telemetry/routing.jsonl

{"t": 0, "ts": 1788178366.407959, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": 12868.3515625, "attn_entropy": 0.6930617809295654, "load_balance": 0.00019552120938897132, "g_entropy": 0.6929812431335449, "tau_bias": 0.6566851735115051, "beta": 0.10165545080477993, "train_val_corr": 0.0}
{"t": 1, "ts": 1788178371.3149278, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1387.3046875, "attn_entropy": 0.6779214382171631, "load_balance": 0.0001967317657545209, "g_entropy": 0.6929794788360596, "tau_bias": 0.6627488136291504, "beta": 0.24947835256656012, "train_val_corr": 0.0}
{"t": 2, "ts": 1788178376.3098598, "step": 20, "action": "PGO_UPDATE", "entropy_pi": 0.0004814649000763893, "advantage": 1385.4244384765625, "attn_entropy": 0.586170768737793, "load_balance": 0.00019957406911998987, "g_entropy": 0.6929859161376953, "tau_bias": 0.6691250801086426, "beta": 0.34635888785123825, "train_val_corr": 0.0}
{"t": 3, "ts": 1788178381.782675, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1241.598388671875, "attn_entropy": 0.5052870273590088, "load_balance": 0.0002019283128902316, "g_entropy": 0.692961311340332, "tau_bias": 0.6719432473182678, "beta": 0.35505929961800575, "train_val_corr": 0.0}
{"t": 4, "ts": 1788178386.8094068, "step": 40, "action": "PGO_UPDATE", "entropy_pi": 2.0724431237795216e-07, "advantage": 1174.00146484375, "attn_entropy": 0.39550352096557617, "load_balance": 0.0002032599179074168, "g_entropy": 0.6929834842681885, "tau_bias": 0.6731102466583252, "beta": 0.37952368954817456, "train_val_corr": 0.0}
{"t": 5, "ts": 1788178391.8301163, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1110.7191162109375, "attn_entropy": 0.36009671688079836, "load_balance": 0.00020297914743423463, "g_entropy": 0.6929580211639405, "tau_bias": 0.6735446453094482, "beta": 0.3864029472072919, "train_val_corr": 0.0}
{"t": 6, "ts": 1788178403.8709505, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1115.9197998046875, "attn_entropy": 0.3665889501571655, "load_balance": 0.0002029221737757325, "g_entropy": 0.6929827213287354, "tau_bias": 0.6737087965011597, "beta": 0.41289402668674785, "train_val_corr": 0.0}
{"t": 7, "ts": 1788178409.275826, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.46967950463294983, "advantage": -975.4447631835938, "attn_entropy": 0.38614845275878906, "load_balance": 0.0002023019827902317, "g_entropy": 0.6929869174957275, "tau_bias": 0.6737749576568604, "beta": 0.15963734864878157, "train_val_corr": 0.0}
{"t": 8, "ts": 1788178415.1191752, "step": 80, "action": "PGO_UPDATE", "entropy_pi": 0.3779993951320648, "advantage": -928.6014404296875, "attn_entropy": 0.3135722875595093, "load_balance": 0.00020557998213917018, "g_entropy": 0.6929802894592285, "tau_bias": 0.6738572716712952, "beta": 0.21937189375360808, "train_val_corr": 0.0}
{"t": 9, "ts": 1788178421.6968908, "step": 90, "action": "PGO_UPDATE", "entropy_pi": 0.18319635093212128, "advantage": -880.1729125976562, "attn_entropy": 0.36675052642822265, "load_balance": 0.00019418150186538695, "g_entropy": 0.6929687976837158, "tau_bias": 0.6735787987709045, "beta": 0.2504323997224371, "train_val_corr": 0.0}
{"t": 10, "ts": 1788178427.3062425, "step": 100, "action": "PGO_UPDATE", "entropy_pi": 0.6378157734870911, "advantage": -831.4134521484375, "attn_entropy": 0.30715892314910886, "load_balance": 0.00019335763063281773, "g_entropy": 0.6929649829864502, "tau_bias": 0.6706937551498413, "beta": 0.23039480981727442, "train_val_corr": 0.0}
{"t": 11, "ts": 1788178433.2099814, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723292948332528e-07, "advantage": -792.9902954101562, "attn_entropy": 0.0009267563000321388, "load_balance": 0.00018954472616314889, "g_entropy": 0.6929629802703857, "tau_bias": 0.6625849604606628, "beta": 0.27720856790741283, "train_val_corr": 0.0}
{"t": 12, "ts": 1788178446.126889, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -750.0081176757812, "attn_entropy": 1.849227410275489e-05, "load_balance": 0.00018940167501568794, "g_entropy": 0.692981481552124, "tau_bias": 0.6579103469848633, "beta": 0.3201726023107767, "train_val_corr": 0.0}
{"t": 13, "ts": 1788178451.2960694, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -723.7029418945312, "attn_entropy": 5.710848927265033e-06, "load_balance": 0.00018969890661537647, "g_entropy": 0.6929709911346436, "tau_bias": 0.656201958656311, "beta": 0.28537106948594254, "train_val_corr": 0.0}
{"t": 14, "ts": 1788178456.923049, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -689.4268798828125, "attn_entropy": 4.609845200320706e-06, "load_balance": 0.0001900886301882565, "g_entropy": 0.6929813861846924, "tau_bias": 0.6554948687553406, "beta": 0.2805153677860896, "train_val_corr": 0.0}
{"t": 15, "ts": 1788178462.6844287, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -657.9388427734375, "attn_entropy": 8.91723029781133e-06, "load_balance": 0.00019005921203643083, "g_entropy": 0.6929707527160645, "tau_bias": 0.6552656888961792, "beta": 0.3130148407071829, "train_val_corr": 0.0}
{"t": 16, "ts": 1788178468.226512, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -555.356689453125, "attn_entropy": 3.1237377697834745e-06, "load_balance": 0.00018994406564161183, "g_entropy": 0.6929113388061523, "tau_bias": 0.6547785997390747, "beta": 0.29271868243813515, "train_val_corr": 0.0}
{"t": 17, "ts": 1788178473.5217803, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -583.735107421875, "attn_entropy": 1.43617753565195e-06, "load_balance": 0.0001902328571304679, "g_entropy": 0.6929604530334472, "tau_bias": 0.6540901064872742, "beta": 0.3367416448891163, "train_val_corr": 0.0}
{"t": 18, "ts": 1788178486.1328108, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -556.4239501953125, "attn_entropy": 2.352328738197684e-06, "load_balance": 0.00019133095629513263, "g_entropy": 0.6929306030273438, "tau_bias": 0.6540812253952026, "beta": 0.3434753393133481, "train_val_corr": 0.0}
{"t": 19, "ts": 1788178491.3338227, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -537.8191528320312, "attn_entropy": 9.827852409216576e-07, "load_balance": 0.00018928252393379807, "g_entropy": 0.692998218536377, "tau_bias": 0.6530589461326599, "beta": 0.28024251696964103, "train_val_corr": 1.0}
{"t": 20, "ts": 1788178496.4698207, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -515.1854858398438, "attn_entropy": 2.417752330075018e-06, "load_balance": 0.00018895230023190378, "g_entropy": 0.692992877960205, "tau_bias": 0.6520435214042664, "beta": 0.3374316915869713, "train_val_corr": 1.0}
{"t": 21, "ts": 1788178501.8074424, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -485.3585510253906, "attn_entropy": 5.384099495131523e-08, "load_balance": 0.00019003545166924597, "g_entropy": 0.6929503917694092, "tau_bias": 0.6520366668701172, "beta": 0.29946815470854443, "train_val_corr": 0.9999999999999999}
{"t": 22, "ts": 1788178507.3024359, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -462.08819580078125, "attn_entropy": 2.3294020934372384e-08, "load_balance": 0.00018930651713162662, "g_entropy": 0.692982816696167, "tau_bias": 0.6493883728981018, "beta": 0.32482119525472325, "train_val_corr": 1.0}
{"t": 23, "ts": 1788178512.531983, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -438.4210510253906, "attn_entropy": 1.504749889136292e-07, "load_balance": 0.00018942743772640825, "g_entropy": 0.6929883003234864, "tau_bias": 0.6469226479530334, "beta": 0.3317965902388096, "train_val_corr": 1.0}
{"t": 24, "ts": 1788178526.3667665, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -421.15960693359375, "attn_entropy": 7.828235766282887e-08, "load_balance": 0.00019047929672524332, "g_entropy": 0.6929853916168213, "tau_bias": 0.6460167169570923, "beta": 0.32875822111964226, "train_val_corr": 1.0}
{"t": 25, "ts": 1788178532.011561, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -400.14447021484375, "attn_entropy": 6.666847184533253e-07, "load_balance": 0.00018927661003544926, "g_entropy": 0.6929835319519043, "tau_bias": 0.6480622887611389, "beta": 0.3247829129298528, "train_val_corr": 1.0}
{"t": 26, "ts": 1788178537.783714, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -379.2138366699219, "attn_entropy": 1.0217129329248564e-07, "load_balance": 0.00018883763113990425, "g_entropy": 0.6929812431335449, "tau_bias": 0.6486843228340149, "beta": 0.35290971274177235, "train_val_corr": 1.0}
{"t": 27, "ts": 1788178543.1097164, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -364.5567321777344, "attn_entropy": 3.722116844073753e-07, "load_balance": 0.00018987359944730998, "g_entropy": 0.6929409027099609, "tau_bias": 0.6451990604400635, "beta": 0.3498726189136505, "train_val_corr": 1.0}
{"t": 28, "ts": 1788178548.5090516, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -346.76904296875, "attn_entropy": 9.151389122052933e-08, "load_balance": 0.00018856532406061888, "g_entropy": 0.6929806709289551, "tau_bias": 0.6449784636497498, "beta": 0.36274556815624237, "train_val_corr": 0.9999999999999998}
{"t": 0, "ts": 1788178900.704158, "step": 0, "action": "PGO_UPDATE", "entropy_pi": 0.0015364914434030652, "advantage": 36064.79296875, "attn_entropy": 1.2523126125335693, "load_balance": 0.00038864035159349444, "g_entropy": 1.3860986709594727, "tau_bias": 0.6474132537841797, "beta": 0.024438629707825992, "train_val_corr": 0.0}
{"t": 1, "ts": 1788178907.492578, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.11239994317293167, "advantage": -1701.7691650390625, "attn_entropy": 1.3523218154907226, "load_balance": 0.0003893697867169976, "g_entropy": 1.3860857486724854, "tau_bias": 0.6483026146888733, "beta": 0.3603341256578763, "train_val_corr": 0.0}
{"t": 2, "ts": 1788178914.48781, "step": 20, "action": "PGO_UPDATE", "entropy_pi": 0.2043371945619583, "advantage": -1671.1064453125, "attn_entropy": 1.3621006965637208, "load_balance": 0.000390305498149246, "g_entropy": 1.386049175262451, "tau_bias": 0.6482111215591431, "beta": 0.3706427278618018, "train_val_corr": 0.0}
{"t": 3, "ts": 1788178921.9741967, "step": 30, "action": "PGO_UPDATE", "entropy_pi": 0.16908623278141022, "advantage": -1591.404541015625, "attn_entropy": 1.3715628147125245, "load_balance": 0.000390154158230871, "g_entropy": 1.3860570430755614, "tau_bias": 0.6474868059158325, "beta": 0.4026748985052109, "train_val_corr": 0.0}
{"t": 4, "ts": 1788178930.0140216, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.6916967630386353, "advantage": -1507.166259765625, "attn_entropy": 1.3702711105346679, "load_balance": 0.0003906377940438688, "g_entropy": 1.386045217514038, "tau_bias": 0.6469678282737732, "beta": 0.39630956823627156, "train_val_corr": 0.0}
{"t": 5, "ts": 1788178937.8650794, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.655390739440918, "advantage": -1431.150634765625, "attn_entropy": 1.3692607879638672, "load_balance": 0.0003899229108355939, "g_entropy": 1.386017370223999, "tau_bias": 0.6488896608352661, "beta": 0.337549506376187, "train_val_corr": 0.0}
{"t": 6, "ts": 1788178976.525856, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -1247.82666015625, "attn_entropy": 1.3356966972351074, "load_balance": 0.00039029966574162246, "g_entropy": 1.385791015625, "tau_bias": 0.6544051170349121, "beta": 0.3660322142144044, "train_val_corr": 0.0}
{"t": 7, "ts": 1788178983.4337006, "step": 70, "action": "PGO_UPDATE", "entropy_pi": 0.004339808132499456, "advantage": -1311.59716796875, "attn_entropy": 1.2420931816101075, "load_balance": 0.00038849698612466457, "g_entropy": 1.386100673675537, "tau_bias": 0.6564322113990784, "beta": 0.3056726374973853, "train_val_corr": 0.0}
{"t": 8, "ts": 1788178990.3242345, "step": 80, "action": "PGO_UPDATE", "entropy_pi": 0.006867717951536179, "advantage": -1257.1885986328125, "attn_entropy": 1.2324449062347411, "load_balance": 0.0003891280270181596, "g_entropy": 1.3860434532165526, "tau_bias": 0.65634685754776, "beta": 0.4105929533640544, "train_val_corr": 0.0}
{"t": 9, "ts": 1788178997.5832748, "step": 90, "action": "PGO_UPDATE", "entropy_pi": 0.03699694573879242, "advantage": -1193.1536865234375, "attn_entropy": 1.2169600963592528, "load_balance": 0.00038793843705207107, "g_entropy": 1.3860812187194824, "tau_bias": 0.653478741645813, "beta": 0.3812776890893777, "train_val_corr": 0.0}
{"t": 10, "ts": 1788179004.6967735, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.5205255746841431, "advantage": -1134.18896484375, "attn_entropy": 1.2701565742492675, "load_balance": 0.0003897547489032149, "g_entropy": 1.3860546112060548, "tau_bias": 0.6519203782081604, "beta": 0.3631215902666251, "train_val_corr": 0.0}
{"t": 11, "ts": 1788179011.8896108, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723432214708737e-07, "advantage": -1075.945068359375, "attn_entropy": 1.206559705734253, "load_balance": 0.00038833748549222947, "g_entropy": 1.3860713005065919, "tau_bias": 0.6525622010231018, "beta": 0.37543292964498204, "train_val_corr": 0.0}
{"t": 12, "ts": 1788179051.3250766, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0973305936422548e-07, "advantage": -1019.89306640625, "attn_entropy": 1.2649333000183105, "load_balance": 0.00038917220663279293, "g_entropy": 1.3860469818115235, "tau_bias": 0.6511127352714539, "beta": 0.3482513874769211, "train_val_corr": 0.0}
{"t": 13, "ts": 1788179059.089428, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723483373785712e-07, "advantage": -971.2001953125, "attn_entropy": 1.2942142486572266, "load_balance": 0.00039001263212412596, "g_entropy": 1.3860313892364502, "tau_bias": 0.6507030129432678, "beta": 0.3984876635173957, "train_val_corr": 0.0}
{"t": 14, "ts": 1788179066.197771, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0724050386888848e-07, "advantage": -926.1233520507812, "attn_entropy": 1.2944165229797364, "load_balance": 0.00038972855545580385, "g_entropy": 1.386046838760376, "tau_bias": 0.6491914987564087, "beta": 0.38712696731090546, "train_val_corr": 0.0}
{"t": 15, "ts": 1788179073.1944335, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725532579035644e-07, "advantage": -881.0587768554688, "attn_entropy": 1.2526559352874755, "load_balance": 0.00038835381856188177, "g_entropy": 1.386067533493042, "tau_bias": 0.6479018926620483, "beta": 0.3894285435477893, "train_val_corr": 0.0}
{"t": 16, "ts": 1788179080.1815593, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0729306982048001e-07, "advantage": -837.1729736328125, "attn_entropy": 1.3132203102111817, "load_balance": 0.00039095644606277344, "g_entropy": 1.385984468460083, "tau_bias": 0.6475562453269958, "beta": 0.42371151472131413, "train_val_corr": 0.0}
{"t": 17, "ts": 1788179087.2118511, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0738295347655367e-07, "advantage": -791.7838134765625, "attn_entropy": 1.303174877166748, "load_balance": 0.00039024036377668383, "g_entropy": 1.3860406398773193, "tau_bias": 0.6481539011001587, "beta": 0.424709918598334, "train_val_corr": 0.0}
{"t": 18, "ts": 1788179125.0088198, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -751.6781616210938, "attn_entropy": 1.2866739749908447, "load_balance": 0.0003901423653587699, "g_entropy": 1.3860479831695556, "tau_bias": 0.6478234529495239, "beta": 0.3834240424136321, "train_val_corr": 0.0}
{"t": 19, "ts": 1788179132.2098672, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.081309133927789e-07, "advantage": -720.46240234375, "attn_entropy": 1.3208241939544678, "load_balance": 0.00039031938649713993, "g_entropy": 1.3860390663146973, "tau_bias": 0.6463908553123474, "beta": 0.46650171652436256, "train_val_corr": 0.9999999999999996}
{"t": 20, "ts": 1788179139.1172583, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1019333473759616e-07, "advantage": -683.1724853515625, "attn_entropy": 1.298809862136841, "load_balance": 0.00039001689292490483, "g_entropy": 1.3860779285430909, "tau_bias": 0.6488666534423828, "beta": 0.44408875827987987, "train_val_corr": 1.0}
{"t": 21, "ts": 1788179146.1356928, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.152044800141084e-07, "advantage": -637.0391845703125, "attn_entropy": 1.0986315727233886, "load_balance": 0.00038815203588455917, "g_entropy": 1.3860854148864745, "tau_bias": 0.6481912136077881, "beta": 0.46417410920063656, "train_val_corr": 1.0}
{"t": 22, "ts": 1788179153.0654225, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1854411613730917e-07, "advantage": -603.5464477539062, "attn_entropy": 1.0635560274124145, "load_balance": 0.0003894517198204994, "g_entropy": 1.3860628604888916, "tau_bias": 0.6521714329719543, "beta": 0.4594390007356803, "train_val_corr": 1.0}
{"t": 23, "ts": 1788179159.9901621, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.212099161624792e-07, "advantage": -582.6981811523438, "attn_entropy": 1.043408989906311, "load_balance": 0.00038965987041592596, "g_entropy": 1.3860458374023437, "tau_bias": 0.6557984352111816, "beta": 0.45446009561419487, "train_val_corr": 0.9999999999999997}
{"t": 24, "ts": 1788179198.8354452, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.44424029460788e-07, "advantage": -553.9280395507812, "attn_entropy": 1.0641301631927491, "load_balance": 0.0003869942156597972, "g_entropy": 1.3860570907592773, "tau_bias": 0.6605790853500366, "beta": 0.47069332872827846, "train_val_corr": 0.9999999999999999}
{"t": 25, "ts": 1788179206.018156, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1938301131285698e-07, "advantage": -521.0501708984375, "attn_entropy": 1.1337667942047118, "load_balance": 0.00038795144064351916, "g_entropy": 1.386068630218506, "tau_bias": 0.6623490452766418, "beta": 0.5134575317303339, "train_val_corr": 0.9999999999999994}
{"t": 26, "ts": 1788179213.0363264, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.172623396745621e-07, "advantage": -493.1999816894531, "attn_entropy": 1.1182891845703125, "load_balance": 0.00038724596379324794, "g_entropy": 1.3860388278961182, "tau_bias": 0.6628642082214355, "beta": 0.4874269316593806, "train_val_corr": 1.0}
{"t": 27, "ts": 1788179219.994967, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.153092708567783e-07, "advantage": -468.14471435546875, "attn_entropy": 1.136225986480713, "load_balance": 0.0003895087167620659, "g_entropy": 1.3860129833221435, "tau_bias": 0.6629421710968018, "beta": 0.48906278734405834, "train_val_corr": 1.0}
{"t": 28, "ts": 1788179227.29822, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.131233145519218e-07, "advantage": -441.0782165527344, "attn_entropy": 1.1410066604614257, "load_balance": 0.0003869545995257795, "g_entropy": 1.3859059810638428, "tau_bias": 0.6654624938964844, "beta": 0.4935251424709956, "train_val_corr": 1.0}
{"t": 29, "ts": 1788179266.3059237, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723997806726402e-07, "advantage": -396.5411071777344, "attn_entropy": 1.1386777400970458, "load_balance": 0.00038802355993539095, "g_entropy": 1.3860463142395019, "tau_bias": 0.6668440103530884, "beta": 0.4932488886018594, "train_val_corr": 1.0}
{"t": 30, "ts": 1788179273.3334713, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1107182135438052e-07, "advantage": -405.0900573730469, "attn_entropy": 1.160575532913208, "load_balance": 0.0003881691489368677, "g_entropy": 1.3860664367675781, "tau_bias": 0.6721041202545166, "beta": 0.504356324672699, "train_val_corr": 0.9999999999999998}
{"t": 31, "ts": 1788179281.3116446, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1196991895067185e-07, "advantage": -390.5760803222656, "attn_entropy": 1.0313515186309814, "load_balance": 0.00038825469091534613, "g_entropy": 1.386042833328247, "tau_bias": 0.6730750799179077, "beta": 0.501639428238074, "train_val_corr": 0.9999999999999996}
{"t": 32, "ts": 1788179288.501215, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1077376288758387e-07, "advantage": -372.8155517578125, "attn_entropy": 1.0320204496383667, "load_balance": 0.00038768553640693425, "g_entropy": 1.3860970497131349, "tau_bias": 0.6717257499694824, "beta": 0.509097141524156, "train_val_corr": 1.0}
{"t": 33, "ts": 1788179295.6096199, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1014422202370042e-07, "advantage": -353.93408203125, "attn_entropy": 1.0129740238189697, "load_balance": 0.00038798090536147357, "g_entropy": 1.3860765457153321, "tau_bias": 0.6726704239845276, "beta": 0.5093937839070956, "train_val_corr": 1.0}
{"t": 0, "ts": 1788179486.3715236, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12832.8828125, "attn_entropy": 0.6930617809295654, "load_balance": 0.00019552120938897132, "g_entropy": 0.6929812431335449, "tau_bias": 0.6566849946975708, "beta": 0.10835317894816399, "train_val_corr": 0.0}
{"t": 1, "ts": 1788179491.2174382, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.073445431160508e-06, "advantage": -587.4507446289062, "attn_entropy": 0.666393232345581, "load_balance": 0.00019712578505277635, "g_entropy": 0.6929786205291748, "tau_bias": 0.6658554673194885, "beta": 0.22102038655430079, "train_val_corr": 0.0}
{"t": 2, "ts": 1788179495.616865, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.004785797093063593, "advantage": -573.068115234375, "attn_entropy": 0.5410383701324463, "load_balance": 0.00019967956468462943, "g_entropy": 0.6929838180541992, "tau_bias": 0.6756224036216736, "beta": 0.2527519352734089, "train_val_corr": 0.0}
{"t": 3, "ts": 1788179500.222418, "step": 30, "action": "PGO_UPDATE", "entropy_pi": 0.6779966354370117, "advantage": -546.783935546875, "attn_entropy": 0.44888620376586913, "load_balance": 0.00020008718129247428, "g_entropy": 0.6929582595825196, "tau_bias": 0.67994624376297, "beta": 0.2310744859278202, "train_val_corr": 0.0}
{"t": 4, "ts": 1788179504.5234144, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.6617884635925293, "advantage": -517.9473266601562, "attn_entropy": 0.39251065254211426, "load_balance": 0.00019728613551706076, "g_entropy": 0.6929805278778076, "tau_bias": 0.6816269159317017, "beta": 0.27597904826203984, "train_val_corr": 0.0}
{"t": 5, "ts": 1788179508.8314958, "step": 50, "action": "PGO_UPDATE", "entropy_pi": 0.24778594076633453, "advantage": -494.4080505371094, "attn_entropy": 0.36379902362823485, "load_balance": 0.00019671940244734288, "g_entropy": 0.6929544925689697, "tau_bias": 0.6823055148124695, "beta": 0.29740134999156, "train_val_corr": 0.0}
{"t": 6, "ts": 1788179519.3026712, "step": 60, "action": "PGO_UPDATE", "entropy_pi": 0.29927805066108704, "advantage": -470.0270080566406, "attn_entropy": 0.3934603214263916, "load_balance": 0.00020074844360351562, "g_entropy": 0.6929795742034912, "tau_bias": 0.6827506422996521, "beta": 0.2810777711371581, "train_val_corr": 0.0}
{"t": 7, "ts": 1788179523.9298527, "step": 70, "action": "PGO_UPDATE", "entropy_pi": 0.7769972085952759, "advantage": -446.7344970703125, "attn_entropy": 0.4491696357727051, "load_balance": 0.00019917828030884265, "g_entropy": 0.6929840564727783, "tau_bias": 0.6825464367866516, "beta": 0.28341985990603763, "train_val_corr": 0.0}
{"t": 8, "ts": 1788179528.768603, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.1880919337272644, "advantage": -426.2367858886719, "attn_entropy": 0.40580053329467775, "load_balance": 0.000194878235924989, "g_entropy": 0.6929774284362793, "tau_bias": 0.6821353435516357, "beta": 0.26676356233656406, "train_val_corr": 0.0}
{"t": 9, "ts": 1788179533.3787165, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.2292408049106598, "advantage": -406.10308837890625, "attn_entropy": 0.42179393768310547, "load_balance": 0.0001975283958017826, "g_entropy": 0.6929655075073242, "tau_bias": 0.6826826333999634, "beta": 0.3204428404569626, "train_val_corr": 0.0}
{"t": 10, "ts": 1788179537.9978316, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.7217736840248108, "advantage": -377.84027099609375, "attn_entropy": 0.4225125789642334, "load_balance": 0.0002040559658780694, "g_entropy": 0.6929612159729004, "tau_bias": 0.6805342435836792, "beta": 0.26795591103533906, "train_val_corr": 0.0}
{"t": 11, "ts": 1788179542.5062282, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": -356.6120300292969, "attn_entropy": 0.2905984401702881, "load_balance": 0.00019130089785903692, "g_entropy": 0.692960548400879, "tau_bias": 0.6691895723342896, "beta": 0.253727313131094, "train_val_corr": 0.0}
{"t": 12, "ts": 1788179554.02604, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723297211588942e-07, "advantage": -340.3717041015625, "attn_entropy": 0.11350558996200562, "load_balance": 0.00018992823315784336, "g_entropy": 0.6929787158966064, "tau_bias": 0.6652941107749939, "beta": 0.3028501483301322, "train_val_corr": 0.0}
{"t": 13, "ts": 1788179558.474778, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.07237533800253e-07, "advantage": -328.4404296875, "attn_entropy": 0.014521124958992004, "load_balance": 0.00018970042001456022, "g_entropy": 0.6929686069488525, "tau_bias": 0.6623126268386841, "beta": 0.2769651226699352, "train_val_corr": 0.0}
{"t": 14, "ts": 1788179563.074353, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0729919469886227e-07, "advantage": -314.3342590332031, "attn_entropy": 0.006752745807170868, "load_balance": 0.00019007010851055384, "g_entropy": 0.6929792881011962, "tau_bias": 0.6613898277282715, "beta": 0.303889565790693, "train_val_corr": 0.0}
{"t": 15, "ts": 1788179567.9909256, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0801782341095532e-07, "advantage": -298.79931640625, "attn_entropy": 0.006889504939317703, "load_balance": 0.00019004291389137506, "g_entropy": 0.6929686069488525, "tau_bias": 0.6611568927764893, "beta": 0.27610740686456364, "train_val_corr": 0.0}
{"t": 16, "ts": 1788179572.8664129, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1574129505097517e-07, "advantage": -285.22100830078125, "attn_entropy": 0.005883748456835747, "load_balance": 0.00018991981633007527, "g_entropy": 0.6929079532623291, "tau_bias": 0.6618309020996094, "beta": 0.2828797393788894, "train_val_corr": 0.0}
{"t": 17, "ts": 1788179577.6830761, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.3459539849900466e-07, "advantage": -273.3564453125, "attn_entropy": 0.007390007376670837, "load_balance": 0.00019020743202418088, "g_entropy": 0.6929575443267822, "tau_bias": 0.661678671836853, "beta": 0.34701195855935413, "train_val_corr": 0.0}
{"t": 18, "ts": 1788179590.0934181, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.488309974069125e-06, "advantage": -256.4199523925781, "attn_entropy": 0.0006342418491840363, "load_balance": 0.00019132092129439115, "g_entropy": 0.6929293632507324, "tau_bias": 0.6547577977180481, "beta": 0.33399096379677456, "train_val_corr": 0.0}
{"t": 19, "ts": 1788179594.618089, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0002153002715203911, "advantage": -246.87278747558594, "attn_entropy": 1.111138135456713e-06, "load_balance": 0.00018935316475108266, "g_entropy": 0.693001651763916, "tau_bias": 0.6422330737113953, "beta": 0.3551528329650561, "train_val_corr": 1.0}
{"t": 20, "ts": 1788179599.5314898, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.06642448902130127, "advantage": -237.90432739257812, "attn_entropy": 1.6901960862014675e-07, "load_balance": 0.00018906177720054985, "g_entropy": 0.692998218536377, "tau_bias": 0.6377795934677124, "beta": 0.3795480045179526, "train_val_corr": 1.0}
{"t": 21, "ts": 1788179604.1098557, "step": 210, "action": "ATTN_META_ONLY", "entropy_pi": 0.6930044293403625, "advantage": -222.24302673339844, "attn_entropy": 5.495467689797806e-08, "load_balance": 0.00019011045806109904, "g_entropy": 0.6929559230804443, "tau_bias": 0.6348145008087158, "beta": 0.3020757393290599, "train_val_corr": 1.0}
{"t": 22, "ts": 1788179607.2900112, "step": 220, "action": "ATTN_META_ONLY", "entropy_pi": 0.5952723026275635, "advantage": -205.94235229492188, "attn_entropy": 7.85052407081821e-08, "load_balance": 0.0001893875072710216, "g_entropy": 0.6929872512817383, "tau_bias": 0.6348145008087158, "beta": 0.2755970110495885, "train_val_corr": 1.0}
{"t": 23, "ts": 1788179610.4184763, "step": 230, "action": "ATTN_META_ONLY", "entropy_pi": 0.5790865421295166, "advantage": -194.1869659423828, "attn_entropy": 6.404772648238577e-07, "load_balance": 0.00018950621597468853, "g_entropy": 0.6929925441741943, "tau_bias": 0.6348145008087158, "beta": 0.33187319710850716, "train_val_corr": 1.0}
{"t": 24, "ts": 1788179620.7122421, "step": 240, "action": "ATTN_META_ONLY", "entropy_pi": 0.5902959704399109, "advantage": -185.35525512695312, "attn_entropy": 5.093179424875416e-07, "load_balance": 0.00019054447766393424, "g_entropy": 0.692989730834961, "tau_bias": 0.6348145008087158, "beta": 0.31698501979311305, "train_val_corr": 1.0}
{"t": 25, "ts": 1788179623.8228676, "step": 250, "action": "ATTN_META_ONLY", "entropy_pi": 0.5744796395301819, "advantage": -175.2393035888672, "attn_entropy": 3.0932926165405663e-06, "load_balance": 0.00018935807747766376, "g_entropy": 0.6929880619049072, "tau_bias": 0.6348145008087158, "beta": 0.3205850863208373, "train_val_corr": 1.0}
{"t": 26, "ts": 1788179627.1090066, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.5531904101371765, "advantage": -167.35076904296875, "attn_entropy": 6.590533303096891e-07, "load_balance": 0.00018892436055466532, "g_entropy": 0.692985725402832, "tau_bias": 0.6348145008087158, "beta": 0.3289332489172618, "train_val_corr": 1.0}
{"t": 27, "ts": 1788179630.5312676, "step": 270, "action": "ATTN_META_ONLY", "entropy_pi": 0.4867624044418335, "advantage": -161.38995361328125, "attn_entropy": 2.546683754189871e-06, "load_balance": 0.00018994605634361505, "g_entropy": 0.6929463386535645, "tau_bias": 0.6348145008087158, "beta": 0.3720691427588463, "train_val_corr": 1.0}
{"t": 28, "ts": 1788179633.9658, "step": 280, "action": "ATTN_META_ONLY", "entropy_pi": 0.4441985487937927, "advantage": -151.921875, "attn_entropy": 7.279102646862157e-07, "load_balance": 0.0001886533689685166, "g_entropy": 0.6929850101470947, "tau_bias": 0.6348145008087158, "beta": 0.3868078924715519, "train_val_corr": 1.0}
{"t": 29, "ts": 1788179674.2646914, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.164759381339536e-07, "advantage": 7904.80712890625, "attn_entropy": 1.1136255741119385, "load_balance": 0.00038749801460653545, "g_entropy": 1.386103630065918, "tau_bias": 0.6371592879295349, "beta": 0.14552697477241358, "train_val_corr": 0.0}
{"t": 30, "ts": 1788179681.3791308, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.9826562695234315e-06, "advantage": -165.56915283203125, "attn_entropy": 1.1760175704956055, "load_balance": 0.000386420963332057, "g_entropy": 1.3860925674438476, "tau_bias": 0.63484787940979, "beta": 0.32117705419659615, "train_val_corr": 0.0}
{"t": 31, "ts": 1788179688.6117802, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.569738692021929e-05, "advantage": -150.85105895996094, "attn_entropy": 1.1652937412261963, "load_balance": 0.00038772199768573047, "g_entropy": 1.3860589027404786, "tau_bias": 0.6303730010986328, "beta": 0.32904427871108055, "train_val_corr": 0.0}
{"t": 32, "ts": 1788179695.5992944, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0028985091485083103, "advantage": -216.38185119628906, "attn_entropy": 1.1389159679412841, "load_balance": 0.00038615205558016895, "g_entropy": 1.3860644817352294, "tau_bias": 0.634788453578949, "beta": 0.4136063978075981, "train_val_corr": 0.0}
{"t": 33, "ts": 1788179702.5911102, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.3162873387336731, "advantage": -191.35507202148438, "attn_entropy": 0.78768892288208, "load_balance": 0.00038602972635999323, "g_entropy": 1.3860559463500977, "tau_bias": 0.6281667351722717, "beta": 0.3842403118809064, "train_val_corr": 0.0}
{"t": 34, "ts": 1788179709.89324, "step": 50, "action": "ATTN_META_ONLY", "entropy_pi": 0.2588149607181549, "advantage": -303.317138671875, "attn_entropy": 0.6542773008346557, "load_balance": 0.0003838058793917298, "g_entropy": 1.3860295295715332, "tau_bias": 0.6293936967849731, "beta": 0.3984954059123993, "train_val_corr": 0.0}
{"t": 35, "ts": 1788179744.39228, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072814595521777e-07, "advantage": -285.168212890625, "attn_entropy": 1.182929229736328, "load_balance": 0.0003853112109936774, "g_entropy": 1.3858186721801757, "tau_bias": 0.6293936967849731, "beta": 0.4717285881439845, "train_val_corr": 0.0}
