{ "evaluation_metrics": { "failed_count": 0, "non_social": { "random_chaser": { "average_distance": { "max": 5.101284503936768, "mean": 4.99551248550415, "min": 4.891185760498047, "n": 10, "std": 0.07501372511139348 }, "chaser_new_fields": { "max": 34.13999938964844, "mean": 33.840998840332034, "min": 33.540000915527344, "n": 10, "std": 0.1902341662964249 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": -3.2359988689422607, "mean": -3.2923987865448, "min": -3.3599987030029297, "n": 10, "std": 0.03739303227587511 }, "collisions_per_episode": { "max": 3.319999933242798, "mean": 2.871999907493591, "min": 2.3499999046325684, "n": 10, "std": 0.31625304644177166 }, "degenerate_fraction": { "max": 0.92, "mean": 0.882, "min": 0.85, "n": 10, "std": 0.019390719429665332 }, "explorer_new_fields": { "max": 84.29999542236328, "mean": 74.87399826049804, "min": 66.5, "n": 10, "std": 6.064381072358134 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 75.625, "mean": 61.53249931335449, "min": 48.89500045776367, "n": 10, "std": 9.081414540792789 }, "final_distance": { "max": 5.32371187210083, "mean": 4.861768817901611, "min": 4.595163345336914, "n": 10, "std": 0.20951748305726414 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 5.258778095245361, "mean": 5.19075198173523, "min": 5.1385345458984375, "n": 10, "std": 0.038175016619519275 }, "chaser_new_fields": { "max": 87.00999450683594, "mean": 82.56399841308594, "min": 80.3499984741211, "n": 10, "std": 1.9011097381092157 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 7.177995681762695, "mean": 6.31579623222351, "min": 5.91399621963501, "n": 10, "std": 0.3661105605352478 }, "collisions_per_episode": { "max": 4.190000057220459, "mean": 3.45099995136261, "min": 3.0399999618530273, "n": 10, "std": 0.3247907454733161 }, "degenerate_fraction": { "max": 0.86, "mean": 0.845, "min": 0.82, "n": 10, "std": 0.011180339887498959 }, "explorer_new_fields": { "max": 33.38999938964844, "mean": 32.96999931335449, "min": 32.43000030517578, "n": 10, "std": 0.24556023828860254 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": -0.8299999833106995, "mean": -1.2994999885559082, "min": -2.1050000190734863, "n": 10, "std": 0.33420391387111775 }, "final_distance": { "max": 5.3512139320373535, "mean": 5.005114603042602, "min": 4.763849258422852, "n": 10, "std": 0.18805559273305048 }, "n": 10 } }, "social": { "random_chaser": { "average_distance": { "max": 5.963335037231445, "mean": 5.583923768997193, "min": 5.301150798797607, "n": 10, "std": 0.19349105457542226 }, "chaser_new_fields": { "max": 34.06999969482422, "mean": 33.93899917602539, "min": 33.709999084472656, "n": 10, "std": 0.09710312543039894 }, "chaser_partner_vision": { "max": 0.2995999753475189, "mean": 0.2658799886703491, "min": 0.2022000104188919, "n": 10, "std": 0.031091657013554774 }, "chaser_return": { "max": -2.223999261856079, "mean": -2.6053990840911867, "min": -2.898998975753784, "n": 10, "std": 0.20821288203553742 }, "collisions_per_episode": { "max": 0.9399999976158142, "mean": 0.5719999879598617, "min": 0.28999999165534973, "n": 10, "std": 0.19213536629169936 }, "degenerate_fraction": { "max": 1.0, "mean": 0.925, "min": 0.88, "n": 10, "std": 0.03930648801406708 }, "explorer_new_fields": { "max": 55.22999954223633, "mean": 45.23899908065796, "min": 