results: pass untouched calibrated oral-B-v2 confirmation

1 files changed, 385 insertions, 0 deletions

diff --git a/results/bci_v2_calibrated_confirmation_gate.json b/results/bci_v2_calibrated_confirmation_gate.json
new file mode 100644
index 0000000..90e54e5
--- /dev/null
+++ b/results/bci_v2_calibrated_confirmation_gate.json
@@ -0,0 +1,385 @@
+{
+  "all_prior_failures_preserved": true,
+  "calibration_quantiles": [
+    0.2,
+    0.35,
+    0.5,
+    0.65,
+    0.8
+  ],
+  "checks": {
+    "all_records_finite_paired_and_cost_audited": true,
+    "innovation_and_network_prediction": {
+      "decoder_corr_lower_bound_nonnegative": true,
+      "mean_decoder_corr_at_least_0p05": true,
+      "mean_raw_residual_corr_gap_at_least_0p20": true,
+      "mean_residual_soma_corr_at_most_0p10": true,
+      "mean_surrounding_accuracy_at_least_0p52": true,
+      "mean_velocity_advantage_at_least_0p05": true,
+      "positive_sign_in_at_least_25_of_30": true,
+      "positive_sign_in_every_task_cluster": true,
+      "raw_residual_corr_gap_lower_bound_at_least_0p15": true,
+      "residual_soma_corr_upper_bound_at_most_0p12": true,
+      "surrounding_accuracy_lower_bound_at_least_0p50": true,
+      "velocity_advantage_lower_bound_nonnegative": true
+    },
+    "learning_and_plasticity": {
+      "every_role_cosine_at_least_0p70": true,
+      "every_task_mean_final_at_least_0p60": true,
+      "fixed_role_gap_lower_bound_at_least_0p10": true,
+      "intact_final_lower_bound_at_least_0p60": true,
+      "learning_gain_lower_bound_at_least_0p05": true,
+      "mean_fixed_role_gap_at_least_0p20": true,
+      "mean_intact_final_at_least_0p70": true,
+      "mean_learning_gain_at_least_0p10": true,
+      "mean_oracle_deficit_at_most_0p10": true,
+      "mean_role_cosine_at_least_0p80": true,
+      "oracle_deficit_upper_bound_at_most_0p20": true,
+      "plasticity_half_margin_lower_bound_nonnegative": true
+    },
+    "target_ladder_outcome_surprise": {
+      "critic_expectedness_lower_bound_at_least_0p02": true,
+      "every_challenge_fraction_between_0p10_and_0p90": true,
+      "every_critic_value_corr_at_least_0p95": true,
+      "mean_critic_expectedness_at_least_0p05": true,
+      "mean_outcome_lesion_drop_at_least_0p20": true,
+      "mean_terminal_accuracy_at_least_0p80": true,
+      "mean_terminal_separation_at_least_0p20": true,
+      "outcome_lesion_drop_lower_bound_at_least_0p15": true,
+      "terminal_accuracy_lower_bound_at_least_0p75": true,
+      "terminal_separation_lower_bound_at_least_0p15": true
+    }
+  },
+  "complete_grid": true,
+  "fixed_config": {
+    "critic_eta": 0.03,
+    "forward_eta": 0.1,
+    "gamma": 0.8,
+    "velocity_reward_scale": 1.0
+  },
+  "input_sha256": {
+    "base_dynamics": "d5a373314562af0daedf55baa04b975fe643236c7821be1560d2cfb4359688fc",
+    "common_runner": "4157a57806dedfc2eb70ec84f7a7ca1f73ab684e5d8f0b227cd80465e8e97aa5",
+    "confirmation_analyzer": "2536fa01f62c3b2d87e782b08075cb5948aa26bfdce71a1d8c2446d5b68bb34c",
+    "confirmation_runner": "5e9beba106b4b29c4ee1568052d6e8c15690dbde92917a2497a52e43e7f557d1",
+    "d4_gate": "636c587e47287338cdca7d9558dc3bfb99a2cec615badb23337025d85b8128ef",
+    "development_analyzer": "dde4a2e4d62f5f9da94357b388139c50251b6863f9c991f1a006e71eeb7953e4",
+    "development_runner": "446ad26fc015eb79d28e0bb442c8aef71bdd1d0320910494cac20f92e3c7dc18",
+    "failed_target_gate": "1df37c0b639542bbee3017da54306b0d572cbf96230428f142fafed139ac92f1",
+    "failed_v2_gate": "e53f46cf456d60ce4d33586ace4f4619fb1a31a58082b2b72943d4ed10cd25fd",
+    "old_r2_gate": "4f6f969ceae88afa2523e3472a3373522991f5ecaab69440830c07479d2d3597",
+    "protocol": "a175dc80fdb7218c010953d5d7e3c77fee0690d5f10fabaf9d700bb8505acde8",
+    "r1_gate": "907648edaea040b0d7e6fbf74f96b9c5ff8fc19c0dc1063b054829d5f2aa9fbc",
+    "recovery_confirmation_analyzer": "29a168143266bc7078e5229183bb7ea45283973592fecd1fcd715488f5d6242b",
+    "recovery_development_analyzer": "2768b877132ece6d612d2a5f1e483c495b277e26fdf54fb8a716aa8de6fd67e2",