{"t": 36, "ts": 1788179748.716337, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072832501198718e-07, "advantage": -282.4036560058594, "attn_entropy": 1.1130141735076904, "load_balance": 0.0003926247591152787, "g_entropy": 1.386112356185913, "tau_bias": 0.6293936967849731, "beta": 0.447229340672493, "train_val_corr": 0.0}
{"t": 37, "ts": 1788179752.9688911, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.5834998496065964e-07, "advantage": -295.684814453125, "attn_entropy": 0.5892148494720459, "load_balance": 0.00039107329212129116, "g_entropy": 1.3860586166381836, "tau_bias": 0.6293936967849731, "beta": 0.4320039401451747, "train_val_corr": 0.0}
{"t": 38, "ts": 1788179757.1928835, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.000186342338565737, "advantage": -282.431640625, "attn_entropy": 0.3700226306915283, "load_balance": 0.00038982111727818846, "g_entropy": 1.3860929012298584, "tau_bias": 0.6293936967849731, "beta": 0.42418741434812546, "train_val_corr": 0.0}
{"t": 39, "ts": 1788179761.6709144, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0100243603810668, "advantage": -274.7188720703125, "attn_entropy": 0.3913231253623962, "load_balance": 0.0003885069047100842, "g_entropy": 1.386067008972168, "tau_bias": 0.6293936967849731, "beta": 0.4316918154557546, "train_val_corr": 0.0}
{"t": 40, "ts": 1788179765.918447, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -258.63739013671875, "attn_entropy": 0.4030353307723999, "load_balance": 0.0003900417243130505, "g_entropy": 1.3860830783843994, "tau_bias": 0.6293936967849731, "beta": 0.4366982951760292, "train_val_corr": 0.0}
{"t": 41, "ts": 1788179798.7042878, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -252.3710479736328, "attn_entropy": 0.36400766372680665, "load_balance": 0.0003896420355886221, "g_entropy": 1.386059331893921, "tau_bias": 0.6293936967849731, "beta": 0.4490785350402196, "train_val_corr": 0.0}
{"t": 42, "ts": 1788179802.6896236, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -239.89007568359375, "attn_entropy": 0.3553754687309265, "load_balance": 0.0003883740515448153, "g_entropy": 1.386044406890869, "tau_bias": 0.6293936967849731, "beta": 0.45159995680054027, "train_val_corr": 0.0}
{"t": 43, "ts": 1788179806.7972505, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -228.73211669921875, "attn_entropy": 0.33202069997787476, "load_balance": 0.0003903054515831172, "g_entropy": 1.386058235168457, "tau_bias": 0.6293936967849731, "beta": 0.46003176892797154, "train_val_corr": 0.0}
{"t": 44, "ts": 1788179810.8706791, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326736879404e-07, "advantage": -218.3818359375, "attn_entropy": 0.28203314542770386, "load_balance": 0.00038920275401324036, "g_entropy": 1.3860772609710694, "tau_bias": 0.6293936967849731, "beta": 0.46158457919955254, "train_val_corr": 0.0}
{"t": 45, "ts": 1788179815.089331, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723776117392845e-07, "advantage": -207.78561401367188, "attn_entropy": 0.3266530454158783, "load_balance": 0.0003871291526593268, "g_entropy": 1.385997438430786, "tau_bias": 0.6293936967849731, "beta": 0.4763703979551792, "train_val_corr": 0.0}
{"t": 46, "ts": 1788179819.1092696, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725994431813888e-07, "advantage": -196.2731170654297, "attn_entropy": 0.2962651550769806, "load_balance": 0.0003876624745316803, "g_entropy": 1.386051845550537, "tau_bias": 0.6293936967849731, "beta": 0.4660465915997823, "train_val_corr": 0.0}
{"t": 47, "ts": 1788179851.407991, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0805214262509253e-07, "advantage": -187.72901916503906, "attn_entropy": 0.33208082914352416, "load_balance": 0.0003883961820974946, "g_entropy": 1.3860585689544678, "tau_bias": 0.6293936967849731, "beta": 0.4770907076696555, "train_val_corr": 0.0}
{"t": 48, "ts": 1788179855.9030201, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.4123798425534915e-07, "advantage": -177.976806640625, "attn_entropy": 0.2823523163795471, "load_balance": 0.0003877866780385375, "g_entropy": 1.3860494136810302, "tau_bias": 0.6293936967849731, "beta": 0.47702064365148544, "train_val_corr": 0.9999999999999998}
{"t": 49, "ts": 1788179860.0187979, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.7585038222023286e-05, "advantage": -168.75454711914062, "attn_entropy": 0.28278069496154784, "load_balance": 0.000387614534702152, "g_entropy": 1.3860862255096436, "tau_bias": 0.6293936967849731, "beta": 0.4811093683044116, "train_val_corr": 1.0}
{"t": 50, "ts": 1788179864.5371969, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.878366886870936e-05, "advantage": -159.65069580078125, "attn_entropy": 0.2710563004016876, "load_balance": 0.0003877919865772128, "g_entropy": 1.3860947132110595, "tau_bias": 0.6293936967849731, "beta": 0.48192313810189563, "train_val_corr": 1.0}
{"t": 51, "ts": 1788179868.8186474, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0005388663266785443, "advantage": -150.6539306640625, "attn_entropy": 0.29498658776283265, "load_balance": 0.00038932638708502056, "g_entropy": 1.3860744953155517, "tau_bias": 0.6293936967849731, "beta": 0.4816363739470641, "train_val_corr": 1.0}
{"t": 52, "ts": 1788179873.0150907, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0014557692920789123, "advantage": -144.758056640625, "attn_entropy": 0.3329703986644745, "load_balance": 0.00038855295861139894, "g_entropy": 1.3860602855682373, "tau_bias": 0.6293936967849731, "beta": 0.49056360498070717, "train_val_corr": 1.0}
{"t": 53, "ts": 1788179907.3370545, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00198802025988698, "advantage": -137.65621948242188, "attn_entropy": 0.2753968179225922, "load_balance": 0.0003889063373208046, "g_entropy": 1.3860731601715088, "tau_bias": 0.6293936967849731, "beta": 0.49248257651925087, "train_val_corr": 1.0}
{"t": 54, "ts": 1788179911.6311915, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.004957615863531828, "advantage": -130.6373291015625, "attn_entropy": 0.27271980345249175, "load_balance": 0.00038959898520261047, "g_entropy": 1.386084794998169, "tau_bias": 0.6293936967849731, "beta": 0.4976736543079217, "train_val_corr": 1.0}
{"t": 55, "ts": 1788179915.7167985, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.005345141980797052, "advantage": -123.80609130859375, "attn_entropy": 0.24229592680931092, "load_balance": 0.0003901576157659292, "g_entropy": 1.3860575199127196, "tau_bias": 0.6293936967849731, "beta": 0.503190298875173, "train_val_corr": 0.9999999999999991}
{"t": 56, "ts": 1788179920.010443, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00943847931921482, "advantage": -117.76036071777344, "attn_entropy": 0.28203771114349363, "load_balance": 0.00038818184984847905, "g_entropy": 1.3860331058502198, "tau_bias": 0.6293936967849731, "beta": 0.5070000166694323, "train_val_corr": 1.0}
{"t": 57, "ts": 1788179924.7081008, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00266255927272141, "advantage": -111.32241821289062, "attn_entropy": 0.2502615928649902, "load_balance": 0.00038957953220233323, "g_entropy": 1.3859359264373778, "tau_bias": 0.6293936967849731, "beta": 0.506485273440679, "train_val_corr": 1.0}
{"t": 58, "ts": 1788179959.271735, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00022284509032033384, "advantage": -104.45906829833984, "attn_entropy": 0.276311469078064, "load_balance": 0.0003892984241247177, "g_entropy": 1.3860666275024414, "tau_bias": 0.6293936967849731, "beta": 0.5092988982796669, "train_val_corr": 0.9999999999999994}
{"t": 59, "ts": 1788179963.465541, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.002601990941911936, "advantage": -101.30751037597656, "attn_entropy": 0.2804760217666626, "load_balance": 0.00038885525427758696, "g_entropy": 1.3860867500305176, "tau_bias": 0.6293936967849731, "beta": 0.5078102176388105, "train_val_corr": 0.9999999999999998}
{"t": 60, "ts": 1788179968.5168412, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.005654171574860811, "advantage": -96.36112976074219, "attn_entropy": 0.2945629954338074, "load_balance": 0.00038755763089284303, "g_entropy": 1.3860661506652832, "tau_bias": 0.6293936967849731, "beta": 0.5096404155095419, "train_val_corr": 1.0}
{"t": 61, "ts": 1788179972.9323945, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.002348700538277626, "advantage": -91.47260284423828, "attn_entropy": 0.26977506577968596, "load_balance": 0.00038857853505760433, "g_entropy": 1.3861149311065675, "tau_bias": 0.6293936967849731, "beta": 0.5086887752016386, "train_val_corr": 0.9999999999999999}
{"t": 62, "ts": 1788179977.469911, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00380115257576108, "advantage": -87.02730560302734, "attn_entropy": 0.2806116819381714, "load_balance": 0.0003887085011228919, "g_entropy": 1.386096477508545, "tau_bias": 0.6293936967849731, "beta": 0.5091003229220709, "train_val_corr": 1.0}
{"t": 63, "ts": 1788179981.8686564, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.002956933341920376, "advantage": -82.32838439941406, "attn_entropy": 0.2802480697631836, "load_balance": 0.0003899252158589661, "g_entropy": 1.3860281467437745, "tau_bias": 0.6293936967849731, "beta": 0.510564461350441, "train_val_corr": 1.0}
{"t": 64, "ts": 1788180019.1232805, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.004561976529657841, "advantage": -78.33930206298828, "attn_entropy": 0.29530675411224366, "load_balance": 0.000389321381226182, "g_entropy": 1.3860744953155517, "tau_bias": 0.6293936967849731, "beta": 0.509162945051988, "train_val_corr": 1.0}
{"t": 65, "ts": 1788180023.6736495, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0021101110614836216, "advantage": -74.58673858642578, "attn_entropy": 0.2736815929412842, "load_balance": 0.0003906632889993489, "g_entropy": 1.3860360622406005, "tau_bias": 0.6293936967849731, "beta": 0.5102539137005806, "train_val_corr": 1.0}
{"t": 66, "ts": 1788180028.2201018, "step": 370, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.005315161310136318, "advantage": -70.7929916381836, "attn_entropy": 0.32519404888153075, "load_balance": 0.00038714781403541564, "g_entropy": 1.3859375, "tau_bias": 0.6293936967849731, "beta": 0.508041908343633, "train_val_corr": 1.0}
{"t": 67, "ts": 1788180033.2808058, "step": 380, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.005470882635563612, "advantage": -66.48503875732422, "attn_entropy": 0.29165871143341066, "load_balance": 0.00039098374545574186, "g_entropy": 1.3860793113708496, "tau_bias": 0.6293936967849731, "beta": 0.5092249934871992, "train_val_corr": 1.0}
{"t": 68, "ts": 1788180037.8344035, "step": 390, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.008035046048462391, "advantage": -63.91178512573242, "attn_entropy": 0.2955048680305481, "load_balance": 0.00038887206465005876, "g_entropy": 1.3860939502716065, "tau_bias": 0.6293936967849731, "beta": 0.5095243180791537, "train_val_corr": 1.0}
{"t": 69, "ts": 1788180042.2240207, "step": 400, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.005896601360291243, "advantage": -54.6678352355957, "attn_entropy": 0.3198613405227661, "load_balance": 0.00038741971366107464, "g_entropy": 1.38579421043396, "tau_bias": 0.6293936967849731, "beta": 0.5145345131556193, "train_val_corr": 1.0}
{"t": 0, "ts": 1788180282.735758, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12832.8828125, "attn_entropy": 0.6930617809295654, "load_balance": 0.00019552120938897132, "g_entropy": 0.6929812431335449, "tau_bias": 0.6566849946975708, "beta": 0.10835317894816399, "train_val_corr": 0.0}
{"t": 1, "ts": 1788180287.6200047, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.073445431160508e-06, "advantage": -587.4507446289062, "attn_entropy": 0.666393232345581, "load_balance": 0.00019712578505277635, "g_entropy": 0.6929786205291748, "tau_bias": 0.6658554673194885, "beta": 0.22102038655430079, "train_val_corr": 0.0}
{"t": 2, "ts": 1788180292.3018734, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.004785797093063593, "advantage": -573.068115234375, "attn_entropy": 0.5410383701324463, "load_balance": 0.00019967956468462943, "g_entropy": 0.6929838180541992, "tau_bias": 0.6756224036216736, "beta": 0.2527519352734089, "train_val_corr": 0.0}
{"t": 3, "ts": 1788180297.5310228, "step": 30, "action": "PGO_UPDATE", "entropy_pi": 0.6779966354370117, "advantage": -546.783935546875, "attn_entropy": 0.44888620376586913, "load_balance": 0.00020008718129247428, "g_entropy": 0.6929582595825196, "tau_bias": 0.67994624376297, "beta": 0.2310744859278202, "train_val_corr": 0.0}
{"t": 4, "ts": 1788180302.6909883, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.6617884635925293, "advantage": -517.9473266601562, "attn_entropy": 0.39251065254211426, "load_balance": 0.00019728613551706076, "g_entropy": 0.6929805278778076, "tau_bias": 0.6816269159317017, "beta": 0.27597904826203984, "train_val_corr": 0.0}
{"t": 5, "ts": 1788180307.814157, "step": 50, "action": "PGO_UPDATE", "entropy_pi": 0.24778594076633453, "advantage": -494.4080505371094, "attn_entropy": 0.36379902362823485, "load_balance": 0.00019671940244734288, "g_entropy": 0.6929544925689697, "tau_bias": 0.6823055148124695, "beta": 0.29740134999156, "train_val_corr": 0.0}
{"t": 6, "ts": 1788180319.4338903, "step": 60, "action": "PGO_UPDATE", "entropy_pi": 0.29927805066108704, "advantage": -470.0270080566406, "attn_entropy": 0.3934603214263916, "load_balance": 0.00020074844360351562, "g_entropy": 0.6929795742034912, "tau_bias": 0.6827506422996521, "beta": 0.2810777711371581, "train_val_corr": 0.0}
{"t": 7, "ts": 1788180324.7716153, "step": 70, "action": "PGO_UPDATE", "entropy_pi": 0.7770779132843018, "advantage": -446.60418701171875, "attn_entropy": 0.4484855651855469, "load_balance": 0.00019895108416676522, "g_entropy": 0.6929839611053467, "tau_bias": 0.6825190782546997, "beta": 0.26477867054442566, "train_val_corr": 0.0}
{"t": 8, "ts": 1788180329.2782094, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.1880374699831009, "advantage": -426.3194274902344, "attn_entropy": 0.39160192012786865, "load_balance": 0.00019413278205320238, "g_entropy": 0.6929774761199952, "tau_bias": 0.6820326447486877, "beta": 0.2703086876620849, "train_val_corr": 0.0}
{"t": 9, "ts": 1788180333.7112994, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.230955570936203, "advantage": -405.2984924316406, "attn_entropy": 0.4175122261047363, "load_balance": 0.00019732958171516656, "g_entropy": 0.6929654598236084, "tau_bias": 0.6827709078788757, "beta": 0.3177870971461137, "train_val_corr": 0.0}
{"t": 10, "ts": 1788180338.0301645, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.7187275886535645, "advantage": -384.1015319824219, "attn_entropy": 0.4076129913330078, "load_balance": 0.0002041323808953166, "g_entropy": 0.6929611206054688, "tau_bias": 0.6806187033653259, "beta": 0.32577276478211087, "train_val_corr": 0.0}
{"t": 11, "ts": 1788180342.4651022, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": -364.6973571777344, "attn_entropy": 0.005328371003270149, "load_balance": 0.00018955166451632975, "g_entropy": 0.6929620742797852, "tau_bias": 0.6649991869926453, "beta": 0.21581758294875422, "train_val_corr": 0.0}
{"t": 12, "ts": 1788180353.9338427, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723294369418e-07, "advantage": -342.6893005371094, "attn_entropy": 5.966397002339363e-05, "load_balance": 0.00018938431749120354, "g_entropy": 0.6929804801940918, "tau_bias": 0.6607851982116699, "beta": 0.25734658042589825, "train_val_corr": 0.0}
{"t": 13, "ts": 1788180358.7839947, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072378180173473e-07, "advantage": -329.51434326171875, "attn_entropy": 8.862389950081707e-06, "load_balance": 0.00018968143267557026, "g_entropy": 0.6929698467254639, "tau_bias": 0.6591408848762512, "beta": 0.3110854346305132, "train_val_corr": 0.0}
{"t": 14, "ts": 1788180363.607088, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0730206529151474e-07, "advantage": -315.81744384765625, "attn_entropy": 3.9238999306689944e-06, "load_balance": 0.00019007195951417088, "g_entropy": 0.6929803371429444, "tau_bias": 0.6584917306900024, "beta": 0.23725857057919106, "train_val_corr": 0.0}
{"t": 15, "ts": 1788180368.2217367, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0800375466478727e-07, "advantage": -301.9831237792969, "attn_entropy": 5.39456814294681e-06, "load_balance": 0.00019004230853170156, "g_entropy": 0.6929697036743164, "tau_bias": 0.6582355499267578, "beta": 0.2594692458709081, "train_val_corr": 0.0}
{"t": 16, "ts": 1788180372.5131633, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1552480689024378e-07, "advantage": -285.3312683105469, "attn_entropy": 4.720315337181091e-06, "load_balance": 0.00018992588156834244, "g_entropy": 0.6929097175598145, "tau_bias": 0.6581416726112366, "beta": 0.31782514974474907, "train_val_corr": 0.0}
{"t": 17, "ts": 1788180376.9266894, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.3428270285185135e-07, "advantage": -273.80621337890625, "attn_entropy": 6.298582593444735e-06, "load_balance": 0.0001902130083180964, "g_entropy": 0.6929590702056885, "tau_bias": 0.658105731010437, "beta": 0.2834706449260314, "train_val_corr": 0.0}
{"t": 18, "ts": 1788180387.9746065, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.466907517373329e-06, "advantage": -263.6184997558594, "attn_entropy": 8.867547148838639e-06, "load_balance": 0.0001913142972625792, "g_entropy": 0.6929287910461426, "tau_bias": 0.6580710411071777, "beta": 0.3284165405978759, "train_val_corr": 0.0}
{"t": 19, "ts": 1788180392.3212068, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00020546684390865266, "advantage": -245.6633758544922, "attn_entropy": 4.527476266957819e-06, "load_balance": 0.00018925733165815474, "g_entropy": 0.692996883392334, "tau_bias": 0.6580292582511902, "beta": 0.3404283958176772, "train_val_corr": 0.9999999999999999}
{"t": 20, "ts": 1788180396.7078762, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.06389246135950089, "advantage": -238.0048828125, "attn_entropy": 4.538290158961899e-06, "load_balance": 0.00018892630469053985, "g_entropy": 0.6929916858673095, "tau_bias": 0.6589157581329346, "beta": 0.3366199272374312, "train_val_corr": 1.0}
{"t": 21, "ts": 1788180401.1271398, "step": 210, "action": "ATTN_META_ONLY", "entropy_pi": 0.6931471824645996, "advantage": -222.07473754882812, "attn_entropy": 9.002384899758909e-08, "load_balance": 0.00019003715133294463, "g_entropy": 0.692950439453125, "tau_bias": 0.6511913537979126, "beta": 0.27715980758269626, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788180404.287595, "step": 220, "action": "ATTN_META_ONLY", "entropy_pi": 0.5972248911857605, "advantage": -204.85134887695312, "attn_entropy": 9.553009476803709e-08, "load_balance": 0.00018931220984086394, "g_entropy": 0.6929831504821777, "tau_bias": 0.6511913537979126, "beta": 0.3474626752237479, "train_val_corr": 0.9999999999999998}
{"t": 23, "ts": 1788180407.4726362, "step": 230, "action": "ATTN_META_ONLY", "entropy_pi": 0.5818414688110352, "advantage": -194.49822998046875, "attn_entropy": 5.613042958430014e-07, "load_balance": 0.00018943242030218245, "g_entropy": 0.6929885387420655, "tau_bias": 0.6511913537979126, "beta": 0.32135777920484543, "train_val_corr": 1.0}
{"t": 24, "ts": 1788180417.3131802, "step": 240, "action": "ATTN_META_ONLY", "entropy_pi": 0.5930978059768677, "advantage": -186.23284912109375, "attn_entropy": 5.014954695070628e-07, "load_balance": 0.00019048333633691071, "g_entropy": 0.692985725402832, "tau_bias": 0.6511913537979126, "beta": 0.29831817125280696, "train_val_corr": 0.9999999999999999}
{"t": 25, "ts": 1788180420.5826907, "step": 250, "action": "ATTN_META_ONLY", "entropy_pi": 0.5730817914009094, "advantage": -179.33053588867188, "attn_entropy": 3.277754512964748e-06, "load_balance": 0.00018928240751847625, "g_entropy": 0.6929839134216309, "tau_bias": 0.6511913537979126, "beta": 0.3446580295761426, "train_val_corr": 0.9999999999999998}
{"t": 26, "ts": 1788180423.5817406, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.553450882434845, "advantage": -169.48361206054688, "attn_entropy": 6.246070370252709e-07, "load_balance": 0.00018884340533986688, "g_entropy": 0.6929815292358399, "tau_bias": 0.6511913537979126, "beta": 0.3377867278953393, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788180426.5769534, "step": 270, "action": "ATTN_META_ONLY", "entropy_pi": 0.49316924810409546, "advantage": -162.86090087890625, "attn_entropy": 2.710553235374391e-06, "load_balance": 0.00018987712683156134, "g_entropy": 0.6929412364959717, "tau_bias": 0.6511913537979126, "beta": 0.3291461293896039, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788180429.5768483, "step": 280, "action": "ATTN_META_ONLY", "entropy_pi": 0.4448728859424591, "advantage": -155.6215362548828, "attn_entropy": 8.275545042124577e-07, "load_balance": 0.0001885690726339817, "g_entropy": 0.6929808616638183, "tau_bias": 0.6511913537979126, "beta": 0.3800113337735335, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788180468.8190062, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.7111105939402478e-06, "advantage": 2823.092529296875, "attn_entropy": 1.1467038631439208, "load_balance": 0.0003875201684422791, "g_entropy": 1.3860989093780518, "tau_bias": 0.6537609100341797, "beta": 0.28122237386802834, "train_val_corr": 0.0}
{"t": 30, "ts": 1788180476.285414, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0804937150842306e-07, "advantage": -52.202117919921875, "attn_entropy": 1.2139817237854005, "load_balance": 0.0003864346770569682, "g_entropy": 1.3860869884490967, "tau_bias": 0.6528395414352417, "beta": 0.315460500617822, "train_val_corr": 0.0}
{"t": 31, "ts": 1788180483.0140727, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723675220324367e-07, "advantage": -92.25147247314453, "attn_entropy": 1.0194334030151366, "load_balance": 0.00038587030721828344, "g_entropy": 1.3860511302947998, "tau_bias": 0.6521405577659607, "beta": 0.36765315135320026, "train_val_corr": 0.0}
{"t": 32, "ts": 1788180489.9709141, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723307159187243e-07, "advantage": -52.54505157470703, "attn_entropy": 0.8622724294662476, "load_balance": 0.00038400887278839947, "g_entropy": 1.3860588073730469, "tau_bias": 0.6513991355895996, "beta": 0.3685278246800105, "train_val_corr": 0.0}
{"t": 33, "ts": 1788180496.5687256, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723372529118933e-07, "advantage": 73.63611602783203, "attn_entropy": 0.7935564517974854, "load_balance": 0.0003858341835439205, "g_entropy": 1.3860466003417968, "tau_bias": 0.6501175761222839, "beta": 0.3873196666439374, "train_val_corr": 0.0}
{"t": 34, "ts": 1788180503.3792517, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723273053135927e-07, "advantage": -78.9076156616211, "attn_entropy": 0.9198231220245361, "load_balance": 0.0003853273228742182, "g_entropy": 1.3860178470611573, "tau_bias": 0.6546578407287598, "beta": 0.44540395339330036, "train_val_corr": 0.0}
{"t": 35, "ts": 1788180537.2161894, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.483715057373047, "attn_entropy": 1.0170397758483887, "load_balance": 0.0003834606148302555, "g_entropy": 1.3858046054840087, "tau_bias": 0.6483941078186035, "beta": 0.4238623169561227, "train_val_corr": 0.0}
{"t": 36, "ts": 1788180544.0274181, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -62.88734436035156, "attn_entropy": 0.992647361755371, "load_balance": 0.0003868126077577472, "g_entropy": 1.3861027240753174, "tau_bias": 0.6580358147621155, "beta": 0.4552500993013382, "train_val_corr": 0.0}
{"t": 37, "ts": 1788180550.693961, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.497371673583984, "attn_entropy": 1.2815523624420166, "load_balance": 0.00038783723721280694, "g_entropy": 1.3860494136810302, "tau_bias": 0.6530374884605408, "beta": 0.37658577039837837, "train_val_corr": 0.0}
{"t": 38, "ts": 1788180557.3691542, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.02465057373047, "attn_entropy": 1.3841375827789306, "load_balance": 0.00039073077496141193, "g_entropy": 1.386082077026367, "tau_bias": 0.6582708954811096, "beta": 0.4402708299458027, "train_val_corr": 0.0}
{"t": 39, "ts": 1788180564.78106, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -60.8590202331543, "attn_entropy": 1.2579901695251465, "load_balance": 0.0003892698558047414, "g_entropy": 1.386056613922119, "tau_bias": 0.6567427515983582, "beta": 0.42502398292223614, "train_val_corr": 0.0}
{"t": 40, "ts": 1788180571.7040577, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -40.03256607055664, "attn_entropy": 0.7999596238136292, "load_balance": 0.0003873741603456438, "g_entropy": 1.386070966720581, "tau_bias": 0.6584932804107666, "beta": 0.45767082646489143, "train_val_corr": 0.0}
{"t": 41, "ts": 1788180607.4078155, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 31.22900390625, "attn_entropy": 0.690041446685791, "load_balance": 0.00038456370821222664, "g_entropy": 1.3860507488250733, "tau_bias": 0.6517192125320435, "beta": 0.4073665762941043, "train_val_corr": 0.0}
{"t": 42, "ts": 1788180614.3022318, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -102.52987670898438, "attn_entropy": 0.7313535749912262, "load_balance": 0.00038726163329556585, "g_entropy": 1.3860310554504394, "tau_bias": 0.6576678156852722, "beta": 0.4638050322731336, "train_val_corr": 0.0}
{"t": 43, "ts": 1788180621.603133, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.34838104248047, "attn_entropy": 0.7388055801391602, "load_balance": 0.0003860935918055475, "g_entropy": 1.3860531330108643, "tau_bias": 0.6435304284095764, "beta": 0.4440668386717637, "train_val_corr": 0.0}
{"t": 44, "ts": 1788180628.4387686, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.48822784423828, "attn_entropy": 0.8262648344039917, "load_balance": 0.00038715751143172385, "g_entropy": 1.3860721588134766, "tau_bias": 0.647079586982727, "beta": 0.4285246891280015, "train_val_corr": 0.0}
{"t": 45, "ts": 1788180635.164547, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 177.44378662109375, "attn_entropy": 1.3300235271453857, "load_balance": 0.00039244736544787886, "g_entropy": 1.3859805583953857, "tau_bias": 0.6612085700035095, "beta": 0.44490248585740727, "train_val_corr": 0.0}
{"t": 46, "ts": 1788180641.8194363, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.48180630803108215, "attn_entropy": 1.3008524894714355, "load_balance": 0.00039121920708566903, "g_entropy": 1.386037826538086, "tau_bias": 0.6609336733818054, "beta": 0.45555945858359337, "train_val_corr": 0.0}
{"t": 47, "ts": 1788180675.4821248, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -118.8721694946289, "attn_entropy": 0.9650171995162964, "load_balance": 0.0003889179090037942, "g_entropy": 1.386042833328247, "tau_bias": 0.6649660468101501, "beta": 0.4843709555764993, "train_val_corr": 0.0}
{"t": 48, "ts": 1788180682.2735608, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.1336281299591064, "attn_entropy": 0.8375222206115722, "load_balance": 0.0003870803629979491, "g_entropy": 1.3860376834869386, "tau_bias": 0.6551528573036194, "beta": 0.446382075548172, "train_val_corr": 1.0}
{"t": 49, "ts": 1788180689.4106786, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.403282165527344, "attn_entropy": 0.80538489818573, "load_balance": 0.0003865889389999211, "g_entropy": 1.3860787868499755, "tau_bias": 0.6540045142173767, "beta": 0.4312798082828522, "train_val_corr": 1.0}
{"t": 50, "ts": 1788180696.4972866, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -85.25262451171875, "attn_entropy": 0.6666055262088776, "load_balance": 0.000386626529507339, "g_entropy": 1.3860852241516113, "tau_bias": 0.6562986373901367, "beta": 0.47641928618152934, "train_val_corr": 1.0}
{"t": 51, "ts": 1788180703.3164613, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -66.78575897216797, "attn_entropy": 0.6379572212696075, "load_balance": 0.00038643478183075786, "g_entropy": 1.3860663414001464, "tau_bias": 0.6517891883850098, "beta": 0.4689716349045436, "train_val_corr": 0.9999999999999996}
{"t": 52, "ts": 1788180710.1354675, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -85.94789123535156, "attn_entropy": 0.6532553702592849, "load_balance": 0.0003869178472086787, "g_entropy": 1.3860525131225585, "tau_bias": 0.6494863033294678, "beta": 0.4667714089155197, "train_val_corr": 0.9999999999999998}
{"t": 53, "ts": 1788180746.066376, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -82.49783325195312, "attn_entropy": 0.6495461642742157, "load_balance": 0.00038553866324946283, "g_entropy": 1.3860645771026612, "tau_bias": 0.6527930498123169, "beta": 0.48418660337726277, "train_val_corr": 1.0}
{"t": 54, "ts": 1788180752.9904606, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -74.96221923828125, "attn_entropy": 0.670500323176384, "load_balance": 0.000386599637567997, "g_entropy": 1.3860748767852784, "tau_bias": 0.6564632654190063, "beta": 0.4908122147123019, "train_val_corr": 1.0}
{"t": 55, "ts": 1788180759.8221307, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -72.8960189819336, "attn_entropy": 0.6721865922212601, "load_balance": 0.00038594079669564965, "g_entropy": 1.386042881011963, "tau_bias": 0.6632264852523804, "beta": 0.49953463673591614, "train_val_corr": 1.0}
{"t": 56, "ts": 1788180766.521871, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -71.35345458984375, "attn_entropy": 0.6827040255069733, "load_balance": 0.00038745211204513905, "g_entropy": 1.3860157489776612, "tau_bias": 0.6650304198265076, "beta": 0.4997771630684535, "train_val_corr": 0.9999999999999998}
{"t": 57, "ts": 1788180773.3024707, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -69.78540802001953, "attn_entropy": 0.7005118787288666, "load_balance": 0.0003866377635858953, "g_entropy": 1.3859135627746582, "tau_bias": 0.663004994392395, "beta": 0.5068729917208353, "train_val_corr": 1.0}
{"t": 58, "ts": 1788180808.6957333, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.259042739868164, "attn_entropy": 0.6995348304510116, "load_balance": 0.0003872996661812067, "g_entropy": 1.3860538482666016, "tau_bias": 0.659912109375, "beta": 0.5111945693691572, "train_val_corr": 1.0}
{"t": 59, "ts": 1788180815.386072, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -65.78572845458984, "attn_entropy": 0.6964557528495788, "load_balance": 0.00038676962722092865, "g_entropy": 1.386076307296753, "tau_bias": 0.6573927998542786, "beta": 0.5209370677669843, "train_val_corr": 1.0}
{"t": 60, "ts": 1788180822.8992913, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -60.547691345214844, "attn_entropy": 0.6947936654090882, "load_balance": 0.00038708162028342483, "g_entropy": 1.3860570430755614, "tau_bias": 0.6526563167572021, "beta": 0.5220783799886703, "train_val_corr": 1.0}
{"t": 61, "ts": 1788180830.0167172, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -57.644935607910156, "attn_entropy": 0.692501300573349, "load_balance": 0.0003870142041705549, "g_entropy": 1.3861083030700683, "tau_bias": 0.6513410806655884, "beta": 0.520176554719607, "train_val_corr": 1.0}
{"t": 62, "ts": 1788180836.8301506, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -57.305091857910156, "attn_entropy": 0.6818742871284484, "load_balance": 0.0003867293242365122, "g_entropy": 1.3860894680023192, "tau_bias": 0.6508477926254272, "beta": 0.5187370851635933, "train_val_corr": 1.0}
{"t": 63, "ts": 1788180843.7055802, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -55.573089599609375, "attn_entropy": 0.69111770093441, "load_balance": 0.00038720408920198677, "g_entropy": 1.3860187530517578, "tau_bias": 0.6511348485946655, "beta": 0.521108532945315, "train_val_corr": 1.0}
{"t": 0, "ts": 1788180962.5987606, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12832.8828125, "attn_entropy": 0.6930617809295654, "load_balance": 0.00019552120938897132, "g_entropy": 0.6929812431335449, "tau_bias": 0.6566849946975708, "beta": 0.10835317894816399, "train_val_corr": 0.0}
{"t": 1, "ts": 1788180967.0119135, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.073445431160508e-06, "advantage": -587.4507446289062, "attn_entropy": 0.666393232345581, "load_balance": 0.00019712578505277635, "g_entropy": 0.6929786205291748, "tau_bias": 0.6658554673194885, "beta": 0.22102038655430079, "train_val_corr": 0.0}