14.960000038146973, "n": 10, "std": 12.602143500604898 }, "explorer_partner_vision": { "max": 0.2995999753475189, "mean": 0.2658799886703491, "min": 0.2022000104188919, "n": 10, "std": 0.031091657013554774 }, "explorer_return": { "max": 32.279998779296875, "mean": 17.3399995803833, "min": -27.864999771118164, "n": 10, "std": 18.824057283505237 }, "final_distance": { "max": 5.957443714141846, "mean": 5.436633443832397, "min": 5.081935882568359, "n": 10, "std": 0.2424814082208821 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 5.011444568634033, "mean": 3.77721266746521, "min": 2.1614410877227783, "n": 10, "std": 0.9921745171855741 }, "chaser_new_fields": { "max": 25.049999237060547, "mean": 17.549999618530272, "min": 11.639999389648438, "n": 10, "std": 5.312298840336466 }, "chaser_partner_vision": { "max": 0.9274999499320984, "mean": 0.6051499605178833, "min": 0.3570999801158905, "n": 10, "std": 0.18497380302985528 }, "chaser_return": { "max": 17.19499397277832, "mean": 3.1739994168281553, "min": -4.736997127532959, "n": 10, "std": 7.155279675492714 }, "collisions_per_episode": { "max": 20.56999969482422, "mean": 8.975999784469604, "min": 2.5899999141693115, "n": 10, "std": 5.698691288194493 }, "degenerate_fraction": { "max": 1.0, "mean": 0.9279999999999999, "min": 0.82, "n": 10, "std": 0.05895761189193471 }, "explorer_new_fields": { "max": 32.63999938964844, "mean": 30.95099925994873, "min": 27.3799991607666, "n": 10, "std": 1.594731716402341 }, "explorer_partner_vision": { "max": 0.9274999499320984, "mean": 0.6051499605178833, "min": 0.3570999801158905, "n": 10, "std": 0.18497380302985528 }, "explorer_return": { "max": -2.994999885559082, "mean": -10.191499829292297, "min": -24.045000076293945, "n": 10, "std": 6.451588527758027 }, "final_distance": { "max": 5.539825439453125, "mean": 4.093514323234558, "min": 2.089906930923462, "n": 10, "std": 1.1190480545858918 }, "n": 10 } } }, "failed_attempts": [], "figures": [ { "caption": "Training curves, mean \u00b1 std across 20 valid pairs.", "name": "training_curves", "path": "runs/reports/mouse-run-run-1/figures/training_curves.png" }, { "caption": "Evaluation metrics over 40 successful evaluations, grouped by opponent mode and task.", "name": "evaluation_comparison", "path": "runs/reports/mouse-run-run-1/figures/evaluation_comparison.png" } ], "generated_at": "2026-07-02T00:51:16.148597+00:00", "rollout_exclusions": { "episodes_with_exclusion_records": 225, "excluded_degenerate_episodes": 82, "records": 9 }, "schema_version": 2, "tables_root": "runs/tables/mouse-run-run-1", "training_metrics": { "non_social": { "approx_kl": { "max": 0.0013146022521439704, "mean": 0.0001880441928043375, "min": -3.892740220834412e-05, "n": 10, "std": 0.0003842831091966281 }, "chaser_grad_norm": { "max": 0.30185332149267197, "mean": 0.2964277371764183, "min": 0.2860255092382431, "n": 10, "std": 0.0046928184319175795 }, "chaser_new_fields": { "max": 85.875, "mean": 82.53500061035156, "min": 79.30000305175781, "n": 10, "std": 1.8451689004484348 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 7.040000915527344, "mean": 6.4245007038116455, "min": 5.765000820159912, "n": 10, "std": 0.3523950824759949 }, "collisions_per_episode": { "max": 3.575000047683716, "mean": 2.0200000166893006, "min": 1.25, "n": 10, "std": 0.7450503555823611 }, "entropy": { "max": 