+    "recovery_metrics": "371cb6a7e2192d25ddbfad69c0ce7a1150aa1c21ca98e141bedb2afc2acbdbaa",
+    "recovery_runner": "cea0ef658c698c376b1fda0cf2f7f910ee44f7bcde153f2fbaf18e66a2d3661a",
+    "runner": "5e9beba106b4b29c4ee1568052d6e8c15690dbde92917a2497a52e43e7f557d1",
+    "v2_dynamics": "786c5aed3091644272ae0a740913ac3cc38cdb02a8cacb6dfc8496e36836cdae",
+    "v2_metrics": "421da9287b92e62c742f9eeefbfdccba582c6a4ad5cf92ade4709689120fa1ee"
+  },
+  "metrics": {
+    "challenge_fraction": {
+      "by_task_seed": [
+        0.493125,
+        0.5209375,
+        0.514375,
+        0.4853125,
+        0.489375,
+        0.5025
+      ],
+      "mean": 0.5009375,
+      "one_sided_95pct_lower": 0.4891770555188329,
+      "one_sided_95pct_upper": 0.5126979444811671
+    },
+    "critic_expectedness": {
+      "by_task_seed": [
+        0.3133184602399429,
+        0.32414884320577275,
+        0.35313381879463884,
+        0.35924697778590636,
+        0.32930879064364416,
+        0.2331896407499471
+      ],
+      "mean": 0.3187244219033087,
+      "one_sided_95pct_lower": 0.2813635909438583,
+      "one_sided_95pct_upper": 0.35608525286275905
+    },
+    "critic_training_gap": {
+      "by_task_seed": [
+        0.19921875,
+        0.13984375,
+        0.19921875,
+        0.0,
+        0.0,
+        0.4
+      ],
+      "mean": 0.15638020833333333,
+      "one_sided_95pct_lower": 0.03314670866020128,
+      "one_sided_95pct_upper": 0.27961370800646534
+    },
+    "critic_value_corr": {
+      "by_task_seed": [
+        0.9999999999993657,
+        0.9999999999992956,
+        0.9999999999990888,
+        0.9999999999994579,
+        0.9999999999993324,
+        0.9999999999993532
+      ],
+      "mean": 0.9999999999993157,
+      "one_sided_95pct_lower": 0.9999999999992141,
+      "one_sided_95pct_upper": 0.9999999999994172
+    },
+    "decoder_corr": {
+      "by_task_seed": [
+        0.12762928350655744,
+        0.11863155855659262,
+        0.11915750059779097,
+        0.11494479466762646,
+        0.11894289331265234,
+        0.1260886647366911
+      ],
+      "mean": 0.1208991158963185,
+      "one_sided_95pct_lower": 0.11687435649320946,
+      "one_sided_95pct_upper": 0.12492387529942753
+    },
+    "fixed_role_gap": {
+      "by_task_seed": [
+        1.0,
+        1.0,
+        1.0,
+        0.99921875,
+        0.99921875,
+        1.0
+      ],
+      "mean": 0.9997395833333333,
+      "one_sided_95pct_lower": 0.9994077009138491,
+      "one_sided_95pct_upper": 1.0000714657528176
+    },
+    "intact_final": {
+      "by_task_seed": [
+        1.0,
+        1.0,
+        1.0,
+        1.0,
+        1.0,
+        1.0
+      ],
+      "mean": 1.0,
+      "one_sided_95pct_lower": 1.0,
+      "one_sided_95pct_upper": 1.0
+    },
+    "intact_gain": {
+      "by_task_seed": [
+        0.9885416666666667,
+        0.9958333333333333,
+        0.990625,
+        0.9927083333333333,
+        0.965625,
+        0.9947916666666666
+      ],
+      "mean": 0.9880208333333333,
+      "one_sided_95pct_lower": 0.978732077409237,
+      "one_sided_95pct_upper": 0.9973095892574297
+    },
+    "oracle_deficit": {
+      "by_task_seed": [
+        0.0,
+        0.0,
+        0.0,
+        0.0,
+        0.0,
+        0.0
+      ],
+      "mean": 0.0,
+      "one_sided_95pct_lower": 0.0,
+      "one_sided_95pct_upper": 0.0
+    },
+    "outcome_lesion_drop": {
+      "by_task_seed": [
+        0.39257537041898616,
+        0.3907302442715473,
+        0.3942709986960553,
+        0.392041901233741,
+        0.4265292127084088,
+        0.4039412037872511
+      ],
+      "mean": 0.40001482185266496,
+      "one_sided_95pct_lower": 0.3886365638440749,
+      "one_sided_95pct_upper": 0.411393079861255
+    },
+    "outcome_training_gap": {
+      "by_task_seed": [
+        0.00625,
+        0.0140625,
+        0.00703125,
+        0.0203125,
+        0.0109375,
+        0.00390625
+      ],
+      "mean": 0.010416666666666666,
+      "one_sided_95pct_lower": 0.005443964824466358,