{"t": 2, "ts": 1788180971.4971368, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.004785797093063593, "advantage": -573.068115234375, "attn_entropy": 0.5410383701324463, "load_balance": 0.00019967956468462943, "g_entropy": 0.6929838180541992, "tau_bias": 0.6756224036216736, "beta": 0.2527519352734089, "train_val_corr": 0.0}
{"t": 3, "ts": 1788180976.0047896, "step": 30, "action": "PGO_UPDATE", "entropy_pi": 0.6779966354370117, "advantage": -546.783935546875, "attn_entropy": 0.44888620376586913, "load_balance": 0.00020008718129247428, "g_entropy": 0.6929582595825196, "tau_bias": 0.67994624376297, "beta": 0.2310744859278202, "train_val_corr": 0.0}
{"t": 4, "ts": 1788180980.4266865, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.6617884635925293, "advantage": -517.9473266601562, "attn_entropy": 0.39251065254211426, "load_balance": 0.00019728613551706076, "g_entropy": 0.6929805278778076, "tau_bias": 0.6816269159317017, "beta": 0.27597904826203984, "train_val_corr": 0.0}
{"t": 5, "ts": 1788180985.1756897, "step": 50, "action": "PGO_UPDATE", "entropy_pi": 0.24778594076633453, "advantage": -494.4080505371094, "attn_entropy": 0.36379902362823485, "load_balance": 0.00019671940244734288, "g_entropy": 0.6929544925689697, "tau_bias": 0.6823055148124695, "beta": 0.29740134999156, "train_val_corr": 0.0}
{"t": 6, "ts": 1788180997.378036, "step": 60, "action": "PGO_UPDATE", "entropy_pi": 0.29927805066108704, "advantage": -470.0270080566406, "attn_entropy": 0.3934603214263916, "load_balance": 0.00020074844360351562, "g_entropy": 0.6929795742034912, "tau_bias": 0.6827506422996521, "beta": 0.2810777711371581, "train_val_corr": 0.0}
{"t": 7, "ts": 1788181002.1658614, "step": 70, "action": "PGO_UPDATE", "entropy_pi": 0.7770779132843018, "advantage": -446.60418701171875, "attn_entropy": 0.4484855651855469, "load_balance": 0.00019895108416676522, "g_entropy": 0.6929839611053467, "tau_bias": 0.6825190782546997, "beta": 0.26477867054442566, "train_val_corr": 0.0}
{"t": 8, "ts": 1788181006.4823895, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.1880374699831009, "advantage": -426.3194274902344, "attn_entropy": 0.39160192012786865, "load_balance": 0.00019413278205320238, "g_entropy": 0.6929774761199952, "tau_bias": 0.6820326447486877, "beta": 0.2703086876620849, "train_val_corr": 0.0}
{"t": 9, "ts": 1788181010.7081163, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.230955570936203, "advantage": -405.2984924316406, "attn_entropy": 0.4175122261047363, "load_balance": 0.00019732958171516656, "g_entropy": 0.6929654598236084, "tau_bias": 0.6827709078788757, "beta": 0.3177870971461137, "train_val_corr": 0.0}
{"t": 10, "ts": 1788181014.8360596, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.7187275886535645, "advantage": -384.1015319824219, "attn_entropy": 0.4076129913330078, "load_balance": 0.0002041323808953166, "g_entropy": 0.6929611206054688, "tau_bias": 0.6806187033653259, "beta": 0.32577276478211087, "train_val_corr": 0.0}
{"t": 11, "ts": 1788181019.0247812, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": -364.6973571777344, "attn_entropy": 0.005328371003270149, "load_balance": 0.00018955166451632975, "g_entropy": 0.6929620742797852, "tau_bias": 0.6649991869926453, "beta": 0.21581758294875422, "train_val_corr": 0.0}
{"t": 12, "ts": 1788181029.8152947, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723294369418e-07, "advantage": -342.6893005371094, "attn_entropy": 5.966397002339363e-05, "load_balance": 0.00018938431749120354, "g_entropy": 0.6929804801940918, "tau_bias": 0.6607851982116699, "beta": 0.25734658042589825, "train_val_corr": 0.0}
{"t": 13, "ts": 1788181034.0042832, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072378180173473e-07, "advantage": -329.51434326171875, "attn_entropy": 8.862389950081707e-06, "load_balance": 0.00018968143267557026, "g_entropy": 0.6929698467254639, "tau_bias": 0.6591408848762512, "beta": 0.3110854346305132, "train_val_corr": 0.0}
{"t": 14, "ts": 1788181038.0081139, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0730206529151474e-07, "advantage": -315.81744384765625, "attn_entropy": 3.9238999306689944e-06, "load_balance": 0.00019007195951417088, "g_entropy": 0.6929803371429444, "tau_bias": 0.6584917306900024, "beta": 0.23725857057919106, "train_val_corr": 0.0}
{"t": 15, "ts": 1788181042.1986663, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0800375466478727e-07, "advantage": -301.9831237792969, "attn_entropy": 5.39456814294681e-06, "load_balance": 0.00019004230853170156, "g_entropy": 0.6929697036743164, "tau_bias": 0.6582355499267578, "beta": 0.2594692458709081, "train_val_corr": 0.0}
{"t": 16, "ts": 1788181046.3149972, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1552480689024378e-07, "advantage": -285.3312683105469, "attn_entropy": 4.720315337181091e-06, "load_balance": 0.00018992588156834244, "g_entropy": 0.6929097175598145, "tau_bias": 0.6581416726112366, "beta": 0.31782514974474907, "train_val_corr": 0.0}
{"t": 17, "ts": 1788181050.7344873, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.3428270285185135e-07, "advantage": -273.80621337890625, "attn_entropy": 6.298582593444735e-06, "load_balance": 0.0001902130083180964, "g_entropy": 0.6929590702056885, "tau_bias": 0.658105731010437, "beta": 0.2834706449260314, "train_val_corr": 0.0}
{"t": 18, "ts": 1788181062.4648132, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.466907517373329e-06, "advantage": -263.6184997558594, "attn_entropy": 8.867547148838639e-06, "load_balance": 0.0001913142972625792, "g_entropy": 0.6929287910461426, "tau_bias": 0.6580710411071777, "beta": 0.3284165405978759, "train_val_corr": 0.0}
{"t": 19, "ts": 1788181066.728522, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00020546684390865266, "advantage": -245.6633758544922, "attn_entropy": 4.527476266957819e-06, "load_balance": 0.00018925733165815474, "g_entropy": 0.692996883392334, "tau_bias": 0.6580292582511902, "beta": 0.3404283958176772, "train_val_corr": 0.9999999999999999}
{"t": 20, "ts": 1788181071.2129822, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.06389246135950089, "advantage": -238.0048828125, "attn_entropy": 4.538290158961899e-06, "load_balance": 0.00018892630469053985, "g_entropy": 0.6929916858673095, "tau_bias": 0.6589157581329346, "beta": 0.3366199272374312, "train_val_corr": 1.0}
{"t": 21, "ts": 1788181075.6160014, "step": 210, "action": "ATTN_META_ONLY", "entropy_pi": 0.6931471824645996, "advantage": -222.07473754882812, "attn_entropy": 9.002384899758909e-08, "load_balance": 0.00019003715133294463, "g_entropy": 0.692950439453125, "tau_bias": 0.6511913537979126, "beta": 0.27715980758269626, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788181078.602463, "step": 220, "action": "ATTN_META_ONLY", "entropy_pi": 0.5972248911857605, "advantage": -204.85134887695312, "attn_entropy": 9.553009476803709e-08, "load_balance": 0.00018931220984086394, "g_entropy": 0.6929831504821777, "tau_bias": 0.6511913537979126, "beta": 0.3474626752237479, "train_val_corr": 0.9999999999999998}
{"t": 23, "ts": 1788181081.7118802, "step": 230, "action": "ATTN_META_ONLY", "entropy_pi": 0.5818414688110352, "advantage": -194.49822998046875, "attn_entropy": 5.613042958430014e-07, "load_balance": 0.00018943242030218245, "g_entropy": 0.6929885387420655, "tau_bias": 0.6511913537979126, "beta": 0.32135777920484543, "train_val_corr": 1.0}
{"t": 24, "ts": 1788181091.4860785, "step": 240, "action": "ATTN_META_ONLY", "entropy_pi": 0.5930978059768677, "advantage": -186.23284912109375, "attn_entropy": 5.014954695070628e-07, "load_balance": 0.00019048333633691071, "g_entropy": 0.692985725402832, "tau_bias": 0.6511913537979126, "beta": 0.29831817125280696, "train_val_corr": 0.9999999999999999}
{"t": 25, "ts": 1788181094.6189382, "step": 250, "action": "ATTN_META_ONLY", "entropy_pi": 0.5730817914009094, "advantage": -179.33053588867188, "attn_entropy": 3.277754512964748e-06, "load_balance": 0.00018928240751847625, "g_entropy": 0.6929839134216309, "tau_bias": 0.6511913537979126, "beta": 0.3446580295761426, "train_val_corr": 0.9999999999999998}
{"t": 26, "ts": 1788181097.8061285, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.553450882434845, "advantage": -169.48361206054688, "attn_entropy": 6.246070370252709e-07, "load_balance": 0.00018884340533986688, "g_entropy": 0.6929815292358399, "tau_bias": 0.6511913537979126, "beta": 0.3377867278953393, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788181100.9140992, "step": 270, "action": "ATTN_META_ONLY", "entropy_pi": 0.49316924810409546, "advantage": -162.86090087890625, "attn_entropy": 2.710553235374391e-06, "load_balance": 0.00018987712683156134, "g_entropy": 0.6929412364959717, "tau_bias": 0.6511913537979126, "beta": 0.3291461293896039, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788181104.1257718, "step": 280, "action": "ATTN_META_ONLY", "entropy_pi": 0.4448728859424591, "advantage": -155.6215362548828, "attn_entropy": 8.275545042124577e-07, "load_balance": 0.0001885690726339817, "g_entropy": 0.6929808616638183, "tau_bias": 0.6511913537979126, "beta": 0.3800113337735335, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788181139.9935527, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.7111105939402478e-06, "advantage": 2823.092529296875, "attn_entropy": 1.1467038631439208, "load_balance": 0.0003875201684422791, "g_entropy": 1.3860989093780518, "tau_bias": 0.6537609100341797, "beta": 0.28122237386802834, "train_val_corr": 0.0}
{"t": 30, "ts": 1788181146.4331534, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0804937150842306e-07, "advantage": -52.202117919921875, "attn_entropy": 1.2139817237854005, "load_balance": 0.0003864346770569682, "g_entropy": 1.3860869884490967, "tau_bias": 0.6528395414352417, "beta": 0.315460500617822, "train_val_corr": 0.0}
{"t": 31, "ts": 1788181153.4329638, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723675220324367e-07, "advantage": -92.25147247314453, "attn_entropy": 1.0194334030151366, "load_balance": 0.00038587030721828344, "g_entropy": 1.3860511302947998, "tau_bias": 0.6521405577659607, "beta": 0.36765315135320026, "train_val_corr": 0.0}
{"t": 32, "ts": 1788181160.3746066, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723307159187243e-07, "advantage": -52.54505157470703, "attn_entropy": 0.8622724294662476, "load_balance": 0.00038400887278839947, "g_entropy": 1.3860588073730469, "tau_bias": 0.6513991355895996, "beta": 0.3685278246800105, "train_val_corr": 0.0}
{"t": 33, "ts": 1788181167.312709, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723372529118933e-07, "advantage": 73.63611602783203, "attn_entropy": 0.7935564517974854, "load_balance": 0.0003858341835439205, "g_entropy": 1.3860466003417968, "tau_bias": 0.6501175761222839, "beta": 0.3873196666439374, "train_val_corr": 0.0}
{"t": 34, "ts": 1788181174.119088, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723273053135927e-07, "advantage": -78.9076156616211, "attn_entropy": 0.9198231220245361, "load_balance": 0.0003853273228742182, "g_entropy": 1.3860178470611573, "tau_bias": 0.6546578407287598, "beta": 0.44540395339330036, "train_val_corr": 0.0}
{"t": 35, "ts": 1788181203.7806356, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.483715057373047, "attn_entropy": 1.0170397758483887, "load_balance": 0.0003834606148302555, "g_entropy": 1.3858046054840087, "tau_bias": 0.6483941078186035, "beta": 0.4238623169561227, "train_val_corr": 0.0}
{"t": 36, "ts": 1788181210.3318574, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -62.88734436035156, "attn_entropy": 0.992647361755371, "load_balance": 0.0003868126077577472, "g_entropy": 1.3861027240753174, "tau_bias": 0.6580358147621155, "beta": 0.4552500993013382, "train_val_corr": 0.0}
{"t": 37, "ts": 1788181216.7975512, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.497371673583984, "attn_entropy": 1.2815523624420166, "load_balance": 0.00038783723721280694, "g_entropy": 1.3860494136810302, "tau_bias": 0.6530374884605408, "beta": 0.37658577039837837, "train_val_corr": 0.0}
{"t": 38, "ts": 1788181223.5194983, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.02465057373047, "attn_entropy": 1.3841375827789306, "load_balance": 0.00039073077496141193, "g_entropy": 1.386082077026367, "tau_bias": 0.6582708954811096, "beta": 0.4402708299458027, "train_val_corr": 0.0}
{"t": 39, "ts": 1788181229.9876144, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -60.8590202331543, "attn_entropy": 1.2579901695251465, "load_balance": 0.0003892698558047414, "g_entropy": 1.386056613922119, "tau_bias": 0.6567427515983582, "beta": 0.42502398292223614, "train_val_corr": 0.0}
{"t": 40, "ts": 1788181237.2979107, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -40.03256607055664, "attn_entropy": 0.7999596238136292, "load_balance": 0.0003873741603456438, "g_entropy": 1.386070966720581, "tau_bias": 0.6584932804107666, "beta": 0.45767082646489143, "train_val_corr": 0.0}
{"t": 41, "ts": 1788181267.009961, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 31.22900390625, "attn_entropy": 0.690041446685791, "load_balance": 0.00038456370821222664, "g_entropy": 1.3860507488250733, "tau_bias": 0.6517192125320435, "beta": 0.4073665762941043, "train_val_corr": 0.0}
{"t": 42, "ts": 1788181273.820441, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -102.52987670898438, "attn_entropy": 0.7313535749912262, "load_balance": 0.00038726163329556585, "g_entropy": 1.3860310554504394, "tau_bias": 0.6576678156852722, "beta": 0.4638050322731336, "train_val_corr": 0.0}
{"t": 43, "ts": 1788181280.4881577, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.34838104248047, "attn_entropy": 0.7388055801391602, "load_balance": 0.0003860935918055475, "g_entropy": 1.3860531330108643, "tau_bias": 0.6435304284095764, "beta": 0.4440668386717637, "train_val_corr": 0.0}
{"t": 44, "ts": 1788181287.567813, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.48822784423828, "attn_entropy": 0.8262648344039917, "load_balance": 0.00038715751143172385, "g_entropy": 1.3860721588134766, "tau_bias": 0.647079586982727, "beta": 0.4285246891280015, "train_val_corr": 0.0}
{"t": 45, "ts": 1788181294.8325405, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 177.44378662109375, "attn_entropy": 1.3300235271453857, "load_balance": 0.00039244736544787886, "g_entropy": 1.3859805583953857, "tau_bias": 0.6612085700035095, "beta": 0.44490248585740727, "train_val_corr": 0.0}
{"t": 46, "ts": 1788181301.52564, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.48180630803108215, "attn_entropy": 1.3008524894714355, "load_balance": 0.00039121920708566903, "g_entropy": 1.386037826538086, "tau_bias": 0.6609336733818054, "beta": 0.45555945858359337, "train_val_corr": 0.0}
{"t": 47, "ts": 1788181331.3879294, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -119.00300598144531, "attn_entropy": 0.9636626482009888, "load_balance": 0.00038892420707270503, "g_entropy": 1.386042833328247, "tau_bias": 0.6648950576782227, "beta": 0.4841971198717753, "train_val_corr": 0.0}
{"t": 48, "ts": 1788181338.0880053, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.019122123718262, "attn_entropy": 0.8329877853393555, "load_balance": 0.0003870337619446218, "g_entropy": 1.3860370635986328, "tau_bias": 0.6563909649848938, "beta": 0.4351421669125557, "train_val_corr": 0.9999999999999998}
{"t": 49, "ts": 1788181344.4794428, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.318355560302734, "attn_entropy": 0.7638442754745484, "load_balance": 0.00038633702788501976, "g_entropy": 1.3860764026641845, "tau_bias": 0.6588214635848999, "beta": 0.40662727256615955, "train_val_corr": 1.0}
{"t": 50, "ts": 1788181351.2024243, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -83.91835021972656, "attn_entropy": 0.6389047205448151, "load_balance": 0.0003862491576001048, "g_entropy": 1.386084508895874, "tau_bias": 0.6581639051437378, "beta": 0.4637837919096152, "train_val_corr": 0.9999999999999998}
{"t": 51, "ts": 1788181357.9227135, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -92.51158905029297, "attn_entropy": 0.6223276764154434, "load_balance": 0.0003862282377667725, "g_entropy": 1.3860651016235352, "tau_bias": 0.6540132164955139, "beta": 0.4687024677793185, "train_val_corr": 0.9999999999999996}
{"t": 52, "ts": 1788181364.4345753, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.67906951904297, "attn_entropy": 0.6519095420837402, "load_balance": 0.0003867620136588812, "g_entropy": 1.386051607131958, "tau_bias": 0.651768684387207, "beta": 0.4754183702170849, "train_val_corr": 0.9999999999999996}
{"t": 53, "ts": 1788181396.1784322, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.32022094726562, "attn_entropy": 0.6802802562713623, "load_balance": 0.00038581886328756807, "g_entropy": 1.3860649585723877, "tau_bias": 0.6519516706466675, "beta": 0.47642258430520695, "train_val_corr": 0.9999999999999996}
{"t": 54, "ts": 1788181402.6768174, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -82.2474136352539, "attn_entropy": 0.7060110807418823, "load_balance": 0.00038695065304636953, "g_entropy": 1.3860758781433105, "tau_bias": 0.6547136306762695, "beta": 0.48182273656129837, "train_val_corr": 1.0}
{"t": 55, "ts": 1788181409.2642272, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -79.18009185791016, "attn_entropy": 0.6998830050230026, "load_balance": 0.00038624278968200085, "g_entropy": 1.3860454082489013, "tau_bias": 0.6585575342178345, "beta": 0.49181029200553894, "train_val_corr": 1.0}
{"t": 56, "ts": 1788181415.7781847, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -75.70050811767578, "attn_entropy": 0.6933891326189041, "load_balance": 0.0003875557682476938, "g_entropy": 1.3860198497772216, "tau_bias": 0.6578031778335571, "beta": 0.49514268959561986, "train_val_corr": 1.0}
{"t": 57, "ts": 1788181423.080064, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -72.32968139648438, "attn_entropy": 0.7106742113828659, "load_balance": 0.0003866579849272966, "g_entropy": 1.3859189987182616, "tau_bias": 0.6569044589996338, "beta": 0.49980781599879265, "train_val_corr": 1.0}
{"t": 58, "ts": 1788181454.4286501, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -68.89099884033203, "attn_entropy": 0.7053763300180436, "load_balance": 0.00038722306489944456, "g_entropy": 1.3860556602478027, "tau_bias": 0.6572539210319519, "beta": 0.5031577323873838, "train_val_corr": 1.0}
{"t": 59, "ts": 1788181461.2845047, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -67.31574249267578, "attn_entropy": 0.7040979772806167, "load_balance": 0.0003867311868816614, "g_entropy": 1.3860765457153321, "tau_bias": 0.6570943593978882, "beta": 0.5099475954969724, "train_val_corr": 1.0}
{"t": 60, "ts": 1788181468.5781481, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -63.466129302978516, "attn_entropy": 0.6965733766555786, "load_balance": 0.00038700145669281485, "g_entropy": 1.386056661605835, "tau_bias": 0.6534034609794617, "beta": 0.5144987627863884, "train_val_corr": 1.0}
{"t": 61, "ts": 1788181475.2994735, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -61.38798141479492, "attn_entropy": 0.7012678354978561, "load_balance": 0.00038704320322722194, "g_entropy": 1.3861083507537841, "tau_bias": 0.6513624787330627, "beta": 0.5198185766736666, "train_val_corr": 1.0}
{"t": 62, "ts": 1788181482.1841543, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -58.63174819946289, "attn_entropy": 0.6935345947742462, "load_balance": 0.0003868039231747389, "g_entropy": 1.3860896587371827, "tau_bias": 0.6498277187347412, "beta": 0.5169008001685143, "train_val_corr": 1.0}
{"t": 63, "ts": 1788181488.79455, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -55.636474609375, "attn_entropy": 0.6997991025447845, "load_balance": 0.0003872632049024105, "g_entropy": 1.38601975440979, "tau_bias": 0.6490833163261414, "beta": 0.5164594973127047, "train_val_corr": 0.9999999999999998}
{"t": 64, "ts": 1788181519.029886, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -53.589473724365234, "attn_entropy": 0.6953095078468323, "load_balance": 0.0003870141692459583, "g_entropy": 1.3860682487487792, "tau_bias": 0.6479621529579163, "beta": 0.5218959127863249, "train_val_corr": 1.0}
{"t": 65, "ts": 1788181525.6210594, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -50.9334602355957, "attn_entropy": 0.7131684720516205, "load_balance": 0.00038803506176918743, "g_entropy": 1.38602876663208, "tau_bias": 0.6478629112243652, "beta": 0.5177547136942545, "train_val_corr": 1.0}
{"t": 0, "ts": 1788181587.132801, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.7111105939402478e-06, "advantage": 2823.092529296875, "attn_entropy": 1.1467038631439208, "load_balance": 0.0003875201684422791, "g_entropy": 1.3860989093780518, "tau_bias": 0.6537609100341797, "beta": 0.28122237386802834, "train_val_corr": 0.0}
{"t": 1, "ts": 1788181593.9947407, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0804937150842306e-07, "advantage": -52.202117919921875, "attn_entropy": 1.2139817237854005, "load_balance": 0.0003864346770569682, "g_entropy": 1.3860869884490967, "tau_bias": 0.6528395414352417, "beta": 0.315460500617822, "train_val_corr": 0.0}
{"t": 2, "ts": 1788181600.3870573, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723675220324367e-07, "advantage": -92.25147247314453, "attn_entropy": 1.0194334030151366, "load_balance": 0.00038587030721828344, "g_entropy": 1.3860511302947998, "tau_bias": 0.6521405577659607, "beta": 0.36765315135320026, "train_val_corr": 0.0}
{"t": 3, "ts": 1788181606.9217782, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723307159187243e-07, "advantage": -52.54505157470703, "attn_entropy": 0.8622724294662476, "load_balance": 0.00038400887278839947, "g_entropy": 1.3860588073730469, "tau_bias": 0.6513991355895996, "beta": 0.3685278246800105, "train_val_corr": 0.0}
{"t": 4, "ts": 1788181613.4313183, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723372529118933e-07, "advantage": 73.63611602783203, "attn_entropy": 0.7935564517974854, "load_balance": 0.0003858341835439205, "g_entropy": 1.3860466003417968, "tau_bias": 0.6501175761222839, "beta": 0.3873196666439374, "train_val_corr": 0.0}
{"t": 5, "ts": 1788181620.485748, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723273053135927e-07, "advantage": -78.9076156616211, "attn_entropy": 0.9198231220245361, "load_balance": 0.0003853273228742182, "g_entropy": 1.3860178470611573, "tau_bias": 0.6546578407287598, "beta": 0.44540395339330036, "train_val_corr": 0.0}
{"t": 6, "ts": 1788181651.7942662, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.483715057373047, "attn_entropy": 1.0170397758483887, "load_balance": 0.0003834606148302555, "g_entropy": 1.3858046054840087, "tau_bias": 0.6483941078186035, "beta": 0.4238623169561227, "train_val_corr": 0.0}
{"t": 7, "ts": 1788181658.56678, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -62.88734436035156, "attn_entropy": 0.992647361755371, "load_balance": 0.0003868126077577472, "g_entropy": 1.3861027240753174, "tau_bias": 0.6580358147621155, "beta": 0.4552500993013382, "train_val_corr": 0.0}
{"t": 8, "ts": 1788181665.8153238, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.497371673583984, "attn_entropy": 1.2815523624420166, "load_balance": 0.00038783723721280694, "g_entropy": 1.3860494136810302, "tau_bias": 0.6530374884605408, "beta": 0.37658577039837837, "train_val_corr": 0.0}
{"t": 9, "ts": 1788181672.8797607, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.02465057373047, "attn_entropy": 1.3841375827789306, "load_balance": 0.00039073077496141193, "g_entropy": 1.386082077026367, "tau_bias": 0.6582708954811096, "beta": 0.4402708299458027, "train_val_corr": 0.0}
{"t": 10, "ts": 1788181679.484641, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -60.8590202331543, "attn_entropy": 1.2579901695251465, "load_balance": 0.0003892698558047414, "g_entropy": 1.386056613922119, "tau_bias": 0.6567427515983582, "beta": 0.42502398292223614, "train_val_corr": 0.0}
{"t": 11, "ts": 1788181686.2303932, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -40.03256607055664, "attn_entropy": 0.7999596238136292, "load_balance": 0.0003873741603456438, "g_entropy": 1.386070966720581, "tau_bias": 0.6584932804107666, "beta": 0.45767082646489143, "train_val_corr": 0.0}
{"t": 12, "ts": 1788181715.990117, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 31.22900390625, "attn_entropy": 0.690041446685791, "load_balance": 0.00038456370821222664, "g_entropy": 1.3860507488250733, "tau_bias": 0.6517192125320435, "beta": 0.4073665762941043, "train_val_corr": 0.0}
{"t": 13, "ts": 1788181722.515794, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -102.52987670898438, "attn_entropy": 0.7313535749912262, "load_balance": 0.00038726163329556585, "g_entropy": 1.3860310554504394, "tau_bias": 0.6576678156852722, "beta": 0.4638050322731336, "train_val_corr": 0.0}
{"t": 14, "ts": 1788181728.895259, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.34838104248047, "attn_entropy": 0.7388055801391602, "load_balance": 0.0003860935918055475, "g_entropy": 1.3860531330108643, "tau_bias": 0.6435304284095764, "beta": 0.4440668386717637, "train_val_corr": 0.0}
{"t": 15, "ts": 1788181735.4871275, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.48822784423828, "attn_entropy": 0.8262648344039917, "load_balance": 0.00038715751143172385, "g_entropy": 1.3860721588134766, "tau_bias": 0.647079586982727, "beta": 0.4285246891280015, "train_val_corr": 0.0}
{"t": 16, "ts": 1788181741.6242266, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 177.44378662109375, "attn_entropy": 1.3300235271453857, "load_balance": 0.00039244736544787886, "g_entropy": 1.3859805583953857, "tau_bias": 0.6612085700035095, "beta": 0.44490248585740727, "train_val_corr": 0.0}
{"t": 17, "ts": 1788181748.2317243, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.48180630803108215, "attn_entropy": 1.3008524894714355, "load_balance": 0.00039121920708566903, "g_entropy": 1.386037826538086, "tau_bias": 0.6609336733818054, "beta": 0.45555945858359337, "train_val_corr": 0.0}
{"t": 18, "ts": 1788181777.5301983, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -119.00300598144531, "attn_entropy": 0.9636626482009888, "load_balance": 0.00038892420707270503, "g_entropy": 1.386042833328247, "tau_bias": 0.6648950576782227, "beta": 0.4841971198717753, "train_val_corr": 0.0}
{"t": 19, "ts": 1788181784.2343764, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.019122123718262, "attn_entropy": 0.8329877853393555, "load_balance": 0.0003870337619446218, "g_entropy": 1.3860370635986328, "tau_bias": 0.6563909649848938, "beta": 0.4351421669125557, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788181790.7072644, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.318355560302734, "attn_entropy": 0.7638442754745484, "load_balance": 0.00038633702788501976, "g_entropy": 1.3860764026641845, "tau_bias": 0.6588214635848999, "beta": 0.40662727256615955, "train_val_corr": 1.0}
{"t": 21, "ts": 1788181797.5938187, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -83.91835021972656, "attn_entropy": 0.6389047205448151, "load_balance": 0.0003862491576001048, "g_entropy": 1.386084508895874, "tau_bias": 0.6581639051437378, "beta": 0.4637837919096152, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788181804.172767, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -92.51158905029297, "attn_entropy": 0.6223276764154434, "load_balance": 0.0003862282377667725, "g_entropy": 1.3860651016235352, "tau_bias": 0.6540132164955139, "beta": 0.4687024677793185, "train_val_corr": 0.9999999999999996}
{"t": 23, "ts": 1788181810.7250323, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.67906951904297, "attn_entropy": 0.6519095420837402, "load_balance": 0.0003867620136588812, "g_entropy": 1.386051607131958, "tau_bias": 0.651768684387207, "beta": 0.4754183702170849, "train_val_corr": 0.9999999999999996}
{"t": 0, "ts": 1788181980.1961086, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 3.7111105939402478e-06, "advantage": 2823.092529296875, "attn_entropy": 1.1467038631439208, "load_balance": 0.0003875201684422791, "g_entropy": 1.3860989093780518, "tau_bias": 0.6537609100341797, "beta": 0.28122237386802834, "train_val_corr": 0.0}
{"t": 1, "ts": 1788181986.974905, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0804937150842306e-07, "advantage": -52.202117919921875, "attn_entropy": 1.2139817237854005, "load_balance": 0.0003864346770569682, "g_entropy": 1.3860869884490967, "tau_bias": 0.6528395414352417, "beta": 0.315460500617822, "train_val_corr": 0.0}
{"t": 2, "ts": 1788181993.7355127, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723675220324367e-07, "advantage": -92.25147247314453, "attn_entropy": 1.0194334030151366, "load_balance": 0.00038587030721828344, "g_entropy": 1.3860511302947998, "tau_bias": 0.6521405577659607, "beta": 0.36765315135320026, "train_val_corr": 0.0}
{"t": 3, "ts": 1788182000.492941, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723307159187243e-07, "advantage": -52.54505157470703, "attn_entropy": 0.8622724294662476, "load_balance": 0.00038400887278839947, "g_entropy": 1.3860588073730469, "tau_bias": 0.6513991355895996, "beta": 0.3685278246800105, "train_val_corr": 0.0}
{"t": 4, "ts": 1788182007.5127418, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723372529118933e-07, "advantage": 73.63611602783203, "attn_entropy": 0.7935564517974854, "load_balance": 0.0003858341835439205, "g_entropy": 1.3860466003417968, "tau_bias": 0.6501175761222839, "beta": 0.3873196666439374, "train_val_corr": 0.0}
{"t": 5, "ts": 1788182014.6369178, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723273053135927e-07, "advantage": -78.9076156616211, "attn_entropy": 0.9198231220245361, "load_balance": 0.0003853273228742182, "g_entropy": 1.3860178470611573, "tau_bias": 0.6546578407287598, "beta": 0.44540395339330036, "train_val_corr": 0.0}
{"t": 6, "ts": 1788182046.9900131, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.483715057373047, "attn_entropy": 1.0170397758483887, "load_balance": 0.0003834606148302555, "g_entropy": 1.3858046054840087, "tau_bias": 0.6483941078186035, "beta": 0.4238623169561227, "train_val_corr": 0.0}
{"t": 7, "ts": 1788182053.988459, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -62.88734436035156, "attn_entropy": 0.992647361755371, "load_balance": 0.0003868126077577472, "g_entropy": 1.3861027240753174, "tau_bias": 0.6580358147621155, "beta": 0.4552500993013382, "train_val_corr": 0.0}
{"t": 8, "ts": 1788182060.982006, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.497371673583984, "attn_entropy": 1.2815523624420166, "load_balance": 0.00038783723721280694, "g_entropy": 1.3860494136810302, "tau_bias": 0.6530374884605408, "beta": 0.37658577039837837, "train_val_corr": 0.0}
{"t": 9, "ts": 1788182068.1813233, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.02465057373047, "attn_entropy": 1.3841375827789306, "load_balance": 0.00039073077496141193, "g_entropy": 1.386082077026367, "tau_bias": 0.6582708954811096, "beta": 0.4402708299458027, "train_val_corr": 0.0}
{"t": 10, "ts": 1788182074.6677015, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -60.8590202331543, "attn_entropy": 1.2579901695251465, "load_balance": 0.0003892698558047414, "g_entropy": 1.386056613922119, "tau_bias": 0.6567427515983582, "beta": 0.42502398292223614, "train_val_corr": 0.0}
{"t": 11, "ts": 1788182081.1058834, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -40.03256607055664, "attn_entropy": 0.7999596238136292, "load_balance": 0.0003873741603456438, "g_entropy": 1.386070966720581, "tau_bias": 0.6584932804107666, "beta": 0.45767082646489143, "train_val_corr": 0.0}
{"t": 12, "ts": 1788182110.6354423, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 31.22900390625, "attn_entropy": 0.690041446685791, "load_balance": 0.00038456370821222664, "g_entropy": 1.3860507488250733, "tau_bias": 0.6517192125320435, "beta": 0.4073665762941043, "train_val_corr": 0.0}
{"t": 13, "ts": 1788182117.5315197, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -102.52987670898438, "attn_entropy": 0.7313535749912262, "load_balance": 0.00038726163329556585, "g_entropy": 1.3860310554504394, "tau_bias": 0.6576678156852722, "beta": 0.4638050322731336, "train_val_corr": 0.0}
