.08333333333333333, + "confidence_bias": 0.014726727454514421, + "top_bins": { + "0.9": { + "n": 24, + "errors": 2, + "error_rate": 0.08333333333333333 + }, + "0.95": { + "n": 9, + "errors": 0, + "error_rate": 0.0 }, - { - "group": "d82b9f42323202feb6db7ed45f9990070e72e41705aeb5aa64d60a3fd7549f17", + "0.99": { "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "dfc85bf0674a0000b0dfbcf621419502a19c4fffabf786b4a5e74a69776a1f9d", - "n": 5, - "accuracy": 0.4, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "e35b0e3bc4a28a3bcf49fd9fe5c6e3c6ece59e921ecabf1904c9459cb7b84ad1", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "e6e40718557ace0b1764c2b7ce7021ff2b42e61db43cd59c7a54132f69807b2d", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "f5b769ca29423ada31602cdd07bba0ab057d937256622788fa0872f6fe095865", - "n": 4, - "accuracy": 0.75, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "f72e0a3b72341d46d7eb5ab932fcce07cf34702afa40d217c5e997bef74764be", - "n": 15, - "accuracy": 0.9333333333333333, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "f9476bc6e284b58a4d168d14b188b2ca1636ebe4a0adfcb0c40060bed41781c8", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "f97f37793bfaa1b487125bd7b1f34bff0fed50146a5606962c763220f17bb6de", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fb39d2600c554fe3ac1eddb65704bc34c6102ca0fbb4a77744ac525869d35be4", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fcd64091f029e53dd4a770300ecf446dbf5bca8cc267148f28b198584caeae46", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fe703c5eb241d3542fcc57e11688b3641c4f1f0b9506a720b42a9b9031382154", - "n": 19, - "accuracy": 0.8947368421052632, - "true_positive_n": 2, - "true_positive_correct": 0 + "errors": 0, + "error_rate": 0.0 } - ], - "correct": 623 - }, - "global_calibrated": { - "n": 1064, - "accuracy": 0.5855263157894737, - "nll": 0.7976013493654481, - "brier": 0.46891907838977726, - "macro_f1": 0.5415438082741008, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.4096153846153846, - "recall": 0.7580071174377224, - "f1-score": 0.5318352059925093, - "support": 281.0 - }, - "FalsePositive": { - "precision": 0.7873831775700935, - "recall": 0.6686507936507936, - "f1-score": 0.723175965665236, - "support": 504.0 + }, + "selective": { + "0.5": { + "coverage": 0.5142857142857142, + "accuracy": 0.8888888888888888, + "confidence_cutoff": 0.9360109691977325 }, - "TruePositive": { - "precision": 0.6293103448275862, - "recall": 0.2616487455197133, - "f1-score": 0.369620253164557, - "support": 279.0 + "0.8": { + "coverage": 0.8, + "accuracy": 0.9285714285714286, + "confidence_cutoff": 0.7615630826924717 } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 0, + "correct": 0, + "recall": null }, - "confusion": [ - [ - 213, - 60, - 8 - ], - [ - 132, - 337, - 35 - ], - [ - 175, - 31, - 73 - ] - ], - "equal_organization_accuracy": 0.649889118396736, - "organization_results": [ - { - "group": "0228d088db03da24994da8d12bb80e2925bf424387c12a0cab440661f61e53db", - "n": 2, - "accuracy": 0.0, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "03b2cff809e7692ca15f889834563f866b356cc60a41023c8611dd2b60bcf51a", - "n": 2, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "04eb81bd28a99e8d254c7a0a9c07234f55af1f9e4fe167269f504109495c29a5", - "n": 4, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06646577f2484818ed5044762571197f87d3e79272459df5462e193f43ddd530", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06f295d348152073a6bf3ba5485ca147a0c14bb94baf88391d465d7ea7e6eca9", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "08b767258688d27f07153aa89f0a2c2c6e8e426289043614eb48c2c881591192", - "n": 17, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "09890e550c1a9a583cf9aaa2a997c1bbc61162a5d36c8a891bbbc1a837dffc36", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "0b8649cde9e63fe45d17a7941ef90f01d1b3068c464c013d3ff7e21047164b9c", - "n": 5, - "accuracy": 0.8, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "10f18e39bc3c8e8713158ad66e666edf9118b8b83fb7eb21e5e39476a75a3125", - "n": 12, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14919c2e5da07951c425b98441bab1107d02cbcb0496a9148b184b1fea1eb4d2", - "n": 3, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14a068f7500ef993a7f2e4cfcd5e194c57217338de4e72cf848d710c20392e91", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 18, - "true_positive_correct": 0 - }, - { - "group": "16cb7259665afd83462953cc90bf025352bab6f7210a59e5d86d49561f03fcbe", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "16fe8b7b9a0987bf861cee078b8763e8f75ec46733a6a59b61ffa8e2bd6e49f7", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17cd31c2cd2d2ccbabdfab902aee6e4e759167d126e1bcaf4d638c23b93272ac", - "n": 17, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "18aa7b3932a87bfdfc7e3f032e91b38999f82e037146754bd8bba044870170ac", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "191618c5651b68d639db0f90f67315430ae08a9467ebc761ca10a54847034de5", - "n": 20, + "permitted": { + "n": 35, + "correct": 30, + "recall": 0.8571428571428571 + } + }, + "threshold": 0.8449815374033091, + "n": 35, + "positives": 0, + "negatives": 35, + "confusion": { + "tp": 0, + "fp": 2, + "tn": 33, + "fn": 0 + }, + "recall": null, + "false_alert_rate": 0.05714285714285714, + "precision": 0.0, + "threshold_accuracy": 0.9428571428571428 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": null, + "ci95": null, + "valid_resamples": 0 + }, + "candidate": { + "value": null, + "ci95": null, + "valid_resamples": 0 + }, + "candidate_minus_parent": { + "value": null, + "ci95": null, + "valid_resamples": 0 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.08571428571428572, + "ci95": [ + 0.0, + 0.2 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.05714285714285714, + "ci95": [ + 0.0, + 0.14285714285714285 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": -0.028571428571428574, + "ci95": [ + -0.08571428571428572, + 0.0 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 1918 + }, + "candidate": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 1747 + }, + "candidate_minus_parent": { + "value": 0.0, + "ci95": [ + 0.0, + 0.0 + ], + "valid_resamples": 1747 + } + }, + "threshold_accuracy": { + "parent": { + "value": 0.9142857142857143, + "ci95": [ + 0.8, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.9428571428571428, + "ci95": [ + 0.8571428571428571, + 1.0 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.02857142857142858, + "ci95": [ + 0.0, + 0.08571428571428574 + ], + "valid_resamples": 2000 + } + } + } + } + } + }, + "by_source": { + "sev_swarmtraces_response_policy": { + "always_permit_baseline": { + "n": 175, + "positives": 35, + "negatives": 140, + "confusion": { + "tp": 0, + "fp": 0, + "tn": 140, + "fn": 35 + }, + "recall": 0.0, + "false_alert_rate": 0.0, + "precision": null, + "threshold_accuracy": 0.8 + }, + "models": { + "parent": { + "questions": 175, + "groups": 35, + "raw": { + "n": 175, + "nll": 0.795502582803619, + "acc": 0.7257142857142858, + "ece": 0.1747005023605082, + "brier": 0.4505153855550131, + "mean_conf": 0.8840572965356658, + "confident_error_rate": 0.13714285714285715, + "coverage_at_0_9": 0.6457142857142857, + "accuracy_at_0_9": 0.7876106194690266, + "coverage_at_5pct_error": 0.02857142857142857, + "coverage_at_1pct_error": 0.02857142857142857, + "aurc": 0.1992292401228163, + "error_rate_at_0_9": 0.21238938053097345, + "confidence_bias": 0.15834301082138003, + "top_bins": { + "0.9": { + "n": 113, + "errors": 24, + "error_rate": 0.21238938053097345 + }, + "0.95": { + "n": 84, + "errors": 18, + "error_rate": 0.21428571428571427 + }, + "0.99": { + "n": 26, + "errors": 2, + "error_rate": 0.07692307692307693 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.7613636363636364, + "confidence_cutoff": 0.9457634807080788 + }, + "0.8": { + "coverage": 0.8, "accuracy": 0.75, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "1e7cf0ee593015038b9cf7686c7db7604b7de8b52d0a8b556038b7795a59dd5a", - "n": 19, - "accuracy": 0.7894736842105263, - "true_positive_n": 4, - "true_positive_correct": 0 - }, - { - "group": "23f95dce880a8574d254a19ac7f94f1284a556dfa61126a92844ba979083ede1", - "n": 2, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2413f2cb2e9a29595c0278b868a30250d3573d487f99e35b3fe3836713d79e96", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "280f0d43bcdf087751b7a2b90739860bc02c2cccea845f81fae7555cdd9d7745", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "2af6f963a1301ac7cbf959fb0f63442e6b1b28be622e0d74c6b94c1e459c8bf0", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2d4fd6af7767d8fbcfeddff36a76b5c993a49d5e45a515806ec30b1b53afa2ce", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3229724a48ff16d1ea5cff92fcbc1a29fb636760780576606a1f931cba64c7e6", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "33593514e9eb3b4a2f11fbff9a900e304d7f7088413e1a6ed36210de11e1a7cd", + "confidence_cutoff": 0.7938824081649869 + } + } + }, + "calibrated": { + "n": 175, + "nll": 0.6260474644745716, + "acc": 0.7257142857142858, + "ece": 0.10761147298493057, + "brier": 0.4111063010906543, + "mean_conf": 0.8140932667241441, + "confident_error_rate": 0.06857142857142857, + "coverage_at_0_9": 0.35428571428571426, + "accuracy_at_0_9": 0.8064516129032258, + "coverage_at_5pct_error": 0.02857142857142857, + "coverage_at_1pct_error": 0.02857142857142857, + "aurc": 0.1992292401228163, + "error_rate_at_0_9": 0.1935483870967742, + "confidence_bias": 0.0883789810098583, + "top_bins": { + "0.9": { + "n": 62, + "errors": 12, + "error_rate": 0.1935483870967742 + }, + "0.95": { + "n": 21, + "errors": 2, + "error_rate": 0.09523809523809523 + }, + "0.99": { "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "357e4fcd5f01346ea07a7b6805f5e0e94f11cd39d870ebde997c06b3ca23403b", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "36be30d9fffc6077ec297a0ccde6690eb34709163c58b4b5df24236dd44c0e91", - "n": 18, - "accuracy": 0.9444444444444444, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3788764c156c1d8107b5e65fc226f169b6cfeb28e92341b3776e0323894d1182", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "397381ff8e300cf10164a627a5a6e614e68a7fb2a4dbac6daa5c67e331248cb3", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3a3a67375cabf27ee46ffe17a8e3122e2bc0d77c22a4bc5fd3caf21af2fdf827", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "3ba2e76ba41ace83ac5fc70feb81f7bd790ebcc01cef912db9a520e0659301e6", - "n": 20, - "accuracy": 0.3, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "3c519ebea078526e0c7724d67b71d0e9d928d62adbb68de9cc3ab45caeab0ba6", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "42a20c36367ded75633c56b5468efe4dab76b5242f7fc0fafe5de10ae8b9a078", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "4f4986e7532d0ba637710cb0010206dd22720f9b35cc163fd2f833fc93c25334", - "n": 14, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "52ed406f643d4adbab570be68c48860e0e9f239fe00b7f1bd0c35e00868fc879", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "557c22b05ccf851b94b1fc4571234d237e761ded29b151b3a3adcacd17f9ba00", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "58f3e1e6f3eca7efe52c2f97b888733f6a541ca293e36d6e1a1f22c2332d05fb", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d11eb324b736d8f810b14b708ea1982028cb405fb0fb2e3f20dab9ab5bb3fe1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d2fc2acbf5e2afa16d5ad87f1cce4b784cdd43ca2da54c3eea33521e52a5a81", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 10, - "true_positive_correct": 0 - }, - { - "group": "5f21eec69194bcd2e80220b2701002f2da10e7cd2b04625674cdf69feb57dedd", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61c9e0872a43ed30a90ab06f7a3e8b85d33f54355b964a7fd1ffa361531c5626", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61e7fb88134413ce324703852a01f81dda0cb96d4ae07a44cf845c9b1de8baa0", - "n": 20, - "accuracy": 0.25, - "true_positive_n": 13, - "true_positive_correct": 0 - }, - { - "group": "6201b69130be0b440362fc95b4cc2ef6c4c856543d9beb9584d43899e72d0283", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "629dffdc159a2dbb566dab56a6f17419a63aa2b0815020338d686277ff90f623", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "63b98074727eecae10e68ab106f9c5f2ec98bd2347e0c3eaa72d5557c1ae1b64", - "n": 7, - "accuracy": 0.42857142857142855, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "63deca8572354a01b85154c34b64122772c89023b59c7931064bb347d8389e7c", + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.7613636363636364, + "confidence_cutoff": 0.8531718492565429 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.75, + "confidence_cutoff": 0.6963743924447515 + } + } + }, + "argmax_classes": { + "requires_approval": { + "n": 35, + "correct": 29, + "recall": 0.8285714285714286 + }, + "permitted": { + "n": 140, + "correct": 98, + "recall": 0.7 + } + }, + "threshold": 0.8875371629397495, + "n": 175, + "positives": 35, + "negatives": 140, + "confusion": { + "tp": 14, + "fp": 12, + "tn": 128, + "fn": 21 + }, + "recall": 0.4, + "false_alert_rate": 0.08571428571428572, + "precision": 0.5384615384615384, + "threshold_accuracy": 0.8114285714285714 + }, + "candidate": { + "questions": 175, + "groups": 35, + "raw": { + "n": 175, + "nll": 0.49844721765472333, + "acc": 0.84, + "ece": 0.11735638435435483, + "brier": 0.269672286103042, + "mean_conf": 0.9559888432162478, + "confident_error_rate": 0.10285714285714286, + "coverage_at_0_9": 0.8742857142857143, + "accuracy_at_0_9": 0.8823529411764706, + "coverage_at_5pct_error": 0.6114285714285714, + "coverage_at_1pct_error": 0.44, + "aurc": 0.044632779737602504, + "error_rate_at_0_9": 0.11764705882352941, + "confidence_bias": 0.1159888432162478, + "top_bins": { + "0.9": { + "n": 153, + "errors": 18, + "error_rate": 0.11764705882352941 + }, + "0.95": { + "n": 140, + "errors": 14, + "error_rate": 0.1 + }, + "0.99": { + "n": 92, + "errors": 3, + "error_rate": 0.03260869565217391 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.9772727272727273, + "confidence_cutoff": 0.9904485398763483 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 0.9, + "confidence_cutoff": 0.9516673073011176 + } + } + }, + "calibrated": { + "n": 175, + "nll": 0.37433095036445935, + "acc": 0.84, + "ece": 0.06636288221880407, + "brier": 0.23938834727626476, + "mean_conf": 0.9048661883246669, + "confident_error_rate": 0.05142857142857143, + "coverage_at_0_9": 0.7142857142857143, + "accuracy_at_0_9": 0.928, + "coverage_at_5pct_error": 0.6114285714285714, + "coverage_at_1pct_error": 0.44, + "aurc": 0.044632779737602504, + "error_rate_at_0_9": 0.072, + "confidence_bias": 0.06486618832466695, + "top_bins": { + "0.9": { + "n": 125, + "errors": 9, + "error_rate": 0.072 + }, + "0.95": { + "n": 77, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { "n": 14, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "659e4d7e826303db51376a34ed1ee1187305d8b4fcd2ba86de0f1df79cbe9d09", - "n": 20, - "accuracy": 0.95, - "true_positive_n": 20, - "true_positive_correct": 19 - }, - { - "group": "681a13deadc6006e0d4d996f2b006e35b1d71fdeaaebfe98db36b84ce9e1bb82", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6972efb6c1c8c5719fc446f6b4e9bbf812c8274c04ddce38ae95cf5b06e65e10", - "n": 12, - "accuracy": 0.9166666666666666, - "true_positive_n": 5, - "true_positive_correct": 5 - }, - { - "group": "6d2a8e5968c904fbeceb057dc6d374d5cd1ba38019b8d31da27cc658c12aa476", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6d5ea32830050e4ac2840eed05d6b46ec9d7a78f65fbf85d0742d83cc628fa9c", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "6d74b56adb5791668e05d491f03f083060be7b184f47660a3b9cc14d7a32b511", - "n": 13, - "accuracy": 1.0, - "true_positive_n": 13, - "true_positive_correct": 13 - }, - { - "group": "6e73fd09bd5175004650b33517e33671c77ebb649384e14ffd95309ac3e567a7", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6f07feecb02e317c2db269ed9ec22266aa9195e11612cf7362bcdd1f3c9a7d3c", - "n": 20, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028571428571429, + "accuracy": 0.9772727272727273, + "confidence_cutoff": 0.9404661687170754 + }, + "0.8": { + "coverage": 0.8, "accuracy": 0.9, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6fc0443881957355817ec4b23a2d3027d6ce1ba949d0c313ac5134c5a33d7752", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "725e8bad3bf121140d3f1e8bb8625fac12766a1a7ebea88a432b39f8af246111", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "745320acbe1a562a24246846ea4e7c92fe29683dd9e75636a6183142147be543", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "77af40ba5530d0abf3b141de72a4c48f4d594f8bc50515d71d71bb0ef57ba861", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "798e40bc3bfcfea7bb0935e395494ac4836e1d49f294b2ca803f7e234a478493", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "79e0a2063b1c1395f9032d7bcca4a7987b25c134851f2b7087535b66973f8767", - "n": 3, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "7c2c1e933c21cfd9f0534226991df5542463f0d2c40138378024cbbabc1f205f", - "n": 20, - "accuracy": 0.55, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "7d092ad55211811544dc5993b69d906906fce9350e187508760b25a549bbc9ad", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 5, - "true_positive_correct": 0 - }, - { - "group": "8269675fe1c1fcb6500b8262056237a7eb277d8d7822e63cc1b335c9b3587400", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "8433ad4a51679bb9fc21396e090e351720fb331d25292de538c1e1a8ca524bab", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8d353f585c30ae854c944e56bcab8cfff771273fefcde8d929e6faf7cb75488f", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8da5a2b78094becb52357364f9f1ea558c4052e869e9636c912fa88cce650666", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8e11912bbdeb4ce0b56ca82a5e0ad37a9159b78dd620607541dc93cb9ac8e02e", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8f019450f830cbc42ded96cce0b7b4c1ee38439a9e198406ffb38ba5e0524fd7", - "n": 4, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "90994dcfeb71c7a87a4784556f5bf0bdb0b542c15a5f6b92da96e8020c4f1bf1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "94d54be3b552337f2fc772ce8ff09d2b668ec497130ded8b947506a120acfb4e", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 14, - "true_positive_correct": 14 - }, - { - "group": "97d7c3f3b10a157cdb2791ff320a99a7d1856f8bbdd7ee020ad9ddbe7ec14b8e", - "n": 13, - "accuracy": 0.0, - "true_positive_n": 13, - "true_positive_correct": 0 - }, - { - "group": "97f8ab62aee5d5d53245a565b836e892a9a7cda46748be23097421abb76ee39a", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 6, - "true_positive_correct": 0 - }, - { - "group": "99a049ff4aba4bc20df5f424b24efb9bab9118c192fd17f29b31ec910e497def", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "9b54c928a43ef6ff8b5b53cfeaad5e554106ed4d6718d08bc4f7440e968414af", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "9bcc91cc05b92cb30cead1b81acb706497144e4815da75642e99d3fd7ebfd154", - "n": 11, - "accuracy": 0.9090909090909091, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9d2f045465f59eb2bcb8a37818341ce3635a125bced7bfbcb7bf9acdd80c1a71", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "9e08a63fef975cba0f7fe19b9931144a9480e43b5b78497f97cfe4079c916045", - "n": 10, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9e38ecba2d710c7a9e3c2bef6abd472773b0edae845c5af26f317a4f5a2a3bc2", - "n": 19, - "accuracy": 0.10526315789473684, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "a04b59b050c050568baa9abccd205e73547f1e5111afaabdbcbe5949c3511ec6", - "n": 20, - "accuracy": 0.25, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "a1e75cda474e69786231632ed64953c19d60658191ea63553964e1d27015a7ce", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a52745698cb48cb1a28ac9bf212d724174c320e4648c9bececd83422b3364c1d", - "n": 19, - "accuracy": 0.10526315789473684, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a5a3b2d9f1ca0af015ed0584c90b28830aa11e86e8311c7f8f32f26b6d4f2768", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "a68f2269c6412f5885bdd26d3023dae3411476f5dd5b2e103355d4e1b0a41bb5", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a7b13e8d802fa5d584ad83d610714c858aab70e0c1f005324a2a91e7a369aed1", - "n": 16, - "accuracy": 0.0625, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a94db659229b95e557d461f6de9071da0762515528aa3a77d5cb870fcf0f914b", - "n": 20, - "accuracy": 0.75, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aab21c537c79f9eee1a053d0d9f7fbea9eb1451e166156d59dd6c464104a7992", - "n": 10, - "accuracy": 0.7, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "abd52b07ebdf5d9235ca1239d9c3acbe23eb7cabdbc661e0dc93653479a095f0", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aefb87718e7fc3725adba69a4f789e9de7d638ec0a0d1f12fcff8a8735fb622c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b095632bbe63fbdc009ed98e30a50899f6a8edc165e3be8a924a7f0f600be4c0", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 18, - "true_positive_correct": 6 - }, - { - "group": "b1b47d1978d292c9b80319bd2b531d2f2af206e3bcabf9aa0c46472112638ba9", - "n": 6, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b7fa0bdd60d2f2321a27ad4ef1c84b2258bb60cb55372ea8aa56643d6d7a6d2b", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 15, - "true_positive_correct": 0 - }, - { - "group": "b8b4a7b5769be3d4db661d88d40336ade544fcac1e5bd2ec6606c09f83e5263c", - "n": 7, - "accuracy": 0.7142857142857143, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "b9063c091cb8deaab4ea7fc341aebc169033dfb16792a7f0f253ce89ce556510", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bc9c57f088052889f45c0b709137c494d747464872fca72f16cfdde538281d64", - "n": 15, - "accuracy": 0.8666666666666667, - "true_positive_n": 6, - "true_positive_correct": 4 - }, - { - "group": "bd4aa774324b8affeb489e7b796b6f32a9a2f46442394a8f5042c1967eac05dc", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c0371b16a587352143c383ca9799634e07feb94a2df75ef8830adb86efe153c1", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c045d0efa4ee5ab2be1d352625f16f6fb9e5cfbe90e02574363e83cd27bc5109", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "c98451abd4304660daabdb977564d36fdad533b4458c467bd2b97d80a651955f", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cb85febf11725dacf437b764d64bf097c9c2bab08e74ac2eeadb496ccd51b5e8", - "n": 16, - "accuracy": 0.875, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "cfe165abb396d68223741b950f77ec55bd59d4c200bd310d7e9a956548c1590e", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d530b979376d67460a1a3dcafcc95c91e1a51ce2d97124dc2050c2030ae223b7", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 12, - "true_positive_correct": 0 - }, - { - "group": "d5496d95b9bcd56d2d382354f44af6098e30a6b132c6bd2ccbac429ee48d7d3b", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d6e9a943c91ab8d80d913c89d053cf89f2fe4bcd7df1963f3fd5a70bf56f3deb", - "n": 12, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d794d72182a0c6f8a3009e01f6d607875a60752bf7f05920823cc55729bcbf3c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d82b9f42323202feb6db7ed45f9990070e72e41705aeb5aa64d60a3fd7549f17", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "dfc85bf0674a0000b0dfbcf621419502a19c4fffabf786b4a5e74a69776a1f9d", - "n": 5, - "accuracy": 0.4, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "e35b0e3bc4a28a3bcf49fd9fe5c6e3c6ece59e921ecabf1904c9459cb7b84ad1", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "e6e40718557ace0b1764c2b7ce7021ff2b42e61db43cd59c7a54132f69807b2d", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "f5b769ca29423ada31602cdd07bba0ab057d937256622788fa0872f6fe095865", - "n": 4, - "accuracy": 0.75, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "f72e0a3b72341d46d7eb5ab932fcce07cf34702afa40d217c5e997bef74764be", - "n": 15, - "accuracy": 0.9333333333333333, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "f9476bc6e284b58a4d168d14b188b2ca1636ebe4a0adfcb0c40060bed41781c8", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "f97f37793bfaa1b487125bd7b1f34bff0fed50146a5606962c763220f17bb6de", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fb39d2600c554fe3ac1eddb65704bc34c6102ca0fbb4a77744ac525869d35be4", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fcd64091f029e53dd4a770300ecf446dbf5bca8cc267148f28b198584caeae46", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fe703c5eb241d3542fcc57e11688b3641c4f1f0b9506a720b42a9b9031382154", - "n": 19, - "accuracy": 0.8947368421052632, - "true_positive_n": 2, - "true_positive_correct": 0 + "confidence_cutoff": 0.8547040294961724 } - ], - "correct": 623 + } + }, + "argmax_classes": { + "requires_approval": { + "n": 35, + "correct": 35, + "recall": 1.0 + }, + "permitted": { + "n": 140, + "correct": 112, + "recall": 0.8 + } + }, + "threshold": 0.8449815374033091, + "n": 175, + "positives": 35, + "negatives": 140, + "confusion": { + "tp": 25, + "fp": 14, + "tn": 126, + "fn": 10 + }, + "recall": 0.7142857142857143, + "false_alert_rate": 0.1, + "precision": 0.6410256410256411, + "threshold_accuracy": 0.8628571428571429 + } + }, + "paired_threshold_intervals": { + "groups": 35, + "samples": 2000, + "seed": 20260927, + "unit": "whole source-closure group, shared paired draw; calibration thresholds fixed", + "metrics": { + "recall": { + "parent": { + "value": 0.4, + "ci95": [ + 0.22857142857142856, + 0.5714285714285714 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.7142857142857143, + "ci95": [ + 0.5714285714285714, + 0.8571428571428571 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.3142857142857143, + "ci95": [ + 0.17142857142857143, + 0.4857142857142857 + ], + "valid_resamples": 2000 + } + }, + "false_alert_rate": { + "parent": { + "value": 0.08571428571428572, + "ci95": [ + 0.05, + 0.12857142857142856 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.1, + "ci95": [ + 0.06428571428571428, + 0.14285714285714285 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.01428571428571429, + "ci95": [ + -0.014285714285714277, + 0.04285714285714286 + ], + "valid_resamples": 2000 + } + }, + "precision": { + "parent": { + "value": 0.5384615384615384, + "ci95": [ + 0.3665909090909091, + 0.70979262672811 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.6410256410256411, + "ci95": [ + 0.55, + 0.7435897435897436 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.10256410256410264, + "ci95": [ + 0.005701872682569551, + 0.21591595300834426 + ], + "valid_resamples": 2000 + } + }, + "threshold_accuracy": { + "parent": { + "value": 0.8114285714285714, + "ci95": [ + 0.7655714285714286, + 0.8628571428571429 + ], + "valid_resamples": 2000 + }, + "candidate": { + "value": 0.8628571428571429, + "ci95": [ + 0.8228571428571428, + 0.9028571428571428 + ], + "valid_resamples": 2000 + }, + "candidate_minus_parent": { + "value": 0.05142857142857149, + "ci95": [ + 0.02285714285714291, + 0.08000000000000007 + ], + "valid_resamples": 2000 + } + } + } + } + } + }, + "retained_development_by_source": { + "parent": { + "excytin_file": { + "questions": 80, + "groups": 1, + "raw": { + "n": 80, + "nll": 0.0001342531538525746, + "acc": 1.0, + "ece": 0.0001340575450615722, + "brier": 7.773831743030022e-07, + "mean_conf": 0.9998659424549384, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0001340575450615722, + "top_bins": { + "0.9": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999710759158646 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9999066849313115 + } } }, - "swarm_static": { - "raw": { - "n": 354, - "accuracy": 0.9152542372881356, - "nll": 0.23980562862108395, - "brier": 0.11952531272784916, - "correct": 324 - }, - "global_calibrated": { - "n": 354, - "accuracy": 0.9152542372881356, - "nll": 0.23046829278993344, - "brier": 0.11775650997940981, - "correct": 324 - }, - "families": { - "binding": { - "raw": { - "n": 164, - "accuracy": 0.8292682926829268, - "nll": 0.4973668964039057, - "brier": 0.24305378453073645, - "correct": 136 - }, - "global_calibrated": { - "n": 164, - "accuracy": 0.8292682926829268, - "nll": 0.47756760916281, - "brier": 0.24123129045340497, - "correct": 136 - } + "calibrated": { + "n": 80, + "nll": 0.0030639977354718724, + "acc": 1.0, + "ece": 0.0030486898704090715, + "brier": 5.723205099455454e-05, + "mean_conf": 0.9969513101295909, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0030486898704090715, + "top_bins": { + "0.9": { + "n": 80, + "errors": 0, + "error_rate": 0.0 }, - "direct": { - "raw": { - "n": 134, - "accuracy": 1.0, - "nll": 0.0006701158955142708, - "brier": 3.3165775411476116e-05, - "correct": 134 - }, - "global_calibrated": { - "n": 134, - "accuracy": 1.0, - "nll": 0.0034830280523404688, - "brier": 0.0001921529648151749, - "correct": 134 - } + "0.95": { + "n": 80, + "errors": 0, + "error_rate": 0.0 }, - "flow": { - "raw": { - "n": 56, - "accuracy": 0.9642857142857143, - "nll": 0.05773617842186206, - "brier": 0.04369099694129809, - "correct": 54 - }, - "global_calibrated": { - "n": 56, - "accuracy": 0.9642857142857143, - "nll": 0.04996360689146422, - "brier": 0.037466507161918496, - "correct": 54 - } + "0.99": { + "n": 79, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9980559839388836 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9963993201639499 + } + } + } + }, + "excytin_parent_join": { + "questions": 78, + "groups": 1, + "raw": { + "n": 78, + "nll": 0.0019727127979591126, + "acc": 1.0, + "ece": 0.0019615703387183547, + "brier": 4.023721192309001e-05, + "mean_conf": 0.9980384296612816, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0019615703387183547, + "top_bins": { + "0.9": { + "n": 78, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 78, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 76, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9996318703080018 + }, + "0.8": { + "coverage": 0.8076923076923077, + "accuracy": 1.0, + "confidence_cutoff": 0.9974582021431176 } } }, - "manual": { - "raw": { - "n": 83, - "accuracy": 0.6987951807228916, - "nll": 1.2611921240040693, - "brier": 0.5153410721147468, - "correct": 58 - }, - "global_calibrated": { - "n": 83, - "accuracy": 0.6987951807228916, - "nll": 1.0302111739617255, - "brier": 0.48339882880272644, - "correct": 58 - }, - "families": { - "binding": { - "raw": { - "n": 40, - "accuracy": 0.625, - "nll": 1.3389280289714236, - "brier": 0.6296514707199694, - "correct": 25 - }, - "global_calibrated": { - "n": 40, - "accuracy": 0.625, - "nll": 1.1010460949559158, - "brier": 0.5735795406676166, - "correct": 25 - } + "calibrated": { + "n": 78, + "nll": 0.01903091706838521, + "acc": 1.0, + "ece": 0.018568486423778917, + "brier": 0.0013900388516520708, + "mean_conf": 0.9814315135762209, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9871794871794872, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.018568486423779063, + "top_bins": { + "0.9": { + "n": 77, + "errors": 0, + "error_rate": 0.0 }, - "flow": { - "raw": { - "n": 40, - "accuracy": 0.75, - "nll": 1.2780391309680283, - "brier": 0.4396812526809053, - "correct": 30 - }, - "global_calibrated": { - "n": 40, - "accuracy": 0.75, - "nll": 1.0365517358773104, - "brier": 0.4294728191968137, - "correct": 30 - } + "0.95": { + "n": 69, + "errors": 0, + "error_rate": 0.0 }, - "prerequisite": { - "raw": { - "n": 3, - "accuracy": 1.0, - "nll": 8.663158656148827e-05, - "brier": 1.649633063680738e-08, - "correct": 3 - }, - "global_calibrated": { - "n": 3, - "accuracy": 1.0, - "nll": 0.0012047351647270432, - "brier": 2.7986830271292114e-06, - "correct": 3 - } + "0.99": { + "n": 46, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9916928553567559 + }, + "0.8": { + "coverage": 0.8076923076923077, + "accuracy": 1.0, + "confidence_cutoff": 0.9693893850709232 + } + } + } + }, + "excytin_process": { + "questions": 120, + "groups": 1, + "raw": { + "n": 120, + "nll": 0.0005180266626209783, + "acc": 1.0, + "ece": 0.0005174316239614507, + "brier": 2.3119235778779676e-06, + "mean_conf": 0.9994825683760385, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0005174316239614507, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999134057097643 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.99901689132107 } } }, - "sysmon": { - "raw": { - "n": 64, - "accuracy": 0.96875, - "nll": 0.08465266759433944, - "brier": 0.044922383555011054, - "correct": 62 - }, - "global_calibrated": { - "n": 64, - "accuracy": 0.96875, - "nll": 0.10719175042876139, - "brier": 0.048442258840311225, - "correct": 62 + "calibrated": { + "n": 120, + "nll": 0.007587462315793864, + "acc": 1.0, + "ece": 0.007523761830370401, + "brier": 0.00023474191911792246, + "mean_conf": 0.9924762381696296, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.007523761830370401, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 83, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9964206554980828 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9859276562719884 + } + } + } + }, + "excytin_signin": { + "questions": 120, + "groups": 1, + "raw": { + "n": 120, + "nll": 0.0013203461429938826, + "acc": 1.0, + "ece": 0.0012737143193113606, + "brier": 0.00013106658103070147, + "mean_conf": 0.9987262856806887, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9916666666666667, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0012737143193113365, + "top_bins": { + "0.9": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9998413534198065 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9996110203225037 + } } }, - "general": { - "raw": { - "n": 105, - "accuracy": 0.819047619047619, - "nll": 0.5016458133964976, - "brier": 0.2771177005204226, - "correct": 86 - }, - "global_calibrated": { - "n": 105, - "accuracy": 0.819047619047619, - "nll": 0.4265669988063534, - "brier": 0.24785251439683742, - "correct": 86 + "calibrated": { + "n": 120, + "nll": 0.009411479962491658, + "acc": 1.0, + "ece": 0.00903361007075822, + "brier": 0.0010010974637190823, + "mean_conf": 0.9909663899292418, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9916666666666667, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.009033610070758202, + "top_bins": { + "0.9": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 105, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9950297195670218 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9914608088731087 + } } } - } - }, - "candidate": { - "temperature": 1.6245047927124707, - "domains": { - "excytin": { - "raw": { - "n": 398, - "accuracy": 1.0, - "nll": 0.0009678808724190956, - "brier": 4.8256567178328636e-05, - "correct": 398 - }, - "global_calibrated": { - "n": 398, - "accuracy": 1.0, - "nll": 0.009470865385844369, - "brier": 0.0006565384935901168, - "correct": 398 + }, + "sev_guide_detector": { + "questions": 1064, + "groups": 114, + "raw": { + "n": 1064, + "nll": 0.8276776090882748, + "acc": 0.599624060150376, + "ece": 0.10621007339433168, + "brier": 0.48019952548397726, + "mean_conf": 0.7041447618951532, + "confident_error_rate": 0.012218045112781954, + "coverage_at_0_9": 0.32706766917293234, + "accuracy_at_0_9": 0.9626436781609196, + "coverage_at_5pct_error": 0.3524436090225564, + "coverage_at_1pct_error": 0.0, + "aurc": 0.19304798592127517, + "error_rate_at_0_9": 0.03735632183908046, + "confidence_bias": 0.10452070174477723, + "top_bins": { + "0.9": { + "n": 348, + "errors": 13, + "error_rate": 0.03735632183908046 + }, + "0.95": { + "n": 327, + "errors": 13, + "error_rate": 0.039755351681957186 + }, + "0.99": { + "n": 319, + "errors": 12, + "error_rate": 0.03761755485893417 + } + }, + "selective": { + "0.5": { + "coverage": 0.5075187969924813, + "accuracy": 0.8074074074074075, + "confidence_cutoff": 0.6273336504002459 + }, + "0.8": { + "coverage": 0.8016917293233082, + "accuracy": 0.6518171160609613, + "confidence_cutoff": 