+      "one_sided_95pct_upper": 0.015389368508866973
+    },
+    "plasticity_half_margin": {
+      "by_task_seed": [
+        0.4942708333333333,
+        0.49895833333333334,
+        0.4953125,
+        0.49635416666666665,
+        0.4828125,
+        0.4973958333333333
+      ],
+      "mean": 0.49418402777777776,
+      "one_sided_95pct_lower": 0.48940971207433226,
+      "one_sided_95pct_upper": 0.49895834348122325
+    },
+    "raw_residual_corr_gap": {
+      "by_task_seed": [
+        0.9343675823071695,
+        0.9346327542596294,
+        0.9351443461556045,
+        0.9371585744517263,
+        0.9368228292363849,
+        0.936621880026107
+      ],
+      "mean": 0.935791327739437,
+      "one_sided_95pct_lower": 0.9347897922928049,
+      "one_sided_95pct_upper": 0.936792863186069
+    },
+    "residual_soma_corr": {
+      "by_task_seed": [
+        0.06455947485925856,
+        0.06426190360381064,
+        0.06372971991387162,
+        0.06172674968637641,
+        0.061977758036086475,
+        0.06227044605503726
+      ],
+      "mean": 0.06308767535907349,
+      "one_sided_95pct_lower": 0.062066201125334404,
+      "one_sided_95pct_upper": 0.06410914959281258
+    },
+    "role_cosine": {
+      "by_task_seed": [
+        0.9775842428207397,
+        0.984451150894165,
+        0.9840954661369323,
+        0.9847705602645874,
+        0.9766013145446777,
+        0.9492486834526062
+      ],
+      "mean": 0.9761252363522848,
+      "one_sided_95pct_lower": 0.9648921083365222,
+      "one_sided_95pct_upper": 0.9873583643680474
+    },
+    "sign_inversion": {
+      "by_task_seed": [
+        0.0417874453823463,
+        0.04286598907057731,
+        0.04230713275613263,
+        0.04317331085758009,
+        0.04380628295437846,
+        0.042331079502170854
+      ],
+      "mean": 0.04271187342053094,
+      "one_sided_95pct_lower": 0.042118910888504044,
+      "one_sided_95pct_upper": 0.04330483595255784
+    },
+    "surrounding_accuracy": {
+      "by_task_seed": [
+        0.5426530288446466,
+        0.5446535266803437,
+        0.5430696812667446,
+        0.5439424567783643,
+        0.5420650374573571,
+        0.5457265455649936
+      ],
+      "mean": 0.5436850460987417,
+      "one_sided_95pct_lower": 0.5425667337998379,
+      "one_sided_95pct_upper": 0.5448033583976455
+    },
+    "terminal_accuracy": {
+      "by_task_seed": [
+        0.9987403100775194,
+        0.998997277676951,
+        0.9996865203761756,
+        1.0,
+        0.9975757575757576,
+        0.9949594968259532
+      ],
+      "mean": 0.9983265604220595,
+      "one_sided_95pct_lower": 0.9968020433207192,
+      "one_sided_95pct_upper": 0.9998510775233997
+    },
+    "terminal_separation": {
+      "by_task_seed": [
+        0.4020353802157061,
+        0.3913365663582604,
+        0.39170622720181636,
+        0.38793115544193,
+        0.43856566447380196,
+        0.4586232591265383
+      ],
+      "mean": 0.4116997088030089,
+      "one_sided_95pct_lower": 0.38730641450983544,
+      "one_sided_95pct_upper": 0.4360930030961823
+    },
+    "velocity_advantage": {
+      "by_task_seed": [
+        0.6370962829746177,
+        0.6361040647361883,
+        0.6326329660812359,
+        0.644287749154193,
+        0.6409354970810485,
+        0.639110211307452
+      ],
+      "mean": 0.6383611285557892,
+      "one_sided_95pct_lower": 0.6350333620000043,
+      "one_sided_95pct_upper": 0.6416888951115741
+    }
+  },
+  "new_oral_a_v2_protocol_may_be_frozen": true,
+  "old_oral_a_gate_remains_closed": true,
+  "oral_b_v2_outcome_surprise_established": true,
+  "positive_sign_count": 30,
+  "protocol": "oral_b_v2_calibrated_recovery_confirmation_v1",
+  "review_score_after": 8,
+  "review_score_before": 7,
+  "score_change_rule": "only a complete untouched calibrated R2 pass establishes role-vectorized TD outcome surprise; prior failures and the old oral-A gate remain unchanged",
+  "source_commit": "70e180c5ef5f78679f2e163ed3ee30873ee523bf",
+  "source_sha256": {