{"t": 14, "ts": 1788182124.7944493, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.34838104248047, "attn_entropy": 0.7388055801391602, "load_balance": 0.0003860935918055475, "g_entropy": 1.3860531330108643, "tau_bias": 0.6435304284095764, "beta": 0.4440668386717637, "train_val_corr": 0.0}
{"t": 15, "ts": 1788182131.7680185, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.48822784423828, "attn_entropy": 0.8262648344039917, "load_balance": 0.00038715751143172385, "g_entropy": 1.3860721588134766, "tau_bias": 0.647079586982727, "beta": 0.4285246891280015, "train_val_corr": 0.0}
{"t": 16, "ts": 1788182138.9342113, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 177.44378662109375, "attn_entropy": 1.3300235271453857, "load_balance": 0.00039244736544787886, "g_entropy": 1.3859805583953857, "tau_bias": 0.6612085700035095, "beta": 0.44490248585740727, "train_val_corr": 0.0}
{"t": 17, "ts": 1788182145.7870858, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.48180630803108215, "attn_entropy": 1.3008524894714355, "load_balance": 0.00039121920708566903, "g_entropy": 1.386037826538086, "tau_bias": 0.6609336733818054, "beta": 0.45555945858359337, "train_val_corr": 0.0}
{"t": 18, "ts": 1788182176.8645117, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -119.00300598144531, "attn_entropy": 0.9636626482009888, "load_balance": 0.00038892420707270503, "g_entropy": 1.386042833328247, "tau_bias": 0.6648950576782227, "beta": 0.4841971198717753, "train_val_corr": 0.0}
{"t": 19, "ts": 1788182183.9785142, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.019122123718262, "attn_entropy": 0.8329877853393555, "load_balance": 0.0003870337619446218, "g_entropy": 1.3860370635986328, "tau_bias": 0.6563909649848938, "beta": 0.4351421669125557, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788182190.9826815, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.318355560302734, "attn_entropy": 0.7638442754745484, "load_balance": 0.00038633702788501976, "g_entropy": 1.3860764026641845, "tau_bias": 0.6588214635848999, "beta": 0.40662727256615955, "train_val_corr": 1.0}
{"t": 21, "ts": 1788182197.9876726, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -83.91835021972656, "attn_entropy": 0.6389047205448151, "load_balance": 0.0003862491576001048, "g_entropy": 1.386084508895874, "tau_bias": 0.6581639051437378, "beta": 0.4637837919096152, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788182204.9651904, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -92.51158905029297, "attn_entropy": 0.6223276764154434, "load_balance": 0.0003862282377667725, "g_entropy": 1.3860651016235352, "tau_bias": 0.6540132164955139, "beta": 0.4687024677793185, "train_val_corr": 0.9999999999999996}
{"t": 23, "ts": 1788182211.8813293, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.67906951904297, "attn_entropy": 0.6519095420837402, "load_balance": 0.0003867620136588812, "g_entropy": 1.386051607131958, "tau_bias": 0.651768684387207, "beta": 0.4754183702170849, "train_val_corr": 0.9999999999999996}
{"t": 24, "ts": 1788182243.0782738, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.32022094726562, "attn_entropy": 0.6802802562713623, "load_balance": 0.00038581886328756807, "g_entropy": 1.3860649585723877, "tau_bias": 0.6519516706466675, "beta": 0.47642258430520695, "train_val_corr": 0.9999999999999996}
{"t": 25, "ts": 1788182250.0013862, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -82.2474136352539, "attn_entropy": 0.7060110807418823, "load_balance": 0.00038695065304636953, "g_entropy": 1.3860758781433105, "tau_bias": 0.6547136306762695, "beta": 0.48182273656129837, "train_val_corr": 1.0}
{"t": 26, "ts": 1788182256.7881286, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -79.18009185791016, "attn_entropy": 0.6998830050230026, "load_balance": 0.00038624278968200085, "g_entropy": 1.3860454082489013, "tau_bias": 0.6585575342178345, "beta": 0.49181029200553894, "train_val_corr": 1.0}
{"t": 27, "ts": 1788182263.8264654, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -75.70050811767578, "attn_entropy": 0.6933891326189041, "load_balance": 0.0003875557682476938, "g_entropy": 1.3860198497772216, "tau_bias": 0.6578031778335571, "beta": 0.49514268959561986, "train_val_corr": 1.0}
{"t": 28, "ts": 1788182270.510835, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -72.32968139648438, "attn_entropy": 0.7106742113828659, "load_balance": 0.0003866579849272966, "g_entropy": 1.3859189987182616, "tau_bias": 0.6569044589996338, "beta": 0.49980781599879265, "train_val_corr": 1.0}
{"t": 29, "ts": 1788182300.9852366, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -68.89099884033203, "attn_entropy": 0.7053763300180436, "load_balance": 0.00038722306489944456, "g_entropy": 1.3860556602478027, "tau_bias": 0.6572539210319519, "beta": 0.5031577323873838, "train_val_corr": 1.0}
{"t": 30, "ts": 1788182307.4248843, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -67.31574249267578, "attn_entropy": 0.7040979772806167, "load_balance": 0.0003867311868816614, "g_entropy": 1.3860765457153321, "tau_bias": 0.6570943593978882, "beta": 0.5099475954969724, "train_val_corr": 1.0}
{"t": 31, "ts": 1788182315.2936647, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -63.466129302978516, "attn_entropy": 0.6965733766555786, "load_balance": 0.00038700145669281485, "g_entropy": 1.386056661605835, "tau_bias": 0.6534034609794617, "beta": 0.5144987627863884, "train_val_corr": 1.0}
{"t": 32, "ts": 1788182322.407416, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -61.38798141479492, "attn_entropy": 0.7012678354978561, "load_balance": 0.00038704320322722194, "g_entropy": 1.3861083507537841, "tau_bias": 0.6513624787330627, "beta": 0.5198185766736666, "train_val_corr": 1.0}
{"t": 33, "ts": 1788182329.5007422, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -58.63174819946289, "attn_entropy": 0.6935345947742462, "load_balance": 0.0003868039231747389, "g_entropy": 1.3860896587371827, "tau_bias": 0.6498277187347412, "beta": 0.5169008001685143, "train_val_corr": 1.0}
{"t": 34, "ts": 1788182337.0242615, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -55.636474609375, "attn_entropy": 0.6997991025447845, "load_balance": 0.0003872632049024105, "g_entropy": 1.38601975440979, "tau_bias": 0.6490833163261414, "beta": 0.5164594973127047, "train_val_corr": 0.9999999999999998}
{"t": 35, "ts": 1788182368.9766183, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -53.589473724365234, "attn_entropy": 0.6953095078468323, "load_balance": 0.0003870141692459583, "g_entropy": 1.3860682487487792, "tau_bias": 0.6479621529579163, "beta": 0.5218959127863249, "train_val_corr": 1.0}
{"t": 36, "ts": 1788182376.0131845, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -50.9334602355957, "attn_entropy": 0.7131684720516205, "load_balance": 0.00038803506176918743, "g_entropy": 1.38602876663208, "tau_bias": 0.6478629112243652, "beta": 0.5177547136942545, "train_val_corr": 1.0}
{"t": 37, "ts": 1788182383.2935793, "step": 370, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -50.18406295776367, "attn_entropy": 0.6953937828540802, "load_balance": 0.00038697727723047137, "g_entropy": 1.3859283924102783, "tau_bias": 0.6468679904937744, "beta": 0.5123449017604192, "train_val_corr": 1.0}
{"t": 0, "ts": 1788182440.2107203, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0817061852085317e-07, "advantage": 13.713120460510254, "attn_entropy": 0.6970125883817673, "load_balance": 0.0003870825283229351, "g_entropy": 1.3861013889312743, "tau_bias": 0.6440906524658203, "beta": 0.5174995884299278, "train_val_corr": 0.0}
{"t": 1, "ts": 1788182446.8832994, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 5.897528581044753e-07, "advantage": 125.71961212158203, "attn_entropy": 0.6890047758817672, "load_balance": 0.0003868252737447619, "g_entropy": 1.3860909461975097, "tau_bias": 0.644088864326477, "beta": 0.5484483515222868, "train_val_corr": 0.0}
{"t": 2, "ts": 1788182453.6745646, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.079557361867046e-07, "advantage": 81.70286560058594, "attn_entropy": 0.6822265446186065, "load_balance": 0.00038617302197963, "g_entropy": 1.3860549449920654, "tau_bias": 0.6444970965385437, "beta": 0.5121008480588595, "train_val_corr": 0.0}
{"t": 3, "ts": 1788182460.8932483, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0726228910916689e-07, "advantage": 34.16429138183594, "attn_entropy": 0.6713385701179504, "load_balance": 0.0003856858937069774, "g_entropy": 1.386064386367798, "tau_bias": 0.6400582194328308, "beta": 0.5106812690695127, "train_val_corr": 0.0}
{"t": 4, "ts": 1788182467.6754627, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723824434298876e-07, "advantage": 26.726747512817383, "attn_entropy": 0.7071383655071258, "load_balance": 0.00038621226558461786, "g_entropy": 1.3860540390014648, "tau_bias": 0.6374273896217346, "beta": 0.5178808346390724, "train_val_corr": 0.0}
{"t": 5, "ts": 1788182474.6995103, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723474847272882e-07, "advantage": 19.262868881225586, "attn_entropy": 0.7006723761558533, "load_balance": 0.0003862655023112893, "g_entropy": 1.386030673980713, "tau_bias": 0.6332598924636841, "beta": 0.5126136963566145, "train_val_corr": 0.0}
{"t": 6, "ts": 1788182507.2069132, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723364002606104e-07, "advantage": 17.189977645874023, "attn_entropy": 0.7090333223342895, "load_balance": 0.0003867812221869826, "g_entropy": 1.3858230113983154, "tau_bias": 0.6306216716766357, "beta": 0.5151266008615494, "train_val_corr": 0.0}
{"t": 7, "ts": 1788182514.3133416, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723315685700072e-07, "advantage": 15.986739158630371, "attn_entropy": 0.699987941980362, "load_balance": 0.0003867850289680064, "g_entropy": 1.3861147880554199, "tau_bias": 0.6289072632789612, "beta": 0.5199589058756828, "train_val_corr": 0.0}
{"t": 8, "ts": 1788182521.638959, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723292948332528e-07, "advantage": 15.691190719604492, "attn_entropy": 0.6922041207551957, "load_balance": 0.00038665077881887554, "g_entropy": 1.3860624313354493, "tau_bias": 0.6275807023048401, "beta": 0.5180975794792175, "train_val_corr": 0.0}
{"t": 9, "ts": 1788182529.1259656, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723283000734227e-07, "advantage": 13.052018165588379, "attn_entropy": 0.684047532081604, "load_balance": 0.0003861729986965656, "g_entropy": 1.386097002029419, "tau_bias": 0.6255034804344177, "beta": 0.5235174397627512, "train_val_corr": 0.0}
{"t": 10, "ts": 1788182536.2154927, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072327731639234e-07, "advantage": 10.318466186523438, "attn_entropy": 0.6790194392204285, "load_balance": 0.00038621515268459915, "g_entropy": 1.3860722064971924, "tau_bias": 0.6242652535438538, "beta": 0.5220654383301735, "train_val_corr": 0.0}
{"t": 11, "ts": 1788182543.293753, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.836060523986816, "attn_entropy": 0.695192739367485, "load_balance": 0.0003872448345646262, "g_entropy": 1.3860883712768555, "tau_bias": 0.6234869360923767, "beta": 0.5171400929490725, "train_val_corr": 0.0}
{"t": 12, "ts": 1788182576.4270127, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.887282371520996, "attn_entropy": 0.6843207359313965, "load_balance": 0.00038620824925601483, "g_entropy": 1.3860651016235352, "tau_bias": 0.6236617565155029, "beta": 0.5141017610828081, "train_val_corr": 0.0}
{"t": 13, "ts": 1788182582.9224262, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 15.37614631652832, "attn_entropy": 0.6871477574110031, "load_balance": 0.0003866247367113829, "g_entropy": 1.3860503673553466, "tau_bias": 0.623771607875824, "beta": 0.5078916052977244, "train_val_corr": 0.0}
{"t": 14, "ts": 1788182589.2310572, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.313080787658691, "attn_entropy": 0.6853887110948562, "load_balance": 0.00038673675153404473, "g_entropy": 1.3860641479492188, "tau_bias": 0.6234549880027771, "beta": 0.5077720830837885, "train_val_corr": 0.0}
{"t": 15, "ts": 1788182595.7044375, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.20047664642334, "attn_entropy": 0.6918005615472793, "load_balance": 0.0003868610132485628, "g_entropy": 1.3860831260681152, "tau_bias": 0.6224879622459412, "beta": 0.504498283068339, "train_val_corr": 0.0}
{"t": 16, "ts": 1788182602.698613, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.637356281280518, "attn_entropy": 0.6888689965009689, "load_balance": 0.0003866719198413193, "g_entropy": 1.386005973815918, "tau_bias": 0.6218122839927673, "beta": 0.5045974751313528, "train_val_corr": 0.0}
{"t": 17, "ts": 1788182609.9031413, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.190775871276855, "attn_entropy": 0.6820834189653396, "load_balance": 0.00038661927683278916, "g_entropy": 1.3860588073730469, "tau_bias": 0.6217818260192871, "beta": 0.5036956618229548, "train_val_corr": 0.0}
{"t": 18, "ts": 1788182641.075888, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.650878429412842, "attn_entropy": 0.6910134315490722, "load_balance": 0.00038724575424566867, "g_entropy": 1.3860651493072509, "tau_bias": 0.6220742464065552, "beta": 0.5025332992275556, "train_val_corr": 0.0}
{"t": 19, "ts": 1788182647.6236966, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.104732513427734, "attn_entropy": 0.6804836839437485, "load_balance": 0.00038647023029625416, "g_entropy": 1.3860562801361085, "tau_bias": 0.6220806837081909, "beta": 0.5018108909328779, "train_val_corr": 1.0}
{"t": 20, "ts": 1788182654.077605, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.89315128326416, "attn_entropy": 0.6843858182430267, "load_balance": 0.0003862053039483726, "g_entropy": 1.3860922813415528, "tau_bias": 0.6245635747909546, "beta": 0.5017737473050753, "train_val_corr": 1.0}
{"t": 21, "ts": 1788182661.1214032, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.0768723487854, "attn_entropy": 0.6830544233322143, "load_balance": 0.0003866925137117505, "g_entropy": 1.3860962390899658, "tau_bias": 0.6328462362289429, "beta": 0.5410831347107887, "train_val_corr": 1.0}
{"t": 22, "ts": 1788182668.190664, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.5755202770233154, "attn_entropy": 0.6872394800186157, "load_balance": 0.00038676229305565355, "g_entropy": 1.3860705852508546, "tau_bias": 0.6442608833312988, "beta": 0.5358409707744917, "train_val_corr": 0.9999999999999998}
{"t": 23, "ts": 1788182675.1134293, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.9130146503448486, "attn_entropy": 0.7064634889364243, "load_balance": 0.00038724465994164345, "g_entropy": 1.3860516548156738, "tau_bias": 0.6525736451148987, "beta": 0.5439841076731682, "train_val_corr": 0.9999999999999997}
{"t": 0, "ts": 1788183140.4871194, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.4056746294954792e-05, "advantage": 74.98120880126953, "attn_entropy": 1.5088031947612763, "load_balance": 0.0004739070369396359, "g_entropy": 2.1600157260894775, "tau_bias": 0.6557140350341797, "beta": 0.5537065689762434, "train_val_corr": 0.0}
{"t": 0, "ts": 1788183408.8845775, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00016502193466294557, "advantage": 233.86448669433594, "attn_entropy": 1.489292985200882, "load_balance": 0.00047873075236566366, "g_entropy": 2.1514075756073, "tau_bias": 0.6509265899658203, "beta": 0.5577196727196375, "train_val_corr": 0.0}
{"t": 1, "ts": 1788183415.4957328, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0732113625854254e-07, "advantage": 135.4520721435547, "attn_entropy": 1.504406327009201, "load_balance": 0.0005040016258135438, "g_entropy": 2.115595817565918, "tau_bias": 0.6485937237739563, "beta": 0.534373015165329, "train_val_corr": 0.0}
{"t": 2, "ts": 1788183424.2704005, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 106.90690612792969, "attn_entropy": 1.4948223650455474, "load_balance": 0.0004752575827296823, "g_entropy": 2.1568811893463136, "tau_bias": 0.6451319456100464, "beta": 0.5370425631602606, "train_val_corr": 0.0}
{"t": 3, "ts": 1788183432.9363961, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 90.00999450683594, "attn_entropy": 1.4930770754814149, "load_balance": 0.00048624624032527207, "g_entropy": 2.138133096694946, "tau_bias": 0.6502702236175537, "beta": 0.5410975863536199, "train_val_corr": 0.0}
{"t": 4, "ts": 1788183441.6999872, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 79.22921752929688, "attn_entropy": 1.4895644664764405, "load_balance": 0.00047609547036699953, "g_entropy": 2.1553314685821534, "tau_bias": 0.6526888012886047, "beta": 0.5445701579252878, "train_val_corr": 0.0}
{"t": 5, "ts": 1788183450.8757617, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.597869873046875, "attn_entropy": 1.476958093047142, "load_balance": 0.0004801563394721597, "g_entropy": 2.147358751296997, "tau_bias": 0.6595838069915771, "beta": 0.5430855378508568, "train_val_corr": 0.0}
{"t": 6, "ts": 1788183459.9369783, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.57062911987305, "attn_entropy": 1.4702332824468614, "load_balance": 0.000479798304149881, "g_entropy": 2.148047399520874, "tau_bias": 0.6685022115707397, "beta": 0.543850320080916, "train_val_corr": 0.0}
{"t": 7, "ts": 1788183483.9670415, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 83.86402893066406, "attn_entropy": 1.4705988615751266, "load_balance": 0.0004767290607560426, "g_entropy": 2.151924896240234, "tau_bias": 0.6778479814529419, "beta": 0.5449230074882507, "train_val_corr": 0.0}
{"t": 8, "ts": 1788183492.6307945, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 58.05849838256836, "attn_entropy": 1.487132951617241, "load_balance": 0.00047664226149208845, "g_entropy": 2.1549264907836916, "tau_bias": 0.6788264513015747, "beta": 0.5277107457319895, "train_val_corr": 0.0}
{"t": 9, "ts": 1788183501.6102118, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.9108772277832, "attn_entropy": 1.4734902918338775, "load_balance": 0.0004830178862903267, "g_entropy": 2.1445977687835693, "tau_bias": 0.6880024671554565, "beta": 0.5379047989845276, "train_val_corr": 0.0}
{"t": 10, "ts": 1788183510.4039202, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.37331008911133, "attn_entropy": 1.485756081342697, "load_balance": 0.00047561593237333, "g_entropy": 2.156116485595703, "tau_bias": 0.6917641162872314, "beta": 0.5583699494600296, "train_val_corr": 0.0}
{"t": 11, "ts": 1788183519.1373317, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 19.8219051361084, "attn_entropy": 1.488300397992134, "load_balance": 0.00047600738471373914, "g_entropy": 2.155643653869629, "tau_bias": 0.704373300075531, "beta": 0.5196714376409849, "train_val_corr": 0.0}
{"t": 12, "ts": 1788183529.238555, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.291961669921875, "attn_entropy": 1.429060271382332, "load_balance": 0.0004909828654490411, "g_entropy": 2.131763744354248, "tau_bias": 0.7111386060714722, "beta": 0.5366366157929102, "train_val_corr": 0.0}
{"t": 13, "ts": 1788183551.8943312, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 118.08148193359375, "attn_entropy": 1.531869900226593, "load_balance": 0.0004752974899020046, "g_entropy": 2.1547550678253176, "tau_bias": 0.7126057147979736, "beta": 0.529900756975015, "train_val_corr": 0.0}
{"t": 14, "ts": 1788183559.833448, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.223304748535156, "attn_entropy": 1.4580867111682891, "load_balance": 0.00048351999721489846, "g_entropy": 2.1429885387420655, "tau_bias": 0.7204989790916443, "beta": 0.5127113337318102, "train_val_corr": 0.0}
{"t": 15, "ts": 1788183568.6258247, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.29686564207077026, "attn_entropy": 1.433584713935852, "load_balance": 0.0004787827667314559, "g_entropy": 2.150816774368286, "tau_bias": 0.7216870784759521, "beta": 0.526393656929334, "train_val_corr": 0.0}
{"t": 16, "ts": 1788183577.5887163, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.063535690307617, "attn_entropy": 1.4875813454389573, "load_balance": 0.0004803567077033222, "g_entropy": 2.150985670089722, "tau_bias": 0.7251416444778442, "beta": 0.5204427341620127, "train_val_corr": 0.0}
{"t": 17, "ts": 1788183586.2196229, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.3759169578552246, "attn_entropy": 1.4775080963969232, "load_balance": 0.0004759552713949233, "g_entropy": 2.156014251708984, "tau_bias": 0.7343813180923462, "beta": 0.5155022591352463, "train_val_corr": 0.0}
{"t": 18, "ts": 1788183595.3355455, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.894024848937988, "attn_entropy": 1.5333922386169434, "load_balance": 0.0004759374482091516, "g_entropy": 2.157166051864624, "tau_bias": 0.7370041608810425, "beta": 0.5161528637011846, "train_val_corr": 0.0}
{"t": 19, "ts": 1788183618.4713807, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.295143127441406, "attn_entropy": 1.4600316867232324, "load_balance": 0.0004753885034006089, "g_entropy": 2.1552161693573, "tau_bias": 0.7339385747909546, "beta": 0.5131564537684122, "train_val_corr": 1.0}
{"t": 20, "ts": 1788183626.8752363, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.908561706542969, "attn_entropy": 1.4705814749002457, "load_balance": 0.0004749368934426457, "g_entropy": 2.153610849380493, "tau_bias": 0.7318506836891174, "beta": 0.512971505522728, "train_val_corr": 1.0}
{"t": 21, "ts": 1788183636.4805584, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -6.801997661590576, "attn_entropy": 1.3796832587569952, "load_balance": 0.00047217421233654023, "g_entropy": 2.1548173427581787, "tau_bias": 0.7305382490158081, "beta": 0.5466798593600591, "train_val_corr": 0.9999999999999999}
{"t": 22, "ts": 1788183645.9692745, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.03079891204834, "attn_entropy": 1.4015901938080788, "load_balance": 0.000477708742255345, "g_entropy": 2.1523598194122315, "tau_bias": 0.7276144623756409, "beta": 0.5561274240414301, "train_val_corr": 1.0}
{"t": 23, "ts": 1788183655.3235989, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -15.480619430541992, "attn_entropy": 1.4288065746426581, "load_balance": 0.00047539000515826044, "g_entropy": 2.155980634689331, "tau_bias": 0.7232627868652344, "beta": 0.5524898047248522, "train_val_corr": 1.0}
{"t": 24, "ts": 1788183665.1156316, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.485584259033203, "attn_entropy": 1.4170255020260811, "load_balance": 0.0004772830579895526, "g_entropy": 2.150512361526489, "tau_bias": 0.7181065678596497, "beta": 0.5575748731692632, "train_val_corr": 1.0}
{"t": 25, "ts": 1788183688.0870621, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.092126369476318, "attn_entropy": 1.3743138790130616, "load_balance": 0.0004830477817449719, "g_entropy": 2.1420952796936037, "tau_bias": 0.710871696472168, "beta": 0.5404765009880066, "train_val_corr": 1.0}
{"t": 26, "ts": 1788183696.5251737, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.179574966430664, "attn_entropy": 1.433441763371229, "load_balance": 0.00047331815003417434, "g_entropy": 2.1570926189422606, "tau_bias": 0.7001933455467224, "beta": 0.5684436360994974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788183705.82035, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.90776824951172, "attn_entropy": 1.4296802416443826, "load_balance": 0.0004759426286909729, "g_entropy": 2.155164432525635, "tau_bias": 0.6949167251586914, "beta": 0.5599769130349159, "train_val_corr": 1.0}
{"t": 28, "ts": 1788183714.875848, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -25.158462524414062, "attn_entropy": 1.403586426563561, "load_balance": 0.000474717904580757, "g_entropy": 2.15653510093689, "tau_bias": 0.6907883882522583, "beta": 0.5746124361952146, "train_val_corr": 1.0}
{"t": 29, "ts": 1788183723.9234622, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.39894676208496, "attn_entropy": 1.4271481703966855, "load_balance": 0.0004718699085060507, "g_entropy": 2.158915090560913, "tau_bias": 0.6855068802833557, "beta": 0.5619990900158882, "train_val_corr": 1.0}
{"t": 30, "ts": 1788183733.5907676, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.365520477294922, "attn_entropy": 1.4609333246946334, "load_balance": 0.00047501508379355073, "g_entropy": 2.1588019371032714, "tau_bias": 0.6868966221809387, "beta": 0.5394027655323347, "train_val_corr": 0.9999999999999998}
{"t": 31, "ts": 1788183758.1763098, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.48574447631836, "attn_entropy": 1.4370897084474563, "load_balance": 0.0004734694550279528, "g_entropy": 2.1587896823883055, "tau_bias": 0.6813430190086365, "beta": 0.5372083187103271, "train_val_corr": 1.0}
{"t": 32, "ts": 1788183766.275444, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.529634475708008, "attn_entropy": 1.4470097206532955, "load_balance": 0.0004744757490698248, "g_entropy": 2.156731700897217, "tau_bias": 0.6736310124397278, "beta": 0.527604249616464, "train_val_corr": 1.0}
{"t": 33, "ts": 1788183775.630192, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.509750366210938, "attn_entropy": 1.4526858747005462, "load_balance": 0.000475892802933231, "g_entropy": 2.1565074920654297, "tau_bias": 0.6729728579521179, "beta": 0.5385467956463496, "train_val_corr": 1.0}
{"t": 34, "ts": 1788183784.665296, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.434677124023438, "attn_entropy": 1.4572451025247575, "load_balance": 0.00047476053587161007, "g_entropy": 2.1585590839385986, "tau_bias": 0.6744930148124695, "beta": 0.5374716048439344, "train_val_corr": 1.0}
{"t": 35, "ts": 1788183794.1874352, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.720075607299805, "attn_entropy": 1.4355932906270028, "load_balance": 0.00047572029870934784, "g_entropy": 2.155581760406494, "tau_bias": 0.6731449365615845, "beta": 0.5328855787714323, "train_val_corr": 1.0}
{"t": 36, "ts": 1788183803.6061223, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.938020706176758, "attn_entropy": 1.4389197781682015, "load_balance": 0.0004744437930639833, "g_entropy": 2.156613254547119, "tau_bias": 0.6717098355293274, "beta": 0.5283562491337458, "train_val_corr": 0.9999999999999998}
{"t": 0, "ts": 1788183937.121734, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.5381751836684998e-06, "advantage": 21.83655548095703, "attn_entropy": 1.458758207410574, "load_balance": 0.000473482528468594, "g_entropy": 2.1583048343658446, "tau_bias": 0.6737804412841797, "beta": 0.5326875448226929, "train_val_corr": 0.0}
{"t": 1, "ts": 1788183943.6851563, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0724074545341864e-07, "advantage": 35.55029296875, "attn_entropy": 1.4766897767782212, "load_balance": 0.00047712993109598755, "g_entropy": 2.151626396179199, "tau_bias": 0.6684935688972473, "beta": 0.5393403023481369, "train_val_corr": 0.0}
{"t": 2, "ts": 1788183951.8328888, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 45.0902214050293, "attn_entropy": 1.4671686708927154, "load_balance": 0.00047357137664221226, "g_entropy": 2.1595471858978272, "tau_bias": 0.6655518412590027, "beta": 0.5446360607941946, "train_val_corr": 0.0}
{"t": 3, "ts": 1788183960.1774368, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 51.9793701171875, "attn_entropy": 1.4673122435808181, "load_balance": 0.00047714381944388153, "g_entropy": 2.1546775817871096, "tau_bias": 0.6628429889678955, "beta": 0.5768052190542221, "train_val_corr": 0.0}
{"t": 4, "ts": 1788183969.7667797, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 44.91798400878906, "attn_entropy": 1.4388526171445846, "load_balance": 0.0004745180660393089, "g_entropy": 2.1589651584625242, "tau_bias": 0.6660708785057068, "beta": 0.541225661834081, "train_val_corr": 0.0}
{"t": 5, "ts": 1788183979.6823015, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 68.2281723022461, "attn_entropy": 1.513240784406662, "load_balance": 0.00047418494941666724, "g_entropy": 2.157308578491211, "tau_bias": 0.6720103025436401, "beta": 0.5501775766412417, "train_val_corr": 0.0}
{"t": 6, "ts": 1788183989.7125971, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 179.09043884277344, "attn_entropy": 1.6780077457427978, "load_balance": 0.0004754280380439013, "g_entropy": 2.1575939655303955, "tau_bias": 0.6715010404586792, "beta": 0.581094262500604, "train_val_corr": 0.0}
{"t": 7, "ts": 1788184014.679546, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 40.36177062988281, "attn_entropy": 1.4472683846950531, "load_balance": 0.00047626493615098297, "g_entropy": 2.1519195079803466, "tau_bias": 0.6777308583259583, "beta": 0.5139541029930115, "train_val_corr": 0.0}
{"t": 8, "ts": 1788184023.7168622, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.379065990447998, "attn_entropy": 1.7159887313842774, "load_balance": 0.000475768483011052, "g_entropy": 2.158986282348633, "tau_bias": 0.684191107749939, "beta": 0.5327391624450684, "train_val_corr": 0.0}
{"t": 9, "ts": 1788184032.5854282, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.74173355102539, "attn_entropy": 1.4558795809745788, "load_balance": 0.00047360589378513395, "g_entropy": 2.158113956451416, "tau_bias": 0.6833652853965759, "beta": 0.5676582033435503, "train_val_corr": 0.0}
{"t": 10, "ts": 1788184041.6763973, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 32.061771392822266, "attn_entropy": 1.4476660430431365, "load_balance": 0.000472807529149577, "g_entropy": 2.159612226486206, "tau_bias": 0.6899245977401733, "beta": 0.5003211026390394, "train_val_corr": 0.0}
{"t": 11, "ts": 1788184051.3208249, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.568561553955078, "attn_entropy": 1.466874298453331, "load_balance": 0.0004736186645459384, "g_entropy": 2.1593819141387938, "tau_bias": 0.69566410779953, "beta": 0.5028317595521609, "train_val_corr": 0.0}
{"t": 12, "ts": 1788184060.9872694, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.482948303222656, "attn_entropy": 1.438286417722702, "load_balance": 0.0004765709629282355, "g_entropy": 2.154563617706299, "tau_bias": 0.6977018117904663, "beta": 0.4894258740047614, "train_val_corr": 0.0}
{"t": 0, "ts": 1788184184.8950887, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00016502193466294557, "advantage": 233.86448669433594, "attn_entropy": 1.489292985200882, "load_balance": 0.00047873075236566366, "g_entropy": 2.1514075756073, "tau_bias": 0.6509265899658203, "beta": 0.5577196727196375, "train_val_corr": 0.0}
{"t": 1, "ts": 1788184192.2659261, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0732113625854254e-07, "advantage": 135.4520721435547, "attn_entropy": 1.504406327009201, "load_balance": 0.0005040016258135438, "g_entropy": 2.115595817565918, "tau_bias": 0.6485937237739563, "beta": 0.534373015165329, "train_val_corr": 0.0}
{"t": 2, "ts": 1788184202.065547, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 106.90690612792969, "attn_entropy": 1.4948223650455474, "load_balance": 0.0004752575827296823, "g_entropy": 2.1568811893463136, "tau_bias": 0.6451319456100464, "beta": 0.5370425631602606, "train_val_corr": 0.0}
{"t": 3, "ts": 1788184211.887971, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 90.00999450683594, "attn_entropy": 1.4930770754814149, "load_balance": 0.00048624624032527207, "g_entropy": 2.138133096694946, "tau_bias": 0.6502702236175537, "beta": 0.5410975863536199, "train_val_corr": 0.0}
{"t": 4, "ts": 1788184221.3373666, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 79.22921752929688, "attn_entropy": 1.4895644664764405, "load_balance": 0.00047609547036699953, "g_entropy": 2.1553314685821534, "tau_bias": 0.6526888012886047, "beta": 0.5445701579252878, "train_val_corr": 0.0}
{"t": 5, "ts": 1788184230.274409, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.597869873046875, "attn_entropy": 1.476958093047142, "load_balance": 0.0004801563394721597, "g_entropy": 2.147358751296997, "tau_bias": 0.6595838069915771, "beta": 0.5430855378508568, "train_val_corr": 0.0}
{"t": 6, "ts": 1788184240.0053642, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.57062911987305, "attn_entropy": 1.4702332824468614, "load_balance": 0.000479798304149881, "g_entropy": 2.148047399520874, "tau_bias": 0.6685022115707397, "beta": 0.543850320080916, "train_val_corr": 0.0}