0.48816800774395525 + } } }, - "guide": { - "raw": { - "n": 231, - "accuracy": 0.5627705627705628, - "nll": 0.8563265185203435, - "brier": 0.5255142828500304, - "macro_f1": 0.4613423970886119, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.3333333333333333, - "recall": 0.4915254237288136, - "f1-score": 0.3972602739726027, - "support": 59.0 - }, - "FalsePositive": { - "precision": 0.7175572519083969, - "recall": 0.6962962962962963, - "f1-score": 0.706766917293233, - "support": 135.0 - }, - "TruePositive": { - "precision": 0.5384615384615384, - "recall": 0.1891891891891892, - "f1-score": 0.28, - "support": 37.0 - } + "calibrated": { + "n": 1064, + "nll": 0.785071927880167, + "acc": 0.599624060150376, + "ece": 0.05266334727942813, + "brier": 0.46075730837789786, + "mean_conf": 0.6304966279496853, + "confident_error_rate": 0.011278195488721804, + "coverage_at_0_9": 0.3007518796992481, + "accuracy_at_0_9": 0.9625, + "coverage_at_5pct_error": 0.3524436090225564, + "coverage_at_1pct_error": 0.0, + "aurc": 0.19295849854549843, + "error_rate_at_0_9": 0.0375, + "confidence_bias": 0.03087256779930936, + "top_bins": { + "0.9": { + "n": 320, + "errors": 12, + "error_rate": 0.0375 }, - "confusion": [ - [ - 29, - 27, - 3 - ], - [ - 38, - 94, - 3 - ], - [ - 20, - 10, - 7 - ] - ], - "equal_organization_accuracy": 0.6510869565217392, - "organization_results": [ - { - "group": "0d1b7eba9c4d5a5852ef2805e5f1c1018c827e36bfb922fefa4d0a92d76f322c", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "108482877637fd1bbe809b2faddd12c9affef276cdb170667c0234a670019cd4", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "13a6e1ec428dca68e3cc729ac008e19870e5caa60b3f2d8d013eeb5a4f25aa71", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17515941d0c83a2c0515166a1503db5e3d24c63b53c0398a9c4f3345d3ef5eda", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "22e3ca4ff2df3a13f6db322d247b7b7f8742181d456be68cb84fb4dfbda7b993", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5667ea75698bc671a6165fb7604bf6d7b7874ba2ef04005f82edf2811b406b53", - "n": 20, - "accuracy": 0.65, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "5cd724f6f8c622cf83c9ee56eeb3e956e4c0428f4bb398608581d51b9f3bd4cf", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5fd9d56568dd7e69c4871b023d3df17203f2eb6d9aaf7a808371a391407f9026", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "66288757ecaf36bf368cd8d4c43f454658be762a12957c52eae438ca3e727ad8", - "n": 8, - "accuracy": 0.375, - "true_positive_n": 6, - "true_positive_correct": 1 - }, - { - "group": "6d53f455fa642b810781930d63db6b35443da067812f1bf329a37de938179aed", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "79707d8a1f7240a758d67ba5212ab21ef286647a4b71ba71c892f88341f3f4e6", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 1 - }, - { - "group": "7b4cb8f8c017bd94da14e9460c49cdf44fa47f6eab9726918b70ebb7b2695ee1", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "7f55e216b81a67b2fcebf627b534d1d18d9d734f31cb41477725aaa041f6d42b", - "n": 20, - "accuracy": 0.6, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "974cf500c91fb4e80273489c70f89803fe8eaf2fc9d4e68a6aac899d0e580551", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 18, - "true_positive_correct": 0 - }, - { - "group": "97dbafe0dc6d23395bb2008b78ccf190cf0710c18154cbbfc664b3d388086baf", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a756d4fa124cf854b444faed605f6fee522c5418d960b8cd60a8a06c44ef56b0", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "adcffe87f88661c654325a95ccbc252777e27ff83f947ff85ea77cf167ad57c0", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b4edbf01a09fe17e2fc4eba076213094ab09b40f0ff73c9d3e5c9592f0f59353", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bfe24bf96507627d9c2e0eb183e0cb607aa50a5195abded30b612c252216e7b0", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cbdba8017701da737b2f23ee11598f766b99054550311935a5730acb5f7e57d9", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cc3696bd6f2c8a2d65d8376338102471db2565326ab124811ce53c428e54425e", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "d592c0767849012441fe14ab7451f4371421ba152f26c3587463adcf58b911e5", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "f562314a5da425b50c820fa25a6756b09f54b042ff1a3a4b580647dd4c20aa7d", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - } - ], - "correct": 130 - }, - "global_calibrated": { - "n": 231, - "accuracy": 0.5627705627705628, - "nll": 0.7793526965750157, - "brier": 0.4718856823661165, - "macro_f1": 0.4613423970886119, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.3333333333333333, - "recall": 0.4915254237288136, - "f1-score": 0.3972602739726027, - "support": 59.0 - }, - "FalsePositive": { - "precision": 0.7175572519083969, - "recall": 0.6962962962962963, - "f1-score": 0.706766917293233, - "support": 135.0 - }, - "TruePositive": { - "precision": 0.5384615384615384, - "recall": 0.1891891891891892, - "f1-score": 0.28, - "support": 37.0 - } + "0.95": { + "n": 213, + "errors": 11, + "error_rate": 0.051643192488262914 }, - "confusion": [ - [ - 29, - 27, - 3 - ], - [ - 38, - 94, - 3 - ], - [ - 20, - 10, - 7 - ] - ], - "equal_organization_accuracy": 0.6510869565217392, - "organization_results": [ - { - "group": "0d1b7eba9c4d5a5852ef2805e5f1c1018c827e36bfb922fefa4d0a92d76f322c", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "108482877637fd1bbe809b2faddd12c9affef276cdb170667c0234a670019cd4", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "13a6e1ec428dca68e3cc729ac008e19870e5caa60b3f2d8d013eeb5a4f25aa71", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17515941d0c83a2c0515166a1503db5e3d24c63b53c0398a9c4f3345d3ef5eda", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "22e3ca4ff2df3a13f6db322d247b7b7f8742181d456be68cb84fb4dfbda7b993", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5667ea75698bc671a6165fb7604bf6d7b7874ba2ef04005f82edf2811b406b53", - "n": 20, - "accuracy": 0.65, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "5cd724f6f8c622cf83c9ee56eeb3e956e4c0428f4bb398608581d51b9f3bd4cf", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5fd9d56568dd7e69c4871b023d3df17203f2eb6d9aaf7a808371a391407f9026", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "66288757ecaf36bf368cd8d4c43f454658be762a12957c52eae438ca3e727ad8", - "n": 8, - "accuracy": 0.375, - "true_positive_n": 6, - "true_positive_correct": 1 - }, - { - "group": "6d53f455fa642b810781930d63db6b35443da067812f1bf329a37de938179aed", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "79707d8a1f7240a758d67ba5212ab21ef286647a4b71ba71c892f88341f3f4e6", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 1 - }, - { - "group": "7b4cb8f8c017bd94da14e9460c49cdf44fa47f6eab9726918b70ebb7b2695ee1", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "7f55e216b81a67b2fcebf627b534d1d18d9d734f31cb41477725aaa041f6d42b", - "n": 20, - "accuracy": 0.6, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "974cf500c91fb4e80273489c70f89803fe8eaf2fc9d4e68a6aac899d0e580551", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 18, - "true_positive_correct": 0 - }, - { - "group": "97dbafe0dc6d23395bb2008b78ccf190cf0710c18154cbbfc664b3d388086baf", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a756d4fa124cf854b444faed605f6fee522c5418d960b8cd60a8a06c44ef56b0", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "adcffe87f88661c654325a95ccbc252777e27ff83f947ff85ea77cf167ad57c0", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b4edbf01a09fe17e2fc4eba076213094ab09b40f0ff73c9d3e5c9592f0f59353", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bfe24bf96507627d9c2e0eb183e0cb607aa50a5195abded30b612c252216e7b0", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cbdba8017701da737b2f23ee11598f766b99054550311935a5730acb5f7e57d9", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cc3696bd6f2c8a2d65d8376338102471db2565326ab124811ce53c428e54425e", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "d592c0767849012441fe14ab7451f4371421ba152f26c3587463adcf58b911e5", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "f562314a5da425b50c820fa25a6756b09f54b042ff1a3a4b580647dd4c20aa7d", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - } - ], - "correct": 130 + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5075187969924813, + "accuracy": 0.8074074074074075, + "confidence_cutoff": 0.5144221593760727 + }, + "0.8": { + "coverage": 0.8016917293233082, + "accuracy": 0.6553341148886284, + "confidence_cutoff": 0.4322968853489006 + } + } + } + }, + "sev_guide_triage": { + "questions": 231, + "groups": 23, + "raw": { + "n": 231, + "nll": 0.8563265185203435, + "acc": 0.5627705627705628, + "ece": 0.1895007716411903, + "brier": 0.5255142828500304, + "mean_conf": 0.7444269666893102, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.35064935064935066, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.36363636363636365, + "coverage_at_1pct_error": 0.36363636363636365, + "aurc": 0.22468166386345326, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.18165640391874738, + "top_bins": { + "0.9": { + "n": 81, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 79, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5021645021645021, + "accuracy": 0.75, + "confidence_cutoff": 0.7894512944300596 + }, + "0.8": { + "coverage": 0.8095238095238095, + "accuracy": 0.5775401069518716, + "confidence_cutoff": 0.49009219177135965 + } } }, - "native": { - "raw": { - "n": 410, - "accuracy": 0.9975609756097561, - "nll": 0.008457485545022507, - "brier": 0.0035344210430228854, - "correct": 409 - }, - "global_calibrated": { - "n": 410, - "accuracy": 0.9975609756097561, - "nll": 0.028769043116345983, - "brier": 0.006284804103393485, - "correct": 409 + "calibrated": { + "n": 231, + "nll": 0.7793526965750157, + "acc": 0.5627705627705628, + "ece": 0.1664752401619534, + "brier": 0.4718856823661165, + "mean_conf": 0.6653812407670776, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.3463203463203463, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.36363636363636365, + "coverage_at_1pct_error": 0.36363636363636365, + "aurc": 0.22610167628363687, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.1026106779965148, + "top_bins": { + "0.9": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 14, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5021645021645021, + "accuracy": 0.75, + "confidence_cutoff": 0.6343620681380909 + }, + "0.8": { + "coverage": 0.8008658008658008, + "accuracy": 0.5675675675675675, + "confidence_cutoff": 0.4381830283799813 + } + } + } + }, + "sev_swarmtraces_static": { + "questions": 354, + "groups": 176, + "raw": { + "n": 354, + "nll": 0.06671770193368179, + "acc": 0.9745762711864406, + "ece": 0.008417033573491902, + "brier": 0.03589778460324393, + "mean_conf": 0.9740214622331307, + "confident_error_rate": 0.002824858757062147, + "coverage_at_0_9": 0.9322033898305084, + "accuracy_at_0_9": 0.996969696969697, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9745762711864406, + "aurc": 0.0009317569638462815, + "error_rate_at_0_9": 0.0030303030303030303, + "confidence_bias": -0.0005548089533099487, + "top_bins": { + "0.9": { + "n": 330, + "errors": 1, + "error_rate": 0.0030303030303030303 + }, + "0.95": { + "n": 317, + "errors": 1, + "error_rate": 0.0031545741324921135 + }, + "0.99": { + "n": 292, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9998280676315141 + }, + "0.8": { + "coverage": 0.8022598870056498, + "accuracy": 1.0, + "confidence_cutoff": 0.9930714713128618 + } } }, - "guide_detector": { - "raw": { - "n": 1064, - "accuracy": 0.599624060150376, - "nll": 0.8276776090882748, - "brier": 0.48019952548397726, - "macro_f1": 0.5681608349992006, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.44772727272727275, - "recall": 0.701067615658363, - "f1-score": 0.5464632454923717, - "support": 281.0 - }, - "FalsePositive": { - "precision": 0.7606263982102909, - "recall": 0.6746031746031746, - "f1-score": 0.7150368033648791, - "support": 504.0 - }, - "TruePositive": { - "precision": 0.5706214689265536, - "recall": 0.36200716845878134, - "f1-score": 0.44298245614035087, - "support": 279.0 - } + "calibrated": { + "n": 354, + "nll": 0.08603305952328294, + "acc": 0.9745762711864406, + "ece": 0.031999170395289905, + "brier": 0.03970483443698841, + "mean_conf": 0.9467552592594386, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.8615819209039548, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9745762711864406, + "aurc": 0.0009516834613723162, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.027821011927002037, + "top_bins": { + "0.9": { + "n": 305, + "errors": 0, + "error_rate": 0.0 }, - "confusion": [ - [ - 197, - 55, - 29 - ], - [ - 117, - 340, - 47 - ], - [ - 126, - 52, - 101 - ] - ], - "equal_organization_accuracy": 0.6743365093994473, - "organization_results": [ - { - "group": "0228d088db03da24994da8d12bb80e2925bf424387c12a0cab440661f61e53db", - "n": 2, - "accuracy": 0.5, - "true_positive_n": 2, - "true_positive_correct": 1 - }, - { - "group": "03b2cff809e7692ca15f889834563f866b356cc60a41023c8611dd2b60bcf51a", - "n": 2, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "04eb81bd28a99e8d254c7a0a9c07234f55af1f9e4fe167269f504109495c29a5", - "n": 4, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06646577f2484818ed5044762571197f87d3e79272459df5462e193f43ddd530", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06f295d348152073a6bf3ba5485ca147a0c14bb94baf88391d465d7ea7e6eca9", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "08b767258688d27f07153aa89f0a2c2c6e8e426289043614eb48c2c881591192", - "n": 17, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "09890e550c1a9a583cf9aaa2a997c1bbc61162a5d36c8a891bbbc1a837dffc36", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "0b8649cde9e63fe45d17a7941ef90f01d1b3068c464c013d3ff7e21047164b9c", - "n": 5, - "accuracy": 0.8, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "10f18e39bc3c8e8713158ad66e666edf9118b8b83fb7eb21e5e39476a75a3125", - "n": 12, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14919c2e5da07951c425b98441bab1107d02cbcb0496a9148b184b1fea1eb4d2", - "n": 3, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14a068f7500ef993a7f2e4cfcd5e194c57217338de4e72cf848d710c20392e91", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 18, - "true_positive_correct": 0 - }, - { - "group": "16cb7259665afd83462953cc90bf025352bab6f7210a59e5d86d49561f03fcbe", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "16fe8b7b9a0987bf861cee078b8763e8f75ec46733a6a59b61ffa8e2bd6e49f7", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17cd31c2cd2d2ccbabdfab902aee6e4e759167d126e1bcaf4d638c23b93272ac", - "n": 17, - "accuracy": 0.9411764705882353, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "18aa7b3932a87bfdfc7e3f032e91b38999f82e037146754bd8bba044870170ac", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "191618c5651b68d639db0f90f67315430ae08a9467ebc761ca10a54847034de5", - "n": 20, - "accuracy": 0.65, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "1e7cf0ee593015038b9cf7686c7db7604b7de8b52d0a8b556038b7795a59dd5a", - "n": 19, - "accuracy": 0.9473684210526315, - "true_positive_n": 4, - "true_positive_correct": 4 - }, - { - "group": "23f95dce880a8574d254a19ac7f94f1284a556dfa61126a92844ba979083ede1", - "n": 2, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2413f2cb2e9a29595c0278b868a30250d3573d487f99e35b3fe3836713d79e96", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "280f0d43bcdf087751b7a2b90739860bc02c2cccea845f81fae7555cdd9d7745", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "2af6f963a1301ac7cbf959fb0f63442e6b1b28be622e0d74c6b94c1e459c8bf0", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2d4fd6af7767d8fbcfeddff36a76b5c993a49d5e45a515806ec30b1b53afa2ce", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3229724a48ff16d1ea5cff92fcbc1a29fb636760780576606a1f931cba64c7e6", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "33593514e9eb3b4a2f11fbff9a900e304d7f7088413e1a6ed36210de11e1a7cd", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "357e4fcd5f01346ea07a7b6805f5e0e94f11cd39d870ebde997c06b3ca23403b", - "n": 20, - "accuracy": 0.3, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "36be30d9fffc6077ec297a0ccde6690eb34709163c58b4b5df24236dd44c0e91", - "n": 18, - "accuracy": 0.9444444444444444, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3788764c156c1d8107b5e65fc226f169b6cfeb28e92341b3776e0323894d1182", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "397381ff8e300cf10164a627a5a6e614e68a7fb2a4dbac6daa5c67e331248cb3", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3a3a67375cabf27ee46ffe17a8e3122e2bc0d77c22a4bc5fd3caf21af2fdf827", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "3ba2e76ba41ace83ac5fc70feb81f7bd790ebcc01cef912db9a520e0659301e6", - "n": 20, - "accuracy": 0.25, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "3c519ebea078526e0c7724d67b71d0e9d928d62adbb68de9cc3ab45caeab0ba6", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "42a20c36367ded75633c56b5468efe4dab76b5242f7fc0fafe5de10ae8b9a078", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "4f4986e7532d0ba637710cb0010206dd22720f9b35cc163fd2f833fc93c25334", - "n": 14, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "52ed406f643d4adbab570be68c48860e0e9f239fe00b7f1bd0c35e00868fc879", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "557c22b05ccf851b94b1fc4571234d237e761ded29b151b3a3adcacd17f9ba00", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "58f3e1e6f3eca7efe52c2f97b888733f6a541ca293e36d6e1a1f22c2332d05fb", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d11eb324b736d8f810b14b708ea1982028cb405fb0fb2e3f20dab9ab5bb3fe1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d2fc2acbf5e2afa16d5ad87f1cce4b784cdd43ca2da54c3eea33521e52a5a81", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 