{ "evaluation_metrics": { "failed_count": 0, "non_social": { "random_chaser": { "average_distance": { "max": 5.129418849945068, "mean": 5.0796972751617435, "min": 5.043961048126221, "n": 10, "std": 0.02873995140186124 }, "chaser_new_fields": { "max": 34.36000061035156, "mean": 34.020999908447266, "min": 33.72999954223633, "n": 10, "std": 0.21644684003085007 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": -3.1859986782073975, "mean": -3.2501988649368285, "min": -3.3139989376068115, "n": 10, "std": 0.04577726762110156 }, "collisions_per_episode": { "max": 3.059999942779541, "mean": 2.796999955177307, "min": 2.319999933242798, "n": 10, "std": 0.2553840545416018 }, "degenerate_fraction": { "max": 0.91, "mean": 0.8789999999999999, "min": 0.86, "n": 10, "std": 0.01813835714721707 }, "explorer_new_fields": { "max": 96.38999938964844, "mean": 94.39299774169922, "min": 92.62999725341797, "n": 10, "std": 1.267440183318429 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 93.50499725341797, "mean": 90.61149673461914, "min": 87.93999481201172, "n": 10, "std": 1.9294984069099073 }, "final_distance": { "max": 4.635434627532959, "mean": 4.085444545745849, "min": 3.837200880050659, "n": 10, "std": 0.22845185221708492 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 5.154397487640381, "mean": 5.11141185760498, "min": 5.0569047927856445, "n": 10, "std": 0.028723257963978274 }, "chaser_new_fields": { "max": 93.91999816894531, "mean": 92.8499969482422, "min": 90.43000030517578, "n": 10, "std": 0.9074680661154696 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 8.579997062683105, "mean": 8.349396276474, "min": 7.805994510650635, "n": 10, "std": 0.20263148293859806 }, "collisions_per_episode": { "max": 4.210000038146973, "mean": 3.394999933242798, "min": 2.6599998474121094, "n": 10, "std": 0.4708556415010507 }, "degenerate_fraction": { "max": 0.87, "mean": 0.853, "min": 0.82, "n": 10, "std": 0.014177446878757837 }, "explorer_new_fields": { "max": 33.21999740600586, "mean": 33.09699897766113, "min": 32.939998626708984, "n": 10, "std": 0.08809638710169938 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": -0.8899999856948853, "mean": -1.149499958753586, "min": -1.59499990940094, "n": 10, "std": 0.21923101569002812 }, "final_distance": { "max": 5.0355939865112305, "mean": 4.388345527648926, "min": 4.020629405975342, "n": 10, "std": 0.3294317958141699 }, "n": 10 } }, "social": { "random_chaser": { "average_distance": { "max": 6.819681167602539, "mean": 6.528078365325928, "min": 6.161418914794922, "n": 10, "std": 0.20855749231374748 }, "chaser_new_fields": { "max": 34.11000061035156, "mean": 34.029999542236325, "min": 33.939998626708984, "n": 10, "std": 0.057620230332741544 }, "chaser_partner_vision": { "max": 0.20059996843338013, "mean": 0.163489992916584, "min": 0.13729999959468842, "n": 10, "std": 0.021343070859131442 }, "chaser_return": { "max": -2.780999183654785, "mean": -2.9034990072250366, "min": -2.978998899459839, "n": 10, "std": 0.06094620981282601 }, "collisions_per_episode": { "max": 0.3700000047683716, "mean": 0.27699999809265136, "min": 0.20999999344348907, "n": 10, "std": 0.05158488367146182 }, "degenerate_fraction": { "max": 1.0, "mean": 0.96, "min": 0.93, "n": 10, "std": 0.020493901531919198 }, "explorer_new_fields": { "max": 46.5, "mean": 37.600999450683595, "min": 30.35999870300293, "n": 10, "std": 5.235516137790574 }, "explorer_partner_vision": { "max": 0.20059996843338013, "mean": 0.163489992916584, "min": 0.13729999959468842, "n": 10, "std": 0.021343070859131442 }, "explorer_return": { "max": 19.404998779296875, "mean": 6.096499764919281, "min": -4.670000076293945, "n": 10, "std": 7.816309965617632 }, "final_distance": { "max": 6.965867042541504, "mean": 6.706823682785034, "min": 6.317955017089844, "n": 10, "std": 0.18157252453346814 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 1.667794942855835, "mean": 1.5712934255599975, "min": 1.4947947263717651, "n": 10, "std": 0.05184039584079168 }, "chaser_new_fields": { "max": 28.6299991607666, "mean": 26.841999435424803, "min": 25.19999885559082, "n": 10, "std": 1.0724253811717708 }, "chaser_partner_vision": { "max": 0.9711999893188477, "mean": 0.9609299838542938, "min": 0.9440000057220459, "n": 10, "std": 0.006767867998884705 }, "chaser_return": { "max": 44.46000289916992, "mean": 40.052001571655275, "min": 34.46500015258789, "n": 10, "std": 3.4813239901279833 }, "collisions_per_episode": { "max": 45.36000061035156, "mean": 41.30799903869629, "min": 36.189998626708984, "n": 10, "std": 3.2882451909664763 }, "degenerate_fraction": { "max": 1.0, "mean": 0.982, "min": 0.96, "n": 10, "std": 0.01536229149573723 }, "explorer_new_fields": { "max": 25.18000030517578, "mean": 23.325999641418456, "min": 21.5, "n": 10, "std": 1.1533272603517895 }, "explorer_partner_vision": { "max": 0.9711999893188477, "mean": 0.9609299838542938, "min": 