{"t": 7, "ts": 1788184263.4922082, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 83.86402893066406, "attn_entropy": 1.4705988615751266, "load_balance": 0.0004767290607560426, "g_entropy": 2.151924896240234, "tau_bias": 0.6778479814529419, "beta": 0.5449230074882507, "train_val_corr": 0.0}
{"t": 8, "ts": 1788184272.2149196, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 58.05849838256836, "attn_entropy": 1.487132951617241, "load_balance": 0.00047664226149208845, "g_entropy": 2.1549264907836916, "tau_bias": 0.6788264513015747, "beta": 0.5277107457319895, "train_val_corr": 0.0}
{"t": 9, "ts": 1788184281.1313004, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.9108772277832, "attn_entropy": 1.4734902918338775, "load_balance": 0.0004830178862903267, "g_entropy": 2.1445977687835693, "tau_bias": 0.6880024671554565, "beta": 0.5379047989845276, "train_val_corr": 0.0}
{"t": 10, "ts": 1788184290.8293276, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.37331008911133, "attn_entropy": 1.485756081342697, "load_balance": 0.00047561593237333, "g_entropy": 2.156116485595703, "tau_bias": 0.6917641162872314, "beta": 0.5583699494600296, "train_val_corr": 0.0}
{"t": 11, "ts": 1788184300.1170363, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 19.8219051361084, "attn_entropy": 1.488300397992134, "load_balance": 0.00047600738471373914, "g_entropy": 2.155643653869629, "tau_bias": 0.704373300075531, "beta": 0.5196714376409849, "train_val_corr": 0.0}
{"t": 12, "ts": 1788184309.1107776, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.291961669921875, "attn_entropy": 1.429060271382332, "load_balance": 0.0004909828654490411, "g_entropy": 2.131763744354248, "tau_bias": 0.7111386060714722, "beta": 0.5366366157929102, "train_val_corr": 0.0}
{"t": 13, "ts": 1788184333.4983935, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 118.08148193359375, "attn_entropy": 1.531869900226593, "load_balance": 0.0004752974899020046, "g_entropy": 2.1547550678253176, "tau_bias": 0.7126057147979736, "beta": 0.529900756975015, "train_val_corr": 0.0}
{"t": 14, "ts": 1788184342.4897523, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.223304748535156, "attn_entropy": 1.4580867111682891, "load_balance": 0.00048351999721489846, "g_entropy": 2.1429885387420655, "tau_bias": 0.7204989790916443, "beta": 0.5127113337318102, "train_val_corr": 0.0}
{"t": 15, "ts": 1788184352.4922884, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.29686564207077026, "attn_entropy": 1.433584713935852, "load_balance": 0.0004787827667314559, "g_entropy": 2.150816774368286, "tau_bias": 0.7216870784759521, "beta": 0.526393656929334, "train_val_corr": 0.0}
{"t": 16, "ts": 1788184362.2949617, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.063535690307617, "attn_entropy": 1.4875813454389573, "load_balance": 0.0004803567077033222, "g_entropy": 2.150985670089722, "tau_bias": 0.7251416444778442, "beta": 0.5204427341620127, "train_val_corr": 0.0}
{"t": 17, "ts": 1788184371.2288687, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.3759169578552246, "attn_entropy": 1.4775080963969232, "load_balance": 0.0004759552713949233, "g_entropy": 2.156014251708984, "tau_bias": 0.7343813180923462, "beta": 0.5155022591352463, "train_val_corr": 0.0}
{"t": 18, "ts": 1788184380.193827, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.894024848937988, "attn_entropy": 1.5333922386169434, "load_balance": 0.0004759374482091516, "g_entropy": 2.157166051864624, "tau_bias": 0.7370041608810425, "beta": 0.5161528637011846, "train_val_corr": 0.0}
{"t": 19, "ts": 1788184404.4010327, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.295143127441406, "attn_entropy": 1.4600316867232324, "load_balance": 0.0004753885034006089, "g_entropy": 2.1552161693573, "tau_bias": 0.7339385747909546, "beta": 0.5131564537684122, "train_val_corr": 1.0}
{"t": 20, "ts": 1788184412.897324, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.908561706542969, "attn_entropy": 1.4705814749002457, "load_balance": 0.0004749368934426457, "g_entropy": 2.153610849380493, "tau_bias": 0.7318506836891174, "beta": 0.512971505522728, "train_val_corr": 1.0}
{"t": 21, "ts": 1788184422.2674484, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -6.801997661590576, "attn_entropy": 1.3796832587569952, "load_balance": 0.00047217421233654023, "g_entropy": 2.1548173427581787, "tau_bias": 0.7305382490158081, "beta": 0.5466798593600591, "train_val_corr": 0.9999999999999999}
{"t": 22, "ts": 1788184431.73506, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.03079891204834, "attn_entropy": 1.4015901938080788, "load_balance": 0.000477708742255345, "g_entropy": 2.1523598194122315, "tau_bias": 0.7276144623756409, "beta": 0.5561274240414301, "train_val_corr": 1.0}
{"t": 23, "ts": 1788184440.8201911, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -15.480619430541992, "attn_entropy": 1.4288065746426581, "load_balance": 0.00047539000515826044, "g_entropy": 2.155980634689331, "tau_bias": 0.7232627868652344, "beta": 0.5524898047248522, "train_val_corr": 1.0}
{"t": 24, "ts": 1788184449.8098593, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.485584259033203, "attn_entropy": 1.4170255020260811, "load_balance": 0.0004772830579895526, "g_entropy": 2.150512361526489, "tau_bias": 0.7181065678596497, "beta": 0.5575748731692632, "train_val_corr": 1.0}
{"t": 25, "ts": 1788184474.4692514, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.092126369476318, "attn_entropy": 1.3743138790130616, "load_balance": 0.0004830477817449719, "g_entropy": 2.1420952796936037, "tau_bias": 0.710871696472168, "beta": 0.5404765009880066, "train_val_corr": 1.0}
{"t": 26, "ts": 1788184483.4822981, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.179574966430664, "attn_entropy": 1.433441763371229, "load_balance": 0.00047331815003417434, "g_entropy": 2.1570926189422606, "tau_bias": 0.7001933455467224, "beta": 0.5684436360994974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788184492.8656235, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.90776824951172, "attn_entropy": 1.4296802416443826, "load_balance": 0.0004759426286909729, "g_entropy": 2.155164432525635, "tau_bias": 0.6949167251586914, "beta": 0.5599769130349159, "train_val_corr": 1.0}
{"t": 28, "ts": 1788184501.9833527, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -25.158462524414062, "attn_entropy": 1.403586426563561, "load_balance": 0.000474717904580757, "g_entropy": 2.15653510093689, "tau_bias": 0.6907883882522583, "beta": 0.5746124361952146, "train_val_corr": 1.0}
{"t": 29, "ts": 1788184511.137238, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.39894676208496, "attn_entropy": 1.4271481703966855, "load_balance": 0.0004718699085060507, "g_entropy": 2.158915090560913, "tau_bias": 0.6855068802833557, "beta": 0.5619990900158882, "train_val_corr": 1.0}
{"t": 30, "ts": 1788184520.2996242, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.365520477294922, "attn_entropy": 1.4609333246946334, "load_balance": 0.00047501508379355073, "g_entropy": 2.1588019371032714, "tau_bias": 0.6868966221809387, "beta": 0.5394027655323347, "train_val_corr": 0.9999999999999998}
{"t": 31, "ts": 1788184545.0892565, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.48574447631836, "attn_entropy": 1.4370897084474563, "load_balance": 0.0004734694550279528, "g_entropy": 2.1587896823883055, "tau_bias": 0.6813430190086365, "beta": 0.5372083187103271, "train_val_corr": 1.0}
{"t": 32, "ts": 1788184552.7967873, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.529634475708008, "attn_entropy": 1.4470097206532955, "load_balance": 0.0004744757490698248, "g_entropy": 2.156731700897217, "tau_bias": 0.6736310124397278, "beta": 0.527604249616464, "train_val_corr": 1.0}
{"t": 33, "ts": 1788184562.4683363, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.509750366210938, "attn_entropy": 1.4526858747005462, "load_balance": 0.000475892802933231, "g_entropy": 2.1565074920654297, "tau_bias": 0.6729728579521179, "beta": 0.5385467956463496, "train_val_corr": 1.0}
{"t": 34, "ts": 1788184572.9172087, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.434677124023438, "attn_entropy": 1.4572451025247575, "load_balance": 0.00047476053587161007, "g_entropy": 2.1585590839385986, "tau_bias": 0.6744930148124695, "beta": 0.5374716048439344, "train_val_corr": 1.0}
{"t": 35, "ts": 1788184583.587973, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.720075607299805, "attn_entropy": 1.4355932906270028, "load_balance": 0.00047572029870934784, "g_entropy": 2.155581760406494, "tau_bias": 0.6731449365615845, "beta": 0.5328855787714323, "train_val_corr": 1.0}
{"t": 0, "ts": 1788184636.032913, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.59635226282262e-06, "advantage": 23.1477108001709, "attn_entropy": 1.4606423109769822, "load_balance": 0.0004740119038615376, "g_entropy": 2.1574799537658693, "tau_bias": 0.6752452850341797, "beta": 0.5333390633265177, "train_val_corr": 0.0}
{"t": 1, "ts": 1788184643.1047618, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725367733120947e-07, "advantage": 59.84957504272461, "attn_entropy": 1.4653254687786101, "load_balance": 0.0004814799001906067, "g_entropy": 2.1472036361694338, "tau_bias": 0.6778368949890137, "beta": 0.5448142811655998, "train_val_corr": 0.0}
{"t": 2, "ts": 1788184651.9942553, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.755023956298828, "attn_entropy": 1.4372749254107475, "load_balance": 0.0004734310612548143, "g_entropy": 2.1591992378234863, "tau_bias": 0.6804002523422241, "beta": 0.5807090451320013, "train_val_corr": 0.0}
{"t": 3, "ts": 1788184660.8359199, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 23.025407791137695, "attn_entropy": 1.4498254999518394, "load_balance": 0.0004778597969561815, "g_entropy": 2.152260971069336, "tau_bias": 0.6886326670646667, "beta": 0.5659464821219444, "train_val_corr": 0.0}
{"t": 4, "ts": 1788184669.328868, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 83.60436248779297, "attn_entropy": 1.4566673040390015, "load_balance": 0.00047397856251336635, "g_entropy": 2.158377742767334, "tau_bias": 0.695793092250824, "beta": 0.555230105916659, "train_val_corr": 0.0}
{"t": 5, "ts": 1788184678.2945454, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 19.73830795288086, "attn_entropy": 1.4603836223483087, "load_balance": 0.0004736044502351433, "g_entropy": 2.155826139450073, "tau_bias": 0.7007213234901428, "beta": 0.5374620457490286, "train_val_corr": 0.0}
{"t": 6, "ts": 1788184687.3080494, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.501568794250488, "attn_entropy": 1.4335948888212442, "load_balance": 0.00047435371088795365, "g_entropy": 2.156221866607666, "tau_bias": 0.7003889679908752, "beta": 0.5407674858967463, "train_val_corr": 0.0}
{"t": 7, "ts": 1788184711.7301967, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.949477195739746, "attn_entropy": 1.3947192879393697, "load_balance": 0.0004776886838953942, "g_entropy": 2.1514840602874754, "tau_bias": 0.6992394328117371, "beta": 0.5250424146652222, "train_val_corr": 0.0}
{"t": 8, "ts": 1788184720.6352491, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.995326042175293, "attn_entropy": 1.4428929455578328, "load_balance": 0.0004748848092276603, "g_entropy": 2.1584208488464354, "tau_bias": 0.6989234089851379, "beta": 0.5408421158790588, "train_val_corr": 0.0}
{"t": 9, "ts": 1788184729.5737636, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.599002838134766, "attn_entropy": 1.441123227030039, "load_balance": 0.0004757033719215542, "g_entropy": 2.15661039352417, "tau_bias": 0.6962786912918091, "beta": 0.5362558588385582, "train_val_corr": 0.0}
{"t": 10, "ts": 1788184738.4851246, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.079965591430664, "attn_entropy": 1.5184394717216492, "load_balance": 0.00047332929098047315, "g_entropy": 2.1592691421508787, "tau_bias": 0.6928916573524475, "beta": 0.5328443422913551, "train_val_corr": 0.0}
{"t": 11, "ts": 1788184747.3222933, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.483515739440918, "attn_entropy": 1.4400692511349917, "load_balance": 0.00047461376525461676, "g_entropy": 2.1589924335479735, "tau_bias": 0.6907348036766052, "beta": 0.5306099827090899, "train_val_corr": 0.0}
{"t": 12, "ts": 1788184756.3048208, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.405046463012695, "attn_entropy": 1.4295033050701023, "load_balance": 0.00047805611975491045, "g_entropy": 2.152081251144409, "tau_bias": 0.6872621178627014, "beta": 0.5252406795819601, "train_val_corr": 0.0}
{"t": 13, "ts": 1788184778.801413, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 11.944075584411621, "attn_entropy": 1.4493017248809337, "load_balance": 0.0004728379077278078, "g_entropy": 2.1590957641601562, "tau_bias": 0.6818945407867432, "beta": 0.5281124661366144, "train_val_corr": 0.0}
{"t": 14, "ts": 1788184787.2977874, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.756349086761475, "attn_entropy": 1.4306704388931393, "load_balance": 0.00047336223651655017, "g_entropy": 2.157182741165161, "tau_bias": 0.6783570051193237, "beta": 0.5308752804994583, "train_val_corr": 0.0}
{"t": 15, "ts": 1788184797.2877412, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.754570007324219, "attn_entropy": 1.436234007589519, "load_balance": 0.0004744874371681362, "g_entropy": 2.1590680122375487, "tau_bias": 0.6805090308189392, "beta": 0.5239301696419716, "train_val_corr": 0.0}
{"t": 16, "ts": 1788184806.428878, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.262256622314453, "attn_entropy": 1.4460862688720226, "load_balance": 0.0004748825856950134, "g_entropy": 2.1582345008850097, "tau_bias": 0.6782267689704895, "beta": 0.5268162240584692, "train_val_corr": 0.0}
{"t": 17, "ts": 1788184815.2782001, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.640336513519287, "attn_entropy": 1.4493039280176163, "load_balance": 0.00047350533423013986, "g_entropy": 2.159375715255737, "tau_bias": 0.6773768663406372, "beta": 0.523301919301351, "train_val_corr": 0.0}
{"t": 18, "ts": 1788184824.9819193, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.965078592300415, "attn_entropy": 1.4630116283893586, "load_balance": 0.00047428793623112143, "g_entropy": 2.1596417903900145, "tau_bias": 0.675847053527832, "beta": 0.5233423734704653, "train_val_corr": 0.0}
{"t": 0, "ts": 1788184915.1146898, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723268789879512e-07, "advantage": 26515.271484375, "attn_entropy": 0.8292756378650665, "load_balance": 0.00029337917803786697, "g_entropy": 1.4584457397460937, "tau_bias": 0.6509332060813904, "beta": 0.6679541940490404, "train_val_corr": 0.0}
{"t": 1, "ts": 1788184921.0048187, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 9.332492481917143e-06, "advantage": -40.530662536621094, "attn_entropy": 0.7648630499839782, "load_balance": 0.0003172042546793818, "g_entropy": 1.4231884956359864, "tau_bias": 0.6419163346290588, "beta": 0.5466359828909239, "train_val_corr": 0.0}
{"t": 2, "ts": 1788184927.8267632, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.01438158005475998, "advantage": -918.5765380859375, "attn_entropy": 0.7302364751696586, "load_balance": 0.00029004550888203087, "g_entropy": 1.4639623165130615, "tau_bias": 0.6485690474510193, "beta": 0.5320525343219439, "train_val_corr": 0.0}
{"t": 3, "ts": 1788184935.7725117, "step": 30, "action": "PGO_UPDATE", "entropy_pi": 0.22799795866012573, "advantage": -1011.9041137695312, "attn_entropy": 0.6388656735420227, "load_balance": 0.0003862828831188381, "g_entropy": 1.3860947608947753, "tau_bias": 0.6533349752426147, "beta": 0.5378654698530833, "train_val_corr": 0.0}
{"t": 4, "ts": 1788184943.7940814, "step": 40, "action": "PGO_UPDATE", "entropy_pi": 0.44710609316825867, "advantage": -966.4281616210938, "attn_entropy": 0.7759586989879608, "load_balance": 0.00028555025928653774, "g_entropy": 1.462371015548706, "tau_bias": 0.6489294767379761, "beta": 0.5343726972738901, "train_val_corr": 0.0}
{"t": 5, "ts": 1788184950.7271428, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.4025556743144989, "advantage": -927.5804443359375, "attn_entropy": 0.7685485776513815, "load_balance": 0.0002897815022151917, "g_entropy": 1.4545952796936035, "tau_bias": 0.6503220200538635, "beta": 0.5366460134585699, "train_val_corr": 0.0}
{"t": 6, "ts": 1788184957.5260687, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.38413384556770325, "advantage": -944.99267578125, "attn_entropy": 0.7306944880634546, "load_balance": 0.0002947004453744739, "g_entropy": 1.4555233001708985, "tau_bias": 0.6487107276916504, "beta": 0.5280633692940077, "train_val_corr": 0.0}
{"t": 7, "ts": 1788184979.1178446, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 9.819827937462833e-07, "advantage": -540.4423828125, "attn_entropy": 1.5187800765037536, "load_balance": 0.00047664611483924093, "g_entropy": 2.152502107620239, "tau_bias": 0.6466034650802612, "beta": 0.530386209487915, "train_val_corr": 0.0}
{"t": 8, "ts": 1788184985.9033198, "step": 80, "action": "PGO_UPDATE", "entropy_pi": 0.5418862104415894, "advantage": -849.2002563476562, "attn_entropy": 0.7635170858353376, "load_balance": 0.0002873168617952615, "g_entropy": 1.4622771739959717, "tau_bias": 0.6473996043205261, "beta": 0.5216666211684545, "train_val_corr": 0.0}
{"t": 9, "ts": 1788184992.8897142, "step": 90, "action": "PGO_UPDATE", "entropy_pi": 0.6534676551818848, "advantage": -761.9830932617188, "attn_entropy": 0.7574376590549946, "load_balance": 0.0002960059326142073, "g_entropy": 1.4527968883514404, "tau_bias": 0.6493622064590454, "beta": 0.5384581064184507, "train_val_corr": 0.0}
{"t": 10, "ts": 1788184999.7950795, "step": 100, "action": "PGO_UPDATE", "entropy_pi": 0.691128134727478, "advantage": -834.6319580078125, "attn_entropy": 0.7044542610645295, "load_balance": 0.00038622724823653696, "g_entropy": 1.3859049320220946, "tau_bias": 0.6477775573730469, "beta": 0.5286632254719734, "train_val_corr": 0.0}
{"t": 11, "ts": 1788185006.790414, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1546514972214936e-07, "advantage": -663.3646240234375, "attn_entropy": 0.6610115349292756, "load_balance": 0.00038606530288234355, "g_entropy": 1.386028242111206, "tau_bias": 0.6487581133842468, "beta": 0.541649209956328, "train_val_corr": 0.0}
{"t": 12, "ts": 1788185013.3986297, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.3463275056201383e-07, "advantage": -606.9111328125, "attn_entropy": 0.6659295409917831, "load_balance": 0.00038616411620751023, "g_entropy": 1.386027193069458, "tau_bias": 0.647495448589325, "beta": 0.5274332265059153, "train_val_corr": 0.0}
{"t": 13, "ts": 1788185039.8219209, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.2477311745205952e-07, "advantage": -661.5302124023438, "attn_entropy": 0.7262288547237403, "load_balance": 0.00028826346970163287, "g_entropy": 1.462565040588379, "tau_bias": 0.6504720449447632, "beta": 0.5305079395572344, "train_val_corr": 0.0}
{"t": 14, "ts": 1788185046.7331393, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.436090653645806e-07, "advantage": -671.7464599609375, "attn_entropy": 0.7562291789799929, "load_balance": 0.000288994942093268, "g_entropy": 1.4523688793182372, "tau_bias": 0.6536614894866943, "beta": 0.5383176629741987, "train_val_corr": 0.0}
{"t": 15, "ts": 1788185054.172308, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 4.3344962818991917e-07, "advantage": -648.3153686523438, "attn_entropy": 0.7476737760007381, "load_balance": 0.00029253524844534694, "g_entropy": 1.4591244220733643, "tau_bias": 0.6542097926139832, "beta": 0.5318203320105871, "train_val_corr": 0.0}
{"t": 16, "ts": 1788185061.170512, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.747886699376977e-06, "advantage": -630.909423828125, "attn_entropy": 0.7336709935218095, "load_balance": 0.00029532311600632966, "g_entropy": 1.4591253280639649, "tau_bias": 0.6544932723045349, "beta": 0.5116402730345726, "train_val_corr": 0.0}
{"t": 17, "ts": 1788185068.0260026, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.503054803237319e-06, "advantage": -607.0411376953125, "attn_entropy": 0.693779372703284, "load_balance": 0.00028960973140783607, "g_entropy": 1.4637132167816163, "tau_bias": 0.6514052152633667, "beta": 0.52294293542703, "train_val_corr": 0.0}
{"t": 18, "ts": 1788185074.7978344, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 8.219885785365477e-05, "advantage": -587.931884765625, "attn_entropy": 0.7439456166699528, "load_balance": 0.0002885935886297375, "g_entropy": 1.4647128582000732, "tau_bias": 0.6504489183425903, "beta": 0.5303420747319857, "train_val_corr": 0.0}
{"t": 19, "ts": 1788185095.68704, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.008200949989259243, "advantage": -501.46966552734375, "attn_entropy": 0.753854826465249, "load_balance": 0.00028513206634670494, "g_entropy": 1.4630837440490723, "tau_bias": 0.6512283086776733, "beta": 0.5051550368467966, "train_val_corr": 0.9999999999999999}
{"t": 20, "ts": 1788185102.4889138, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.483627051115036, "advantage": -556.2119140625, "attn_entropy": 0.7430488221347332, "load_balance": 0.0002851528115570545, "g_entropy": 1.4616518497467041, "tau_bias": 0.6540138125419617, "beta": 0.5136482566595078, "train_val_corr": 1.0}
{"t": 21, "ts": 1788185109.5027893, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723280158563284e-07, "advantage": -509.8738098144531, "attn_entropy": 0.750447179377079, "load_balance": 0.0002848305506631732, "g_entropy": 1.462707233428955, "tau_bias": 0.655134916305542, "beta": 0.5339209338029226, "train_val_corr": 1.0}
{"t": 22, "ts": 1788185116.3762424, "step": 220, "action": "PGO_UPDATE", "entropy_pi": 0.2104247808456421, "advantage": -528.4754028320312, "attn_entropy": 0.7477819889783859, "load_balance": 0.00029757301090285184, "g_entropy": 1.4606199741363526, "tau_bias": 0.6523676514625549, "beta": 0.5176251406470934, "train_val_corr": 1.0}
{"t": 23, "ts": 1788185123.283255, "step": 230, "action": "PGO_UPDATE", "entropy_pi": 0.16282272338867188, "advantage": -484.37017822265625, "attn_entropy": 0.6186775982379913, "load_balance": 0.0003861060016788542, "g_entropy": 1.3860753536224366, "tau_bias": 0.6499617099761963, "beta": 0.5398266365130743, "train_val_corr": 1.0}
{"t": 24, "ts": 1788185130.3340032, "step": 240, "action": "PGO_UPDATE", "entropy_pi": 0.19322915375232697, "advantage": -489.45965576171875, "attn_entropy": 0.7492279171943664, "load_balance": 0.00028571945149451493, "g_entropy": 1.458821439743042, "tau_bias": 0.6502867937088013, "beta": 0.5223274007439613, "train_val_corr": 0.9999999999999998}
{"t": 25, "ts": 1788185156.3722339, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -414.9049377441406, "attn_entropy": 1.4324228465557098, "load_balance": 0.0004814153595361859, "g_entropy": 2.143503522872925, "tau_bias": 0.6508514881134033, "beta": 0.5256901979446411, "train_val_corr": 1.0}
{"t": 26, "ts": 1788185163.226334, "step": 260, "action": "PGO_UPDATE", "entropy_pi": 0.18686266243457794, "advantage": -442.2037353515625, "attn_entropy": 0.7895808458328247, "load_balance": 0.0002860961714759469, "g_entropy": 1.4644089698791505, "tau_bias": 0.6598299145698547, "beta": 0.5245953301588694, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788185170.832742, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.17128144204616547, "advantage": -413.0153503417969, "attn_entropy": 0.7698263108730317, "load_balance": 0.0002878268773201853, "g_entropy": 1.4625204086303711, "tau_bias": 0.6639866232872009, "beta": 0.5302988688151041, "train_val_corr": 1.0}
{"t": 28, "ts": 1788185178.4302428, "step": 280, "action": "PGO_UPDATE", "entropy_pi": 0.17285360395908356, "advantage": -412.14007568359375, "attn_entropy": 0.6796444118022918, "load_balance": 0.0003873888636007905, "g_entropy": 1.3860005378723144, "tau_bias": 0.6657844185829163, "beta": 0.5240728681286176, "train_val_corr": 1.0}
{"t": 29, "ts": 1788185186.0694258, "step": 290, "action": "PGO_UPDATE", "entropy_pi": 0.16444341838359833, "advantage": -395.5089111328125, "attn_entropy": 0.7847230553627014, "load_balance": 0.0002817947359289974, "g_entropy": 1.46602463722229, "tau_bias": 0.660631000995636, "beta": 0.525507889688015, "train_val_corr": 1.0}
{"t": 30, "ts": 1788185193.6709678, "step": 300, "action": "PGO_UPDATE", "entropy_pi": 0.1528007835149765, "advantage": -375.4992370605469, "attn_entropy": 0.7647671818733215, "load_balance": 0.0002850690914783627, "g_entropy": 1.4659056663513184, "tau_bias": 0.6585508584976196, "beta": 0.5186015715201696, "train_val_corr": 1.0}
{"t": 31, "ts": 1788185216.7343137, "step": 310, "action": "PGO_UPDATE", "entropy_pi": 0.10132104903459549, "advantage": -360.3843994140625, "attn_entropy": 0.7865539491176605, "load_balance": 0.00028270576149225235, "g_entropy": 1.4659008026123046, "tau_bias": 0.6556937098503113, "beta": 0.5236234093705813, "train_val_corr": 1.0}
{"t": 32, "ts": 1788185223.112437, "step": 320, "action": "PGO_UPDATE", "entropy_pi": 0.09041833132505417, "advantage": -336.2904357910156, "attn_entropy": 0.7966763615608216, "load_balance": 0.0002839347813278437, "g_entropy": 1.4639573574066163, "tau_bias": 0.6527218818664551, "beta": 0.5196109761794409, "train_val_corr": 1.0}
{"t": 33, "ts": 1788185230.4218554, "step": 330, "action": "PGO_UPDATE", "entropy_pi": 0.08239232748746872, "advantage": -319.23486328125, "attn_entropy": 0.7808690756559372, "load_balance": 0.00028650995227508247, "g_entropy": 1.4637136936187745, "tau_bias": 0.6518802642822266, "beta": 0.5167375927170118, "train_val_corr": 1.0}
{"t": 34, "ts": 1788185237.5954015, "step": 340, "action": "PGO_UPDATE", "entropy_pi": 0.07815944403409958, "advantage": -309.6928405761719, "attn_entropy": 0.8062925159931182, "load_balance": 0.0002843012975063175, "g_entropy": 1.4656609058380128, "tau_bias": 0.6510321497917175, "beta": 0.5167885273694992, "train_val_corr": 1.0}
{"t": 35, "ts": 1788185244.6099844, "step": 350, "action": "PGO_UPDATE", "entropy_pi": 0.07009848207235336, "advantage": -293.4467468261719, "attn_entropy": 0.8390251636505127, "load_balance": 0.0003873849054798484, "g_entropy": 1.3860865116119385, "tau_bias": 0.6491847634315491, "beta": 0.5175875425338745, "train_val_corr": 1.0}
{"t": 36, "ts": 1788185251.6863146, "step": 360, "action": "PGO_UPDATE", "entropy_pi": 0.07037197798490524, "advantage": -286.55523681640625, "attn_entropy": 0.7999843120574951, "load_balance": 0.0003875089343637228, "g_entropy": 1.386040449142456, "tau_bias": 0.649778425693512, "beta": 0.5207868566115698, "train_val_corr": 0.9999999999999998}
{"t": 37, "ts": 1788185278.402113, "step": 370, "action": "PGO_UPDATE", "entropy_pi": 0.061872173100709915, "advantage": -266.043212890625, "attn_entropy": 0.7847936630249024, "load_balance": 0.00038764408091083167, "g_entropy": 1.3860329151153565, "tau_bias": 0.647911012172699, "beta": 0.5128170723716418, "train_val_corr": 1.0}
{"t": 38, "ts": 1788185284.4960814, "step": 380, "action": "PGO_UPDATE", "entropy_pi": 0.05752599239349365, "advantage": -254.19825744628906, "attn_entropy": 0.7942236244678498, "load_balance": 0.0002826321171596646, "g_entropy": 1.4641545772552491, "tau_bias": 0.6422389149665833, "beta": 0.523783286412557, "train_val_corr": 0.9999999999999998}
{"t": 39, "ts": 1788185291.7223504, "step": 390, "action": "PGO_UPDATE", "entropy_pi": 0.05468388646841049, "advantage": -245.3748779296875, "attn_entropy": 0.7774310290813446, "load_balance": 0.00028587447595782576, "g_entropy": 1.4617913246154786, "tau_bias": 0.6395074725151062, "beta": 0.5253883798917135, "train_val_corr": 1.0}
{"t": 40, "ts": 1788185299.2323673, "step": 400, "action": "PGO_UPDATE", "entropy_pi": 0.05085393786430359, "advantage": -234.3141632080078, "attn_entropy": 0.7315290316939353, "load_balance": 0.0002847152587492019, "g_entropy": 1.4654592990875244, "tau_bias": 0.6347759962081909, "beta": 0.5517642671863238, "train_val_corr": 0.9999999999999999}
{"t": 41, "ts": 1788185306.7092152, "step": 410, "action": "PGO_UPDATE", "entropy_pi": 0.03854118287563324, "advantage": -198.21456909179688, "attn_entropy": 0.7489118456840516, "load_balance": 0.00028195870690979065, "g_entropy": 1.4607249736785888, "tau_bias": 0.6339575052261353, "beta": 0.5675064673026403, "train_val_corr": 0.9999999999999998}
{"t": 0, "ts": 1788185438.1238217, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 7.400409685942577e-06, "advantage": 60.79299545288086, "attn_entropy": 0.7800679415464401, "load_balance": 0.00028412913088686763, "g_entropy": 1.463716173171997, "tau_bias": 0.6327648162841797, "beta": 0.5433004995187124, "train_val_corr": 0.0}
{"t": 1, "ts": 1788185443.8752494, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0732632322051359e-07, "advantage": 85.18016052246094, "attn_entropy": 0.8308367431163788, "load_balance": 0.0002936553850304335, "g_entropy": 1.4501043796539306, "tau_bias": 0.6300535798072815, "beta": 0.5814666723211607, "train_val_corr": 0.0}
{"t": 2, "ts": 1788185451.0963163, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723273053135927e-07, "advantage": 127.54791259765625, "attn_entropy": 0.8348769962787628, "load_balance": 0.0002845934825018048, "g_entropy": 1.4660126209259032, "tau_bias": 0.6309282183647156, "beta": 0.5969614038864771, "train_val_corr": 0.0}
{"t": 3, "ts": 1788185458.4964342, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 56.4415397644043, "attn_entropy": 0.8251064419746399, "load_balance": 0.00028888281085528435, "g_entropy": 1.4575326919555665, "tau_bias": 0.6309438347816467, "beta": 0.5692184517780939, "train_val_corr": 0.0}
{"t": 4, "ts": 1788185465.7638361, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 89.35346221923828, "attn_entropy": 0.8255286693572998, "load_balance": 0.0002845609560608864, "g_entropy": 1.465078353881836, "tau_bias": 0.6305059790611267, "beta": 0.5869104390343031, "train_val_corr": 0.0}
{"t": 5, "ts": 1788185472.8940814, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 83.66268157958984, "attn_entropy": 0.8275206208229064, "load_balance": 0.0002845724404323846, "g_entropy": 1.4619431018829345, "tau_bias": 0.6258172988891602, "beta": 0.6622922345995903, "train_val_corr": 0.0}
{"t": 6, "ts": 1788185480.00379, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 123.61279296875, "attn_entropy": 0.8131971478462219, "load_balance": 0.0002881149121094495, "g_entropy": 1.462412452697754, "tau_bias": 0.6264534592628479, "beta": 0.5397590051094691, "train_val_corr": 0.0}
{"t": 7, "ts": 1788185504.6986754, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072333273872573e-07, "advantage": 92.86007690429688, "attn_entropy": 1.4827744513750076, "load_balance": 0.0004733048554044217, "g_entropy": 2.1528353691101074, "tau_bias": 0.628013014793396, "beta": 0.5941548943519592, "train_val_corr": 0.0}
{"t": 8, "ts": 1788185511.9281063, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.459577560424805, "attn_entropy": 0.8231762886047364, "load_balance": 0.0002862169698346406, "g_entropy": 1.4649159908294678, "tau_bias": 0.6229037046432495, "beta": 0.5853058472275734, "train_val_corr": 0.0}
{"t": 9, "ts": 1788185519.3238754, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 69.6794662475586, "attn_entropy": 0.8257243692874908, "load_balance": 0.0002861742686945945, "g_entropy": 1.461679744720459, "tau_bias": 0.6230704188346863, "beta": 0.5309455518921217, "train_val_corr": 0.0}
{"t": 10, "ts": 1788185526.7814894, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 48.761680603027344, "attn_entropy": 0.5702465772628784, "load_balance": 0.0003838359727524221, "g_entropy": 1.3859297275543212, "tau_bias": 0.6152085661888123, "beta": 0.5575546671946844, "train_val_corr": 0.0}
{"t": 11, "ts": 1788185534.2103827, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.5788688659668, "attn_entropy": 0.5147317871451378, "load_balance": 0.00038406047970056533, "g_entropy": 1.3860400199890137, "tau_bias": 0.6293071508407593, "beta": 0.5477651506662369, "train_val_corr": 0.0}
{"t": 12, "ts": 1788185541.5853035, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.70039367675781, "attn_entropy": 0.5417974591255188, "load_balance": 0.00038434368325397373, "g_entropy": 1.3860401153564452, "tau_bias": 0.6237093210220337, "beta": 0.5688629398743311, "train_val_corr": 0.0}