10, - "true_positive_correct": 0 - }, - { - "group": "5f21eec69194bcd2e80220b2701002f2da10e7cd2b04625674cdf69feb57dedd", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61c9e0872a43ed30a90ab06f7a3e8b85d33f54355b964a7fd1ffa361531c5626", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61e7fb88134413ce324703852a01f81dda0cb96d4ae07a44cf845c9b1de8baa0", - "n": 20, - "accuracy": 0.7, - "true_positive_n": 13, - "true_positive_correct": 9 - }, - { - "group": "6201b69130be0b440362fc95b4cc2ef6c4c856543d9beb9584d43899e72d0283", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "629dffdc159a2dbb566dab56a6f17419a63aa2b0815020338d686277ff90f623", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "63b98074727eecae10e68ab106f9c5f2ec98bd2347e0c3eaa72d5557c1ae1b64", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "63deca8572354a01b85154c34b64122772c89023b59c7931064bb347d8389e7c", - "n": 14, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "659e4d7e826303db51376a34ed1ee1187305d8b4fcd2ba86de0f1df79cbe9d09", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 20, - "true_positive_correct": 20 - }, - { - "group": "681a13deadc6006e0d4d996f2b006e35b1d71fdeaaebfe98db36b84ce9e1bb82", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6972efb6c1c8c5719fc446f6b4e9bbf812c8274c04ddce38ae95cf5b06e65e10", - "n": 12, - "accuracy": 0.9166666666666666, - "true_positive_n": 5, - "true_positive_correct": 5 - }, - { - "group": "6d2a8e5968c904fbeceb057dc6d374d5cd1ba38019b8d31da27cc658c12aa476", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6d5ea32830050e4ac2840eed05d6b46ec9d7a78f65fbf85d0742d83cc628fa9c", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "6d74b56adb5791668e05d491f03f083060be7b184f47660a3b9cc14d7a32b511", - "n": 13, - "accuracy": 1.0, - "true_positive_n": 13, - "true_positive_correct": 13 - }, - { - "group": "6e73fd09bd5175004650b33517e33671c77ebb649384e14ffd95309ac3e567a7", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6f07feecb02e317c2db269ed9ec22266aa9195e11612cf7362bcdd1f3c9a7d3c", - "n": 20, - "accuracy": 0.95, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6fc0443881957355817ec4b23a2d3027d6ce1ba949d0c313ac5134c5a33d7752", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "725e8bad3bf121140d3f1e8bb8625fac12766a1a7ebea88a432b39f8af246111", - "n": 20, - "accuracy": 0.95, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "745320acbe1a562a24246846ea4e7c92fe29683dd9e75636a6183142147be543", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "77af40ba5530d0abf3b141de72a4c48f4d594f8bc50515d71d71bb0ef57ba861", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "798e40bc3bfcfea7bb0935e395494ac4836e1d49f294b2ca803f7e234a478493", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "79e0a2063b1c1395f9032d7bcca4a7987b25c134851f2b7087535b66973f8767", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "7c2c1e933c21cfd9f0534226991df5542463f0d2c40138378024cbbabc1f205f", - "n": 20, - "accuracy": 0.55, - "true_positive_n": 7, - "true_positive_correct": 1 - }, - { - "group": "7d092ad55211811544dc5993b69d906906fce9350e187508760b25a549bbc9ad", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 5, - "true_positive_correct": 0 - }, - { - "group": "8269675fe1c1fcb6500b8262056237a7eb277d8d7822e63cc1b335c9b3587400", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "8433ad4a51679bb9fc21396e090e351720fb331d25292de538c1e1a8ca524bab", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8d353f585c30ae854c944e56bcab8cfff771273fefcde8d929e6faf7cb75488f", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8da5a2b78094becb52357364f9f1ea558c4052e869e9636c912fa88cce650666", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8e11912bbdeb4ce0b56ca82a5e0ad37a9159b78dd620607541dc93cb9ac8e02e", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8f019450f830cbc42ded96cce0b7b4c1ee38439a9e198406ffb38ba5e0524fd7", - "n": 4, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "90994dcfeb71c7a87a4784556f5bf0bdb0b542c15a5f6b92da96e8020c4f1bf1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "94d54be3b552337f2fc772ce8ff09d2b668ec497130ded8b947506a120acfb4e", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 14, - "true_positive_correct": 14 - }, - { - "group": "97d7c3f3b10a157cdb2791ff320a99a7d1856f8bbdd7ee020ad9ddbe7ec14b8e", - "n": 13, - "accuracy": 0.15384615384615385, - "true_positive_n": 13, - "true_positive_correct": 2 - }, - { - "group": "97f8ab62aee5d5d53245a565b836e892a9a7cda46748be23097421abb76ee39a", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 6, - "true_positive_correct": 0 - }, - { - "group": "99a049ff4aba4bc20df5f424b24efb9bab9118c192fd17f29b31ec910e497def", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "9b54c928a43ef6ff8b5b53cfeaad5e554106ed4d6718d08bc4f7440e968414af", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "9bcc91cc05b92cb30cead1b81acb706497144e4815da75642e99d3fd7ebfd154", - "n": 11, - "accuracy": 0.9090909090909091, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9d2f045465f59eb2bcb8a37818341ce3635a125bced7bfbcb7bf9acdd80c1a71", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "9e08a63fef975cba0f7fe19b9931144a9480e43b5b78497f97cfe4079c916045", - "n": 10, - "accuracy": 0.5, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9e38ecba2d710c7a9e3c2bef6abd472773b0edae845c5af26f317a4f5a2a3bc2", - "n": 19, - "accuracy": 0.42105263157894735, - "true_positive_n": 7, - "true_positive_correct": 6 - }, - { - "group": "a04b59b050c050568baa9abccd205e73547f1e5111afaabdbcbe5949c3511ec6", - "n": 20, - "accuracy": 0.05, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "a1e75cda474e69786231632ed64953c19d60658191ea63553964e1d27015a7ce", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a52745698cb48cb1a28ac9bf212d724174c320e4648c9bececd83422b3364c1d", - "n": 19, - "accuracy": 0.10526315789473684, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a5a3b2d9f1ca0af015ed0584c90b28830aa11e86e8311c7f8f32f26b6d4f2768", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "a68f2269c6412f5885bdd26d3023dae3411476f5dd5b2e103355d4e1b0a41bb5", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a7b13e8d802fa5d584ad83d610714c858aab70e0c1f005324a2a91e7a369aed1", - "n": 16, - "accuracy": 0.0625, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a94db659229b95e557d461f6de9071da0762515528aa3a77d5cb870fcf0f914b", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aab21c537c79f9eee1a053d0d9f7fbea9eb1451e166156d59dd6c464104a7992", - "n": 10, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "abd52b07ebdf5d9235ca1239d9c3acbe23eb7cabdbc661e0dc93653479a095f0", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aefb87718e7fc3725adba69a4f789e9de7d638ec0a0d1f12fcff8a8735fb622c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b095632bbe63fbdc009ed98e30a50899f6a8edc165e3be8a924a7f0f600be4c0", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 18, - "true_positive_correct": 7 - }, - { - "group": "b1b47d1978d292c9b80319bd2b531d2f2af206e3bcabf9aa0c46472112638ba9", - "n": 6, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b7fa0bdd60d2f2321a27ad4ef1c84b2258bb60cb55372ea8aa56643d6d7a6d2b", - "n": 20, - "accuracy": 0.05, - "true_positive_n": 15, - "true_positive_correct": 0 - }, - { - "group": "b8b4a7b5769be3d4db661d88d40336ade544fcac1e5bd2ec6606c09f83e5263c", - "n": 7, - "accuracy": 0.7142857142857143, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "b9063c091cb8deaab4ea7fc341aebc169033dfb16792a7f0f253ce89ce556510", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bc9c57f088052889f45c0b709137c494d747464872fca72f16cfdde538281d64", - "n": 15, - "accuracy": 1.0, - "true_positive_n": 6, - "true_positive_correct": 6 - }, - { - "group": "bd4aa774324b8affeb489e7b796b6f32a9a2f46442394a8f5042c1967eac05dc", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c0371b16a587352143c383ca9799634e07feb94a2df75ef8830adb86efe153c1", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c045d0efa4ee5ab2be1d352625f16f6fb9e5cfbe90e02574363e83cd27bc5109", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "c98451abd4304660daabdb977564d36fdad533b4458c467bd2b97d80a651955f", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cb85febf11725dacf437b764d64bf097c9c2bab08e74ac2eeadb496ccd51b5e8", - "n": 16, - "accuracy": 0.875, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "cfe165abb396d68223741b950f77ec55bd59d4c200bd310d7e9a956548c1590e", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d530b979376d67460a1a3dcafcc95c91e1a51ce2d97124dc2050c2030ae223b7", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 12, - "true_positive_correct": 0 - }, - { - "group": "d5496d95b9bcd56d2d382354f44af6098e30a6b132c6bd2ccbac429ee48d7d3b", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d6e9a943c91ab8d80d913c89d053cf89f2fe4bcd7df1963f3fd5a70bf56f3deb", - "n": 12, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d794d72182a0c6f8a3009e01f6d607875a60752bf7f05920823cc55729bcbf3c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d82b9f42323202feb6db7ed45f9990070e72e41705aeb5aa64d60a3fd7549f17", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "dfc85bf0674a0000b0dfbcf621419502a19c4fffabf786b4a5e74a69776a1f9d", - "n": 5, - "accuracy": 0.4, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "e35b0e3bc4a28a3bcf49fd9fe5c6e3c6ece59e921ecabf1904c9459cb7b84ad1", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "e6e40718557ace0b1764c2b7ce7021ff2b42e61db43cd59c7a54132f69807b2d", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "f5b769ca29423ada31602cdd07bba0ab057d937256622788fa0872f6fe095865", - "n": 4, - "accuracy": 0.75, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "f72e0a3b72341d46d7eb5ab932fcce07cf34702afa40d217c5e997bef74764be", - "n": 15, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "f9476bc6e284b58a4d168d14b188b2ca1636ebe4a0adfcb0c40060bed41781c8", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "f97f37793bfaa1b487125bd7b1f34bff0fed50146a5606962c763220f17bb6de", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fb39d2600c554fe3ac1eddb65704bc34c6102ca0fbb4a77744ac525869d35be4", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fcd64091f029e53dd4a770300ecf446dbf5bca8cc267148f28b198584caeae46", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fe703c5eb241d3542fcc57e11688b3641c4f1f0b9506a720b42a9b9031382154", - "n": 19, - "accuracy": 0.7368421052631579, - "true_positive_n": 2, - "true_positive_correct": 0 - } - ], - "correct": 638 - }, - "global_calibrated": { - "n": 1064, - "accuracy": 0.599624060150376, - "nll": 0.785071927880167, - "brier": 0.46075730837789786, - "macro_f1": 0.5681608349992006, - "class_order": [ - "BenignPositive", - "FalsePositive", - "TruePositive" - ], - "per_class": { - "BenignPositive": { - "precision": 0.44772727272727275, - "recall": 0.701067615658363, - "f1-score": 0.5464632454923717, - "support": 281.0 - }, - "FalsePositive": { - "precision": 0.7606263982102909, - "recall": 0.6746031746031746, - "f1-score": 0.7150368033648791, - "support": 504.0 - }, - "TruePositive": { - "precision": 0.5706214689265536, - "recall": 0.36200716845878134, - "f1-score": 0.44298245614035087, - "support": 279.0 - } - }, - "confusion": [ - [ - 197, - 55, - 29 - ], - [ - 117, - 340, - 47 - ], - [ - 126, - 52, - 101 - ] - ], - "equal_organization_accuracy": 0.6743365093994473, - "organization_results": [ - { - "group": "0228d088db03da24994da8d12bb80e2925bf424387c12a0cab440661f61e53db", - "n": 2, - "accuracy": 0.5, - "true_positive_n": 2, - "true_positive_correct": 1 - }, - { - "group": "03b2cff809e7692ca15f889834563f866b356cc60a41023c8611dd2b60bcf51a", - "n": 2, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "04eb81bd28a99e8d254c7a0a9c07234f55af1f9e4fe167269f504109495c29a5", - "n": 4, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06646577f2484818ed5044762571197f87d3e79272459df5462e193f43ddd530", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "06f295d348152073a6bf3ba5485ca147a0c14bb94baf88391d465d7ea7e6eca9", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "08b767258688d27f07153aa89f0a2c2c6e8e426289043614eb48c2c881591192", - "n": 17, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "09890e550c1a9a583cf9aaa2a997c1bbc61162a5d36c8a891bbbc1a837dffc36", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "0b8649cde9e63fe45d17a7941ef90f01d1b3068c464c013d3ff7e21047164b9c", - "n": 5, - "accuracy": 0.8, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "10f18e39bc3c8e8713158ad66e666edf9118b8b83fb7eb21e5e39476a75a3125", - "n": 12, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14919c2e5da07951c425b98441bab1107d02cbcb0496a9148b184b1fea1eb4d2", - "n": 3, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "14a068f7500ef993a7f2e4cfcd5e194c57217338de4e72cf848d710c20392e91", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 18, - "true_positive_correct": 0 - }, - { - "group": "16cb7259665afd83462953cc90bf025352bab6f7210a59e5d86d49561f03fcbe", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "16fe8b7b9a0987bf861cee078b8763e8f75ec46733a6a59b61ffa8e2bd6e49f7", - "n": 5, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "17cd31c2cd2d2ccbabdfab902aee6e4e759167d126e1bcaf4d638c23b93272ac", - "n": 17, - "accuracy": 0.9411764705882353, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "18aa7b3932a87bfdfc7e3f032e91b38999f82e037146754bd8bba044870170ac", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "191618c5651b68d639db0f90f67315430ae08a9467ebc761ca10a54847034de5", - "n": 20, - "accuracy": 0.65, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "1e7cf0ee593015038b9cf7686c7db7604b7de8b52d0a8b556038b7795a59dd5a", - "n": 19, - "accuracy": 0.9473684210526315, - "true_positive_n": 4, - "true_positive_correct": 4 - }, - { - "group": "23f95dce880a8574d254a19ac7f94f1284a556dfa61126a92844ba979083ede1", - "n": 2, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2413f2cb2e9a29595c0278b868a30250d3573d487f99e35b3fe3836713d79e96", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "280f0d43bcdf087751b7a2b90739860bc02c2cccea845f81fae7555cdd9d7745", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "2af6f963a1301ac7cbf959fb0f63442e6b1b28be622e0d74c6b94c1e459c8bf0", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "2d4fd6af7767d8fbcfeddff36a76b5c993a49d5e45a515806ec30b1b53afa2ce", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3229724a48ff16d1ea5cff92fcbc1a29fb636760780576606a1f931cba64c7e6", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "33593514e9eb3b4a2f11fbff9a900e304d7f7088413e1a6ed36210de11e1a7cd", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "357e4fcd5f01346ea07a7b6805f5e0e94f11cd39d870ebde997c06b3ca23403b", - "n": 20, - "accuracy": 0.3, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "36be30d9fffc6077ec297a0ccde6690eb34709163c58b4b5df24236dd44c0e91", - "n": 18, - "accuracy": 0.9444444444444444, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3788764c156c1d8107b5e65fc226f169b6cfeb28e92341b3776e0323894d1182", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "397381ff8e300cf10164a627a5a6e614e68a7fb2a4dbac6daa5c67e331248cb3", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "3a3a67375cabf27ee46ffe17a8e3122e2bc0d77c22a4bc5fd3caf21af2fdf827", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "3ba2e76ba41ace83ac5fc70feb81f7bd790ebcc01cef912db9a520e0659301e6", - "n": 20, - "accuracy": 0.25, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "3c519ebea078526e0c7724d67b71d0e9d928d62adbb68de9cc3ab45caeab0ba6", - "n": 20, - "accuracy": 0.35, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "42a20c36367ded75633c56b5468efe4dab76b5242f7fc0fafe5de10ae8b9a078", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "4f4986e7532d0ba637710cb0010206dd22720f9b35cc163fd2f833fc93c25334", - "n": 14, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "52ed406f643d4adbab570be68c48860e0e9f239fe00b7f1bd0c35e00868fc879", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "557c22b05ccf851b94b1fc4571234d237e761ded29b151b3a3adcacd17f9ba00", - "n": 5, - "accuracy": 0.6, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "58f3e1e6f3eca7efe52c2f97b888733f6a541ca293e36d6e1a1f22c2332d05fb", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d11eb324b736d8f810b14b708ea1982028cb405fb0fb2e3f20dab9ab5bb3fe1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "5d2fc2acbf5e2afa16d5ad87f1cce4b784cdd43ca2da54c3eea33521e52a5a81", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 10, - "true_positive_correct": 0 - }, - { - "group": "5f21eec69194bcd2e80220b2701002f2da10e7cd2b04625674cdf69feb57dedd", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61c9e0872a43ed30a90ab06f7a3e8b85d33f54355b964a7fd1ffa361531c5626", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "61e7fb88134413ce324703852a01f81dda0cb96d4ae07a44cf845c9b1de8baa0", - "n": 20, - "accuracy": 0.7, - "true_positive_n": 13, - "true_positive_correct": 9 - }, - { - "group": "6201b69130be0b440362fc95b4cc2ef6c4c856543d9beb9584d43899e72d0283", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "629dffdc159a2dbb566dab56a6f17419a63aa2b0815020338d686277ff90f623", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "63b98074727eecae10e68ab106f9c5f2ec98bd2347e0c3eaa72d5557c1ae1b64", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "63deca8572354a01b85154c34b64122772c89023b59c7931064bb347d8389e7c", - "n": 14, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "659e4d7e826303db51376a34ed1ee1187305d8b4fcd2ba86de0f1df79cbe9d09", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 20, - "true_positive_correct": 20 - }, - { - "group": "681a13deadc6006e0d4d996f2b006e35b1d71fdeaaebfe98db36b84ce9e1bb82", - "n": 9, - "accuracy": 0.8888888888888888, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6972efb6c1c8c5719fc446f6b4e9bbf812c8274c04ddce38ae95cf5b06e65e10", - "n": 