{ "evaluation_metrics": { "failed_count": 0, "non_social": { "random_chaser": { "average_distance": { "max": 5.199133396148682, "mean": 5.105685377120972, "min": 4.98706579208374, "n": 10, "std": 0.06848503924382211 }, "chaser_new_fields": { "max": 33.97999954223633, "mean": 33.77499961853027, "min": 33.43000030517578, "n": 10, "std": 0.18618548859771447 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": -3.2539987564086914, "mean": -3.3035988092422484, "min": -3.3739986419677734, "n": 10, "std": 0.037113861963604215 }, "collisions_per_episode": { "max": 2.919999837875366, "mean": 2.306999957561493, "min": 1.8600000143051147, "n": 10, "std": 0.27513812098917756 }, "degenerate_fraction": { "max": 0.97, "mean": 0.8880000000000001, "min": 0.85, "n": 10, "std": 0.030594117081556706 }, "explorer_new_fields": { "max": 80.68999481201172, "mean": 70.52699851989746, "min": 60.68000030517578, "n": 10, "std": 6.243129016079577 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 70.2699966430664, "mean": 55.076498413085936, "min": 40.404998779296875, "n": 10, "std": 9.31878213892601 }, "final_distance": { "max": 5.588272571563721, "mean": 5.037810468673706, "min": 4.463276386260986, "n": 10, "std": 0.311963295971499 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 5.262848854064941, "mean": 5.140882539749145, "min": 5.038296699523926, "n": 10, "std": 0.0614441857456308 }, "chaser_new_fields": { "max": 86.43999481201172, "mean": 80.3719985961914, "min": 76.8699951171875, "n": 10, "std": 2.755952134734959 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 7.12199592590332, "mean": 5.922596454620361, "min": 5.221996784210205, "n": 10, "std": 0.5440438394983717 }, "collisions_per_episode": { "max": 3.049999952316284, "mean": 2.68099992275238, "min": 2.309999942779541, "n": 10, "std": 0.24341115086841367 }, "degenerate_fraction": { "max": 0.87, "mean": 0.845, "min": 0.83, "n": 10, "std": 0.013601470508735456 }, "explorer_new_fields": { "max": 33.040000915527344, "mean": 32.88699913024902, "min": 32.790000915527344, "n": 10, "std": 0.08391110303705251 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": -0.9799999594688416, "mean": -1.282999974489212, "min": -1.5199999809265137, "n": 10, "std": 0.14951924109474032 }, "final_distance": { "max": 5.392402648925781, "mean": 4.9553193092346195, "min": 4.301638603210449, "n": 10, "std": 0.2753013389573719 }, "n": 10 } }, "social": { "random_chaser": { "average_distance": { "max": 6.055323123931885, "mean": 5.832867050170899, "min": 5.418818473815918, "n": 10, "std": 0.1926549744871535 }, "chaser_new_fields": { "max": 33.95000076293945, "mean": 33.88299980163574, "min": 33.73999786376953, "n": 10, "std": 0.05728056247021595 }, "chaser_partner_vision": { "max": 0.2913999855518341, "mean": 0.24062999188899994, "min": 0.204599991440773, "n": 10, "std": 0.028305234515883754 }, "chaser_return": { "max": -1.4439992904663086, "mean": -2.2341991543769835, "min": -2.5089991092681885, "n": 10, "std": 0.3334365212524313 }, "collisions_per_episode": { "max": 1.659999966621399, "mean": 0.9239999949932098, "min": 0.6699999570846558, "n": 10, "std": 0.3079350544139646 }, "degenerate_fraction": { "max": 0.91, "mean": 0.892, "min": 0.87, "n": 10, "std": 0.013266499161421611 }, "explorer_new_fields": { "max": 