{"t": 13, "ts": 1788185572.289191, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 45.29948425292969, "attn_entropy": 0.7677601310890168, "load_balance": 0.00028995216707699, "g_entropy": 1.465374517440796, "tau_bias": 0.6186417937278748, "beta": 0.6048264478643736, "train_val_corr": 0.0}
{"t": 14, "ts": 1788185579.686783, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.650504112243652, "attn_entropy": 0.8182293117046356, "load_balance": 0.0002863924775738269, "g_entropy": 1.4614808082580566, "tau_bias": 0.6225220561027527, "beta": 0.5519735738635063, "train_val_corr": 0.0}
{"t": 15, "ts": 1788185587.4839764, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 20.069316864013672, "attn_entropy": 0.7998610824346543, "load_balance": 0.0002870653581339866, "g_entropy": 1.4644794464111328, "tau_bias": 0.6289250254631042, "beta": 0.5277998621265093, "train_val_corr": 0.0}
{"t": 16, "ts": 1788185595.5738654, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 59.49628448486328, "attn_entropy": 0.7813403218984604, "load_balance": 0.0002890263975132257, "g_entropy": 1.4636032581329346, "tau_bias": 0.6291022300720215, "beta": 0.6045025984446207, "train_val_corr": 0.0}
{"t": 17, "ts": 1788185603.5252485, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.406618118286133, "attn_entropy": 0.7896462112665177, "load_balance": 0.000286051124567166, "g_entropy": 1.4657378673553467, "tau_bias": 0.6289530396461487, "beta": 0.5311192398269972, "train_val_corr": 0.0}
{"t": 18, "ts": 1788185611.2192051, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.1129913330078125, "attn_entropy": 0.8237005710601807, "load_balance": 0.0002857891202438623, "g_entropy": 1.4661868572235108, "tau_bias": 0.6283185482025146, "beta": 0.5354098230600357, "train_val_corr": 0.0}
{"t": 19, "ts": 1788185636.0209112, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.564875602722168, "attn_entropy": 0.8059044003486633, "load_balance": 0.0002855296246707439, "g_entropy": 1.4656651020050049, "tau_bias": 0.6288058161735535, "beta": 0.538608027001222, "train_val_corr": 1.0}
{"t": 20, "ts": 1788185643.5013638, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -6.848313808441162, "attn_entropy": 0.7917820245027543, "load_balance": 0.0002845569455530494, "g_entropy": 1.465065097808838, "tau_bias": 0.6380885243415833, "beta": 0.5596097161372503, "train_val_corr": 1.0}
{"t": 21, "ts": 1788185651.8944535, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 33.77821350097656, "attn_entropy": 0.81202632188797, "load_balance": 0.000285240076482296, "g_entropy": 1.4656100273132324, "tau_bias": 0.626155436038971, "beta": 0.5343370363116264, "train_val_corr": 1.0}
{"t": 22, "ts": 1788185660.0360146, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1.6238824129104614, "attn_entropy": 0.787935508787632, "load_balance": 0.00028613425674848256, "g_entropy": 1.4648627758026123, "tau_bias": 0.6195394992828369, "beta": 0.5633299003044764, "train_val_corr": 1.0}
{"t": 23, "ts": 1788185667.988666, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -8.069734573364258, "attn_entropy": 0.6257252812385559, "load_balance": 0.000384907575789839, "g_entropy": 1.3860900402069092, "tau_bias": 0.617311954498291, "beta": 0.5406699627637863, "train_val_corr": 0.9999999999999999}
{"t": 24, "ts": 1788185676.0164852, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.00897216796875, "attn_entropy": 0.7747444681823253, "load_balance": 0.00028500736807473003, "g_entropy": 1.4630655288696288, "tau_bias": 0.6114614009857178, "beta": 0.5373064577579498, "train_val_corr": 0.9999999999999998}
{"t": 25, "ts": 1788185709.7824228, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3998.843017578125, "attn_entropy": 1.495440673828125, "load_balance": 0.00048668733797967433, "g_entropy": 2.144509124755859, "tau_bias": 0.6064755916595459, "beta": 0.6021270751953125, "train_val_corr": 1.0}
{"t": 26, "ts": 1788185717.5345075, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -199.98667907714844, "attn_entropy": 0.8794123291969299, "load_balance": 0.00028528093243949113, "g_entropy": 1.466221237182617, "tau_bias": 0.6056629419326782, "beta": 0.5263560463984808, "train_val_corr": 1.0}
{"t": 27, "ts": 1788185725.8059585, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -203.42703247070312, "attn_entropy": 0.8564549684524536, "load_balance": 0.0002852251229342073, "g_entropy": 1.4651257991790771, "tau_bias": 0.6055827736854553, "beta": 0.5258180623253187, "train_val_corr": 1.0}
{"t": 28, "ts": 1788185733.9235065, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -195.18453979492188, "attn_entropy": 0.5108907650841502, "load_balance": 0.000385539885610342, "g_entropy": 1.3860358715057373, "tau_bias": 0.6056044697761536, "beta": 0.5256927559773127, "train_val_corr": 1.0}
{"t": 29, "ts": 1788185742.314553, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -187.550048828125, "attn_entropy": 0.8579438865184784, "load_balance": 0.0002829693665262312, "g_entropy": 1.46680645942688, "tau_bias": 0.6031451225280762, "beta": 0.5297710448503494, "train_val_corr": 1.0}
{"t": 30, "ts": 1788185750.3354542, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -176.79010009765625, "attn_entropy": 0.8432735681533814, "load_balance": 0.0002848444215487689, "g_entropy": 1.4667584896087646, "tau_bias": 0.6031692624092102, "beta": 0.5161740357677141, "train_val_corr": 1.0}
{"t": 31, "ts": 1788185778.7929053, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -144.10818481445312, "attn_entropy": 0.852590161561966, "load_balance": 0.00028270724578760564, "g_entropy": 1.4668796062469482, "tau_bias": 0.6048827171325684, "beta": 0.5620639796058337, "train_val_corr": 0.9999999999999999}
{"t": 32, "ts": 1788185786.2936454, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -153.91441345214844, "attn_entropy": 0.8738239169120788, "load_balance": 0.0002818381239194423, "g_entropy": 1.4657196044921874, "tau_bias": 0.6061394810676575, "beta": 0.542254830400149, "train_val_corr": 0.9999999999999998}
{"t": 33, "ts": 1788185794.9827993, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -144.43020629882812, "attn_entropy": 0.8735870003700257, "load_balance": 0.00028421457391232253, "g_entropy": 1.4659023761749268, "tau_bias": 0.6064969897270203, "beta": 0.5419706876079241, "train_val_corr": 1.0}
{"t": 34, "ts": 1788185803.3836684, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -142.05699157714844, "attn_entropy": 0.8412885904312134, "load_balance": 0.0002827222866471857, "g_entropy": 1.4666731357574463, "tau_bias": 0.605866014957428, "beta": 0.53933667888244, "train_val_corr": 1.0}
{"t": 35, "ts": 1788185812.0899694, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -130.4031524658203, "attn_entropy": 0.5464083343744278, "load_balance": 0.000384897505864501, "g_entropy": 1.3861059665679931, "tau_bias": 0.6034478545188904, "beta": 0.5465619017680486, "train_val_corr": 1.0}
{"t": 36, "ts": 1788185820.6750546, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -133.2315216064453, "attn_entropy": 0.5329980954527855, "load_balance": 0.0003849281114526093, "g_entropy": 1.3860649108886718, "tau_bias": 0.602962076663971, "beta": 0.5373459880550703, "train_val_corr": 1.0}
{"t": 0, "ts": 1788185998.871624, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.4348186141432961e-06, "advantage": 599.0568237304688, "attn_entropy": 1.5055978715419769, "load_balance": 0.00048133453819900753, "g_entropy": 2.1467979907989503, "tau_bias": 0.6509265899658203, "beta": 0.5490318660934767, "train_val_corr": 0.0}
{"t": 1, "ts": 1788186005.7163982, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1043356923655665e-07, "advantage": 266.82391357421875, "attn_entropy": 1.4676829725503922, "load_balance": 0.0005325290490873158, "g_entropy": 2.0771127700805665, "tau_bias": 0.6566330194473267, "beta": 0.5521855354309082, "train_val_corr": 0.0}
{"t": 2, "ts": 1788186014.5995207, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0727956950850057e-07, "advantage": 194.95960998535156, "attn_entropy": 1.4972401291131974, "load_balance": 0.0004754547670017928, "g_entropy": 2.1563781261444093, "tau_bias": 0.6638267636299133, "beta": 0.5428364301721255, "train_val_corr": 0.0}
{"t": 3, "ts": 1788186023.602905, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723467741845525e-07, "advantage": 226.34017944335938, "attn_entropy": 1.4835683792829513, "load_balance": 0.0004822212504222989, "g_entropy": 2.143466329574585, "tau_bias": 0.6596271991729736, "beta": 0.5472139815489451, "train_val_corr": 0.0}
{"t": 4, "ts": 1788186032.6294513, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723278737477813e-07, "advantage": 177.79095458984375, "attn_entropy": 1.477675986289978, "load_balance": 0.00048101115389727057, "g_entropy": 2.148718738555908, "tau_bias": 0.6538527011871338, "beta": 0.5503670151034991, "train_val_corr": 0.0}
{"t": 5, "ts": 1788186041.3768604, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 140.633056640625, "attn_entropy": 1.4584775567054749, "load_balance": 0.000483423174591735, "g_entropy": 2.1424437046051024, "tau_bias": 0.6587620377540588, "beta": 0.5300281470020612, "train_val_corr": 0.0}
{"t": 6, "ts": 1788186050.0128667, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.3560333251953, "attn_entropy": 1.4789179682731628, "load_balance": 0.0004942999570630491, "g_entropy": 2.1273443698883057, "tau_bias": 0.655666172504425, "beta": 0.538783073425293, "train_val_corr": 0.0}
{"t": 7, "ts": 1788186073.7853327, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072334126523856e-07, "advantage": 83.16289520263672, "attn_entropy": 1.4178901851177215, "load_balance": 0.0004760083800647408, "g_entropy": 2.1522231578826903, "tau_bias": 0.6635648012161255, "beta": 0.517788290977478, "train_val_corr": 0.0}
{"t": 8, "ts": 1788186082.7195997, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 137.16683959960938, "attn_entropy": 1.4611757278442383, "load_balance": 0.000480431760661304, "g_entropy": 2.1519487857818604, "tau_bias": 0.6625792384147644, "beta": 0.5268450304865837, "train_val_corr": 0.0}
{"t": 9, "ts": 1788186091.7773142, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 200.62782287597656, "attn_entropy": 1.4216911315917968, "load_balance": 0.0004921886720694602, "g_entropy": 2.11695237159729, "tau_bias": 0.6560797095298767, "beta": 0.530722680191199, "train_val_corr": 0.0}
{"t": 10, "ts": 1788186100.7971087, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 53.1185417175293, "attn_entropy": 1.485696229338646, "load_balance": 0.0004740906821098179, "g_entropy": 2.1568515300750732, "tau_bias": 0.6597229838371277, "beta": 0.5008959099650383, "train_val_corr": 0.0}
{"t": 11, "ts": 1788186109.9147935, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.90252685546875, "attn_entropy": 1.4534144885838032, "load_balance": 0.0004709144064690918, "g_entropy": 2.158797788619995, "tau_bias": 0.6676686406135559, "beta": 0.5153869936863581, "train_val_corr": 0.0}
{"t": 12, "ts": 1788186119.5826864, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.87871170043945, "attn_entropy": 1.4190070460084825, "load_balance": 0.00047640383709222076, "g_entropy": 2.1525025367736816, "tau_bias": 0.6668230295181274, "beta": 0.519769474864006, "train_val_corr": 0.0}
{"t": 13, "ts": 1788186148.708504, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 257.27325439453125, "attn_entropy": 1.31828983258456, "load_balance": 0.0004860638815443963, "g_entropy": 2.150734519958496, "tau_bias": 0.675154983997345, "beta": 0.5093301311135292, "train_val_corr": 0.0}
{"t": 14, "ts": 1788186157.3346071, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 100.74104309082031, "attn_entropy": 1.3456658490002156, "load_balance": 0.0004792645981069654, "g_entropy": 2.150221824645996, "tau_bias": 0.6854615211486816, "beta": 0.508221206565698, "train_val_corr": 0.0}
{"t": 15, "ts": 1788186167.096795, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 18.650981903076172, "attn_entropy": 1.4292087719310076, "load_balance": 0.000485490815481171, "g_entropy": 2.1480048656463624, "tau_bias": 0.6926027536392212, "beta": 0.5147445400555929, "train_val_corr": 0.0}
{"t": 16, "ts": 1788186176.6133351, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 413.0403747558594, "attn_entropy": 1.2950055003166199, "load_balance": 0.0004939052509143948, "g_entropy": 2.1398456573486326, "tau_bias": 0.6891778707504272, "beta": 0.51402647793293, "train_val_corr": 0.0}
{"t": 17, "ts": 1788186186.1032476, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.621177673339844, "attn_entropy": 1.4083624958992005, "load_balance": 0.0004795385408215225, "g_entropy": 2.1540215492248533, "tau_bias": 0.6822358965873718, "beta": 0.5011413022875786, "train_val_corr": 0.0}
{"t": 18, "ts": 1788186195.5980237, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -55.10371780395508, "attn_entropy": 1.4446635484695434, "load_balance": 0.0004792632826138288, "g_entropy": 2.1564876556396486, "tau_bias": 0.6794387102127075, "beta": 0.4999256407221158, "train_val_corr": 0.0}
{"t": 19, "ts": 1788186220.496827, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.144454956054688, "attn_entropy": 1.4459418879821897, "load_balance": 0.00047863476793281735, "g_entropy": 2.1547357559204103, "tau_bias": 0.683122456073761, "beta": 0.49972417826453847, "train_val_corr": 1.0}
{"t": 20, "ts": 1788186229.010136, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -39.06784439086914, "attn_entropy": 1.4244419813156128, "load_balance": 0.0004897302598692477, "g_entropy": 2.1440284729003904, "tau_bias": 0.685064435005188, "beta": 0.4991164306799571, "train_val_corr": 1.0}
{"t": 21, "ts": 1788186237.875258, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.42458152770996, "attn_entropy": 1.4445943981409073, "load_balance": 0.0004777545516844839, "g_entropy": 2.1544378757476808, "tau_bias": 0.6845418810844421, "beta": 0.5112601071596146, "train_val_corr": 1.0}
{"t": 22, "ts": 1788186247.277132, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.22417449951172, "attn_entropy": 1.2620552770793438, "load_balance": 0.0004856241459492594, "g_entropy": 2.145492696762085, "tau_bias": 0.681154191493988, "beta": 0.5157825872302055, "train_val_corr": 1.0}
{"t": 23, "ts": 1788186256.9088094, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -41.28977584838867, "attn_entropy": 1.3213276982307434, "load_balance": 0.00047365109203383325, "g_entropy": 2.1580783843994142, "tau_bias": 0.6784226298332214, "beta": 0.5216052383184433, "train_val_corr": 0.9999999999999999}
{"t": 24, "ts": 1788186266.0695941, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -44.803688049316406, "attn_entropy": 1.3907989640720189, "load_balance": 0.0004934421507641673, "g_entropy": 2.1361435890197753, "tau_bias": 0.67706698179245, "beta": 0.5268291781346003, "train_val_corr": 1.0}
{"t": 25, "ts": 1788186293.8798308, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.56611251831055, "attn_entropy": 1.277533913962543, "load_balance": 0.000481176603352651, "g_entropy": 2.14298357963562, "tau_bias": 0.6735971570014954, "beta": 0.5267601013183594, "train_val_corr": 1.0}
{"t": 26, "ts": 1788186302.2976496, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -69.73224639892578, "attn_entropy": 1.4124260872602463, "load_balance": 0.00047720440197736026, "g_entropy": 2.1559309005737304, "tau_bias": 0.6794676780700684, "beta": 0.5280350049336752, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788186311.390778, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -76.29232788085938, "attn_entropy": 1.333499512821436, "load_balance": 0.00047582106781192125, "g_entropy": 2.154751682281494, "tau_bias": 0.6843888163566589, "beta": 0.5278861001133919, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788186320.592725, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -79.59697723388672, "attn_entropy": 1.1647369906306266, "load_balance": 0.000474688271060586, "g_entropy": 2.1559574604034424, "tau_bias": 0.6927041411399841, "beta": 0.5333874498804411, "train_val_corr": 1.0}
{"t": 29, "ts": 1788186330.728892, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -71.95927429199219, "attn_entropy": 1.3872828103601933, "load_balance": 0.00047158711240626874, "g_entropy": 2.158252811431885, "tau_bias": 0.6934722661972046, "beta": 0.5227975298961004, "train_val_corr": 1.0}
{"t": 30, "ts": 1788186340.5170174, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -56.59892654418945, "attn_entropy": 1.2660109900403769, "load_balance": 0.00047486289986409247, "g_entropy": 2.1586379528045656, "tau_bias": 0.7070410251617432, "beta": 0.5412052248915037, "train_val_corr": 1.0}
{"t": 31, "ts": 1788186365.6638665, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -39.59022521972656, "attn_entropy": 1.2854751811828464, "load_balance": 0.0004738228861242533, "g_entropy": 2.157383155822754, "tau_bias": 0.7006568312644958, "beta": 0.5326943397521973, "train_val_corr": 1.0}
{"t": 32, "ts": 1788186373.9802322, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -59.029518127441406, "attn_entropy": 1.3639380924403668, "load_balance": 0.0004755221016239375, "g_entropy": 2.154789686203003, "tau_bias": 0.7042104601860046, "beta": 0.528799536327521, "train_val_corr": 0.9999999999999997}
{"t": 33, "ts": 1788186384.0945468, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -72.53294372558594, "attn_entropy": 1.4145977038890123, "load_balance": 0.0004811233375221491, "g_entropy": 2.1532734870910644, "tau_bias": 0.7116537094116211, "beta": 0.5336480240027109, "train_val_corr": 1.0}
{"t": 34, "ts": 1788186394.600637, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.723888397216797, "attn_entropy": 1.3518692523241043, "load_balance": 0.0004757182381581515, "g_entropy": 2.1570244789123536, "tau_bias": 0.7175579071044922, "beta": 0.5330763906240463, "train_val_corr": 1.0}
{"t": 0, "ts": 1788186595.276636, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788186602.2036242, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788186611.0707474, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788186619.8101163, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788186629.290004, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788186638.2241871, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788186647.1785233, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788186670.3998456, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072568747735204e-07, "advantage": 265.6076965332031, "attn_entropy": 1.5030881226062776, "load_balance": 0.0004761439689900726, "g_entropy": 2.1543633460998537, "tau_bias": 0.6877700686454773, "beta": 0.5814792513847351, "train_val_corr": 0.0}
{"t": 8, "ts": 1788186679.6377695, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.08036422729492, "attn_entropy": 1.3145545154809952, "load_balance": 0.0004747225611936301, "g_entropy": 2.1542311191558836, "tau_bias": 0.6906642913818359, "beta": 0.528576043744882, "train_val_corr": 0.0}
{"t": 9, "ts": 1788186689.0765598, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.843027114868164, "attn_entropy": 1.3495253011584283, "load_balance": 0.00047779259039089085, "g_entropy": 2.152908754348755, "tau_bias": 0.7005805969238281, "beta": 0.5125961527228355, "train_val_corr": 0.0}
{"t": 10, "ts": 1788186697.9843664, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.59051513671875, "attn_entropy": 1.3137300699949264, "load_balance": 0.0004715623043011874, "g_entropy": 2.159061622619629, "tau_bias": 0.6974457502365112, "beta": 0.5270653938253721, "train_val_corr": 0.0}
{"t": 11, "ts": 1788186706.9839811, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 63.567413330078125, "attn_entropy": 1.3178742796182632, "load_balance": 0.0004757127200718969, "g_entropy": 2.1570900440216065, "tau_bias": 0.7036505341529846, "beta": 0.5037190566460291, "train_val_corr": 0.0}
{"t": 12, "ts": 1788186716.497191, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 81.36640167236328, "attn_entropy": 1.3078213602304458, "load_balance": 0.00047625720617361367, "g_entropy": 2.15496187210083, "tau_bias": 0.7044507265090942, "beta": 0.5062859182556471, "train_val_corr": 0.0}
{"t": 13, "ts": 1788186747.969521, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 139.10911560058594, "attn_entropy": 1.3155339658260345, "load_balance": 0.0004734063579235226, "g_entropy": 2.157094430923462, "tau_bias": 0.7016552090644836, "beta": 0.5043448482950529, "train_val_corr": 0.0}
{"t": 14, "ts": 1788186756.207375, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.1669921875, "attn_entropy": 1.3176018953323365, "load_balance": 0.0004729832347948104, "g_entropy": 2.1582408428192137, "tau_bias": 0.7084656953811646, "beta": 0.4924413189291954, "train_val_corr": 0.0}
{"t": 15, "ts": 1788186765.1983864, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.67552375793457, "attn_entropy": 1.3095583647489548, "load_balance": 0.0004747989762108773, "g_entropy": 2.156223726272583, "tau_bias": 0.7248103022575378, "beta": 0.50504020601511, "train_val_corr": 0.0}
{"t": 16, "ts": 1788186774.1296585, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.507124900817871, "attn_entropy": 1.3049489289522171, "load_balance": 0.0004726510320324451, "g_entropy": 2.1587456703186034, "tau_bias": 0.727159857749939, "beta": 0.5168147633473078, "train_val_corr": 0.0}
{"t": 17, "ts": 1788186783.3276792, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.640141487121582, "attn_entropy": 1.2987197622656823, "load_balance": 0.00047588840243406594, "g_entropy": 2.154458427429199, "tau_bias": 0.7372605204582214, "beta": 0.509162075817585, "train_val_corr": 0.0}
{"t": 18, "ts": 1788186792.6177855, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.425833702087402, "attn_entropy": 1.29538391828537, "load_balance": 0.00047366743674501775, "g_entropy": 2.158130407333374, "tau_bias": 0.7522925138473511, "beta": 0.5207645917932192, "train_val_corr": 0.0}
{"t": 19, "ts": 1788186816.9095871, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.008971214294434, "attn_entropy": 1.3026934191584587, "load_balance": 0.00047535528428852556, "g_entropy": 2.15622820854187, "tau_bias": 0.7648354172706604, "beta": 0.5240735610326132, "train_val_corr": 1.0}
{"t": 20, "ts": 1788186825.2128527, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.0892333984375, "attn_entropy": 1.288326808810234, "load_balance": 0.0004718317068181932, "g_entropy": 2.1587572574615477, "tau_bias": 0.7841517925262451, "beta": 0.5151468689242998, "train_val_corr": 1.0}
{"t": 21, "ts": 1788186834.211467, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.89907455444336, "attn_entropy": 1.3007862269878387, "load_balance": 0.0004738116404041648, "g_entropy": 2.1569839000701903, "tau_bias": 0.77274489402771, "beta": 0.5318196639418602, "train_val_corr": 1.0}
{"t": 22, "ts": 1788186843.633891, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.476198196411133, "attn_entropy": 1.2932611420750617, "load_balance": 0.00047460654168389737, "g_entropy": 2.1563860893249513, "tau_bias": 0.7668711543083191, "beta": 0.5303046852350235, "train_val_corr": 1.0}
{"t": 23, "ts": 1788186853.188133, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.281461715698242, "attn_entropy": 1.3086626946926116, "load_balance": 0.00047534722252748904, "g_entropy": 2.1567180156707764, "tau_bias": 0.7632516026496887, "beta": 0.5222538237770399, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788186862.8372722, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.884878158569336, "attn_entropy": 1.2891508758068084, "load_balance": 0.0004754542780574411, "g_entropy": 2.1558223724365235, "tau_bias": 0.7628621459007263, "beta": 0.5259591713547707, "train_val_corr": 1.0}
{"t": 25, "ts": 1788186896.6152701, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 112.73678588867188, "attn_entropy": 1.529216033220291, "load_balance": 0.0004814352316316217, "g_entropy": 2.147814464569092, "tau_bias": 0.7611204385757446, "beta": 0.5668929815292358, "train_val_corr": 1.0}
{"t": 26, "ts": 1788186905.2691016, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.884578704833984, "attn_entropy": 1.314683187007904, "load_balance": 0.0004732374974992126, "g_entropy": 2.1584380149841307, "tau_bias": 0.7609435319900513, "beta": 0.5235250890254974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788186914.7336051, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.4482307434082, "attn_entropy": 1.3079102873802184, "load_balance": 0.0004728054569568485, "g_entropy": 2.1587852954864504, "tau_bias": 0.7613591551780701, "beta": 0.5282758449514707, "train_val_corr": 1.0}
{"t": 28, "ts": 1788186924.4822254, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.322853088378906, "attn_entropy": 1.3001174703240395, "load_balance": 0.0004734332964289933, "g_entropy": 2.159209394454956, "tau_bias": 0.7568161487579346, "beta": 0.5196411783496538, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788186934.0850885, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.142391204833984, "attn_entropy": 1.2946814641356468, "load_balance": 0.00047475415049120785, "g_entropy": 2.1572482109069826, "tau_bias": 0.752494752407074, "beta": 0.515541118880113, "train_val_corr": 0.9999999999999999}
{"t": 30, "ts": 1788186943.0680633, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.949480056762695, "attn_entropy": 1.300105407834053, "load_balance": 0.0004751511150971055, "g_entropy": 2.156321716308594, "tau_bias": 0.7509691715240479, "beta": 0.5224061657985052, "train_val_corr": 0.9999999999999999}
{"t": 31, "ts": 1788186967.2675598, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.932716369628906, "attn_entropy": 1.278603382408619, "load_balance": 0.0004737613082397729, "g_entropy": 2.1578856468200684, "tau_bias": 0.7508198022842407, "beta": 0.5157291491826376, "train_val_corr": 0.9999999999999999}
{"t": 32, "ts": 1788186974.9212418, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.253692626953125, "attn_entropy": 1.2791134685277938, "load_balance": 0.00047297976561821997, "g_entropy": 2.1593931674957276, "tau_bias": 0.7560440301895142, "beta": 0.5224737003445625, "train_val_corr": 1.0}
{"t": 33, "ts": 1788186984.0233748, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.127723693847656, "attn_entropy": 1.252753832563758, "load_balance": 0.00047304381732828917, "g_entropy": 2.15961332321167, "tau_bias": 0.7601215243339539, "beta": 0.5210563242435455, "train_val_corr": 1.0}
{"t": 0, "ts": 1788187151.083228, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788187158.3341272, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788187167.8715527, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788187177.8740885, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788187187.6190898, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788187196.7998698, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788187207.5157373, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788187232.0808153, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072568747735204e-07, "advantage": 265.6076965332031, "attn_entropy": 1.5030881226062776, "load_balance": 0.0004761439689900726, "g_entropy": 2.1543633460998537, "tau_bias": 0.6877700686454773, "beta": 0.5814792513847351, "train_val_corr": 0.0}
{"t": 8, "ts": 1788187241.8167124, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.08036422729492, "attn_entropy": 1.3145545154809952, "load_balance": 0.0004747225611936301, "g_entropy": 2.1542311191558836, "tau_bias": 0.6906642913818359, "beta": 0.528576043744882, "train_val_corr": 0.0}
{"t": 9, "ts": 1788187251.7322261, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.843027114868164, "attn_entropy": 1.3495253011584283, "load_balance": 0.00047779259039089085, "g_entropy": 2.152908754348755, "tau_bias": 0.7005805969238281, "beta": 0.5125961527228355, "train_val_corr": 0.0}
{"t": 10, "ts": 1788187261.4938598, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.59051513671875, "attn_entropy": 1.3137300699949264, "load_balance": 0.0004715623043011874, "g_entropy": 2.159061622619629, "tau_bias": 0.6974457502365112, "beta": 0.5270653938253721, "train_val_corr": 0.0}
{"t": 11, "ts": 1788187270.5893788, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 63.567413330078125, "attn_entropy": 1.3178742796182632, "load_balance": 0.0004757127200718969, "g_entropy": 2.1570900440216065, "tau_bias": 0.7036505341529846, "beta": 0.5037190566460291, "train_val_corr": 0.0}
{"t": 12, "ts": 1788187280.189438, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 81.36640167236328, "attn_entropy": 1.3078213602304458, "load_balance": 0.00047625720617361367, "g_entropy": 2.15496187210083, "tau_bias": 0.7044507265090942, "beta": 0.5062859182556471, "train_val_corr": 0.0}
{"t": 13, "ts": 1788187311.6104627, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 139.10911560058594, "attn_entropy": 1.3155339658260345, "load_balance": 0.0004734063579235226, "g_entropy": 2.157094430923462, "tau_bias": 0.7016552090644836, "beta": 0.5043448482950529, "train_val_corr": 0.0}
{"t": 14, "ts": 1788187320.5696986, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.1669921875, "attn_entropy": 1.3176018953323365, "load_balance": 0.0004729832347948104, "g_entropy": 2.1582408428192137, "tau_bias": 0.7084656953811646, "beta": 0.4924413189291954, "train_val_corr": 0.0}
{"t": 15, "ts": 1788187330.68815, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.67552375793457, "attn_entropy": 1.3095583647489548, "load_balance": 0.0004747989762108773, "g_entropy": 2.156223726272583, "tau_bias": 0.7248103022575378, "beta": 0.50504020601511, "train_val_corr": 0.0}
{"t": 16, "ts": 1788187340.3314955, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.507124900817871, "attn_entropy": 1.3049489289522171, "load_balance": 0.0004726510320324451, "g_entropy": 2.1587456703186034, "tau_bias": 0.727159857749939, "beta": 0.5168147633473078, "train_val_corr": 0.0}
{"t": 17, "ts": 1788187351.5662198, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.640141487121582, "attn_entropy": 1.2987197622656823, "load_balance": 0.00047588840243406594, "g_entropy": 2.154458427429199, "tau_bias": 0.7372605204582214, "beta": 0.509162075817585, "train_val_corr": 0.0}
{"t": 18, "ts": 1788187360.7903626, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.425833702087402, "attn_entropy": 1.29538391828537, "load_balance": 0.00047366743674501775, "g_entropy": 2.158130407333374, "tau_bias": 0.7522925138473511, "beta": 0.5207645917932192, "train_val_corr": 0.0}
{"t": 19, "ts": 1788187383.2838426, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.008971214294434, "attn_entropy": 1.3026934191584587, "load_balance": 0.00047535528428852556, "g_entropy": 2.15622820854187, "tau_bias": 0.7648354172706604, "beta": 0.5240735610326132, "train_val_corr": 1.0}
{"t": 20, "ts": 1788187392.3732817, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.0892333984375, "attn_entropy": 1.288326808810234, "load_balance": 0.0004718317068181932, "g_entropy": 2.1587572574615477, "tau_bias": 0.7841517925262451, "beta": 0.5151468689242998, "train_val_corr": 1.0}
{"t": 21, "ts": 1788187402.182254, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.89907455444336, "attn_entropy": 1.3007862269878387, "load_balance": 0.0004738116404041648, "g_entropy": 2.1569839000701903, "tau_bias": 0.77274489402771, "beta": 0.5318196639418602, "train_val_corr": 1.0}
{"t": 22, "ts": 1788187412.1359262, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.476198196411133, "attn_entropy": 1.2932611420750617, "load_balance": 0.00047460654168389737, "g_entropy": 2.1563860893249513, "tau_bias": 0.7668711543083191, "beta": 0.5303046852350235, "train_val_corr": 1.0}
{"t": 23, "ts": 1788187421.8677268, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.281461715698242, "attn_entropy": 1.3086626946926116, "load_balance": 0.00047534722252748904, "g_entropy": 2.1567180156707764, "tau_bias": 0.7632516026496887, "beta": 0.5222538237770399, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788187431.792922, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.884878158569336, "attn_entropy": 1.2891508758068084, "load_balance": 0.0004754542780574411, "g_entropy": 2.1558223724365235, "tau_bias": 0.7628621459007263, "beta": 0.5259591713547707, "train_val_corr": 1.0}