12, - "accuracy": 0.9166666666666666, - "true_positive_n": 5, - "true_positive_correct": 5 - }, - { - "group": "6d2a8e5968c904fbeceb057dc6d374d5cd1ba38019b8d31da27cc658c12aa476", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6d5ea32830050e4ac2840eed05d6b46ec9d7a78f65fbf85d0742d83cc628fa9c", - "n": 20, - "accuracy": 0.4, - "true_positive_n": 7, - "true_positive_correct": 0 - }, - { - "group": "6d74b56adb5791668e05d491f03f083060be7b184f47660a3b9cc14d7a32b511", - "n": 13, - "accuracy": 1.0, - "true_positive_n": 13, - "true_positive_correct": 13 - }, - { - "group": "6e73fd09bd5175004650b33517e33671c77ebb649384e14ffd95309ac3e567a7", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6f07feecb02e317c2db269ed9ec22266aa9195e11612cf7362bcdd1f3c9a7d3c", - "n": 20, - "accuracy": 0.95, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "6fc0443881957355817ec4b23a2d3027d6ce1ba949d0c313ac5134c5a33d7752", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "725e8bad3bf121140d3f1e8bb8625fac12766a1a7ebea88a432b39f8af246111", - "n": 20, - "accuracy": 0.95, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "745320acbe1a562a24246846ea4e7c92fe29683dd9e75636a6183142147be543", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "77af40ba5530d0abf3b141de72a4c48f4d594f8bc50515d71d71bb0ef57ba861", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "798e40bc3bfcfea7bb0935e395494ac4836e1d49f294b2ca803f7e234a478493", - "n": 4, - "accuracy": 0.5, - "true_positive_n": 4, - "true_positive_correct": 2 - }, - { - "group": "79e0a2063b1c1395f9032d7bcca4a7987b25c134851f2b7087535b66973f8767", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "7c2c1e933c21cfd9f0534226991df5542463f0d2c40138378024cbbabc1f205f", - "n": 20, - "accuracy": 0.55, - "true_positive_n": 7, - "true_positive_correct": 1 - }, - { - "group": "7d092ad55211811544dc5993b69d906906fce9350e187508760b25a549bbc9ad", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 5, - "true_positive_correct": 0 - }, - { - "group": "8269675fe1c1fcb6500b8262056237a7eb277d8d7822e63cc1b335c9b3587400", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "8433ad4a51679bb9fc21396e090e351720fb331d25292de538c1e1a8ca524bab", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8d353f585c30ae854c944e56bcab8cfff771273fefcde8d929e6faf7cb75488f", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8da5a2b78094becb52357364f9f1ea558c4052e869e9636c912fa88cce650666", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8e11912bbdeb4ce0b56ca82a5e0ad37a9159b78dd620607541dc93cb9ac8e02e", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "8f019450f830cbc42ded96cce0b7b4c1ee38439a9e198406ffb38ba5e0524fd7", - "n": 4, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "90994dcfeb71c7a87a4784556f5bf0bdb0b542c15a5f6b92da96e8020c4f1bf1", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "94d54be3b552337f2fc772ce8ff09d2b668ec497130ded8b947506a120acfb4e", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 14, - "true_positive_correct": 14 - }, - { - "group": "97d7c3f3b10a157cdb2791ff320a99a7d1856f8bbdd7ee020ad9ddbe7ec14b8e", - "n": 13, - "accuracy": 0.15384615384615385, - "true_positive_n": 13, - "true_positive_correct": 2 - }, - { - "group": "97f8ab62aee5d5d53245a565b836e892a9a7cda46748be23097421abb76ee39a", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 6, - "true_positive_correct": 0 - }, - { - "group": "99a049ff4aba4bc20df5f424b24efb9bab9118c192fd17f29b31ec910e497def", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "9b54c928a43ef6ff8b5b53cfeaad5e554106ed4d6718d08bc4f7440e968414af", - "n": 20, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "9bcc91cc05b92cb30cead1b81acb706497144e4815da75642e99d3fd7ebfd154", - "n": 11, - "accuracy": 0.9090909090909091, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9d2f045465f59eb2bcb8a37818341ce3635a125bced7bfbcb7bf9acdd80c1a71", - "n": 20, - "accuracy": 0.0, - "true_positive_n": 20, - "true_positive_correct": 0 - }, - { - "group": "9e08a63fef975cba0f7fe19b9931144a9480e43b5b78497f97cfe4079c916045", - "n": 10, - "accuracy": 0.5, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "9e38ecba2d710c7a9e3c2bef6abd472773b0edae845c5af26f317a4f5a2a3bc2", - "n": 19, - "accuracy": 0.42105263157894735, - "true_positive_n": 7, - "true_positive_correct": 6 - }, - { - "group": "a04b59b050c050568baa9abccd205e73547f1e5111afaabdbcbe5949c3511ec6", - "n": 20, - "accuracy": 0.05, - "true_positive_n": 3, - "true_positive_correct": 0 - }, - { - "group": "a1e75cda474e69786231632ed64953c19d60658191ea63553964e1d27015a7ce", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a52745698cb48cb1a28ac9bf212d724174c320e4648c9bececd83422b3364c1d", - "n": 19, - "accuracy": 0.10526315789473684, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a5a3b2d9f1ca0af015ed0584c90b28830aa11e86e8311c7f8f32f26b6d4f2768", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "a68f2269c6412f5885bdd26d3023dae3411476f5dd5b2e103355d4e1b0a41bb5", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a7b13e8d802fa5d584ad83d610714c858aab70e0c1f005324a2a91e7a369aed1", - "n": 16, - "accuracy": 0.0625, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "a94db659229b95e557d461f6de9071da0762515528aa3a77d5cb870fcf0f914b", - "n": 20, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aab21c537c79f9eee1a053d0d9f7fbea9eb1451e166156d59dd6c464104a7992", - "n": 10, - "accuracy": 0.5, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "abd52b07ebdf5d9235ca1239d9c3acbe23eb7cabdbc661e0dc93653479a095f0", - "n": 7, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "aefb87718e7fc3725adba69a4f789e9de7d638ec0a0d1f12fcff8a8735fb622c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b095632bbe63fbdc009ed98e30a50899f6a8edc165e3be8a924a7f0f600be4c0", - "n": 20, - "accuracy": 0.45, - "true_positive_n": 18, - "true_positive_correct": 7 - }, - { - "group": "b1b47d1978d292c9b80319bd2b531d2f2af206e3bcabf9aa0c46472112638ba9", - "n": 6, - "accuracy": 0.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "b7fa0bdd60d2f2321a27ad4ef1c84b2258bb60cb55372ea8aa56643d6d7a6d2b", - "n": 20, - "accuracy": 0.05, - "true_positive_n": 15, - "true_positive_correct": 0 - }, - { - "group": "b8b4a7b5769be3d4db661d88d40336ade544fcac1e5bd2ec6606c09f83e5263c", - "n": 7, - "accuracy": 0.7142857142857143, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "b9063c091cb8deaab4ea7fc341aebc169033dfb16792a7f0f253ce89ce556510", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "bc9c57f088052889f45c0b709137c494d747464872fca72f16cfdde538281d64", - "n": 15, - "accuracy": 1.0, - "true_positive_n": 6, - "true_positive_correct": 6 - }, - { - "group": "bd4aa774324b8affeb489e7b796b6f32a9a2f46442394a8f5042c1967eac05dc", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c0371b16a587352143c383ca9799634e07feb94a2df75ef8830adb86efe153c1", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "c045d0efa4ee5ab2be1d352625f16f6fb9e5cfbe90e02574363e83cd27bc5109", - "n": 1, - "accuracy": 0.0, - "true_positive_n": 1, - "true_positive_correct": 0 - }, - { - "group": "c98451abd4304660daabdb977564d36fdad533b4458c467bd2b97d80a651955f", - "n": 3, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "cb85febf11725dacf437b764d64bf097c9c2bab08e74ac2eeadb496ccd51b5e8", - "n": 16, - "accuracy": 0.875, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "cfe165abb396d68223741b950f77ec55bd59d4c200bd310d7e9a956548c1590e", - "n": 8, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d530b979376d67460a1a3dcafcc95c91e1a51ce2d97124dc2050c2030ae223b7", - "n": 20, - "accuracy": 0.1, - "true_positive_n": 12, - "true_positive_correct": 0 - }, - { - "group": "d5496d95b9bcd56d2d382354f44af6098e30a6b132c6bd2ccbac429ee48d7d3b", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d6e9a943c91ab8d80d913c89d053cf89f2fe4bcd7df1963f3fd5a70bf56f3deb", - "n": 12, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d794d72182a0c6f8a3009e01f6d607875a60752bf7f05920823cc55729bcbf3c", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "d82b9f42323202feb6db7ed45f9990070e72e41705aeb5aa64d60a3fd7549f17", - "n": 2, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "dfc85bf0674a0000b0dfbcf621419502a19c4fffabf786b4a5e74a69776a1f9d", - "n": 5, - "accuracy": 0.4, - "true_positive_n": 3, - "true_positive_correct": 1 - }, - { - "group": "e35b0e3bc4a28a3bcf49fd9fe5c6e3c6ece59e921ecabf1904c9459cb7b84ad1", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "e6e40718557ace0b1764c2b7ce7021ff2b42e61db43cd59c7a54132f69807b2d", - "n": 7, - "accuracy": 0.5714285714285714, - "true_positive_n": 1, - "true_positive_correct": 1 - }, - { - "group": "f5b769ca29423ada31602cdd07bba0ab057d937256622788fa0872f6fe095865", - "n": 4, - "accuracy": 0.75, - "true_positive_n": 2, - "true_positive_correct": 2 - }, - { - "group": "f72e0a3b72341d46d7eb5ab932fcce07cf34702afa40d217c5e997bef74764be", - "n": 15, - "accuracy": 0.6666666666666666, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "f9476bc6e284b58a4d168d14b188b2ca1636ebe4a0adfcb0c40060bed41781c8", - "n": 3, - "accuracy": 0.3333333333333333, - "true_positive_n": 2, - "true_positive_correct": 0 - }, - { - "group": "f97f37793bfaa1b487125bd7b1f34bff0fed50146a5606962c763220f17bb6de", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fb39d2600c554fe3ac1eddb65704bc34c6102ca0fbb4a77744ac525869d35be4", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fcd64091f029e53dd4a770300ecf446dbf5bca8cc267148f28b198584caeae46", - "n": 1, - "accuracy": 1.0, - "true_positive_n": 0, - "true_positive_correct": 0 - }, - { - "group": "fe703c5eb241d3542fcc57e11688b3641c4f1f0b9506a720b42a9b9031382154", - "n": 19, - "accuracy": 0.7368421052631579, - "true_positive_n": 2, - "true_positive_correct": 0 - } - ], - "correct": 638 + "0.95": { + "n": 277, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 194, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9929179700515357 + }, + "0.8": { + "coverage": 0.8022598870056498, + "accuracy": 1.0, + "confidence_cutoff": 0.9423198026275108 + } + } + } + }, + "sev_sysmon_native_file": { + "questions": 62, + "groups": 4, + "raw": { + "n": 62, + "nll": 0.010792211612586852, + "acc": 1.0, + "ece": 0.009954006393107204, + "brier": 0.002287040487090685, + "mean_conf": 0.9900459936068928, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9838709677419355, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.009954006393107173, + "top_bins": { + "0.9": { + "n": 61, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 59, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 54, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9986117420701618 + }, + "0.8": { + "coverage": 0.8064516129032258, + "accuracy": 1.0, + "confidence_cutoff": 0.9936233886125292 + } } }, - "swarm_static": { - "raw": { - "n": 354, - "accuracy": 0.9745762711864406, - "nll": 0.06671770193368179, - "brier": 0.03589778460324393, - "correct": 345 - }, - "global_calibrated": { - "n": 354, - "accuracy": 0.9745762711864406, - "nll": 0.08603305952328294, - "brier": 0.03970483443698841, - "correct": 345 - }, - "families": { - "binding": { - "raw": { - "n": 164, - "accuracy": 0.9512195121951219, - "nll": 0.12986993405061373, - "brier": 0.06690797882642434, - "correct": 156 - }, - "global_calibrated": { - "n": 164, - "accuracy": 0.9512195121951219, - "nll": 0.17000841292483287, - "brier": 0.07668135319778648, - "correct": 156 - } + "calibrated": { + "n": 62, + "nll": 0.04080521144437412, + "acc": 1.0, + "ece": 0.03730996348948351, + "brier": 0.008675036111100168, + "mean_conf": 0.9626900365105165, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9516129032258065, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.037309963489483455, + "top_bins": { + "0.9": { + "n": 59, + "errors": 0, + "error_rate": 0.0 }, - "direct": { - "raw": { - "n": 134, - "accuracy": 1.0, - "nll": 0.0002463532494033971, - "brier": 5.956582865744346e-06, - "correct": 134 - }, - "global_calibrated": { - "n": 134, - "accuracy": 1.0, - "nll": 0.003880609414537556, - "brier": 0.00015479616373253736, - "correct": 134 - } + "0.95": { + "n": 48, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 23, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9803392906572079 + }, + "0.8": { + "coverage": 0.8064516129032258, + "accuracy": 1.0, + "confidence_cutoff": 0.9477741924481837 + } + } + } + }, + "sev_sysmon_native_join": { + "questions": 120, + "groups": 3, + "raw": { + "n": 120, + "nll": 0.0008810538663341702, + "acc": 1.0, + "ece": 0.0008800625080441771, + "brier": 3.5216739059086607e-06, + "mean_conf": 0.9991199374919558, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0008800625080441771, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9994848628990232 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.998641259167216 + } + } + }, + "calibrated": { + "n": 120, + "nll": 0.013824573239669817, + "acc": 1.0, + "ece": 0.013685808798396426, + "brier": 0.00044603592410257366, + "mean_conf": 0.9863141912016036, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.013685808798396426, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 118, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 55, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9888123032356468 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.980357236597309 + } + } + } + }, + "sev_sysmon_native_other_effect": { + "questions": 88, + "groups": 4, + "raw": { + "n": 88, + "nll": 0.007295006715728625, + "acc": 1.0, + "ece": 0.006888792886758961, + "brier": 0.001385599246512425, + "mean_conf": 0.9931112071132411, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9886363636363636, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.006888792886758921, + "top_bins": { + "0.9": { + "n": 87, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 86, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 79, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9993634031772723 + }, + "0.8": { + "coverage": 0.8068181818181818, + "accuracy": 1.0, + "confidence_cutoff": 0.9952397949435474 + } + } + }, + "calibrated": { + "n": 88, + "nll": 0.03669008514897713, + "acc": 1.0, + "ece": 0.03458520458981973, + "brier": 0.006053607090618847, + "mean_conf": 0.9654147954101805, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9431818181818182, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.03458520458981951, + "top_bins": { + "0.9": { + "n": 83, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 67, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 41, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9865664502206583 + }, + "0.8": { + "coverage": 0.8068181818181818, + "accuracy": 1.0, + "confidence_cutoff": 0.9454838788991825 + } + } + } + }, + "sev_sysmon_native_process": { + "questions": 74, + "groups": 4, + "raw": { + "n": 74, + "nll": 0.005360095133334598, + "acc": 1.0, + "ece": 0.005183972707978367, + "brier": 0.0006452896028548641, + "mean_conf": 0.9948160272920217, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9864864864864865, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.005183972707978324, + "top_bins": { + "0.9": { + "n": 73, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 72, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 67, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9998754065974452 + }, + "0.8": { + "coverage": 0.8108108108108109, + "accuracy": 1.0, + "confidence_cutoff": 0.9964080071736937 + } + } + }, + "calibrated": { + "n": 74, + "nll": 0.02430026482825407, + "acc": 1.0, + "ece": 0.023069255552173623, + "brier": 0.004038757822471681, + "mean_conf": 0.9769307444478267, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9324324324324325, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.023069255552173318, + "top_bins": { + "0.9": { + "n": 69, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 65, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 43, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9959686665627235 + }, + "0.8": { + "coverage": 0.8108108108108109, + "accuracy": 1.0, + "confidence_cutoff": 0.9671912008530428 + } + } + } + }, + "sev_sysmon_native_registry": { + "questions": 66, + "groups": 4, + "raw": { + "n": 66, + "nll": 0.02506239179799794, + "acc": 0.9848484848484849, + "ece": 0.005781042813278618, + "brier": 0.017230444731313608, + "mean_conf": 0.9868495288218273, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9696969696969697, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9696969696969697, + "aurc": 0.00046266864448682633, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.0020010439733424423, + "top_bins": { + "0.9": { + "n": 64, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 64, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 61, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9996958591947809 + }, + "0.8": { + "coverage": 0.803030303030303, + "accuracy": 1.0, + "confidence_cutoff": 0.9967594448763499 + } + } + }, + "calibrated": { + "n": 66, + "nll": 0.03908316195438624, + "acc": 0.9848484848484849, + "ece": 0.018983776220710902, + "brier": 0.01748193378444616, + "mean_conf": 0.9706927611865057, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9545454545454546, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9696969696969697, + "aurc": 0.00046266864448682633, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.014155723661979147, + "top_bins": { + "0.9": { + "n": 63, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 58, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 36, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9919306919122572 + }, + "0.8": { + "coverage": 0.803030303030303, + "accuracy": 1.0, + "confidence_cutoff": 0.9688350009585398 + } + } + } + } + }, + "candidate": { + "excytin_file": { + "questions": 80, + "groups": 1, + "raw": { + "n": 80, + "nll": 8.416562133899781e-05, + "acc": 1.0, + "ece": 