57.55999755859375, "mean": 52.3989990234375, "min": 43.37999725341797, "n": 10, "std": 4.266789586753233 }, "explorer_partner_vision": { "max": 0.2913999855518341, "mean": 0.24062999188899994, "min": 0.204599991440773, "n": 10, "std": 0.028305234515883754 }, "explorer_return": { "max": 35.744998931884766, "mean": 27.815499019622802, "min": 14.429999351501465, "n": 10, "std": 6.396563521591426 }, "final_distance": { "max": 6.047131061553955, "mean": 5.773358011245728, "min": 5.25391149520874, "n": 10, "std": 0.2512269355601858 }, "n": 10 }, "random_explorer": { "average_distance": { "max": 2.13496994972229, "mean": 1.921411967277527, "min": 1.7290464639663696, "n": 10, "std": 0.11278872658258984 }, "chaser_new_fields": { "max": 33.89999771118164, "mean": 31.42699909210205, "min": 28.920000076293945, "n": 10, "std": 1.3234575205271881 }, "chaser_partner_vision": { "max": 0.9477999806404114, "mean": 0.9316199898719788, "min": 0.8927000164985657, "n": 10, "std": 0.017225497745840646 }, "chaser_return": { "max": 36.003997802734375, "mean": 25.334495544433594, "min": 20.284996032714844, "n": 10, "std": 4.144235438399579 }, "collisions_per_episode": { "max": 37.13999938964844, "mean": 26.934999465942383, "min": 21.850000381469727, "n": 10, "std": 4.003446545522908 }, "degenerate_fraction": { "max": 0.98, "mean": 0.95, "min": 0.92, "n": 10, "std": 0.020493901531919184 }, "explorer_new_fields": { "max": 28.149999618530273, "mean": 26.528999519348144, "min": 23.78999900817871, "n": 10, "std": 1.2791048421652784 }, "explorer_partner_vision": { "max": 0.9477999806404114, "mean": 0.9316199898719788, "min": 0.8927000164985657, "n": 10, "std": 0.017225497745840646 }, "explorer_return": { "max": -23.260000228881836, "mean": -28.943499755859374, "min": -39.125, "n": 10, "std": 4.167097957623246 }, "final_distance": { "max": 1.8503373861312866, "mean": 1.7153441309928894, "min": 1.5262326002120972, "n": 10, "std": 0.09933532411070188 }, "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-mlp/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-mlp/figures/evaluation_comparison.png" } ], "generated_at": "2026-07-02T11:51:02.813569+00:00", "rollout_exclusions": { "episodes_with_exclusion_records": 0, "excluded_degenerate_episodes": 0, "records": 0 }, "schema_version": 2, "tables_root": "runs/tables/mouse-run-run-2-mlp", "training_metrics": { "non_social": { "approx_kl": { "max": 0.0007710583850932262, "mean": 0.00027836359963407043, "min": 3.927184380664306e-06, "n": 10, "std": 0.00022781666240450588 }, "chaser_grad_norm": { "max": 0.11076641827821732, "mean": 0.07840223964303732, "min": 0.04017048981040716, "n": 10, "std": 0.023169039813528615 }, "chaser_new_fields": { "max": 87.07500457763672, "mean": 82.87750015258788, "min": 80.4749984741211, "n": 10, "std": 1.7213541677267494 }, "chaser_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "chaser_return": { "max": 7.300001621246338, "mean": 6.471501016616822, "min": 6.015000820159912, "n": 10, "std": 0.34321339840677706 }, "collisions_per_episode": { "max": 2.7250001430511475, "mean": 1.690000021457672, "min": 0.9750000238418579, "n": 10, "std": 0.5471745917052043 }, "entropy": { "max": 0.13418130949139595, "mean": 0.05714834872633219, "min": 0.02094564540311694, "n": 10, "std": 0.030692080850040007 }, "explorer_grad_norm": { "max": 1.1887003779411316, "mean": 0.583899623900652, "min": 0.2844117097556591, "n": 10, "std": 0.27718636321364126 }, "explorer_new_fields": { "max": 82.75, "mean": 71.42000007629395, "min": 60.57500076293945, "n": 10, "std": 6.900640552089833 }, "explorer_partner_vision": { "max": 0.0, "mean": 0.0, "min": 0.0, "n": 10, "std": 0.0 }, "explorer_return": { "max": 73.30000305175781, "mean": 56.473750305175784, "min": 40.07500076293945, "n": 10, "std": 10.319657580882787 }, "final_distance": { "max": 5.683727741241455, "mean": 4.834419059753418, "min": 4.144174098968506, "n": 10, "std": 0.4431673642962756 }, "max_param_abs": { "max": 1.7384926080703735, "mean": 1.452017331123352, "min": 1.2401388883590698, "n": 10, "std": 0.15908309762880601 }, "n": 10, "policy_loss": { "max": -0.0002411644566997495, "mean": -0.000661178331059098, "min": -0.0011177757633866747, "n": 10, "std": 0.00025768936399840804 }, "value_loss": { "max": 4.014994144439697, "mean": 3.651918852329254, "min": 3.322889506816864, "n": 10, "std": 0.2633674319523318 } }, "social": { "approx_kl": { "max": 0.002805363851033249, "mean": 0.00087992062623089, "min": -0.0002899035942244277, "n": 10, "std": 0.0009453947863305684 }, "chaser_grad_norm": { "max": 0.9264960587024689, "mean": 0.41003721952438354, "min": 0.188689436763525, "n": 10, "std": 0.2119915449421527 }, "chaser_new_fields": { "max": 52.92499923706055, "mean": 49.65500030517578, "min": 45.57500076293945, "n": 10, "std": 2.4984694774842433 }, "chaser_partner_vision": { "max": 0.9552500247955322, "mean": 0.9360250353813171, "min": 0.9205000400543213, "n": 10, "std": 0.012655263263627304 }, "chaser_return": { "max": 24.285001754760742, "mean": 21.551751327514648, "min": 18.48750114440918, "n": 10, "std": 1.796639770827262 }, "collisions_per_episode": { "max": 23.200000762939453, "mean": 20.35250053405762, "min": 17.625, "n": 10, "std": 1.6785577964252976 }, "entropy": { "max": 0.19220001995563507, "mean": 0.15101509019732476, "min": 0.12773804366588593, "n": 10, "std": 0.020022485274942464 }, "explorer_grad_norm": { "max": 1.1695344895124435, "mean": 0.5494304157793521, "min": 0.23243309557437897, "n": 10, "std": 0.2504391396265317 }, "explorer_new_fields": { "max": 65.55000305175781, "mean": 58.827500915527345, "min": 52.775001525878906, "n": 10, "std": 3.201532040899581 }, "explorer_partner_vision": { "max": 0.9552500247955322, "mean": 0.9360250353813171, "min": 0.9205000400543213, "n": 10, "std": 0.012655263263627304 }, "explorer_return": { "max": 31.037500381469727, "mean": 21.787500286102294, "min": 12.71250057220459, "n": 10, "std": 4.746788332300563 }, "final_distance": { "max": 1.8305482864379883, "mean": 1.670472300052643, "min": 1.4895528554916382, "n": 10, "std": 0.1115461884593616 }, "max_param_abs": { "max": 1.6354095935821533, "mean": 1.2778974533081056, "min": 1.0662928819656372, "n": 10, "std": 0.16398531671135025 }, "n": 10, "policy_loss": { "max": -0.0027591757117768623, "mean": -0.00390199398393738, "min": -0.005259661337355226, "n": 10, "std": 0.0008555915976877569 }, "value_loss": { "max": 5.285007953643799, "mean": 4.877697789669037, "min": 4.4600828886032104, "n": 10, "std": 0.27699674303611843 } } }, "valid_pairs_by_task": { "non_social": 10, "social": 10 } }