{"t": 25, "ts": 1788187466.3128855, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 112.73678588867188, "attn_entropy": 1.529216033220291, "load_balance": 0.0004814352316316217, "g_entropy": 2.147814464569092, "tau_bias": 0.7611204385757446, "beta": 0.5668929815292358, "train_val_corr": 1.0}
{"t": 26, "ts": 1788187475.4859295, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.884578704833984, "attn_entropy": 1.314683187007904, "load_balance": 0.0004732374974992126, "g_entropy": 2.1584380149841307, "tau_bias": 0.7609435319900513, "beta": 0.5235250890254974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788187485.323601, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.4482307434082, "attn_entropy": 1.3079102873802184, "load_balance": 0.0004728054569568485, "g_entropy": 2.1587852954864504, "tau_bias": 0.7613591551780701, "beta": 0.5282758449514707, "train_val_corr": 1.0}
{"t": 28, "ts": 1788187495.2692654, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.322853088378906, "attn_entropy": 1.3001174703240395, "load_balance": 0.0004734332964289933, "g_entropy": 2.159209394454956, "tau_bias": 0.7568161487579346, "beta": 0.5196411783496538, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788187505.2779272, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.142391204833984, "attn_entropy": 1.2946814641356468, "load_balance": 0.00047475415049120785, "g_entropy": 2.1572482109069826, "tau_bias": 0.752494752407074, "beta": 0.515541118880113, "train_val_corr": 0.9999999999999999}
{"t": 30, "ts": 1788187515.2381504, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.949480056762695, "attn_entropy": 1.300105407834053, "load_balance": 0.0004751511150971055, "g_entropy": 2.156321716308594, "tau_bias": 0.7509691715240479, "beta": 0.5224061657985052, "train_val_corr": 0.9999999999999999}
{"t": 31, "ts": 1788187540.5854976, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.932716369628906, "attn_entropy": 1.278603382408619, "load_balance": 0.0004737613082397729, "g_entropy": 2.1578856468200684, "tau_bias": 0.7508198022842407, "beta": 0.5157291491826376, "train_val_corr": 0.9999999999999999}
{"t": 0, "ts": 1788187595.967387, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.7605805169296218e-06, "advantage": 25.795976638793945, "attn_entropy": 1.2769749358296394, "load_balance": 0.0004723635909613222, "g_entropy": 2.1588505268096925, "tau_bias": 0.7485828399658203, "beta": 0.5183995515108109, "train_val_corr": 0.0}
{"t": 1, "ts": 1788187603.210479, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.073454936635244e-07, "advantage": 109.83809661865234, "attn_entropy": 1.277151470631361, "load_balance": 0.0004704176739323884, "g_entropy": 2.1588789939880373, "tau_bias": 0.7468885779380798, "beta": 0.546175387998422, "train_val_corr": 0.0}
{"t": 2, "ts": 1788187612.066695, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": 74.46682739257812, "attn_entropy": 1.285459078103304, "load_balance": 0.0004729449574369937, "g_entropy": 2.1595772743225097, "tau_bias": 0.7428246140480042, "beta": 0.5663055156668028, "train_val_corr": 0.0}
{"t": 3, "ts": 1788187620.8831797, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 11.853382110595703, "attn_entropy": 1.270102073252201, "load_balance": 0.0004734263347927481, "g_entropy": 2.1595442295074463, "tau_bias": 0.7469478845596313, "beta": 0.5341440911094347, "train_val_corr": 0.0}
{"t": 4, "ts": 1788187629.9103644, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 78.81743621826172, "attn_entropy": 1.262566977739334, "load_balance": 0.00047272552619688214, "g_entropy": 2.1592212677001954, "tau_bias": 0.7429232001304626, "beta": 0.6086501578489939, "train_val_corr": 0.0}
{"t": 5, "ts": 1788187639.4993653, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 56.10427474975586, "attn_entropy": 1.2604252785444259, "load_balance": 0.00047390773543156686, "g_entropy": 2.157368040084839, "tau_bias": 0.7503805756568909, "beta": 0.5404845227797827, "train_val_corr": 0.0}
{"t": 6, "ts": 1788187649.0968084, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 21.078815460205078, "attn_entropy": 1.2729432702064514, "load_balance": 0.0004731423745397478, "g_entropy": 2.1586989402770995, "tau_bias": 0.7485483884811401, "beta": 0.5818909530838331, "train_val_corr": 0.0}
{"t": 7, "ts": 1788187680.3878133, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072332279112743e-07, "advantage": 157.56222534179688, "attn_entropy": 1.496801894903183, "load_balance": 0.0004814775369595736, "g_entropy": 2.147589588165283, "tau_bias": 0.7478504776954651, "beta": 0.5416291356086731, "train_val_corr": 0.0}
{"t": 8, "ts": 1788187689.5164957, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 30.414806365966797, "attn_entropy": 1.2749261945486068, "load_balance": 0.00047222999855875967, "g_entropy": 2.1593515872955322, "tau_bias": 0.7579589486122131, "beta": 0.5375691105922064, "train_val_corr": 0.0}
{"t": 9, "ts": 1788187698.025954, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 17.850589752197266, "attn_entropy": 1.261942657828331, "load_balance": 0.00047199606779031457, "g_entropy": 2.159786033630371, "tau_bias": 0.761876106262207, "beta": 0.5741494869192442, "train_val_corr": 0.0}
{"t": 10, "ts": 1788187707.016322, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.73984909057617, "attn_entropy": 1.2669324070215224, "load_balance": 0.0004726758925244212, "g_entropy": 2.158695125579834, "tau_bias": 0.7615060210227966, "beta": 0.5408812786142031, "train_val_corr": 0.0}
{"t": 11, "ts": 1788187715.8642678, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.19796085357666, "attn_entropy": 1.2702761225402355, "load_balance": 0.0004723931837361306, "g_entropy": 2.159236478805542, "tau_bias": 0.7710251212120056, "beta": 0.5209901432196299, "train_val_corr": 0.0}
{"t": 12, "ts": 1788187724.9015925, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.47568130493164, "attn_entropy": 1.2663661047816277, "load_balance": 0.0004722140089143068, "g_entropy": 2.159245729446411, "tau_bias": 0.7741991877555847, "beta": 0.5344529350598654, "train_val_corr": 0.0}
{"t": 13, "ts": 1788187748.6084714, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 27.720062255859375, "attn_entropy": 1.2536179058253765, "load_balance": 0.0004721729957964271, "g_entropy": 2.157918643951416, "tau_bias": 0.7812412977218628, "beta": 0.5251122663418452, "train_val_corr": 0.0}
{"t": 14, "ts": 1788187757.0212815, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.290483474731445, "attn_entropy": 1.2481800734996795, "load_balance": 0.0004722143756225705, "g_entropy": 2.1590296745300295, "tau_bias": 0.7943588495254517, "beta": 0.511164036889871, "train_val_corr": 0.0}
{"t": 15, "ts": 1788187766.975163, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.464431285858154, "attn_entropy": 1.2651068449020386, "load_balance": 0.000474543176824227, "g_entropy": 2.159158229827881, "tau_bias": 0.81340491771698, "beta": 0.5158639873067538, "train_val_corr": 0.0}
{"t": 16, "ts": 1788187777.1383626, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.073097229003906, "attn_entropy": 1.1869518220424653, "load_balance": 0.0004705667786765844, "g_entropy": 2.1598388671875, "tau_bias": 0.8110300898551941, "beta": 0.5402356684207916, "train_val_corr": 0.0}
{"t": 17, "ts": 1788187787.5743113, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.032845497131348, "attn_entropy": 1.2683328941464425, "load_balance": 0.00047109443112276496, "g_entropy": 2.158835744857788, "tau_bias": 0.8183087110519409, "beta": 0.5068948542078336, "train_val_corr": 0.0}
{"t": 18, "ts": 1788187797.9310262, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -6.624350547790527, "attn_entropy": 1.2542282029986382, "load_balance": 0.000471277954056859, "g_entropy": 2.159781980514526, "tau_bias": 0.8290609121322632, "beta": 0.5180466820796331, "train_val_corr": 0.0}
{"t": 0, "ts": 1788187925.2703633, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788187932.5964642, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788187942.476692, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788187951.6718645, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788187960.7910285, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788187969.8274949, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788187978.833888, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788188003.4876668, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072568747735204e-07, "advantage": 265.6076965332031, "attn_entropy": 1.5030881226062776, "load_balance": 0.0004761439689900726, "g_entropy": 2.1543633460998537, "tau_bias": 0.6877700686454773, "beta": 0.5814792513847351, "train_val_corr": 0.0}
{"t": 8, "ts": 1788188013.3284113, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.08036422729492, "attn_entropy": 1.3145545154809952, "load_balance": 0.0004747225611936301, "g_entropy": 2.1542311191558836, "tau_bias": 0.6906642913818359, "beta": 0.528576043744882, "train_val_corr": 0.0}
{"t": 9, "ts": 1788188023.217421, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.843027114868164, "attn_entropy": 1.3495253011584283, "load_balance": 0.00047779259039089085, "g_entropy": 2.152908754348755, "tau_bias": 0.7005805969238281, "beta": 0.5125961527228355, "train_val_corr": 0.0}
{"t": 10, "ts": 1788188032.2154124, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.59051513671875, "attn_entropy": 1.3137300699949264, "load_balance": 0.0004715623043011874, "g_entropy": 2.159061622619629, "tau_bias": 0.6974457502365112, "beta": 0.5270653938253721, "train_val_corr": 0.0}
{"t": 11, "ts": 1788188041.3065155, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 63.567413330078125, "attn_entropy": 1.3178742796182632, "load_balance": 0.0004757127200718969, "g_entropy": 2.1570900440216065, "tau_bias": 0.7036505341529846, "beta": 0.5037190566460291, "train_val_corr": 0.0}
{"t": 12, "ts": 1788188050.978173, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 81.36640167236328, "attn_entropy": 1.3078213602304458, "load_balance": 0.00047625720617361367, "g_entropy": 2.15496187210083, "tau_bias": 0.7044507265090942, "beta": 0.5062859182556471, "train_val_corr": 0.0}
{"t": 13, "ts": 1788188076.475996, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 139.10911560058594, "attn_entropy": 1.3155339658260345, "load_balance": 0.0004734063579235226, "g_entropy": 2.157094430923462, "tau_bias": 0.7016552090644836, "beta": 0.5043448482950529, "train_val_corr": 0.0}
{"t": 14, "ts": 1788188085.4112704, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.1669921875, "attn_entropy": 1.3176018953323365, "load_balance": 0.0004729832347948104, "g_entropy": 2.1582408428192137, "tau_bias": 0.7084656953811646, "beta": 0.4924413189291954, "train_val_corr": 0.0}
{"t": 15, "ts": 1788188094.7870047, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.67552375793457, "attn_entropy": 1.3095583647489548, "load_balance": 0.0004747989762108773, "g_entropy": 2.156223726272583, "tau_bias": 0.7248103022575378, "beta": 0.50504020601511, "train_val_corr": 0.0}
{"t": 16, "ts": 1788188103.9703276, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.507124900817871, "attn_entropy": 1.3049489289522171, "load_balance": 0.0004726510320324451, "g_entropy": 2.1587456703186034, "tau_bias": 0.727159857749939, "beta": 0.5168147633473078, "train_val_corr": 0.0}
{"t": 17, "ts": 1788188114.0955122, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.640141487121582, "attn_entropy": 1.2987197622656823, "load_balance": 0.00047588840243406594, "g_entropy": 2.154458427429199, "tau_bias": 0.7372605204582214, "beta": 0.509162075817585, "train_val_corr": 0.0}
{"t": 18, "ts": 1788188122.9222507, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.425833702087402, "attn_entropy": 1.29538391828537, "load_balance": 0.00047366743674501775, "g_entropy": 2.158130407333374, "tau_bias": 0.7522925138473511, "beta": 0.5207645917932192, "train_val_corr": 0.0}
{"t": 19, "ts": 1788188145.5091953, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.008971214294434, "attn_entropy": 1.3026934191584587, "load_balance": 0.00047535528428852556, "g_entropy": 2.15622820854187, "tau_bias": 0.7648354172706604, "beta": 0.5240735610326132, "train_val_corr": 1.0}
{"t": 20, "ts": 1788188153.879983, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.0892333984375, "attn_entropy": 1.288326808810234, "load_balance": 0.0004718317068181932, "g_entropy": 2.1587572574615477, "tau_bias": 0.7841517925262451, "beta": 0.5151468689242998, "train_val_corr": 1.0}
{"t": 21, "ts": 1788188163.1839936, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.89907455444336, "attn_entropy": 1.3007862269878387, "load_balance": 0.0004738116404041648, "g_entropy": 2.1569839000701903, "tau_bias": 0.77274489402771, "beta": 0.5318196639418602, "train_val_corr": 1.0}
{"t": 22, "ts": 1788188172.9184816, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.476198196411133, "attn_entropy": 1.2932611420750617, "load_balance": 0.00047460654168389737, "g_entropy": 2.1563860893249513, "tau_bias": 0.7668711543083191, "beta": 0.5303046852350235, "train_val_corr": 1.0}
{"t": 23, "ts": 1788188182.2903192, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.281461715698242, "attn_entropy": 1.3086626946926116, "load_balance": 0.00047534722252748904, "g_entropy": 2.1567180156707764, "tau_bias": 0.7632516026496887, "beta": 0.5222538237770399, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788188191.6346354, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.884878158569336, "attn_entropy": 1.2891508758068084, "load_balance": 0.0004754542780574411, "g_entropy": 2.1558223724365235, "tau_bias": 0.7628621459007263, "beta": 0.5259591713547707, "train_val_corr": 1.0}
{"t": 25, "ts": 1788188216.483831, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 112.73678588867188, "attn_entropy": 1.529216033220291, "load_balance": 0.0004814352316316217, "g_entropy": 2.147814464569092, "tau_bias": 0.7611204385757446, "beta": 0.5668929815292358, "train_val_corr": 1.0}
{"t": 26, "ts": 1788188225.1366181, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.884578704833984, "attn_entropy": 1.314683187007904, "load_balance": 0.0004732374974992126, "g_entropy": 2.1584380149841307, "tau_bias": 0.7609435319900513, "beta": 0.5235250890254974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788188234.324072, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.4482307434082, "attn_entropy": 1.3079102873802184, "load_balance": 0.0004728054569568485, "g_entropy": 2.1587852954864504, "tau_bias": 0.7613591551780701, "beta": 0.5282758449514707, "train_val_corr": 1.0}
{"t": 28, "ts": 1788188243.5932884, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.322853088378906, "attn_entropy": 1.3001174703240395, "load_balance": 0.0004734332964289933, "g_entropy": 2.159209394454956, "tau_bias": 0.7568161487579346, "beta": 0.5196411783496538, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788188252.6157017, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.142391204833984, "attn_entropy": 1.2946814641356468, "load_balance": 0.00047475415049120785, "g_entropy": 2.1572482109069826, "tau_bias": 0.752494752407074, "beta": 0.515541118880113, "train_val_corr": 0.9999999999999999}
{"t": 30, "ts": 1788188261.8159149, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.949480056762695, "attn_entropy": 1.300105407834053, "load_balance": 0.0004751511150971055, "g_entropy": 2.156321716308594, "tau_bias": 0.7509691715240479, "beta": 0.5224061657985052, "train_val_corr": 0.9999999999999999}
{"t": 31, "ts": 1788188287.9681, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.932716369628906, "attn_entropy": 1.278603382408619, "load_balance": 0.0004737613082397729, "g_entropy": 2.1578856468200684, "tau_bias": 0.7508198022842407, "beta": 0.5157291491826376, "train_val_corr": 0.9999999999999999}
{"t": 32, "ts": 1788188296.6735933, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.253692626953125, "attn_entropy": 1.2791134685277938, "load_balance": 0.00047297976561821997, "g_entropy": 2.1593931674957276, "tau_bias": 0.7560440301895142, "beta": 0.5224737003445625, "train_val_corr": 1.0}
{"t": 33, "ts": 1788188306.4342666, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.127723693847656, "attn_entropy": 1.252753832563758, "load_balance": 0.00047304381732828917, "g_entropy": 2.15961332321167, "tau_bias": 0.7601215243339539, "beta": 0.5210563242435455, "train_val_corr": 1.0}
{"t": 0, "ts": 1788188364.8939598, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.7458684169469052e-06, "advantage": 25.78700065612793, "attn_entropy": 1.2587751515209675, "load_balance": 0.0004732281900942326, "g_entropy": 2.158819007873535, "tau_bias": 0.7578601837158203, "beta": 0.5181377256910006, "train_val_corr": 0.0}
{"t": 1, "ts": 1788188372.2740827, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0738585249091557e-07, "advantage": 113.50968170166016, "attn_entropy": 1.2186762800440192, "load_balance": 0.00046942867920733987, "g_entropy": 2.158850622177124, "tau_bias": 0.7553792595863342, "beta": 0.5568753679593405, "train_val_corr": 0.0}
{"t": 2, "ts": 1788188381.9716935, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 55.54576873779297, "attn_entropy": 1.2434848250821233, "load_balance": 0.00047274031094275416, "g_entropy": 2.1595603466033935, "tau_bias": 0.7518952488899231, "beta": 0.5411183660229048, "train_val_corr": 0.0}
{"t": 3, "ts": 1788188390.9646964, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 15.889249801635742, "attn_entropy": 1.2370203781872988, "load_balance": 0.00047232012148015203, "g_entropy": 2.159529113769531, "tau_bias": 0.7542800903320312, "beta": 0.5437234689791998, "train_val_corr": 0.0}
{"t": 4, "ts": 1788188400.274937, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 93.42957305908203, "attn_entropy": 1.2187264343723654, "load_balance": 0.00047178353415802123, "g_entropy": 2.159207344055176, "tau_bias": 0.7478312253952026, "beta": 0.5672908325990041, "train_val_corr": 0.0}
{"t": 5, "ts": 1788188409.9342008, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 79.682861328125, "attn_entropy": 1.2170932799577714, "load_balance": 0.00047158916131593285, "g_entropy": 2.1573222160339354, "tau_bias": 0.7580018639564514, "beta": 0.519164003431797, "train_val_corr": 0.0}
{"t": 6, "ts": 1788188420.379662, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 27.12491226196289, "attn_entropy": 1.2273634843528272, "load_balance": 0.00047166267759166656, "g_entropy": 2.1586500644683837, "tau_bias": 0.7620314359664917, "beta": 0.5431018273035685, "train_val_corr": 0.0}
{"t": 7, "ts": 1788188444.7803893, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4119.74365234375, "attn_entropy": 1.8731523036956788, "load_balance": 0.0004884035850409418, "g_entropy": 2.147256278991699, "tau_bias": 0.7598875164985657, "beta": 0.5543126463890076, "train_val_corr": 0.0}
{"t": 8, "ts": 1788188453.6206024, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -157.47552490234375, "attn_entropy": 1.2263733707368374, "load_balance": 0.00047070596483536067, "g_entropy": 2.1593297004699705, "tau_bias": 0.7666877508163452, "beta": 0.5099132830897967, "train_val_corr": 0.0}
{"t": 9, "ts": 1788188462.8845003, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -185.26902770996094, "attn_entropy": 1.2057709664106369, "load_balance": 0.0004705355153419077, "g_entropy": 2.1597659587860107, "tau_bias": 0.7759506106376648, "beta": 0.5153335332870483, "train_val_corr": 0.0}
{"t": 10, "ts": 1788188471.6697319, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -148.4285888671875, "attn_entropy": 1.2117539204657077, "load_balance": 0.0004705455794464797, "g_entropy": 2.1586417675018312, "tau_bias": 0.7764886617660522, "beta": 0.550000381966432, "train_val_corr": 0.0}
{"t": 11, "ts": 1788188480.8142436, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -134.01446533203125, "attn_entropy": 1.204718487150967, "load_balance": 0.0004712004389148206, "g_entropy": 2.1592030048370363, "tau_bias": 0.7841323614120483, "beta": 0.49463725835084915, "train_val_corr": 0.0}
{"t": 12, "ts": 1788188490.4221635, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -136.0907745361328, "attn_entropy": 1.192477452941239, "load_balance": 0.0004741683253087103, "g_entropy": 2.1591843605041503, "tau_bias": 0.7999434471130371, "beta": 0.5072264944513639, "train_val_corr": 0.0}
{"t": 13, "ts": 1788188514.2295418, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -119.8198013305664, "attn_entropy": 0.9700731949880719, "load_balance": 0.00046823714510537686, "g_entropy": 2.157814121246338, "tau_bias": 0.7994530200958252, "beta": 0.5250617191195488, "train_val_corr": 0.0}
{"t": 14, "ts": 1788188522.52685, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -124.06698608398438, "attn_entropy": 1.170475934445858, "load_balance": 0.0004736065689940006, "g_entropy": 2.159014081954956, "tau_bias": 0.7978566288948059, "beta": 0.5233488778273264, "train_val_corr": 0.0}
{"t": 15, "ts": 1788188531.6946356, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -127.26642608642578, "attn_entropy": 1.2288680285215379, "load_balance": 0.0004744491307064891, "g_entropy": 2.1591484546661377, "tau_bias": 0.8195734620094299, "beta": 0.5051519672075907, "train_val_corr": 0.0}
{"t": 16, "ts": 1788188541.2141676, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -147.28416442871094, "attn_entropy": 1.2579207994043826, "load_balance": 0.000472750881453976, "g_entropy": 2.1598021030426025, "tau_bias": 0.839677095413208, "beta": 0.5118981078267097, "train_val_corr": 0.0}
{"t": 17, "ts": 1788188551.204239, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -34.49756622314453, "attn_entropy": 0.5873400367796421, "load_balance": 0.00048094798694364727, "g_entropy": 2.1587607860565186, "tau_bias": 0.842115581035614, "beta": 0.5207289805014929, "train_val_corr": 0.0}
{"t": 18, "ts": 1788188560.7966087, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -84.90771484375, "attn_entropy": 0.5525470890104771, "load_balance": 0.0004750160092953593, "g_entropy": 2.159753704071045, "tau_bias": 0.8480371832847595, "beta": 0.4963963230450948, "train_val_corr": 0.0}
{"t": 0, "ts": 1788188644.8133588, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788188651.382603, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788188660.1266751, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788188669.1036398, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788188677.9643366, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788188686.478399, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788188695.8755589, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788188721.0872416, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072568747735204e-07, "advantage": 265.6076965332031, "attn_entropy": 1.5030881226062776, "load_balance": 0.0004761439689900726, "g_entropy": 2.1543633460998537, "tau_bias": 0.6877700686454773, "beta": 0.5814792513847351, "train_val_corr": 0.0}
{"t": 8, "ts": 1788188730.3019266, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.08036422729492, "attn_entropy": 1.3145545154809952, "load_balance": 0.0004747225611936301, "g_entropy": 2.1542311191558836, "tau_bias": 0.6906642913818359, "beta": 0.528576043744882, "train_val_corr": 0.0}
{"t": 9, "ts": 1788188739.5259671, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.843027114868164, "attn_entropy": 1.3495253011584283, "load_balance": 0.00047779259039089085, "g_entropy": 2.152908754348755, "tau_bias": 0.7005805969238281, "beta": 0.5125961527228355, "train_val_corr": 0.0}
{"t": 10, "ts": 1788188748.7694767, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.59051513671875, "attn_entropy": 1.3137300699949264, "load_balance": 0.0004715623043011874, "g_entropy": 2.159061622619629, "tau_bias": 0.6974457502365112, "beta": 0.5270653938253721, "train_val_corr": 0.0}
{"t": 11, "ts": 1788188758.06602, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 63.567413330078125, "attn_entropy": 1.3178742796182632, "load_balance": 0.0004757127200718969, "g_entropy": 2.1570900440216065, "tau_bias": 0.7036505341529846, "beta": 0.5037190566460291, "train_val_corr": 0.0}
{"t": 12, "ts": 1788188767.418107, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 81.36640167236328, "attn_entropy": 1.3078213602304458, "load_balance": 0.00047625720617361367, "g_entropy": 2.15496187210083, "tau_bias": 0.7044507265090942, "beta": 0.5062859182556471, "train_val_corr": 0.0}
{"t": 13, "ts": 1788188789.9641218, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 139.10911560058594, "attn_entropy": 1.3155339658260345, "load_balance": 0.0004734063579235226, "g_entropy": 2.157094430923462, "tau_bias": 0.7016552090644836, "beta": 0.5043448482950529, "train_val_corr": 0.0}
{"t": 14, "ts": 1788188798.2010918, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.1669921875, "attn_entropy": 1.3176018953323365, "load_balance": 0.0004729832347948104, "g_entropy": 2.1582408428192137, "tau_bias": 0.7084656953811646, "beta": 0.4924413189291954, "train_val_corr": 0.0}
{"t": 15, "ts": 1788188807.4893813, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.67552375793457, "attn_entropy": 1.3095583647489548, "load_balance": 0.0004747989762108773, "g_entropy": 2.156223726272583, "tau_bias": 0.7248103022575378, "beta": 0.50504020601511, "train_val_corr": 0.0}
{"t": 16, "ts": 1788188816.8769546, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.507124900817871, "attn_entropy": 1.3049489289522171, "load_balance": 0.0004726510320324451, "g_entropy": 2.1587456703186034, "tau_bias": 0.727159857749939, "beta": 0.5168147633473078, "train_val_corr": 0.0}
{"t": 17, "ts": 1788188826.0690765, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.640141487121582, "attn_entropy": 1.2987197622656823, "load_balance": 0.00047588840243406594, "g_entropy": 2.154458427429199, "tau_bias": 0.7372605204582214, "beta": 0.509162075817585, "train_val_corr": 0.0}
{"t": 18, "ts": 1788188835.4849672, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.425833702087402, "attn_entropy": 1.29538391828537, "load_balance": 0.00047366743674501775, "g_entropy": 2.158130407333374, "tau_bias": 0.7522925138473511, "beta": 0.5207645917932192, "train_val_corr": 0.0}
{"t": 19, "ts": 1788188859.234516, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.008971214294434, "attn_entropy": 1.3026934191584587, "load_balance": 0.00047535528428852556, "g_entropy": 2.15622820854187, "tau_bias": 0.7648354172706604, "beta": 0.5240735610326132, "train_val_corr": 1.0}
{"t": 20, "ts": 1788188867.5839925, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.0892333984375, "attn_entropy": 1.288326808810234, "load_balance": 0.0004718317068181932, "g_entropy": 2.1587572574615477, "tau_bias": 0.7841517925262451, "beta": 0.5151468689242998, "train_val_corr": 1.0}
{"t": 21, "ts": 1788188877.6684942, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.89907455444336, "attn_entropy": 1.3007862269878387, "load_balance": 0.0004738116404041648, "g_entropy": 2.1569839000701903, "tau_bias": 0.77274489402771, "beta": 0.5318196639418602, "train_val_corr": 1.0}
{"t": 22, "ts": 1788188887.1209595, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.476198196411133, "attn_entropy": 1.2932611420750617, "load_balance": 0.00047460654168389737, "g_entropy": 2.1563860893249513, "tau_bias": 0.7668711543083191, "beta": 0.5303046852350235, "train_val_corr": 1.0}
{"t": 23, "ts": 1788188895.9775167, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.281461715698242, "attn_entropy": 1.3086626946926116, "load_balance": 0.00047534722252748904, "g_entropy": 2.1567180156707764, "tau_bias": 0.7632516026496887, "beta": 0.5222538237770399, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788188905.6775768, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.884878158569336, "attn_entropy": 1.2891508758068084, "load_balance": 0.0004754542780574411, "g_entropy": 2.1558223724365235, "tau_bias": 0.7628621459007263, "beta": 0.5259591713547707, "train_val_corr": 1.0}
{"t": 25, "ts": 1788188931.08319, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 112.73678588867188, "attn_entropy": 1.529216033220291, "load_balance": 0.0004814352316316217, "g_entropy": 2.147814464569092, "tau_bias": 0.7611204385757446, "beta": 0.5668929815292358, "train_val_corr": 1.0}
{"t": 26, "ts": 1788188939.5295234, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.884578704833984, "attn_entropy": 1.314683187007904, "load_balance": 0.0004732374974992126, "g_entropy": 2.1584380149841307, "tau_bias": 0.7609435319900513, "beta": 0.5235250890254974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788188949.8188093, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.4482307434082, "attn_entropy": 1.3079102873802184, "load_balance": 0.0004728054569568485, "g_entropy": 2.1587852954864504, "tau_bias": 0.7613591551780701, "beta": 0.5282758449514707, "train_val_corr": 1.0}
{"t": 28, "ts": 1788188959.3022356, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.322853088378906, "attn_entropy": 1.3001174703240395, "load_balance": 0.0004734332964289933, "g_entropy": 2.159209394454956, "tau_bias": 0.7568161487579346, "beta": 0.5196411783496538, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788188968.3214028, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.142391204833984, "attn_entropy": 1.2946814641356468, "load_balance": 0.00047475415049120785, "g_entropy": 2.1572482109069826, "tau_bias": 0.752494752407074, "beta": 0.515541118880113, "train_val_corr": 0.9999999999999999}
{"t": 30, "ts": 1788188977.4087348, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.949480056762695, "attn_entropy": 1.300105407834053, "load_balance": 0.0004751511150971055, "g_entropy": 2.156321716308594, "tau_bias": 0.7509691715240479, "beta": 0.5224061657985052, "train_val_corr": 0.9999999999999999}
{"t": 31, "ts": 1788189002.225475, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.932716369628906, "attn_entropy": 1.278603382408619, "load_balance": 0.0004737613082397729, "g_entropy": 2.1578856468200684, "tau_bias": 0.7508198022842407, "beta": 0.5157291491826376, "train_val_corr": 0.9999999999999999}
{"t": 32, "ts": 1788189009.9718223, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.253692626953125, "attn_entropy": 1.2791134685277938, "load_balance": 0.00047297976561821997, "g_entropy": 2.1593931674957276, "tau_bias": 0.7560440301895142, "beta": 0.5224737003445625, "train_val_corr": 1.0}
{"t": 33, "ts": 1788189019.0287595, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.127723693847656, "attn_entropy": 1.252753832563758, "load_balance": 0.00047304381732828917, "g_entropy": 2.15961332321167, "tau_bias": 0.7601215243339539, "beta": 0.5210563242435455, "train_val_corr": 1.0}
{"t": 34, "ts": 1788189028.4875615, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -27.147001266479492, "attn_entropy": 1.265871847420931, "load_balance": 0.0004732055414933711, "g_entropy": 2.158197212219238, "tau_bias": 0.7602447271347046, "beta": 0.5213276048501333, "train_val_corr": 1.0}
{"t": 35, "ts": 1788189037.587792, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.643104553222656, "attn_entropy": 1.265810776501894, "load_balance": 0.0004733260546345264, "g_entropy": 2.159141492843628, "tau_bias": 0.7569579482078552, "beta": 0.5221047525604566, "train_val_corr": 1.0}
{"t": 0, "ts": 1788189082.58859, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.7848218476501643e-06, "advantage": 26.24952507019043, "attn_entropy": 1.2625275120139121, "load_balance": 0.00047298515564762054, "g_entropy": 2.1586628437042235, "tau_bias": 0.7544422149658203, "beta": 0.5187829484542211, "train_val_corr": 0.0}
{"t": 1, "ts": 1788189089.6687233, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0732439054427232e-07, "advantage": 103.2655258178711, "attn_entropy": 1.2540753727778793, "load_balance": 0.00046935887658037243, "g_entropy": 2.158974361419678, "tau_bias": 0.7526364326477051, "beta": 0.5492624913652738, "train_val_corr": 0.0}
{"t": 2, "ts": 1788189099.2111707, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.55939483642578, "attn_entropy": 1.2650707267224788, "load_balance": 0.000472945423098281, "g_entropy": 2.159547758102417, "tau_bias": 0.7502713799476624, "beta": 0.5420457596580187, "train_val_corr": 0.0}