8.407504508700647e-05, + "brier": 3.608306194877327e-07, + "mean_conf": 0.999915924954913, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -8.407504508700647e-05, + "top_bins": { + "0.9": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999870465451609 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9999445146924221 + } + } + }, + "calibrated": { + "n": 80, + "nll": 0.0025800477441721577, + "acc": 1.0, + "ece": 0.0025682388226171415, + "brier": 4.4624375778219304e-05, + "mean_conf": 0.9974317611773829, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0025682388226171415, + "top_bins": { + "0.9": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 79, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9984996790217291 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9968816753727368 + } + } + } + }, + "excytin_parent_join": { + "questions": 78, + "groups": 1, + "raw": { + "n": 78, + "nll": 0.0010931532029286512, + "acc": 1.0, + "ece": 0.0010890559164937752, + "brier": 1.524968968159761e-05, + "mean_conf": 0.9989109440835062, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0010890559164937752, + "top_bins": { + "0.9": { + "n": 78, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 78, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 77, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9997808486535813 + }, + "0.8": { + "coverage": 0.8076923076923077, + "accuracy": 1.0, + "confidence_cutoff": 0.9985521003494622 + } + } + }, + "calibrated": { + "n": 78, + "nll": 0.014932533018017379, + "acc": 1.0, + "ece": 0.014655484687648102, + "brier": 0.0008621856358382366, + "mean_conf": 0.9853445153123519, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.014655484687648102, + "top_bins": { + "0.9": { + "n": 78, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 73, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 49, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9924833351572758 + }, + "0.8": { + "coverage": 0.8076923076923077, + "accuracy": 1.0, + "confidence_cutoff": 0.9749372136312703 + } + } + } + }, + "excytin_process": { + "questions": 120, + "groups": 1, + "raw": { + "n": 120, + "nll": 0.00038924968205708544, + "acc": 1.0, + "ece": 0.00038892368944332123, + "brier": 1.2813112890567744e-06, + "mean_conf": 0.9996110763105567, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.00038892368944332123, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999547557532524 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9992749850755753 + } + } + }, + "calibrated": { + "n": 120, + "nll": 0.007136704365819422, + "acc": 1.0, + "ece": 0.007078397430680172, + "brier": 0.00021856930435733632, + "mean_conf": 0.9929216025693198, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.007078397430680172, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 84, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9971072222116408 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9865275469490526 + } + } + } + }, + "excytin_signin": { + "questions": 120, + "groups": 1, + "raw": { + "n": 120, + "nll": 0.0009466221977245799, + "acc": 1.0, + "ece": 0.0009222842581878554, + "brier": 7.013154556645113e-05, + "mean_conf": 0.9990777157418121, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0009222842581878554, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999016374480455 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9997216681533668 + } + } + }, + "calibrated": { + "n": 120, + "nll": 0.00847100086283192, + "acc": 1.0, + "ece": 0.008170507143524745, + "brier": 0.0008193865392318428, + "mean_conf": 0.9918294928564755, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9916666666666667, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.008170507143524519, + "top_bins": { + "0.9": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 119, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 105, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9955464217668417 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9921417222813483 + } + } + } + }, + "sev_guide_detector": { + "questions": 1064, + "groups": 114, + "raw": { + "n": 1064, + "nll": 0.8605571101399087, + "acc": 0.6005639097744361, + "ece": 0.14734385206430384, + "brier": 0.5008109312564021, + "mean_conf": 0.7382136284198534, + "confident_error_rate": 0.017857142857142856, + "coverage_at_0_9": 0.33270676691729323, + "accuracy_at_0_9": 0.9463276836158192, + "coverage_at_5pct_error": 0.3082706766917293, + "coverage_at_1pct_error": 0.0, + "aurc": 0.19350209200158913, + "error_rate_at_0_9": 0.05367231638418079, + "confidence_bias": 0.1376497186454173, + "top_bins": { + "0.9": { + "n": 354, + "errors": 19, + "error_rate": 0.05367231638418079 + }, + "0.95": { + "n": 327, + "errors": 13, + "error_rate": 0.039755351681957186 + }, + "0.99": { + "n": 319, + "errors": 12, + "error_rate": 0.03761755485893417 + } + }, + "selective": { + "0.5": { + "coverage": 0.5028195488721805, + "accuracy": 0.8205607476635514, + "confidence_cutoff": 0.7059991149306561 + }, + "0.8": { + "coverage": 0.8035714285714286, + "accuracy": 0.6596491228070176, + "confidence_cutoff": 0.5473316703985754 + } + } + }, + "calibrated": { + "n": 1064, + "nll": 0.7966730081647482, + "acc": 0.6005639097744361, + "ece": 0.06373435814748656, + "brier": 0.46843654552918773, + "mean_conf": 0.6455031255616036, + "confident_error_rate": 0.011278195488721804, + "coverage_at_0_9": 0.299812030075188, + "accuracy_at_0_9": 0.9623824451410659, + "coverage_at_5pct_error": 0.3082706766917293, + "coverage_at_1pct_error": 0.0, + "aurc": 0.1938629191063446, + "error_rate_at_0_9": 0.03761755485893417, + "confidence_bias": 0.04493921578716753, + "top_bins": { + "0.9": { + "n": 319, + "errors": 12, + "error_rate": 0.03761755485893417 + }, + "0.95": { + "n": 0, + "errors": 0, + "error_rate": null + }, + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5028195488721805, + "accuracy": 0.8205607476635514, + "confidence_cutoff": 0.5612814391210313 + }, + "0.8": { + "coverage": 0.8007518796992481, + "accuracy": 0.6619718309859155, + "confidence_cutoff": 0.4656043556130704 + } + } + } + }, + "sev_guide_triage": { + "questions": 231, + "groups": 23, + "raw": { + "n": 231, + "nll": 0.8396428451107271, + "acc": 0.5670995670995671, + "ece": 0.1889188281653689, + "brier": 0.5162927524049745, + "mean_conf": 0.7421494516023849, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.3593073593073593, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.3722943722943723, + "coverage_at_1pct_error": 0.36363636363636365, + "aurc": 0.20987091476964936, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.17504988450281778, + "top_bins": { + "0.9": { + "n": 83, + "errors": 0, + "error_rate": 0.0 }, - "flow": { - "raw": { - "n": 56, - "accuracy": 0.9821428571428571, - "nll": 0.040828320800047294, - "brier": 0.030966232855549104, - "correct": 55 - }, - "global_calibrated": { - "n": 56, - "accuracy": 0.9821428571428571, - "nll": 0.03668431589324172, - "brier": 0.02605404964851352, - "correct": 55 - } + "0.95": { + "n": 80, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 14, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5627705627705628, + "accuracy": 0.7153846153846154, + "confidence_cutoff": 0.7564145922660828 + }, + "0.8": { + "coverage": 0.8051948051948052, + "accuracy": 0.5806451612903226, + "confidence_cutoff": 0.51023405030021 } } }, - "manual": { - "raw": { - "n": 83, - "accuracy": 0.7831325301204819, - "nll": 1.1015953331868638, - "brier": 0.38672041228910636, - "correct": 65 - }, - "global_calibrated": { - "n": 83, - "accuracy": 0.7831325301204819, - "nll": 0.8197108651023933, - "brier": 0.3693842986364894, - "correct": 65 - }, - "families": { - "binding": { - "raw": { - "n": 40, - "accuracy": 0.8, - "nll": 0.906880605987712, - "brier": 0.3722274315061049, - "correct": 32 - }, - "global_calibrated": { - "n": 40, - "accuracy": 0.8, - "nll": 0.7146033829092314, - "brier": 0.3514071574500127, - "correct": 32 - } + "calibrated": { + "n": 231, + "nll": 0.7761249864817235, + "acc": 0.5670995670995671, + "ece": 0.16797778410789055, + "brier": 0.46407372894941223, + "mean_conf": 0.6510059283006658, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.341991341991342, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 0.37662337662337664, + "coverage_at_1pct_error": 0.36363636363636365, + "aurc": 0.20742662880964652, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.08390636120109873, + "top_bins": { + "0.9": { + "n": 79, + "errors": 0, + "error_rate": 0.0 }, - "flow": { - "raw": { - "n": 40, - "accuracy": 0.75, - "nll": 1.3789279125484732, - "brier": 0.43021742389448914, - "correct": 30 - }, - "global_calibrated": { - "n": 40, - "accuracy": 0.75, - "nll": 0.9861678633098097, - "brier": 0.4150648418384071, - "correct": 30 - } + "0.95": { + "n": 14, + "errors": 0, + "error_rate": 0.0 }, - "prerequisite": { - "raw": { - "n": 3, - "accuracy": 1.0, - "nll": 2.3971020760583907e-05, - "brier": 1.3240221828953951e-09, - "correct": 3 - }, - "global_calibrated": { - "n": 3, - "accuracy": 1.0, - "nll": 0.0017173182456664555, - "brier": 5.60509727845392e-06, - "correct": 3 - } + "0.99": { + "n": 0, + "errors": 0, + "error_rate": null + } + }, + "selective": { + "0.5": { + "coverage": 0.5584415584415584, + "accuracy": 0.7209302325581395, + "confidence_cutoff": 0.6016800953517527 + }, + "0.8": { + "coverage": 0.8095238095238095, + "accuracy": 0.5989304812834224, + "confidence_cutoff": 0.4481415540755688 } } - }, - "sysmon": { - "raw": { - "n": 64, - "accuracy": 0.96875, - "nll": 0.08322377416424186, - "brier": 0.045695299016738855, - "correct": 62 - }, - "global_calibrated": { - "n": 64, - "accuracy": 0.96875, - "nll": 0.11024750472054996, - "brier": 0.04954806836760281, - "correct": 62 + } + }, + "sev_swarmtraces_static": { + "questions": 354, + "groups": 176, + "raw": { + "n": 354, + "nll": 0.04944532768977841, + "acc": 0.980225988700565, + "ece": 0.01487544035693287, + "brier": 0.024608535342050798, + "mean_conf": 0.9802643779708107, + "confident_error_rate": 0.002824858757062147, + "coverage_at_0_9": 0.9463276836158192, + "accuracy_at_0_9": 0.9970149253731343, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9858757062146892, + "aurc": 0.0005032049402250012, + "error_rate_at_0_9": 0.0029850746268656717, + "confidence_bias": 3.8389270245686724e-05, + "top_bins": { + "0.9": { + "n": 335, + "errors": 1, + "error_rate": 0.0029850746268656717 + }, + "0.95": { + "n": 330, + "errors": 1, + "error_rate": 0.0030303030303030303 + }, + "0.99": { + "n": 308, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999276209511099 + }, + "0.8": { + "coverage": 0.8022598870056498, + "accuracy": 1.0, + "confidence_cutoff": 0.997796467038603 + } } }, - "general": { - "raw": { - "n": 105, - "accuracy": 0.819047619047619, - "nll": 0.5574328669578856, - "brier": 0.29243113448917374, - "correct": 86 - }, - "global_calibrated": { - "n": 105, - "accuracy": 0.819047619047619, - "nll": 0.42292499037031256, - "brier": 0.24649933996243034, - "correct": 86 + "calibrated": { + "n": 354, + "nll": 0.06588860080731399, + "acc": 0.980225988700565, + "ece": 0.029025934670617953, + "brier": 0.028907916842541088, + "mean_conf": 0.9580211141684397, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.8898305084745762, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9887005649717514, + "aurc": 0.0005463830895640532, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.022204874532125274, + "top_bins": { + "0.9": { + "n": 315, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 297, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 221, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9952110216327251 + }, + "0.8": { + "coverage": 0.8022598870056498, + "accuracy": 1.0, + "confidence_cutoff": 0.9673839577051934 + } } } - } - } - }, - "admissions": { - "parent": { - "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-swarmtraces-4b-v2/00-checkpoint", - "files": { - "result.json": "46b3ee2a68f250ba1d433c25612083bb8235a03e617d90707fa796f86ae2cbf4", - "calibration/rows.json": "9a33b44bee9cb63a635ac1516bcab9f67fbe56b15a72995a8c2b54a321ad4dd6", - "development/rows.json": "d882c91290398ba96cf39de9128ea5a973cf8d79d20ca84f5249f9f1c1cd5300", - "provenance.json": "39898e5e6ef87112f653436a9b6b80555da10a01b0d203b3f5eb29000dddae3a", - "calibration/report.json": "15dc07e79e0121e7345ece52b028220fbbf9123609a2eee5b331c9d372370383", - "development/report.json": "02fc22461d5b44f88173b90e0fffe97905b4eeb132eec79088c9a8396ce73134", - "transfer/rows.json": "5b624c0b0f91940332045523703dd5d52761a5ff931f2c9f1751185883976368", - "transfer/report.json": "f44bc7be71a3b37ded9e35f6fa382c1907c70fc959c15bc712b76f024ac726cc" }, - "provenance": { - "config": {}, - "config_sha256": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a", - "suite_sha256": "61e3c1898b8a03eb178355d3c8c5050e85bd217285c700df9b4506ec9e3680f8", - "source_hashes": { - "kev/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", - "kev/anchors.py": "6963eafdb276db5a6c94d939eae675c761555553b8d0f963ffd803448beaabbb", - "kev/api.py": "7bffacfb762c626b8bc2f670f350295af5ccb0883dbe90239c7d2f8e5ef56582", - "kev/autoresearch.py": "f7d7fabb9c0df7065bee3fec4aa4bee7028c5d1565847aefe2d74e3d7d40ed8a", - "kev/benchmark.py": "8d486659a9036102386b077fcbd197fc0d85ae7752a4711666d6b3a95622b26f", - "kev/calibrate.py": "c1eff4744fabd349e8abca86777a7aa0cbea44904c8223d3b61fca3d43741519", - "kev/checkpoint.py": "9f1dbc5d3d7d3050aca4ee34e0578c9ad76b7a2a7467dfc7a4ad65e3c471db59", - "kev/compare.py": "3606dbf98bb305d66420158498cb837304d2edb8dab01cdaf2ec42964bef3435", - "kev/composition.py": "f335ed17e18e0a544893db5e22b9a059e6ce1b2e14dbb863ac7d7bbf8f3e0536", - "kev/contrastive.py": "cbb979aa5d40265ad0e64695f94b281d91751fa405811ddfa8212fede111edcf", - "kev/data.py": "a5a5e200f56b60e4bf30a195a4cc87a74afddae15999ec48c7543a4f9b59f71c", - "kev/device.py": "d1677fd98ec0979c7284546306e34e0d09298ca4042fc39eb2b32a74f4e975c4", - "kev/evaluate.py": "999264f2837dcfbf2ec93601aa4745e701698ab850a43ad89324672a6604965c", - "kev/experiment.py": "1a5ca47d2b3d236e53c87e8c4197704f162c6692aa008521d31ea491945693f0", - "kev/jev.py": "acd4cc3f1844e438cc83a8d409c15ef78a5ef64d39ea10f583d75a6c7646b243", - "kev/metrics.py": "db0930d470c0ce16f16dcaf726d49a7a103973a5296cefc5259d1e68707d882b", - "kev/mlx_model.py": "f582428796faf6bf962b772a69a6227ac3259ccbd3215c0c8b55ce28dbbd0909", - "kev/model.py": "eac0f56c807b85b5c4077483d922e6bb03f07890de2835efd9372cf20cd01b8e", - "kev/plot.py": "0874bcff2885d8155a1cceade0de8a275a7163d3c3fd6aa0c294e8ccf5ff7e02", - "kev/predictors.py": "b2a66dd9f0f0a8026bc94603383bacfdd622e75d1c935173997175606626a9b3", - "kev/publish.py": "8abc9bcf4a01365697b05cf3c5ad0013462bd92d005f5616954e40ab1100c7aa", - "kev/serve.py": "c210b1f4598e64b49c399f73603ed372f4314c165ceda50e1340c0466ec1c735", - "kev/study_v3.py": "7891150aa623e4479c7c789d4dc4662f186132b37b7bac1e1f072b9e08ee13e3", - "kev/suite.py": "c5e5e27f2fed871b547a021d180b0a7f4a0542a488cce4ee02ad936f433441db", - "kev/train.py": "a2d85632d73760fb3efe51627ff4fde6af3100ed1c34e75d290c1b00c9899aa9", - "kev/transfer_v9.py": "0902409742151250a28af1fd8f42258b70fcc161f3deed3ee35fe3f473f3c763", - "modal_app.py": "639577258f4d581daefa51dad639961280ce92de9da256316ab4124f005a53fb", - "pyproject.toml": "39ec40f14c88598dee7ac98fb859a5a44e22337235a3064ebd5e89036ce81e30", - "uv.lock": "a9922dbb89acdef78299fd2b4a8c3f7f0fa1b2bc08b55595b6926fa785a9c466" + "sev_sysmon_native_file": { + "questions": 62, + "groups": 4, + "raw": { + "n": 62, + "nll": 0.008428780795422771, + "acc": 1.0, + "ece": 0.007810434776248104, + "brier": 0.0017443201233063612, + "mean_conf": 0.9921895652237519, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9838709677419355, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0078104347762480986, + "top_bins": { + "0.9": { + "n": 61, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 59, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 57, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9991686260052061 + }, + "0.8": { + "coverage": 0.8064516129032258, + "accuracy": 1.0, + "confidence_cutoff": 0.9962482122843107 + } + } }, - "git_commit": "449101a53d4385bac3aef7df4fee5a0d93248953", - "platform": "Linux-4.19.0-gvisor-x86_64-with-glibc2.36", - "torch": "2.8.0+cu128", - "device": "cuda", - "gpu": "NVIDIA H100 80GB HBM3", - "legacy_checkpoint": true, - "measured_checkpoint": { - "requested": "/runs/sev-r2-curriculum-repair-4b-v1/02-trial-2/checkpoint", - "resolved": "/runs/sev-r2-curriculum-repair-4b-v1/02-trial-2/checkpoint", - "head_sha256": "5a91ed82f11c6d0c8ab0a910e0c85770119e0034039379aa601d17a3ec7664d6", - "adapter_sha256": "0452010abdcc58f7c39a47a1562a1c812bf85f025d4a6c105c93d486051765de", - "inference_temperature": 1.0 + "calibrated": { + "n": 62, + "nll": 0.03509983435601263, + "acc": 1.0, + "ece": 0.032160212545650484, + "brier": 0.007419485561600889, + "mean_conf": 0.9678397874543494, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9516129032258065, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0321602125456506, + "top_bins": { + "0.9": { + "n": 59, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 55, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 26, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9838915386931907 + }, + "0.8": { + "coverage": 0.8064516129032258, + "accuracy": 1.0, + "confidence_cutoff": 0.9583052893470861 + } + } } }, - "experiment_gates": { - "passed": true, - "checks": { - "complete_coverage": true, - "isolation_and_packing": true, - "transfer_complete": true, - "heldout_pairs_at_least_70pct": true, - "transfer_confident_errors_below_10pct": true - }, - "policy": "Research screening only: 0.001 isolation; 5pp task regression; 70% heldout pair correctness; <=10% confident errors; no automatic