{ "evaluation_metrics": { "failed_count": 0, "non_social": { "random_chaser": { "average_distance": { "max": 5.149752616882324, "mean": 5.07270040512085, "min": 4.984620094299316, "n": 10, "std": 0.052581313876808256 }, "chaser_new_fields": { "max": 34.119998931884766, "mean": 33.85199890136719, "min": 33.57999801635742, "n": 10, "std": 0.1823073325500152 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": -3.21799898147583, "mean": -3.277198886871338, "min": -3.339998722076416, "n": 10, "std": 0.038659484893794616 }, "collisions_per_episode": { "max": 3.069999933242798, "mean": 2.6709999561309816, "min": 2.2200000286102295, "n": 10, "std": 0.22823014297265515 }, "degenerate_fraction": { "max": 0.9, "mean": 0.877, "min": 0.84, "n": 10, "std": 0.018466185312619406 }, "explorer_new_fields": { "max": 93.8499984741211, "mean": 89.89299926757812, "min": 87.11000061035156, "n": 10, "std": 2.045321700944324 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 89.7699966430664, "mean": 83.92299880981446, "min": 79.95999908447266, "n": 10, "std": 3.0531824523312636 }, "final_distance": { "max": 5.063153266906738, "mean": 4.689916372299194, "min": 4.29344367980957, "n": 10, "std": 0.24075538114003284 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 5.187206268310547, "mean": 5.081867837905884, "min": 4.789496898651123, "n": 10, "std": 0.11045187425055718 }, "chaser_new_fields": { "max": 92.52999877929688, "mean": 85.69199981689454, "min": 64.63999938964844, "n": 10, "std": 7.5101531060067055 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 8.309996604919434, "mean": 6.941596150398254, "min": 2.80199933052063, "n": 10, "std": 1.4797752292220605 }, "collisions_per_episode": { "max": 3.93999981880188, "mean": 3.153999948501587, "min": 2.549999952316284, "n": 10, "std": 0.3422045718542642 }, "degenerate_fraction": { "max": 0.88, "mean": 0.849, "min": 0.82, "n": 10, "std": 0.017000000000000015 }, "explorer_new_fields": { "max": 33.25, "mean": 32.99299964904785, "min": 32.70000076293945, "n": 10, "std": 0.18857623473784849 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": -0.7099999785423279, "mean": -1.2379999816417695, "min": -1.7599999904632568, "n": 10, "std": 0.32022804770173824 }, "final_distance": { "max": 5.472009181976318, "mean": 4.858643102645874, "min": 4.226670742034912, "n": 10, "std": 0.34587100538558063 }, "n": 10 } }, "social": { "random_chaser": { "average_distance": { "max": 5.879758358001709, "mean": 5.504114437103271, "min": 5.189486980438232, "n": 10, "std": 0.20798645050228695 }, "chaser_new_fields": { "max": 34.0, "mean": 33.920999145507814, "min": 33.86000061035156, "n": 10, "std": 0.053188073779345785 }, "chaser_partner_vision": { "max": 0.28529998660087585, "mean": 0.2600799843668938, "min": 0.22279998660087585, "n": 10, "std": 0.02106593583620516 }, "chaser_return": { "max": -2.607998847961426, "mean": -2.740798997879028, "min": -2.90799880027771, "n": 10, "std": 0.09688315177175998 }, "collisions_per_episode": { "max": 0.5699999928474426, "mean": 0.4459999918937683, "min": 0.2800000011920929, "n": 10, "std": 0.09024411318064167 }, "degenerate_fraction": { "max": 0.95, "mean": 0.9019999999999999, "min": 0.88, "n": 10, "std": 0.02039607805437113 }, "explorer_new_fields": { "max": 64.8699951171875, "mean": 57.02199821472168, "min": 44.84000015258789, "n": 10, "std": 5.721276559798167 }, "explorer_partner_vision": { "max": 0.28529998660087585, "mean": 0.2600799843668938, "min": 0.22279998660087585, "n": 10, "std": 0.02106593583620516 }, "explorer_return": { "max": 46.69499969482422, "mean": 35.07899894714355, "min": 17.014999389648438, "n": 10, "std": 8.519186425470997 }, "final_distance": { "max": 6.539805889129639, "mean": 5.76410174369812, "min": 5.206815242767334, "n": 10, "std": 0.3719581486248411 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 2.132615566253662, "mean": 1.7829328179359436, "min": 1.632234811782837, "n": 10, "std": 0.14898319724470568 }, "chaser_new_fields": { "max": 27.899999618530273, "mean": 25.913999557495117, "min": 24.229999542236328, "n": 10, "std": 1.084123517153727 }, "chaser_partner_vision": { "max": 0.9590999484062195, "mean": 0.9385299682617188, "min": 0.8789999485015869, "n": 10, "std": 0.022973079625888917 }, "chaser_return": { "max": 39.76400375366211, "mean": 34.08119926452637, "min": 28.073993682861328, "n": 10, "std": 3.871035340753223 }, "collisions_per_episode": { "max": 41.20000076293945, "mean": 35.917999839782716, "min": 30.399999618530273, "n": 10, "std": 3.558077732067056 }, "degenerate_fraction": { "max": 0.99, "mean": 0.9719999999999999, "min": 0.94, "n": 10, "std": 0.018867962264113223 }, "explorer_new_fields": { "max": 25.0, "mean": 23.14699935913086, "min": 21.170000076293945, "n": 10, "std": 1.0500862030540252 }, "explorer_partner_vision": { "max": 0.9590999484062195, "mean": 0.9385299682617188, "min": 0.8789999485015869, "n": 10, "std": 0.022973079625888917 }, "explorer_return": { "max": -34.52499771118164, "mean": -39.57749862670899, "min": -45.099998474121094, "n": 10, "std": 3.3728608394975317 }, "final_distance": { "max": 2.1192781925201416, "mean": 1.6397449612617492, "min": 1.4150844812393188, "n": 10, "std": 0.19734149561495856 }, "n": 10 } } }, "failed_attempts": [], "figures": [ { "caption": "Training curves, mean \u00b1 std across 20 valid pairs.", "name": "training_curves", "path": "runs/reports/mouse-run-run-2-ssm/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-2-ssm/figures/evaluation_comparison.png" } ], "generated_at": "2026-07-03T11:35:04.839739+00:00", "rollout_exclusions": { "episodes_with_exclusion_records": 325, "excluded_degenerate_episodes": 174, "records": 13 }, "schema_version": 2, "tables_root": "runs/tables/mouse-run-run-2-ssm", "training_metrics": { "non_social": { "approx_kl": { "max": 0.000728657579465164, "mean": 0.000300020712514204, "min": 5.2467806275657836e-05, "n": 10, "std": 0.00019725783076226578 }, "chaser_grad_norm": { "max": 1.2475789934396744, "mean": 0.26381482994183897, "min": 0.05319501273334026, "n": 10, "std": 0.3342593565966424 }, "chaser_new_fields": { "max": 94.625, "mean": 87.26500091552734, "min": 65.0999984741211, "n": 10, "std": 7.99176343972919 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 8.90000057220459, "mean": 7.414500594139099, "min": 3.000000238418579, "n": 10, "std": 1.5921644085885014 }, "collisions_per_episode": { "max": 1.7000000476837158, "mean": 0.8100000083446502, "min": 0.42500001192092896, "n": 10, "std": 0.4228179313976203 }, "entropy": { "max": 0.007616071961820126, "mean": 0.0036673748720204456, "min": 0.0007984262483660132, "n": 10, "std": 0.002200403007573944 }, "explorer_grad_norm": { "max": 3.9367371797561646, "mean": 3.30723537504673, "min": 2.3192933201789856, "n": 10, "std": 0.5007569113691553 }, "explorer_new_fields": { "max": 95.57500457763672, "mean": 91.47750320434571, "min": 88.45000457763672, "n": 10, "std": 1.9656890552082262 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 92.95000457763672, "mean": 86.84125137329102, "min": 82.2249984741211, "n": 10, "std": 2.9704857891339618 }, "final_distance": { "max": 5.970248699188232, "mean": 4.79477469921112, "min": 3.6594765186309814, "n": 10, "std": 0.6786811046554324 }, "max_param_abs": { "max": 1.983300805091858, "mean": 1.774276888370514, "min": 1.5573863983154297, "n": 10, "std": 0.12230022258023632 }, "n": 10, "policy_loss": { "max": 3.472161620265979e-05, "mean": -0.0003661589439929003, "min": -0.0015874679654643842, "n": 10, "std": 0.00044941529510602994 }, "value_loss": { "max": 0.32997676730155945, "mean": 0.11813394324854017, "min": 0.05447797849774361, "n": 10, "std": 0.0825904616637552 } }, "social": { "approx_kl": { "max": 0.05561700484361154, "mean": 0.008216197857820618, "min": 0.00037949177203699946, "n": 10, "std": 0.0158652728054278 }, "chaser_grad_norm": { "max": 3.413390338420868, "mean": 1.138201667368412, "min": 0.43530818074941635, "n": 10, "std": 1.011093652124495 }, "chaser_new_fields": { "max": 60.67499923706055, "mean": 54.04250068664551, "min": 48.775001525878906, "n": 10, "std": 4.090003145352089 }, "chaser_partner_vision": { "max": 0.9892500638961792, "mean": 0.9794250428676605, "min": 0.968500018119812, "n": 10, "std": 0.0065240518577340005 }, "chaser_return": { "max": 29.962499618530273, "mean": 18.85100030899048, "min": 13.307501792907715, "n": 10, "std": 4.570849016960882 }, "collisions_per_episode": { "max": 27.625, "mean": 16.935000133514404, "min": 12.475000381469727, "n": 10, "std": 4.3139482144779056 }, "entropy": { "max": 0.19277657940983772, "mean": 0.11876806290820241, "min": 0.0607997989282012, "n": 10, "std": 0.038334725206743116 }, "explorer_grad_norm": { "max": 5.088832139968872, "mean": 1.8970485970377922, "min": 1.111926943063736, "n": 10, "std": 1.1092405254795958 }, "explorer_new_fields": { "max": 76.42500305175781, "mean": 64.69000091552735, "min": 55.95000076293945, "n": 10, "std": 5.872448338151377 }, "explorer_partner_vision": { "max": 0.9892500638961792, "mean": 0.9794250428676605, "min": 0.968500018119812, "n": 10, "std": 0.0065240518577340005 }, "explorer_return": { "max": 49.82500076293945, "mean": 33.88375072479248, "min": 16.024999618530273, "n": 10, "std": 9.15257681083699 }, "final_distance": { "max": 1.8760557174682617, "mean": 1.542952036857605, "min": 1.3376797437667847, "n": 10, "std": 0.14775578438701342 }, "max_param_abs": { "max": 1.8514448404312134, "mean": 1.6233562350273132, "min": 1.5146597623825073, "n": 10, "std": 0.09383932884842139 }, "n": 10, "policy_loss": { "max": -0.0006575217575459646, "mean": -0.004132653732619928, "min": -0.015071312334384857, "n": 10, "std": 0.0037863677234218625 }, "value_loss": { "max": 4.675158381462097, "mean": 3.026606521010399, "min": 1.0913494527339935, "n": 10, "std": 1.0434881912269929 } } }, "valid_pairs_by_task": { "non_social": 10, "social": 10 } }