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t30_m0.json": "2909c54021e9d451ef54ae5de6b521478119b9bffaca06f58f711b24431c4d06",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t30_m1.json": "775b835a8d48f6d8dea2cb0078446552fb7cfab2d3e437aa9631c311cd6f40e2",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t30_m2.json": "0f95bdeb1a156ef4a97a59748453dfa1f872b490e68e13df926e37057a42eb36",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t30_m3.json": "fec845763b8402c8abf021000041d3b0790b5b50979ef381c1e2b016a9c5fee1",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t30_m4.json": "110fa43b9b0fcbcef89ed8a2332e871777c116690024ed2fa4bedea155fbb246",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t31_m0.json": "1f1c1fa49dd56743081d6420193b56ecba5f11cb7b2a40300f26e666bc4fb821",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t31_m1.json": "4efde69c884057c73d68a712625f5d940fb152e0ad7f5e301ffbecf9c9b8379a",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t31_m2.json": "10e16869c8b23007b840bec177378abe18c26f3fbabaa9dc09d112d4655b140e",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t31_m3.json": "a8be20d34915f95f7c278c4a4a3c2eaaf775d280b42614dfad8256cd877fe789",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t31_m4.json": "38e58907ad43ab06361a2946935855ea9d059a0ffd51d3cbd0ad596e1b923271",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t32_m0.json": "0254fc9d1945d11c5f178b9267ecbf8fe7f7ac2af8099b84095b4e931e1bb719",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t32_m1.json": "359c7a46d1f3bdde2e9394b5f7dceea01511b4a1714688976be3d69b4312a370",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t32_m2.json": "b2e1228d511fffd9753aacfd2b8f534652612d348900d44584f3c8f7e9ce7c05",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t32_m3.json": "e34d2a5e285184c6e60d088b67d8658d6dc78c62b2fc42c06acdb78f55bd6803",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t32_m4.json": "033dd12ed1e2efacc07ae14297bb0cc8fc74dc9114cd57431de9794bee7c7695",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t33_m0.json": "fba80b70c40169efcb60144e8315b5837184463601c86f3e3bda21361b3318ed",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t33_m1.json": "be952f8068e07612ef469dab0c495788d5b5b9dbc1c198b84ef5f9a17be3f62e",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t33_m2.json": "1d144e27aff463aacbc2e418bb5567016690477ba6f1e9e0e84f241cc8f24105",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t33_m3.json": "f81a7942503b1768c5819aff76e4525c1f0425c70b5a9f1ba397a30b0d5b31f2",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t33_m4.json": "42cc8ca59a3239c351bd3436bd8662e08c706f0861fce07266958e8e55d6dcd4",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t34_m0.json": "b798f59f61ff25c1f6d8e06b35727492a901ee87d8798700d2d7ff111cff0ecb",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t34_m1.json": "f5d4d84beeefd5c2e46187ccb61ef4432b0193c4c7f053c40252740a939ee651",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t34_m2.json": "7b8a5e144f71c5c6476be92612b8924dcdf1ba1d8f3083ce80618099b756adc1",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t34_m3.json": "c75f0c42a28ee1b764a7181e5fb7f449fecf84a3a63f69bf0663ed29f4dd1846",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t34_m4.json": "eb3de9a7edb8d3283a8762c1202310775d634af9a8ea5a25f90f314412f8f615",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t35_m0.json": "c026bcf06003c4129fe2d444f30331187b872ea2d28d55e41e5752cf2e3acb15",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t35_m1.json": "42452e471e10d9b17916f906b6b205632aa1f1de8335b1ffe962e855c53ee4e1",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t35_m2.json": "37ecff67bc1f7899b95f81c5587eb826602c1053b6777d76076bb891742161df",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t35_m3.json": "e53a422eff2bb2961ea396f071ea040a71d921e348a35bcb2d7e54d0952b3c23",
+    "results/bci_v2_calibrated_confirmation/bci_v2_calibrated_confirm_t35_m4.json": "4d9d35cb6d9bc30d240c85a0a5887c7d983e81acdb4a94b613bc633a11ea3178"
+  },
+  "status": "passed"
+}