release." - }, - "weight_verification": "Recorded checkpoint digests; model weights were not opened by this report" - }, - "candidate": { - "directory": "/Users/mac/Sev-security-20260925/runs/sev-r2-swarmtraces-4b-retention-v1/00-trial-0", - "files": { - "result.json": "9e6143f4cf6055c8644458f6920b7138a1a348cc254422e9d3173c268b97c4c6", - "calibration/rows.json": "5dc5633f59c9da863561e1ae34fbc1c87199053cf1397d4592b921ec5ff69790", - "development/rows.json": "9c969c91604dbe733626246efcdecee0c3706d84d5e7504d07f586ee2b518542", - "provenance.json": "5712631ff496a274abe8abecc84f8f259c8e58d6bf1f4a73638d969f2cce8227", - "calibration/report.json": "4f0cae5eb45b35029fe8b65241b5a8976c6a11a7bb5453b8111a6da995c4c40f", - "development/report.json": "f12c5bd74787f4c0daac0f93807f6f2e7c537509a6cc05fb98828e31a58aab20", - "checkpoint/training_config.json": "daf81a93394c6cd9e38cc4b0e049f54910048ffebf4de34f6ded2339d85478cb", - "transfer/rows.json": "2987ed490118f383c96f590852e4a4a4ef72721d50887f8fe4ff7e406670ba37", - "transfer/report.json": "31345b7b6f60a2036039d233f0c484dc7abac1285e1b5db0fe7444d82c9014e4" - }, - "provenance": { - "config": { - "epochs": 1, - "seed": 4, - "lr": 2.5e-06, - "lora": 16, - "accum": 2, - "batch": 4, - "perm_kl": 0, - "perm_frac": 0.3, - "ord_w": 0.0, - "p_none": 0, - "p_none_distract": 0, - "p_distract": 0, - "p_none_pair": 0, - "synthetic_repeat": 1, - "public_frac": 1, - "head_lr": 0.0, - "weight_decay": 0.01, - "anchor_w": 0.0, - "label_smoothing": 0.0, - "brier_w": 0.0, - "focal_gamma": 0.0, - "base": "Qwen/Qwen3.5-4B-Base", - "base_revision": "1001bb4d826a52d1f399e183466143f4da7b741b", - "init_from": "/runs/sev-r2-curriculum-repair-4b-v1/02-trial-2/checkpoint", - "lora_targets": "all", - "head_dim": 256, - "weights_dtype": "fp32", - "dtype": "bf16", - "checkpointing": 1, - "option_isolation": 0, - "special_embeddings": 0 - }, - "config_sha256": "958d0b20b3ffd1c572e0abf221383a450e0f8e8d3fff3cb745235a371e6a025d", - "suite_sha256": "61e3c1898b8a03eb178355d3c8c5050e85bd217285c700df9b4506ec9e3680f8", - "source_hashes": { - "kev/__init__.py": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", - "kev/anchors.py": "6963eafdb276db5a6c94d939eae675c761555553b8d0f963ffd803448beaabbb", - "kev/api.py": "7bffacfb762c626b8bc2f670f350295af5ccb0883dbe90239c7d2f8e5ef56582", - "kev/autoresearch.py": "f7d7fabb9c0df7065bee3fec4aa4bee7028c5d1565847aefe2d74e3d7d40ed8a", - "kev/benchmark.py": "8d486659a9036102386b077fcbd197fc0d85ae7752a4711666d6b3a95622b26f", - "kev/calibrate.py": "c1eff4744fabd349e8abca86777a7aa0cbea44904c8223d3b61fca3d43741519", - "kev/checkpoint.py": "9f1dbc5d3d7d3050aca4ee34e0578c9ad76b7a2a7467dfc7a4ad65e3c471db59", - "kev/compare.py": "3606dbf98bb305d66420158498cb837304d2edb8dab01cdaf2ec42964bef3435", - "kev/composition.py": "f335ed17e18e0a544893db5e22b9a059e6ce1b2e14dbb863ac7d7bbf8f3e0536", - "kev/contrastive.py": "cbb979aa5d40265ad0e64695f94b281d91751fa405811ddfa8212fede111edcf", - "kev/data.py": "a5a5e200f56b60e4bf30a195a4cc87a74afddae15999ec48c7543a4f9b59f71c", - "kev/device.py": "d1677fd98ec0979c7284546306e34e0d09298ca4042fc39eb2b32a74f4e975c4", - "kev/evaluate.py": "999264f2837dcfbf2ec93601aa4745e701698ab850a43ad89324672a6604965c", - "kev/experiment.py": "1a5ca47d2b3d236e53c87e8c4197704f162c6692aa008521d31ea491945693f0", - "kev/jev.py": "acd4cc3f1844e438cc83a8d409c15ef78a5ef64d39ea10f583d75a6c7646b243", - "kev/metrics.py": "db0930d470c0ce16f16dcaf726d49a7a103973a5296cefc5259d1e68707d882b", - "kev/mlx_model.py": "f582428796faf6bf962b772a69a6227ac3259ccbd3215c0c8b55ce28dbbd0909", - "kev/model.py": "eac0f56c807b85b5c4077483d922e6bb03f07890de2835efd9372cf20cd01b8e", - "kev/plot.py": "0874bcff2885d8155a1cceade0de8a275a7163d3c3fd6aa0c294e8ccf5ff7e02", - "kev/predictors.py": "b2a66dd9f0f0a8026bc94603383bacfdd622e75d1c935173997175606626a9b3", - "kev/publish.py": "8abc9bcf4a01365697b05cf3c5ad0013462bd92d005f5616954e40ab1100c7aa", - "kev/serve.py": "c210b1f4598e64b49c399f73603ed372f4314c165ceda50e1340c0466ec1c735", - "kev/study_v3.py": "7891150aa623e4479c7c789d4dc4662f186132b37b7bac1e1f072b9e08ee13e3", - "kev/suite.py": "c5e5e27f2fed871b547a021d180b0a7f4a0542a488cce4ee02ad936f433441db", - "kev/train.py": "a2d85632d73760fb3efe51627ff4fde6af3100ed1c34e75d290c1b00c9899aa9", - "kev/transfer_v9.py": "0902409742151250a28af1fd8f42258b70fcc161f3deed3ee35fe3f473f3c763", - "modal_app.py": "639577258f4d581daefa51dad639961280ce92de9da256316ab4124f005a53fb", - "pyproject.toml": "39ec40f14c88598dee7ac98fb859a5a44e22337235a3064ebd5e89036ce81e30", - "uv.lock": "a9922dbb89acdef78299fd2b4a8c3f7f0fa1b2bc08b55595b6926fa785a9c466" + "sev_sysmon_native_join": { + "questions": 120, + "groups": 3, + "raw": { + "n": 120, + "nll": 0.00039767294618876087, + "acc": 1.0, + "ece": 0.0003974740966455892, + "brier": 7.310901229709924e-07, + "mean_conf": 0.9996025259033544, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.0003974740966455892, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9997660582574378 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9994606035447323 + } + } }, - "git_commit": "76e180880160721082a065bf1882495e756a643a", - "platform": "Linux-4.19.0-gvisor-x86_64-with-glibc2.36", - "torch": "2.8.0+cu128", - "device": "cuda", - "gpu": "NVIDIA H100 80GB HBM3", - "legacy_checkpoint": false, - "measured_checkpoint": { - "requested": "/runs/sev-r2-swarmtraces-4b-retention-v1/00-trial-0/checkpoint", - "resolved": "/runs/sev-r2-swarmtraces-4b-retention-v1/00-trial-0/checkpoint", - "head_sha256": "b29abc817f2340aa510fcd4b2c813455ef38102508120ffc35f7917ce359c27e", - "adapter_sha256": "b5afd73584d1098cccb0a1bd42afc40e2de3c3bc3074a5173a642b6d4b880420", - "inference_temperature": 1.0 + "calibrated": { + "n": 120, + "nll": 0.009879536093519296, + "acc": 1.0, + "ece": 0.009811175359927704, + "brier": 0.00022722202758135087, + "mean_conf": 0.9901888246400723, + "confident_error_rate": 0.0, + "coverage_at_0_9": 1.0, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.009811175359927704, + "top_bins": { + "0.9": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 120, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 67, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9917343480609009 + }, + "0.8": { + "coverage": 0.8, + "accuracy": 1.0, + "confidence_cutoff": 0.9855433361281362 + } + } } }, - "experiment_gates": { - "passed": true, - "checks": { - "complete_coverage": true, - "isolation_and_packing": true, - "transfer_complete": true, - "heldout_pairs_at_least_70pct": true, - "transfer_confident_errors_below_10pct": true - }, - "policy": "Research screening only: 0.001 isolation; 5pp task regression; 70% heldout pair correctness; <=10% confident errors; no automatic release." - }, - "weight_verification": "Recorded checkpoint digests; model weights were not opened by this report" - } - }, - "training_resources": { - "wall_seconds": 1043.7473661899567, - "records_seen": 6438, - "requested_records": 6438, - "truncated_records": 0, - "rejected_records": 0, - "optimizer_steps": 805, - "forward_tokens": 1447855, - "peak_device_bytes": 22227358208, - "device": "cuda", - "dtype": "bf16", - "batch": 4, - "peak_rss_bytes": 31557873664 - }, - "paired_raw": { - "excytin": { - "groups": 1, - "ci95": null, - "point_deltas": { - "acc": 0.0, - "nll": -0.0010619269814233934, - "brier": -8.441910898222422e-05 - }, - "reason": "One incident group; no generalization interval estimated" - }, - "guide": { - "acc": { - "micro_acc_delta": 0.012987012987012991, - "ci95": [ - -0.005102694924123447, - 0.028847543352601156 - ], - "samples": 2000, - "groups": 23, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": 0.0806574651079075, - "ci95": [ - -0.016061369620481, - 0.1993407432639239 - ], - "samples": 2000, - "groups": 23, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "brier": { - "micro_brier_delta": 0.04629405344979143, - "ci95": [ - -0.005474368002895286, - 0.10610974848164327 - ], - "samples": 2000, - "groups": 23, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - }, - "native": { - "acc": { - "micro_acc_delta": 0.0, - "ci95": [ - 0.0, - 0.0 - ], - "samples": 2000, - "groups": 4, - "aggregation": "micro", - "unit": "Whole capture group across source families", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": -0.004541798865699442, - "ci95": [ - -0.007275272740407906, - -0.003258894034076542 - ], - "samples": 2000, + "sev_sysmon_native_other_effect": { + "questions": 88, "groups": 4, - "aggregation": "micro", - "unit": "Whole capture group across source families", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" + "raw": { + "n": 88, + "nll": 0.005466714567741097, + "acc": 1.0, + "ece": 0.005048372154544871, + "brier": 0.0014126265728775101, + "mean_conf": 0.994951627845455, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9886363636363636, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.005048372154545011, + "top_bins": { + "0.9": { + "n": 87, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 86, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 83, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9997677646417501 + }, + "0.8": { + "coverage": 0.8068181818181818, + "accuracy": 1.0, + "confidence_cutoff": 0.9978176224669177 + } + } + }, + "calibrated": { + "n": 88, + "nll": 0.02789592785139411, + "acc": 1.0, + "ece": 0.026282565295571752, + "brier": 0.004783993285232044, + "mean_conf": 0.9737174347044284, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9545454545454546, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.026282565295571603, + "top_bins": { + "0.9": { + "n": 84, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 75, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 46, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9913758403771589 + }, + "0.8": { + "coverage": 0.8068181818181818, + "accuracy": 1.0, + "confidence_cutoff": 0.9627441989296704 + } + } + } }, - "brier": { - "micro_brier_delta": -0.0006570378773440486, - "ci95": [ - -0.0018239016582102339, - -0.00012501403854909632 - ], - "samples": 2000, + "sev_sysmon_native_process": { + "questions": 74, "groups": 4, - "aggregation": "micro", - "unit": "Whole capture group across source families", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - }, - "guide_detector": { - "acc": { - "micro_acc_delta": 0.014097744360902276, - "ci95": [ - -0.010733268142241958, - 0.04307115301628897 - ], - "samples": 2000, - "groups": 114, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": 0.010318565978466143, - "ci95": [ - -0.02562492366637801, - 0.044691969172975764 - ], - "samples": 2000, - "groups": 114, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "brier": { - "micro_brier_delta": 0.005962989482191983, - "ci95": [ - -0.01664672866743052, - 0.028271086382067975 - ], - "samples": 2000, - "groups": 114, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - }, - "swarm_static": { - "acc": { - "micro_acc_delta": 0.05932203389830504, - "ci95": [ - 0.034480288509040564, - 0.08547197640117996 - ], - "samples": 2000, - "groups": 176, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": -0.17308792668740217, - "ci95": [ - -0.23906732648623316, - -0.11508206848561314 - ], - "samples": 2000, - "groups": 176, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "brier": { - "micro_brier_delta": -0.08362752812460522, - "ci95": [ - -0.11395797861253039, - -0.05612652240109466 - ], - "samples": 2000, - "groups": 176, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - }, - "manual": { - "acc": { - "micro_acc_delta": 0.0843373493975903, - "ci95": [ - 0.034628911857593994, - 0.14634146341463417 - ], - "samples": 2000, - "groups": 40, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": -0.15959679081720557, - "ci95": [ - -0.29781254216292613, - -0.037512233422527574 - ], - "samples": 2000, - "groups": 40, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "brier": { - "micro_brier_delta": -0.12862065982564036, - "ci95": [ - -0.20555506872375667, - -0.0596051281217685 - ], - "samples": 2000, - "groups": 40, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - }, - "sysmon": { - "acc": { - "micro_acc_delta": 0.0, - "ci95": [ - 0.0, - 0.0 - ], - "samples": 2000, - "groups": 8, - "aggregation": "micro", - "unit": "Whole capture group across source families", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": -0.0014288934300975553, - "ci95": [ - -0.011110020952993583, - 0.013753809583785515 - ], - "samples": 2000, - "groups": 8, - "aggregation": "micro", - "unit": "Whole capture group across source families", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "brier": { - "micro_brier_delta": 0.0007729154617278078, - "ci95": [ - -0.002342110229997109, - 0.005453425939115126 - ], - "samples": 2000, - "groups": 8, - "aggregation": "micro", - "unit": "Whole capture group across source families", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - }, - "general": { - "acc": { - "micro_acc_delta": 0.0, - "ci95": [ - 0.0, - 0.0 - ], - "samples": 2000, - "groups": 64, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "nll": { - "micro_nll_delta": 0.05578705356138791, - "ci95": [ - 0.02406632980677795, - 0.08982514373984912 - ], - "samples": 2000, - "groups": 64, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - }, - "brier": { - "micro_brier_delta": 0.015313433968751122, - "ci95": [ - 0.004495110348257341, - 0.02583795985424143 - ], - "samples": 2000, - "groups": 64, - "aggregation": "micro", - "unit": "Whole source/group; all original variants included", - "method": "paired cluster percentile bootstrap; full statistic recomputed in each resample" - } - } - }, - "paired_guide_decisions": { - "guide": { - "samples": 2000, - "seed": 20260927, - "groups": 23, - "unit": "Whole organization, sampled with replacement; all incidents remain together.", - "method": "Paired percentile bootstrap; fixed three-class macro-F1, absent-class F1 zero. TP recall excludes draws with no true positives.", - "statistics": { - "macro_f1": { - "delta": 0.013536133799551708, - "ci95": [ - -0.016836824631839598, - 0.05791991746908926 - ], - "valid_samples": 2000 - }, - "true_positive_recall": { - "delta": -0.02702702702702703, - "ci95": [ - -0.047619047619047616, - 0.0 - ], - "valid_samples": 2000 + "raw": { + "n": 74, + "nll": 0.004798899209615666, + "acc": 1.0, + "ece": 0.004644321057477101, + "brier": 0.0005716684001551928, + "mean_conf": 0.995355678942523, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9864864864864865, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.00464432105747703, + "top_bins": { + "0.9": { + "n": 73, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 72, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 68, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9999268900168323 + }, + "0.8": { + "coverage": 0.8108108108108109, + "accuracy": 1.0, + "confidence_cutoff": 0.9973453419434639 + } + } }, - "equal_organization_accuracy": { - "delta": 0.006521739130434745, - "ci95": [ - -0.0021739130434783593, - 0.015217391304347794 - ], - "valid_samples": 2000 + "calibrated": { + "n": 74, + "nll": 0.023528453108139967, + "acc": 1.0, + "ece": 0.022317812691888844, + "brier": 0.004027925361032504, + "mean_conf": 0.9776821873081113, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9324324324324325, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 1.0, + "aurc": 0.0, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.022317812691888705, + "top_bins": { + "0.9": { + "n": 69, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 65, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 45, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9964760714345572 + }, + "0.8": { + "coverage": 0.8108108108108109, + "accuracy": 1.0, + "confidence_cutoff": 0.9703686880072409 + } + } } - } - }, - "guide_detector": { - "samples": 2000, - "seed": 20260927, - "groups": 114, - "unit": "Whole organization, sampled with replacement; all incidents remain together.", - "method": "Paired percentile bootstrap; fixed three-class macro-F1, absent-class F1 zero. TP recall excludes draws with no true positives.", - "statistics": { - "macro_f1": { - "delta": 0.0266170267250998, - "ci95": [ - -0.005222855290995668, - 0.06622974286948383 - ], - "valid_samples": 2000 - }, - "true_positive_recall": { - "delta": 0.10035842293906805, - "ci95": [ - 0.033210332103321055, - 0.194811225925777 - ], - "valid_samples": 2000 + }, + "sev_sysmon_native_registry": { + "questions": 66, + "groups": 4, + "raw": { + "n": 66, + "nll": 0.024015813931122083, + "acc": 0.9848484848484849, + "ece": 0.016913465896023366, + "brier": 0.01690246747159733, + "mean_conf": 0.9866786138448916, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9696969696969697, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9696969696969697, + "aurc": 0.00046266864448682633, + "error_rate_at_0_9": 0.0, + "confidence_bias": 0.0018301289964067768, + "top_bins": { + "0.9": { + "n": 64, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 64, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 62, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9998656142719909 + }, + "0.8": { + "coverage": 0.803030303030303, + "accuracy": 1.0, + "confidence_cutoff": 0.9982153954384291 + } + } }, - "equal_organization_accuracy": { - "delta": 0.024447391002711205, - "ci95": [ - -0.0024186565385442365, - 0.05694784972373788 - ], - "valid_samples": 2000 + "calibrated": { + "n": 66, + "nll": 0.03582391955947498, + "acc": 0.9848484848484849, + "ece": 0.01540577138447916, + "brier": 0.016990289728223678, + "mean_conf": 0.9730644954445979, + "confident_error_rate": 0.0, + "coverage_at_0_9": 0.9696969696969697, + "accuracy_at_0_9": 1.0, + "coverage_at_5pct_error": 1.0, + "coverage_at_1pct_error": 0.9696969696969697, + "aurc": 0.00046266864448682633, + "error_rate_at_0_9": 