0.9440000057220459, "n": 10, "std": 0.006767867998884705 }, "explorer_return": { "max": -38.09000015258789, "mean": -43.3179988861084, "min": -46.93000030517578, "n": 10, "std": 3.1799977794472576 }, "final_distance": { "max": 1.5478826761245728, "mean": 1.403444504737854, "min": 1.2944831848144531, "n": 10, "std": 0.08294436722867861 }, "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-transformer/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-transformer/figures/evaluation_comparison.png" } ], "generated_at": "2026-07-03T01:24:40.539994+00:00", "rollout_exclusions": { "episodes_with_exclusion_records": 300, "excluded_degenerate_episodes": 59, "records": 12 }, "schema_version": 2, "tables_root": "runs/tables/mouse-run-run-2-transformer", "training_metrics": { "non_social": { "approx_kl": { "max": 0.0031268233263772593, "mean": 0.0008274657820166141, "min": 3.0426282704221208e-05, "n": 10, "std": 0.0011119185808850796 }, "chaser_grad_norm": { "max": 0.2260504849255085, "mean": 0.05761161218397319, "min": 0.02223971695639193, "n": 10, "std": 0.06012663616499861 }, "chaser_new_fields": { "max": 95.80000305175781, "mean": 94.1650016784668, "min": 91.82500457763672, "n": 10, "std": 0.9980102893568585 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 9.135000228881836, "mean": 8.787000560760498, "min": 8.360000610351562, "n": 10, "std": 0.19291447540213363 }, "collisions_per_episode": { "max": 1.149999976158142, "mean": 0.7475000083446502, "min": 0.3499999940395355, "n": 10, "std": 0.2667981449728105 }, "entropy": { "max": 0.003422852954827249, "mean": 0.0021465113706653936, "min": 0.0009539475431665778, "n": 10, "std": 0.0007943372788127037 }, "explorer_grad_norm": { "max": 0.6812433376908302, "mean": 0.31067706923931837, "min": 0.1489477176219225, "n": 10, "std": 0.15010061774843952 }, "explorer_new_fields": { "max": 97.95000457763672, "mean": 96.32750244140625, "min": 94.82500457763672, "n": 10, "std": 1.1369062620540298 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 96.0625, "mean": 94.04125213623047, "min": 91.82500457763672, "n": 10, "std": 1.5333274673425037 }, "final_distance": { "max": 4.808191776275635, "mean": 3.247923755645752, "min": 2.2879505157470703, "n": 10, "std": 0.7642974377711885 }, "max_param_abs": { "max": 3.969390392303467, "mean": 2.849128818511963, "min": 2.185072422027588, "n": 10, "std": 0.5202608551067897 }, "n": 10, "policy_loss": { "max": -4.3445531147767724e-05, "mean": -0.0006323951178690334, "min": -0.0035095140663869984, "n": 10, "std": 0.0009944805630664182 }, "value_loss": { "max": 0.2153652273118496, "mean": 0.08045755904167891, "min": 0.040869749151170254, "n": 10, "std": 0.04654358081243757 } }, "social": { "approx_kl": { "max": 0.004598107310812227, "mean": 0.00303017191217893, "min": 0.0012853747145520344, "n": 10, "std": 0.0010207158969125706 }, "chaser_grad_norm": { "max": 1.5083578824996948, "mean": 0.7038940638303757, "min": 0.36728256940841675, "n": 10, "std": 0.3619897429634022 }, "chaser_new_fields": { "max": 63.22500228881836, "mean": 58.885000991821286, "min": 55.75, "n": 10, "std": 2.30868508603447 }, "chaser_partner_vision": { "max": 0.986750066280365, "mean": 0.9699750483036041, "min": 0.9477500319480896, "n": 10, "std": 0.010192195843222708 }, "chaser_return": { "max": 16.552501678466797, "mean": 12.937000370025634, "min": 9.160000801086426, "n": 10, "std": 2.1664790907335263 }, "collisions_per_episode": { "max": 14.475000381469727, "mean": 10.745000123977661, "min": 7.550000190734863, "n": 10, "std": 2.1296654790690113 }, "entropy": { "max": 0.20668550580739975, "mean": 0.15412018559873103, "min": 0.11317355558276176, "n": 10, "std": 0.027327137994586925 }, "explorer_grad_norm": { "max": 1.6371502578258514, "mean": 0.9565056964755059, "min": 0.60902339220047, "n": 10, "std": 0.3402408819474271 }, "explorer_new_fields": { "max": 70.80000305175781, "mean": 66.54500160217285, "min": 62.5, "n": 10, "std": 2.835794556837627 }, "explorer_partner_vision": { "max": 0.986750066280365, "mean": 0.9699750483036041, "min": 0.9477500319480896, "n": 10, "std": 0.010192195843222708 }, "explorer_return": { "max": 47.42499923706055, "mean": 41.021250534057614, "min": 33.775001525878906, "n": 10, "std": 5.261294318837594 }, "final_distance": { "max": 1.8629791736602783, "mean": 1.6734241247177124, "min": 1.4801658391952515, "n": 10, "std": 0.11725492127768533 }, "max_param_abs": { "max": 2.9420251846313477, "mean": 2.7518649816513063, "min": 2.4855704307556152, "n": 10, "std": 0.15851267210951459 }, "n": 10, "policy_loss": { "max": -0.002069776682645852, "mean": -0.004644430192469429, "min": -0.007046535808569843, "n": 10, "std": 0.001406250863722305 }, "value_loss": { "max": 3.7607646584510803, "mean": 2.8793665051460264, "min": 1.877354085445404, "n": 10, "std": 0.4693238146909516 } } }, "valid_pairs_by_task": { "non_social": 10, "social": 10 } }