0.3894049674272537, "mean": 0.2497164834290743, "min": 0.1369955912232399, "n": 10, "std": 0.07146796210122017 }, "explorer_grad_norm": { "max": 0.48010634630918503, "mean": 0.31330167390406133, "min": 0.2777683436870575, "n": 10, "std": 0.05855302773722617 }, "explorer_new_fields": { "max": 84.70000457763672, "mean": 74.98750228881836, "min": 66.92500305175781, "n": 10, "std": 6.284696669500838 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 76.63750457763672, "mean": 61.8962516784668, "min": 49.525001525878906, "n": 10, "std": 9.458782937915352 }, "final_distance": { "max": 6.04818868637085, "mean": 5.044440841674804, "min": 4.513998031616211, "n": 10, "std": 0.4256858202083112 }, "max_param_abs": { "max": 2.794391393661499, "mean": 1.781886065006256, "min": 1.1763957738876343, "n": 10, "std": 0.4974013613049897 }, "n": 10, "policy_loss": { "max": -6.328064564797842e-05, "mean": -0.0005636374810809031, "min": -0.001731324238917864, "n": 10, "std": 0.00045155382786986624 }, "value_loss": { "max": 5.339919090270996, "mean": 5.034486663341522, "min": 4.692311406135559, "n": 10, "std": 0.1927513653139877 } }, "social": { "approx_kl": { "max": 0.021217559756848914, "mean": 0.004209044880608204, "min": 0.0008390653200152087, "n": 10, "std": 0.005810697997759177 }, "chaser_grad_norm": { "max": 4.812156915664673, "mean": 2.5534823894500733, "min": 0.45851049572229385, "n": 10, "std": 1.4746699561338215 }, "chaser_new_fields": { "max": 21.55000114440918, "mean": 15.06250057220459, "min": 9.90000057220459, "n": 10, "std": 4.094955086886805 }, "chaser_partner_vision": { "max": 0.9697500467300415, "mean": 0.3262500174343586, "min": 0.07450000196695328, "n": 10, "std": 0.22845306925201264 }, "chaser_return": { "max": 18.110000610351562, "mean": -1.8387504994869233, "min": -7.507500648498535, "n": 10, "std": 7.419360010252439 }, "collisions_per_episode": { "max": 22.0, "mean": 4.757500104606152, "min": 0.22500000894069672, "n": 10, "std": 6.289555354641951 }, "entropy": { "max": 0.9565532207489014, "mean": 0.6314987547695636, "min": 0.3469325006008148, "n": 10, "std": 0.1663051921358858 }, "explorer_grad_norm": { "max": 2.6690046787261963, "mean": 1.0611258432269097, "min": 0.6042927652597427, "n": 10, "std": 0.5949323554089893 }, "explorer_new_fields": { "max": 64.57500457763672, "mean": 51.49000129699707, "min": 26.225000381469727, "n": 10, "std": 12.230761821815344 }, "explorer_partner_vision": { "max": 0.9697500467300415, "mean": 0.3262500174343586, "min": 0.07450000196695328, "n": 10, "std": 0.22845306925201264 }, "explorer_return": { "max": 46.5625, "mean": 23.843750286102296, "min": -24.512500762939453, "n": 10, "std": 21.32616347922944 }, "final_distance": { "max": 7.126477241516113, "mean": 5.428527927398681, "min": 1.616368293762207, "n": 10, "std": 1.5044527608487566 }, "max_param_abs": { "max": 5.279012680053711, "mean": 2.3758437395095826, "min": 1.6932011842727661, "n": 10, "std": 0.9874779044869594 }, "n": 10, "policy_loss": { "max": 0.014010647196300852, "mean": 0.000734417684680011, "min": -0.002069361328949526, "n": 10, "std": 0.004606471718274356 }, "value_loss": { "max": 8.042522430419922, "mean": 7.117423975467682, "min": 5.209755539894104, "n": 10, "std": 0.7179044963241135 } } }, "valid_pairs_by_task": { "non_social": 10, "social": 10 } }