0.0, + "confidence_bias": -0.011783989403886919, + "top_bins": { + "0.9": { + "n": 64, + "errors": 0, + "error_rate": 0.0 + }, + "0.95": { + "n": 59, + "errors": 0, + "error_rate": 0.0 + }, + "0.99": { + "n": 40, + "errors": 0, + "error_rate": 0.0 + } + }, + "selective": { + "0.5": { + "coverage": 0.5, + "accuracy": 1.0, + "confidence_cutoff": 0.9940913990982823 + }, + "0.8": { + "coverage": 0.803030303030303, + "accuracy": 1.0, + "confidence_cutoff": 0.975504713794386 + } + } } } } }, - "gates": { - "passed": false, - "checks": { - "manual_correct_gain": { - "actual": 7, - "bound": 3, - "operator": ">=", - "passed": true - }, - "native_correct": { - "actual": 409, - "bound": 398, - "operator": ">=", - "passed": true - }, - "excytin_correct": { - "actual": 398, - "bound": 391, - "operator": ">=", - "passed": true - }, - "sysmon_correct": { - "actual": 62, - "bound": 60, - "operator": ">=", - "passed": true - }, - "general_correct": { - "actual": 86, - "bound": 85, - "operator": ">=", - "passed": true - }, - "manual_raw_nll": { - "actual": 1.1015953331868638, - "bound": 1.3111921240040694, - "operator": "<=", - "passed": true - }, - "manual_raw_brier": { - "actual": 0.38672041228910636, - "bound": 0.5353410721147468, - "operator": "<=", - "passed": true - }, - "manual_global_calibrated_nll": { - "actual": 0.8197108651023933, - "bound": 1.0802111739617255, - "operator": "<=", - "passed": true - }, - "manual_global_calibrated_brier": { - "actual": 0.3693842986364894, - "bound": 0.5033988288027265, - "operator": "<=", - "passed": true - }, - "native_raw_nll": { - "actual": 0.008457485545022507, - "bound": 0.06299928441072196, - "operator": "<=", - "passed": true - }, - "native_raw_brier": { - "actual": 0.0035344210430228854, - "bound": 0.024191458920366934, - "operator": "<=", - "passed": true - }, - "native_global_calibrated_nll": { - "actual": 0.028769043116345983, - "bound": 0.08020084130636222, - "operator": "<=", - "passed": true - }, - "native_global_calibrated_brier": { - "actual": 0.006284804103393485, - "bound": 0.02669768749600822, - "operator": "<=", - "passed": true - }, - "excytin_raw_nll": { - "actual": 0.0009678808724190956, - "bound": 0.052029807853842494, - "operator": "<=", - "passed": true - }, - "excytin_raw_brier": { - "actual": 4.8256567178328636e-05, - "bound": 0.020132675676160554, - "operator": "<=", - "passed": true - }, - "excytin_global_calibrated_nll": { - "actual": 0.009470865385844369, - "bound": 0.05820502390654013, - "operator": "<=", - "passed": true - }, - "excytin_global_calibrated_brier": { - "actual": 0.0006565384935901168, - "bound": 0.02062640684122579, - "operator": "<=", - "passed": true - }, - "sysmon_raw_nll": { - "actual": 0.08322377416424186, - "bound": 0.13465266759433944, - "operator": "<=", - "passed": true - }, - "sysmon_raw_brier": { - "actual": 0.045695299016738855, - "bound": 0.06492238355501105, - "operator": "<=", - "passed": true - }, - "sysmon_global_calibrated_nll": { - "actual": 0.11024750472054996, - "bound": 0.15719175042876138, - "operator": "<=", - "passed": true - }, - "sysmon_global_calibrated_brier": { - "actual": 0.04954806836760281, - "bound": 0.06844225884031123, - "operator": "<=", - "passed": true - }, - "general_raw_nll": { - "actual": 0.5574328669578856, - "bound": 0.5516458133964977, - "operator": "<=", - "passed": false - }, - "general_raw_brier": { - "actual": 0.29243113448917374, - "bound": 0.29711770052042263, - "operator": "<=", - "passed": true - }, - "general_global_calibrated_nll": { - "actual": 0.42292499037031256, - "bound": 0.4765669988063534, - "operator": "<=", - "passed": true - }, - "general_global_calibrated_brier": { - "actual": 0.24649933996243034, - "bound": 0.26785251439683744, - "operator": "<=", - "passed": true - }, - "guide_raw_nll": { - "actual": 0.8563265185203435, - "bound": 0.8256690534124361, - "operator": "<=", - "passed": false - }, - "guide_raw_brier": { - "actual": 0.5255142828500304, - "bound": 0.499220229400239, - "operator": "<=", - "passed": false - }, - "guide_global_calibrated_nll": { - "actual": 0.7793526965750157, - "bound": 0.8008521265465799, - "operator": "<=", - "passed": true - }, - "guide_global_calibrated_brier": { - "actual": 0.4718856823661165, - "bound": 0.4766593103734014, - "operator": "<=", - "passed": true - }, - "guide_detector_raw_nll": { - "actual": 0.8276776090882748, - "bound": 0.8673590431098086, - "operator": "<=", - "passed": true - }, - "guide_detector_raw_brier": { - "actual": 0.48019952548397726, - "bound": 0.49423653600178535, - "operator": "<=", - "passed": true - }, - "guide_detector_global_calibrated_nll": { - "actual": 0.785071927880167, - "bound": 0.8476013493654482, - "operator": "<=", - "passed": true - }, - "guide_detector_global_calibrated_brier": { - "actual": 0.46075730837789786, - "bound": 0.4889190783897773, - "operator": "<=", - "passed": true - }, - "guide_macro_f1": { - "actual": 0.4613423970886119, - "bound": 0.4278062632890602, - "operator": ">=", - "passed": true - }, - "guide_true_positive_correct": { - "actual": 7, - "bound": 6, - "operator": ">=", - "passed": true - }, - "guide_detector_macro_f1": { - "actual": 0.5681608349992006, - "bound": 0.5215438082741007, - "operator": ">=", - "passed": true - }, - "guide_detector_true_positive_correct": { - "actual": 101, - "bound": 71, - "operator": ">=", - "passed": true - } - }, - "interpretation": "Fixed descriptive engineering gates for a research continuation; not statistical noninferiority or a field detection claim." + "policy": { + "positive": "Requires approval under the supplied policy", + "negative": "Permitted under the supplied policy", + "threshold_split": "calibration", + "temperature": "reuse saved calibration receipt, or benchmark inference temperature; no refit", + "population": "all five conditional policy questions per admitted source program", + "bootstrap": "percentile intervals conditional on the fixed calibration threshold; no threshold reselection" }, - "protocol_sha256": "acc38c3ec612344b8bf732f8bacf60f3ebd58815d735982ad25c748be3480f6a", - "scorer_sha256": "967b5bf646812d95167d3d88e1b4df7405da6d74983f8aceefb283512bc29285", - "protocol": { - "version": 3, - "registered_date": "2026-09-27", - "study": "sev-r2-swarmtraces-4b-retention-v1", - "suite": "evals/sev/swarmtraces-curriculum-v1", - "plan": "experiments/sev-swarmtraces-4b-retention-v1.json", - "transfer": "evals/sev/swarmtraces-transfer-v1", - "original_transfer": "evals/sev/security-transfer-v1", - "checkpoints": { - "parent": "/runs/sev-r2-curriculum-repair-4b-v1/02-trial-2/checkpoint", - "candidate": "/runs/sev-r2-swarmtraces-4b-retention-v1/00-trial-0/checkpoint" - }, - "parent_checkpoint_hashes": { - "head_sha256": "5a91ed82f11c6d0c8ab0a910e0c85770119e0034039379aa601d17a3ec7664d6", - "adapter_sha256": "0452010abdcc58f7c39a47a1562a1c812bf85f025d4a6c105c93d486051765de" - }, - "panels": { - "excytin": { - "suite": "evals/sev/excytin-native-v1" - }, - "guide": { - "suite": "evals/sev/guide-v1" - }, - "native": { - "suite": "evals/sev/sysmon-native-v2" - }, - "guide_detector": { - "suite": "evals/sev/guide-detector-v1" - }, - "swarm_static": { - "suite": "evals/sev/swarmtraces-static-v1" - }, - "manual": { - "suite": "evals/sev/swarmtraces-manual-v1" - } - }, - "gates": { - "manual_min_correct_gain": 3, - "minimum_correct": { - "native": 398, - "excytin": 391, - "sysmon": 60, - "general": 85 - }, - "probability_retention_domains": [ - "manual", - "native", - "excytin", - "sysmon", - "general", - "guide", - "guide_detector" - ], - "nll_margin": 0.05, - "brier_margin": 0.02, - "guide_macro_f1_margin": 0.02, - "guide_tp_count_margin": 2 - }, - "calibration": "One micro-NLL temperature for each model on identical union calibration; unchanged on all evaluation panels. Raw and calibrated metrics reported separately.", - "primary": "Independent manually annotated complete-program panel. Development screen, not final field validation.", - "secondary": "Parser-derived held-out programs, grouped by exact text, source reconstruction and whole-program AST templates.", - "interpretation": "Continuation adoption criteria only. Prior GUIDE shortcomings remain separate; this experiment does not establish a general release. No final test is used.", - "bootstrap": { - "samples": 2000, - "seed": 20260927 - }, - "timeout_seconds": 3600, - "maximum_reservation_usd": "5.68", - "input_sha256": { - "docs/reviews/sev-swarmtraces-curriculum-review.md": "5ba6f9682dbfe1782309c934e05c487825731c544ce23f293d131927b059c309", - "docs/reviews/sev-swarmtraces-scorer-v1.md": "d63c433da881fd9a03a2b3d065533dd19f68f0f905bbe7c3aa042d0ccfb63447", - "docs/sev-swarmtraces-round-v1.md": "bdddd3fe1c8f4776710bff45e1f257c66e48f2bb198ca641aaa51f0db053bf35", - "evals/sev/curriculum-repair-v1/manifest.json": "5ae5dbfd48e9a2717397c0197ed8db016db87d69abc30d51813878ef734beba3", - "evals/sev/excytin-native-v1/manifest.json": "1a537e295dce0df278b7648160674bdabcec5d1e6d8adc80c203c1dae9dcaf5c", - "evals/sev/guide-detector-v1/manifest.json": "507cefca384f2a88aa7172f54757dc06bfd6f815e9501c522fcbf071f05be714", - "evals/sev/guide-v1/manifest.json": "50245db125492792319f4fc477ed5b356819f0bf32567bfc059f5285ae58b069", - "evals/sev/security-transfer-v1/manifest.json": "6b2d546ac337946bfce4c85c7634b6f749f98bff0dc84e6c7f2317a4f6eddaea", - "evals/sev/swarmtraces-curriculum-v1/calibration.jsonl": "8504531b423590392325f9e1d5fd291280122e6ca038483268e05bfe94970f77", - "evals/sev/swarmtraces-curriculum-v1/development.jsonl": "69d4aec321f985901ab0ca3ecccf072c66d60479924111d663d5f71d7e877c84", - "evals/sev/swarmtraces-curriculum-v1/manifest.json": "61e3c1898b8a03eb178355d3c8c5050e85bd217285c700df9b4506ec9e3680f8", - "evals/sev/swarmtraces-curriculum-v1/train.jsonl": "d6faa54db510a9285d4c2c2bc1f0c9095108a42b71460ad7ce83bccfcf199b92", - "evals/sev/swarmtraces-manual-v1/annotations.json": "e5f3412722153a5de7e4ee6ee3aed9844f0c062a8afc028b396e65c794314206", - "evals/sev/swarmtraces-manual-v1/calibration.jsonl": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", - "evals/sev/swarmtraces-manual-v1/development.jsonl": "d0e5832305f1f04e82a16fe3f643a3518061817858736297b6cc3f7423ad4055", - "evals/sev/swarmtraces-manual-v1/manifest.json": "6c1f892e39bd722dea54042f2139514720021faf33698aba9eb7cd399b67324e", - "evals/sev/swarmtraces-manual-v1/protection.json": "abd1e0ebf71e0edd2d94ded564d52f1555667a3fbf6fb828f3e44e1d7feb173d", - "evals/sev/swarmtraces-manual-v1/train.jsonl": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", - "evals/sev/swarmtraces-manual-v1/validation.json": "b4302a4fa05444ac1714c91ad0daf4e63dc66e8e4065fbc52e78bb6d79de0fea", - "evals/sev/swarmtraces-static-v1/calibration.jsonl": "7754bac053cd7a2bfb2303d9e1a591381a2a25925f6f728d7995a6fbb343dc5d", - "evals/sev/swarmtraces-static-v1/development.jsonl": "1847a598a428615a416c2c11d5bce0f6a129bd926684cbbe3bd5ccd7b4ec1321", - "evals/sev/swarmtraces-static-v1/manifest.json": "1f304228dbbcad5b4aab5d9ce582738b8b7109214c1fdfd56c617fa5f1db583b", - "evals/sev/swarmtraces-static-v1/train.jsonl": "c09de8a213c5cd6b3506c9b3ccea60074115052617ce71307a24d99d067ef8bd", - "evals/sev/swarmtraces-transfer-v1/calibration.jsonl": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", - "evals/sev/swarmtraces-transfer-v1/development.jsonl": "e312086e75f6c3b8367be5b949e163f9d48a5634a1f304f555d9f063685c9190", - "evals/sev/swarmtraces-transfer-v1/manifest.json": "1caf2255850930113bc56b708d0881ed22d3df9ba1868e966f9454595c2cefa7", - "evals/sev/swarmtraces-transfer-v1/train.jsonl": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855", - "evals/sev/sysmon-native-v2/manifest.json": "e8c3b5dd306525e88e17b70f410892744bf99f2ed18fb14f42b910db52e558b1", - "experiments/sev-swarmtraces-4b-v1.json": "87e70bce1e76ccddd4a60866f3eb93e946a6f950de062cb2290db58e4ca1aee1", - "experiments/sev-swarmtraces-assembly-v1.json": "1915555716b0c993e1dfebc8928c1a8188265f1869db1884873f28c85d363851", - "runs/sev-curriculum-repair-v1/comparison.json": "623c2c4b2e348eab8a5975e48a94a4e9b0d2c98e7f7a43d6024a87779485b042", - "runs/sev-swarmtraces-review-20260927/draft4-admission.json": "ff1bcfbfb7448d4c60b4d86038ec5c7b6057c3e9e833a3b51d01ee66337cb318", - "runs/sev-swarmtraces-review-20260927/final-admission-receipt.json": "5f037732f74e09e0e46bbebbe95ad45146f89060d96ad9cbccfa59568ac5fad2", - "runs/sev-swarmtraces-review-20260927/panel-protection.json": "abd1e0ebf71e0edd2d94ded564d52f1555667a3fbf6fb828f3e44e1d7feb173d", - "runs/sev-swarmtraces-review-20260927/shared-fingerprints.jsonl": "452985f0efabc3b97a1f40a1e6a3eec57dd881ebeb448049f2de9120b491055b", - "scripts/assemble_sev_curriculum.py": "147b9f2a5133761d3a30acb08c90ba65cf9f4a278fed0e6310d39b5922eb551a", - "scripts/build_sev_swarmtraces.py": "08d60fcdfd9557633b5293e2b5e3b69758756f50ecedfa3e997ef310d396267c", - "scripts/integrate_sev_swarmtraces.py": "08ec5f630da2e8987eae7d6f93cd0cf7c674a0f528e3c2597f6886d01075a8ed", - "scripts/score_sev_curriculum_repair.py": "f9456677fbfc5f97106ce16c8bd31ab8818513778ab527e366a808ba655506b5", - "scripts/score_sev_guide.py": "1fc0d5e94af1fb456c571e48f312b8ff833dc2db53a5d53db1ba1f5b00cb8527", - "scripts/score_sev_mixed_public.py": "c2bbbef53c567065ee7c694f355b26af7b43696cf3244d89bac835dba4081192", - "scripts/score_sev_security.py": "83351d2299aa308611e14679e4bb77845e40519a8a01248bc651547f844f172b", - "scripts/score_sev_swarmtraces.py": "967b5bf646812d95167d3d88e1b4df7405da6d74983f8aceefb283512bc29285", - "scripts/swarmtraces_ast.mjs": "1c1035149778b411eda4ca95a5cf822afd6a0117064c6f6687130fc113ea81e3", - "modal_app.py": "639577258f4d581daefa51dad639961280ce92de9da256316ab4124f005a53fb", - "tests/test_modal_input_mount.py": "6fcc9ccc9da13f72646538e1342728b7e935bf7ebb156af704f0ad0f4843d3e3", - "runs/sev-swarmtraces-intake-20260927/failed-launch-v1.json": "6702c11219de850dba1723cbad354276c0db691bc5ba9125ce49ca1f67e5edfe", - "experiments/sev-swarmtraces-4b-retention-v1.json": "8dea92d4ab89f7aa09bcdae6c1299ea06a8875cfc09d657601b64331ac4377a4", - "docs/sev-swarmtraces-retention-v1.md": "39b57384b91a850420cc624d0081b2d1de7551539145e19af764854ed31b059c", - "experiments/sev-swarmtraces-protocol-v2.json": "62680e8f7965b981efe8d9281fc0b0dbd5b8f0759a871bad68e249365377538a", - "runs/sev-swarmtraces-v2/comparison.json": "df7a4433b1320e98f5a2cec59f792c2ac641aa74fa69a6e5306b7341f3219b57", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/provenance.json": "39898e5e6ef87112f653436a9b6b80555da10a01b0d203b3f5eb29000dddae3a", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/result.json": "46b3ee2a68f250ba1d433c25612083bb8235a03e617d90707fa796f86ae2cbf4", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/calibration/rows.json": "9a33b44bee9cb63a635ac1516bcab9f67fbe56b15a72995a8c2b54a321ad4dd6", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/calibration/report.json": "15dc07e79e0121e7345ece52b028220fbbf9123609a2eee5b331c9d372370383", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/development/rows.json": "d882c91290398ba96cf39de9128ea5a973cf8d79d20ca84f5249f9f1c1cd5300", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/development/report.json": "02fc22461d5b44f88173b90e0fffe97905b4eeb132eec79088c9a8396ce73134", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/transfer/rows.json": "5b624c0b0f91940332045523703dd5d52761a5ff931f2c9f1751185883976368", - "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint/transfer/report.json": "f44bc7be71a3b37ded9e35f6fa382c1907c70fc959c15bc712b76f024ac726cc", - "docs/reviews/sev-swarmtraces-results-v2.md": "bbfc48b93b09dafbdac5d44576d7c852fa07fcb021e587560109c5c1516313f8", - "runs/sev-swarmtraces-review-20260927/results-v2-final-receipt.json": "6d474088bbd9bbe725ea7b0fb17f4e82d31ec2f95bfa07b3a6994b7a93a5ce52", - "runs/sev-swarmtraces-review-20260927/results-v2-review.json": "dc5f2a90997f3d4c13fc65a63e1eb8939b8d9cae8dc278ff5c1707f8e588f397", - "runs/sev-swarmtraces-review-20260927/audit-results-v2.py": "5e9966bd8a51312e85a162e17d3ce71da87b2b9a7000641a407fa56ccb93be87" - }, - "retained_suite": "evals/sev/curriculum-repair-v1", - "admission_limitations": [ - "No earlier-response credential examples survive automatic admission.", - "Five execution-flow and eleven unavailable-context examples are training-only; automatic development does not test these classes.", - "The manual panel is a fixed development screen, not a locked final test or field detection evaluation." - ], - "maximum_total_reservations_including_failed_launch_usd": "28.40", - "preceding_development_screen": { - "protocol": "experiments/sev-swarmtraces-protocol-v2.json", - "protocol_sha256": "62680e8f7965b981efe8d9281fc0b0dbd5b8f0759a871bad68e249365377538a", - "result": "runs/sev-swarmtraces-v2/comparison.json", - "result_sha256": "df7a4433b1320e98f5a2cec59f792c2ac641aa74fa69a6e5306b7341f3219b57", - "decision": "Keep the same parent, data and gates. Test one continuation at half the learning rate after the older GUIDE panel failed raw NLL, raw Brier and true-positive retention checks. The manual panel is reused development data, not fresh confirmation." - }, - "reused_parent_evaluation": "runs/sev-r2-swarmtraces-4b-v2/00-checkpoint", - "expected_training_resources": { - "records_seen": 6438, - "optimizer_steps": 805, - "forward_tokens": 1447855, - "truncated_records": 0, - "rejected_records": 0 - } - } + "limitations": [ + "Authored permissions and static callback dataflow, not historical maliciousness or execution evidence.", + "The selected conditional panel does not estimate deployment prevalence.", + "Actor origin is not measured. No locked test is read and this report does not promote a model.", + "The calibration false-alert budget is empirical, not a deployment guarantee.", + "Rare-operation rates are null without the relevant class; bootstrap draws without that class are omitted." + ] }