{"t": 3, "ts": 1788189108.1801713, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.008769989013672, "attn_entropy": 1.2396377183496952, "load_balance": 0.000472020567394793, "g_entropy": 2.1591211318969727, "tau_bias": 0.7523955702781677, "beta": 0.5533993095159531, "train_val_corr": 0.0}
{"t": 4, "ts": 1788189117.5119488, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 97.91923522949219, "attn_entropy": 1.2210240922868252, "load_balance": 0.0004731551103759557, "g_entropy": 2.158779335021973, "tau_bias": 0.745317816734314, "beta": 0.5676398326953253, "train_val_corr": 0.0}
{"t": 5, "ts": 1788189127.4189215, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 108.3379898071289, "attn_entropy": 1.22932483330369, "load_balance": 0.0004712896014098078, "g_entropy": 2.1577320098876953, "tau_bias": 0.7586233019828796, "beta": 0.5118241036931673, "train_val_corr": 0.0}
{"t": 6, "ts": 1788189136.3884063, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 52.86005783081055, "attn_entropy": 1.2342828646302224, "load_balance": 0.00047167912707664074, "g_entropy": 2.158523511886597, "tau_bias": 0.7622684240341187, "beta": 0.5224277973175049, "train_val_corr": 0.0}
{"t": 7, "ts": 1788189159.996441, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1926.6318359375, "attn_entropy": 1.9484524726867676, "load_balance": 0.00048530561616644263, "g_entropy": 2.1486411571502684, "tau_bias": 0.759388267993927, "beta": 0.5478860139846802, "train_val_corr": 0.0}
{"t": 8, "ts": 1788189169.267185, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.688501358032227, "attn_entropy": 1.2277922548353672, "load_balance": 0.0004705200961325318, "g_entropy": 2.1595641136169434, "tau_bias": 0.7661486864089966, "beta": 0.5203307444850603, "train_val_corr": 0.0}
{"t": 9, "ts": 1788189178.6923993, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.83421325683594, "attn_entropy": 1.2386636644601823, "load_balance": 0.00047110518789850175, "g_entropy": 2.159706687927246, "tau_bias": 0.7743778824806213, "beta": 0.5203105683128039, "train_val_corr": 0.0}
{"t": 10, "ts": 1788189187.9853702, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.67779541015625, "attn_entropy": 1.2391040414571761, "load_balance": 0.00047166626318357886, "g_entropy": 2.158838129043579, "tau_bias": 0.7782093286514282, "beta": 0.550186055401961, "train_val_corr": 0.0}
{"t": 11, "ts": 1788189197.6373982, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -27.802574157714844, "attn_entropy": 1.2393934484571218, "load_balance": 0.00047175140934996307, "g_entropy": 2.159341907501221, "tau_bias": 0.7849997878074646, "beta": 0.5276291146874428, "train_val_corr": 0.0}
{"t": 12, "ts": 1788189207.0169532, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.76366424560547, "attn_entropy": 1.2550477866083383, "load_balance": 0.00047251692158170047, "g_entropy": 2.1591780185699463, "tau_bias": 0.8090412616729736, "beta": 0.5105513135592142, "train_val_corr": 0.0}
{"t": 13, "ts": 1788189230.3797283, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -59.08256149291992, "attn_entropy": 1.2432089395821095, "load_balance": 0.00047337522846646607, "g_entropy": 2.1575868129730225, "tau_bias": 0.8113448023796082, "beta": 0.5326447735230128, "train_val_corr": 0.0}
{"t": 14, "ts": 1788189239.0059385, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.407575845718384, "attn_entropy": 1.2225934974849224, "load_balance": 0.00047355663846246896, "g_entropy": 2.1591264247894286, "tau_bias": 0.813320517539978, "beta": 0.519411489367485, "train_val_corr": 0.0}
{"t": 15, "ts": 1788189248.3700657, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.384193420410156, "attn_entropy": 1.2267195880413055, "load_balance": 0.00047585545107722285, "g_entropy": 2.159132957458496, "tau_bias": 0.8219267725944519, "beta": 0.5078891888260841, "train_val_corr": 0.0}
{"t": 16, "ts": 1788189258.1250393, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -91.5788345336914, "attn_entropy": 1.0024372309446334, "load_balance": 0.00047958386712707577, "g_entropy": 2.1596553325653076, "tau_bias": 0.8374529480934143, "beta": 0.5092045317093531, "train_val_corr": 0.0}
{"t": 17, "ts": 1788189267.9042244, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -63.3342399597168, "attn_entropy": 0.9766600407660008, "load_balance": 0.0004808165715076029, "g_entropy": 2.1588175296783447, "tau_bias": 0.840945303440094, "beta": 0.5458492636680603, "train_val_corr": 0.0}
{"t": 18, "ts": 1788189277.3653662, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -54.423892974853516, "attn_entropy": 1.0923100724816321, "load_balance": 0.0004791697487235069, "g_entropy": 2.159784173965454, "tau_bias": 0.8458116054534912, "beta": 0.5101249491175016, "train_val_corr": 0.0}
{"t": 0, "ts": 1788189431.1204572, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788189438.007429, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788189447.4809458, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788189456.6808403, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788189466.0177002, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788189474.967702, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788189484.0034463, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788189507.3692272, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072568747735204e-07, "advantage": 265.6076965332031, "attn_entropy": 1.5030881226062776, "load_balance": 0.0004761439689900726, "g_entropy": 2.1543633460998537, "tau_bias": 0.6877700686454773, "beta": 0.5814792513847351, "train_val_corr": 0.0}
{"t": 8, "ts": 1788189516.1261384, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.08036422729492, "attn_entropy": 1.3145545154809952, "load_balance": 0.0004747225611936301, "g_entropy": 2.1542311191558836, "tau_bias": 0.6906642913818359, "beta": 0.528576043744882, "train_val_corr": 0.0}
{"t": 9, "ts": 1788189525.1781442, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.843027114868164, "attn_entropy": 1.3495253011584283, "load_balance": 0.00047779259039089085, "g_entropy": 2.152908754348755, "tau_bias": 0.7005805969238281, "beta": 0.5125961527228355, "train_val_corr": 0.0}
{"t": 10, "ts": 1788189534.0823867, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.59051513671875, "attn_entropy": 1.3137300699949264, "load_balance": 0.0004715623043011874, "g_entropy": 2.159061622619629, "tau_bias": 0.6974457502365112, "beta": 0.5270653938253721, "train_val_corr": 0.0}
{"t": 11, "ts": 1788189542.9649098, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 63.567413330078125, "attn_entropy": 1.3178742796182632, "load_balance": 0.0004757127200718969, "g_entropy": 2.1570900440216065, "tau_bias": 0.7036505341529846, "beta": 0.5037190566460291, "train_val_corr": 0.0}
{"t": 12, "ts": 1788189551.7129157, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 81.36640167236328, "attn_entropy": 1.3078213602304458, "load_balance": 0.00047625720617361367, "g_entropy": 2.15496187210083, "tau_bias": 0.7044507265090942, "beta": 0.5062859182556471, "train_val_corr": 0.0}
{"t": 13, "ts": 1788189576.5362287, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 139.10911560058594, "attn_entropy": 1.3155339658260345, "load_balance": 0.0004734063579235226, "g_entropy": 2.157094430923462, "tau_bias": 0.7016552090644836, "beta": 0.5043448482950529, "train_val_corr": 0.0}
{"t": 14, "ts": 1788189585.4855235, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.1669921875, "attn_entropy": 1.3176018953323365, "load_balance": 0.0004729832347948104, "g_entropy": 2.1582408428192137, "tau_bias": 0.7084656953811646, "beta": 0.4924413189291954, "train_val_corr": 0.0}
{"t": 15, "ts": 1788189594.1284668, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.67552375793457, "attn_entropy": 1.3095583647489548, "load_balance": 0.0004747989762108773, "g_entropy": 2.156223726272583, "tau_bias": 0.7248103022575378, "beta": 0.50504020601511, "train_val_corr": 0.0}
{"t": 16, "ts": 1788189602.9705102, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.507124900817871, "attn_entropy": 1.3049489289522171, "load_balance": 0.0004726510320324451, "g_entropy": 2.1587456703186034, "tau_bias": 0.727159857749939, "beta": 0.5168147633473078, "train_val_corr": 0.0}
{"t": 17, "ts": 1788189611.7807577, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.640141487121582, "attn_entropy": 1.2987197622656823, "load_balance": 0.00047588840243406594, "g_entropy": 2.154458427429199, "tau_bias": 0.7372605204582214, "beta": 0.509162075817585, "train_val_corr": 0.0}
{"t": 18, "ts": 1788189620.5086896, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.425833702087402, "attn_entropy": 1.29538391828537, "load_balance": 0.00047366743674501775, "g_entropy": 2.158130407333374, "tau_bias": 0.7522925138473511, "beta": 0.5207645917932192, "train_val_corr": 0.0}
{"t": 19, "ts": 1788189642.664934, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.008971214294434, "attn_entropy": 1.3026934191584587, "load_balance": 0.00047535528428852556, "g_entropy": 2.15622820854187, "tau_bias": 0.7648354172706604, "beta": 0.5240735610326132, "train_val_corr": 1.0}
{"t": 20, "ts": 1788189650.7841587, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.0892333984375, "attn_entropy": 1.288326808810234, "load_balance": 0.0004718317068181932, "g_entropy": 2.1587572574615477, "tau_bias": 0.7841517925262451, "beta": 0.5151468689242998, "train_val_corr": 1.0}
{"t": 21, "ts": 1788189659.995258, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.89907455444336, "attn_entropy": 1.3007862269878387, "load_balance": 0.0004738116404041648, "g_entropy": 2.1569839000701903, "tau_bias": 0.77274489402771, "beta": 0.5318196639418602, "train_val_corr": 1.0}
{"t": 22, "ts": 1788189669.1724956, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.476198196411133, "attn_entropy": 1.2932611420750617, "load_balance": 0.00047460654168389737, "g_entropy": 2.1563860893249513, "tau_bias": 0.7668711543083191, "beta": 0.5303046852350235, "train_val_corr": 1.0}
{"t": 23, "ts": 1788189677.8829372, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.281461715698242, "attn_entropy": 1.3086626946926116, "load_balance": 0.00047534722252748904, "g_entropy": 2.1567180156707764, "tau_bias": 0.7632516026496887, "beta": 0.5222538237770399, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788189686.7202125, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.884878158569336, "attn_entropy": 1.2891508758068084, "load_balance": 0.0004754542780574411, "g_entropy": 2.1558223724365235, "tau_bias": 0.7628621459007263, "beta": 0.5259591713547707, "train_val_corr": 1.0}
{"t": 25, "ts": 1788189710.3761666, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 112.73678588867188, "attn_entropy": 1.529216033220291, "load_balance": 0.0004814352316316217, "g_entropy": 2.147814464569092, "tau_bias": 0.7611204385757446, "beta": 0.5668929815292358, "train_val_corr": 1.0}
{"t": 26, "ts": 1788189718.5272706, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.884578704833984, "attn_entropy": 1.314683187007904, "load_balance": 0.0004732374974992126, "g_entropy": 2.1584380149841307, "tau_bias": 0.7609435319900513, "beta": 0.5235250890254974, "train_val_corr": 1.0}
{"t": 27, "ts": 1788189727.8256974, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.4482307434082, "attn_entropy": 1.3079102873802184, "load_balance": 0.0004728054569568485, "g_entropy": 2.1587852954864504, "tau_bias": 0.7613591551780701, "beta": 0.5282758449514707, "train_val_corr": 1.0}
{"t": 28, "ts": 1788189737.315172, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.322853088378906, "attn_entropy": 1.3001174703240395, "load_balance": 0.0004734332964289933, "g_entropy": 2.159209394454956, "tau_bias": 0.7568161487579346, "beta": 0.5196411783496538, "train_val_corr": 0.9999999999999999}
{"t": 29, "ts": 1788189746.9732146, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.142391204833984, "attn_entropy": 1.2946814641356468, "load_balance": 0.00047475415049120785, "g_entropy": 2.1572482109069826, "tau_bias": 0.752494752407074, "beta": 0.515541118880113, "train_val_corr": 0.9999999999999999}
{"t": 30, "ts": 1788189756.8732357, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.949480056762695, "attn_entropy": 1.300105407834053, "load_balance": 0.0004751511150971055, "g_entropy": 2.156321716308594, "tau_bias": 0.7509691715240479, "beta": 0.5224061657985052, "train_val_corr": 0.9999999999999999}
{"t": 31, "ts": 1788189784.036075, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.932716369628906, "attn_entropy": 1.278603382408619, "load_balance": 0.0004737613082397729, "g_entropy": 2.1578856468200684, "tau_bias": 0.7508198022842407, "beta": 0.5157291491826376, "train_val_corr": 0.9999999999999999}
{"t": 32, "ts": 1788189792.277676, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.253692626953125, "attn_entropy": 1.2791134685277938, "load_balance": 0.00047297976561821997, "g_entropy": 2.1593931674957276, "tau_bias": 0.7560440301895142, "beta": 0.5224737003445625, "train_val_corr": 1.0}
{"t": 33, "ts": 1788189802.3707762, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.127723693847656, "attn_entropy": 1.252753832563758, "load_balance": 0.00047304381732828917, "g_entropy": 2.15961332321167, "tau_bias": 0.7601215243339539, "beta": 0.5210563242435455, "train_val_corr": 1.0}
{"t": 34, "ts": 1788189812.5701747, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -27.147001266479492, "attn_entropy": 1.265871847420931, "load_balance": 0.0004732055414933711, "g_entropy": 2.158197212219238, "tau_bias": 0.7602447271347046, "beta": 0.5213276048501333, "train_val_corr": 1.0}
{"t": 35, "ts": 1788189822.701205, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.643104553222656, "attn_entropy": 1.265810776501894, "load_balance": 0.0004733260546345264, "g_entropy": 2.159141492843628, "tau_bias": 0.7569579482078552, "beta": 0.5221047525604566, "train_val_corr": 1.0}
{"t": 0, "ts": 1788189855.182386, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.7848218476501643e-06, "advantage": 26.24952507019043, "attn_entropy": 1.2625275120139121, "load_balance": 0.00047298515564762054, "g_entropy": 2.1586628437042235, "tau_bias": 0.7544422149658203, "beta": 0.5187829484542211, "train_val_corr": 0.0}
{"t": 1, "ts": 1788189861.9725907, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0732439054427232e-07, "advantage": 103.2655258178711, "attn_entropy": 1.2540753727778793, "load_balance": 0.00046935887658037243, "g_entropy": 2.158974361419678, "tau_bias": 0.7526364326477051, "beta": 0.5492624913652738, "train_val_corr": 0.0}
{"t": 2, "ts": 1788189871.6301825, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.55939483642578, "attn_entropy": 1.2650707267224788, "load_balance": 0.000472945423098281, "g_entropy": 2.159547758102417, "tau_bias": 0.7502713799476624, "beta": 0.5420457596580187, "train_val_corr": 0.0}
{"t": 3, "ts": 1788189881.7937171, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.008769989013672, "attn_entropy": 1.2396377183496952, "load_balance": 0.000472020567394793, "g_entropy": 2.1591211318969727, "tau_bias": 0.7523955702781677, "beta": 0.5533993095159531, "train_val_corr": 0.0}
{"t": 4, "ts": 1788189891.2002122, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 97.91923522949219, "attn_entropy": 1.2210240922868252, "load_balance": 0.0004731551103759557, "g_entropy": 2.158779335021973, "tau_bias": 0.745317816734314, "beta": 0.5676398326953253, "train_val_corr": 0.0}
{"t": 5, "ts": 1788189900.2164626, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 108.3379898071289, "attn_entropy": 1.22932483330369, "load_balance": 0.0004712896014098078, "g_entropy": 2.1577320098876953, "tau_bias": 0.7586233019828796, "beta": 0.5118241036931673, "train_val_corr": 0.0}
{"t": 6, "ts": 1788189910.0921533, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 52.86005783081055, "attn_entropy": 1.2342828646302224, "load_balance": 0.00047167912707664074, "g_entropy": 2.158523511886597, "tau_bias": 0.7622684240341187, "beta": 0.5224277973175049, "train_val_corr": 0.0}
{"t": 7, "ts": 1788189934.1264713, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1926.6318359375, "attn_entropy": 1.9484524726867676, "load_balance": 0.00048530561616644263, "g_entropy": 2.1486411571502684, "tau_bias": 0.759388267993927, "beta": 0.5478860139846802, "train_val_corr": 0.0}
{"t": 8, "ts": 1788189942.5368435, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.688501358032227, "attn_entropy": 1.2277922548353672, "load_balance": 0.0004705200961325318, "g_entropy": 2.1595641136169434, "tau_bias": 0.7661486864089966, "beta": 0.5203307444850603, "train_val_corr": 0.0}
{"t": 9, "ts": 1788189951.4227297, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -86.83421325683594, "attn_entropy": 1.2386636644601823, "load_balance": 0.00047110518789850175, "g_entropy": 2.159706687927246, "tau_bias": 0.7743778824806213, "beta": 0.5203105683128039, "train_val_corr": 0.0}
{"t": 10, "ts": 1788189960.5198455, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.67779541015625, "attn_entropy": 1.2391040414571761, "load_balance": 0.00047166626318357886, "g_entropy": 2.158838129043579, "tau_bias": 0.7782093286514282, "beta": 0.550186055401961, "train_val_corr": 0.0}
{"t": 11, "ts": 1788189969.8272412, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -27.802574157714844, "attn_entropy": 1.2393934484571218, "load_balance": 0.00047175140934996307, "g_entropy": 2.159341907501221, "tau_bias": 0.7849997878074646, "beta": 0.5276291146874428, "train_val_corr": 0.0}
{"t": 12, "ts": 1788189979.1978288, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.76366424560547, "attn_entropy": 1.2550477866083383, "load_balance": 0.00047251692158170047, "g_entropy": 2.1591780185699463, "tau_bias": 0.8090412616729736, "beta": 0.5105513135592142, "train_val_corr": 0.0}
{"t": 13, "ts": 1788190003.8111513, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -59.08256149291992, "attn_entropy": 1.2432089395821095, "load_balance": 0.00047337522846646607, "g_entropy": 2.1575868129730225, "tau_bias": 0.8113448023796082, "beta": 0.5326447735230128, "train_val_corr": 0.0}
{"t": 14, "ts": 1788190012.4831192, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.407575845718384, "attn_entropy": 1.2225934974849224, "load_balance": 0.00047355663846246896, "g_entropy": 2.1591264247894286, "tau_bias": 0.813320517539978, "beta": 0.519411489367485, "train_val_corr": 0.0}
{"t": 15, "ts": 1788190022.6730964, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.384193420410156, "attn_entropy": 1.2267195880413055, "load_balance": 0.00047585545107722285, "g_entropy": 2.159132957458496, "tau_bias": 0.8219267725944519, "beta": 0.5078891888260841, "train_val_corr": 0.0}
{"t": 16, "ts": 1788190032.6094491, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -91.5788345336914, "attn_entropy": 1.0024372309446334, "load_balance": 0.00047958386712707577, "g_entropy": 2.1596553325653076, "tau_bias": 0.8374529480934143, "beta": 0.5092045317093531, "train_val_corr": 0.0}
{"t": 17, "ts": 1788190042.8892548, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -63.3342399597168, "attn_entropy": 0.9766600407660008, "load_balance": 0.0004808165715076029, "g_entropy": 2.1588175296783447, "tau_bias": 0.840945303440094, "beta": 0.5458492636680603, "train_val_corr": 0.0}
{"t": 18, "ts": 1788190052.4868004, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -54.423892974853516, "attn_entropy": 1.0923100724816321, "load_balance": 0.0004791697487235069, "g_entropy": 2.159784173965454, "tau_bias": 0.8458116054534912, "beta": 0.5101249491175016, "train_val_corr": 0.0}
{"t": 0, "ts": 1788190184.9842267, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788190191.8715909, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788190200.9870362, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788190209.9949539, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788190219.3839586, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788190228.676971, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788190237.3786726, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788190260.9818456, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072568747735204e-07, "advantage": 265.6076965332031, "attn_entropy": 1.5030881226062776, "load_balance": 0.0004761439689900726, "g_entropy": 2.1543633460998537, "tau_bias": 0.6877700686454773, "beta": 0.5814792513847351, "train_val_corr": 0.0}
{"t": 8, "ts": 1788190270.328444, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.08036422729492, "attn_entropy": 1.3145545154809952, "load_balance": 0.0004747225611936301, "g_entropy": 2.1542311191558836, "tau_bias": 0.6906642913818359, "beta": 0.528576043744882, "train_val_corr": 0.0}
{"t": 9, "ts": 1788190279.7115476, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.843027114868164, "attn_entropy": 1.3495253011584283, "load_balance": 0.00047779259039089085, "g_entropy": 2.152908754348755, "tau_bias": 0.7005805969238281, "beta": 0.5125961527228355, "train_val_corr": 0.0}
{"t": 10, "ts": 1788190289.021808, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 186.59051513671875, "attn_entropy": 1.3137300699949264, "load_balance": 0.0004715623043011874, "g_entropy": 2.159061622619629, "tau_bias": 0.6974457502365112, "beta": 0.5270653938253721, "train_val_corr": 0.0}
{"t": 11, "ts": 1788190298.220097, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 63.567413330078125, "attn_entropy": 1.3178742796182632, "load_balance": 0.0004757127200718969, "g_entropy": 2.1570900440216065, "tau_bias": 0.7036505341529846, "beta": 0.5037190566460291, "train_val_corr": 0.0}
{"t": 12, "ts": 1788190307.58759, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 81.36640167236328, "attn_entropy": 1.3078213602304458, "load_balance": 0.00047625720617361367, "g_entropy": 2.15496187210083, "tau_bias": 0.7044507265090942, "beta": 0.5062859182556471, "train_val_corr": 0.0}
{"t": 13, "ts": 1788190331.1783712, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 139.10911560058594, "attn_entropy": 1.3155339658260345, "load_balance": 0.0004734063579235226, "g_entropy": 2.157094430923462, "tau_bias": 0.7016552090644836, "beta": 0.5043448482950529, "train_val_corr": 0.0}
{"t": 14, "ts": 1788190339.1346853, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.1669921875, "attn_entropy": 1.3176018953323365, "load_balance": 0.0004729832347948104, "g_entropy": 2.1582408428192137, "tau_bias": 0.7084656953811646, "beta": 0.4924413189291954, "train_val_corr": 0.0}
{"t": 15, "ts": 1788190347.9117155, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -11.67552375793457, "attn_entropy": 1.3095583647489548, "load_balance": 0.0004747989762108773, "g_entropy": 2.156223726272583, "tau_bias": 0.7248103022575378, "beta": 0.50504020601511, "train_val_corr": 0.0}
{"t": 16, "ts": 1788190356.569914, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.507124900817871, "attn_entropy": 1.3049489289522171, "load_balance": 0.0004726510320324451, "g_entropy": 2.1587456703186034, "tau_bias": 0.727159857749939, "beta": 0.5168147633473078, "train_val_corr": 0.0}
{"t": 17, "ts": 1788190365.902772, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.640141487121582, "attn_entropy": 1.2987197622656823, "load_balance": 0.00047588840243406594, "g_entropy": 2.154458427429199, "tau_bias": 0.7372605204582214, "beta": 0.509162075817585, "train_val_corr": 0.0}
{"t": 18, "ts": 1788190375.121476, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.425833702087402, "attn_entropy": 1.29538391828537, "load_balance": 0.00047366743674501775, "g_entropy": 2.158130407333374, "tau_bias": 0.7522925138473511, "beta": 0.5207645917932192, "train_val_corr": 0.0}
{"t": 19, "ts": 1788190398.9204128, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.008971214294434, "attn_entropy": 1.3026934191584587, "load_balance": 0.00047535528428852556, "g_entropy": 2.15622820854187, "tau_bias": 0.7648354172706604, "beta": 0.5240735610326132, "train_val_corr": 1.0}
{"t": 20, "ts": 1788190407.5656238, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.0892333984375, "attn_entropy": 1.288326808810234, "load_balance": 0.0004718317068181932, "g_entropy": 2.1587572574615477, "tau_bias": 0.7841517925262451, "beta": 0.5151468689242998, "train_val_corr": 1.0}
{"t": 21, "ts": 1788190416.6868908, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.89907455444336, "attn_entropy": 1.3007862269878387, "load_balance": 0.0004738116404041648, "g_entropy": 2.1569839000701903, "tau_bias": 0.77274489402771, "beta": 0.5318196639418602, "train_val_corr": 1.0}
{"t": 22, "ts": 1788190425.6715648, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.476198196411133, "attn_entropy": 1.2932611420750617, "load_balance": 0.00047460654168389737, "g_entropy": 2.1563860893249513, "tau_bias": 0.7668711543083191, "beta": 0.5303046852350235, "train_val_corr": 1.0}
{"t": 23, "ts": 1788190434.8057075, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.281461715698242, "attn_entropy": 1.3086626946926116, "load_balance": 0.00047534722252748904, "g_entropy": 2.1567180156707764, "tau_bias": 0.7632516026496887, "beta": 0.5222538237770399, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788190443.8658404, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.884878158569336, "attn_entropy": 1.2891508758068084, "load_balance": 0.0004754542780574411, "g_entropy": 2.1558223724365235, "tau_bias": 0.7628621459007263, "beta": 0.5259591713547707, "train_val_corr": 1.0}
{"t": 25, "ts": 1788190467.9215167, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -38.36581039428711, "attn_entropy": 1.2978039965033532, "load_balance": 0.00047229216434061525, "g_entropy": 2.1600598335266112, "tau_bias": 0.760733425617218, "beta": 0.5402330284317335, "train_val_corr": 0.9999999999999998}
{"t": 26, "ts": 1788190477.3866282, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.08338928222656, "attn_entropy": 1.3049081861972809, "load_balance": 0.0004718662181403488, "g_entropy": 2.159425973892212, "tau_bias": 0.7567146420478821, "beta": 0.5390619387229284, "train_val_corr": 1.0}
{"t": 27, "ts": 1788190486.773858, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.63517379760742, "attn_entropy": 1.299768754839897, "load_balance": 0.0004716946801636368, "g_entropy": 2.1600194454193113, "tau_bias": 0.752423882484436, "beta": 0.5328036298354467, "train_val_corr": 1.0}
{"t": 28, "ts": 1788190496.0084224, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.12898254394531, "attn_entropy": 1.3032237470149994, "load_balance": 0.000472807209007442, "g_entropy": 2.1600162506103517, "tau_bias": 0.749363899230957, "beta": 0.5292484536767006, "train_val_corr": 1.0}
{"t": 29, "ts": 1788190505.1290627, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.873409271240234, "attn_entropy": 1.3075667679309846, "load_balance": 0.00047288836212828756, "g_entropy": 2.1600436210632323, "tau_bias": 0.7471659779548645, "beta": 0.5337525432308515, "train_val_corr": 0.9999999999999999}
{"t": 30, "ts": 1788190514.7650044, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.5652961730957, "attn_entropy": 1.3012290358543397, "load_balance": 0.00047227072063833476, "g_entropy": 2.160126066207886, "tau_bias": 0.7443758845329285, "beta": 0.524654800693194, "train_val_corr": 1.0}
{"t": 31, "ts": 1788190539.364265, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -31.069141387939453, "attn_entropy": 1.3042496681213378, "load_balance": 0.00047308478388004006, "g_entropy": 2.1599988460540773, "tau_bias": 0.7414298057556152, "beta": 0.5223045572638512, "train_val_corr": 1.0}
{"t": 32, "ts": 1788190549.0077372, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.497608184814453, "attn_entropy": 1.3023498415946961, "load_balance": 0.00047223430592566726, "g_entropy": 2.160101127624512, "tau_bias": 0.7400312423706055, "beta": 0.5200078909595808, "train_val_corr": 1.0}
{"t": 33, "ts": 1788190558.336761, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -27.74846649169922, "attn_entropy": 1.3075928479433059, "load_balance": 0.00047284567845053973, "g_entropy": 2.1593887329101564, "tau_bias": 0.7359461784362793, "beta": 0.518788124124209, "train_val_corr": 0.9999999999999999}
{"t": 34, "ts": 1788190567.490503, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.880428314208984, "attn_entropy": 1.3059355080127717, "load_balance": 0.0004730667162220925, "g_entropy": 2.159861373901367, "tau_bias": 0.7350226640701294, "beta": 0.5147443388899168, "train_val_corr": 1.0}
{"t": 0, "ts": 1788190610.576322, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 4.403885213832837e-06, "advantage": 50.137943267822266, "attn_entropy": 1.533861917257309, "load_balance": 0.00047334444825537504, "g_entropy": 2.1600292682647706, "tau_bias": 0.7367687225341797, "beta": 0.5323572183648745, "train_val_corr": 0.0}
{"t": 1, "ts": 1788190619.0895736, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0728867866637302e-07, "advantage": 71.58637237548828, "attn_entropy": 1.5339849710464477, "load_balance": 0.0004731794586405158, "g_entropy": 2.1602336883544924, "tau_bias": 0.7322222590446472, "beta": 0.5695954859256744, "train_val_corr": 0.0}
{"t": 2, "ts": 1788190629.2309947, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.46499252319336, "attn_entropy": 1.5245151698589325, "load_balance": 0.00047298225690610706, "g_entropy": 2.1600883483886717, "tau_bias": 0.7401109933853149, "beta": 0.5219327385226885, "train_val_corr": 0.0}
{"t": 3, "ts": 1788190639.0720932, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 80.06089782714844, "attn_entropy": 1.5300200998783111, "load_balance": 0.00047242407454177736, "g_entropy": 2.160051202774048, "tau_bias": 0.7436720132827759, "beta": 0.5570184166232744, "train_val_corr": 0.0}
{"t": 4, "ts": 1788190648.279455, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.924774169921875, "attn_entropy": 1.8307051181793212, "load_balance": 0.0004745736252516508, "g_entropy": 2.1600045204162597, "tau_bias": 0.7453014850616455, "beta": 0.5210045700271925, "train_val_corr": 0.0}
{"t": 5, "ts": 1788190657.27142, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.616886138916016, "attn_entropy": 2.0215752124786377, "load_balance": 0.00047595933428965507, "g_entropy": 2.1599592208862304, "tau_bias": 0.7561219930648804, "beta": 0.505678283671538, "train_val_corr": 0.0}
{"t": 6, "ts": 1788190679.634555, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 113.580078125, "attn_entropy": 2.06947340965271, "load_balance": 0.00047678978880867363, "g_entropy": 2.159578561782837, "tau_bias": 0.757969081401825, "beta": 0.5149516488114992, "train_val_corr": 0.0}
{"t": 7, "ts": 1788190688.8291373, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 140.23165893554688, "attn_entropy": 1.7929401636123656, "load_balance": 0.0004750640771817416, "g_entropy": 2.160167694091797, "tau_bias": 0.7674220204353333, "beta": 0.502609501282374, "train_val_corr": 0.0}
{"t": 8, "ts": 1788190697.6054235, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 31.773693084716797, "attn_entropy": 1.4675497740507126, "load_balance": 0.00047249665949493646, "g_entropy": 2.1600496768951416, "tau_bias": 0.7664327621459961, "beta": 0.517387256026268, "train_val_corr": 0.0}
{"t": 9, "ts": 1788190706.9005845, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 31.474205017089844, "attn_entropy": 1.4240631222724915, "load_balance": 0.0004740994598250836, "g_entropy": 2.1596930503845213, "tau_bias": 0.7708118557929993, "beta": 0.5002617364128431, "train_val_corr": 0.0}
{"t": 10, "ts": 1788190715.8857358, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 59.2969856262207, "attn_entropy": 1.333017173409462, "load_balance": 0.0004747857281472534, "g_entropy": 2.1599839687347413, "tau_bias": 0.790237307548523, "beta": 0.5042508294185003, "train_val_corr": 0.0}
{"t": 11, "ts": 1788190725.306695, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 55.188316345214844, "attn_entropy": 1.4621624320745468, "load_balance": 0.0004744421865325421, "g_entropy": 2.1599385261535646, "tau_bias": 0.789684534072876, "beta": 0.5178962176044782, "train_val_corr": 0.0}
{"t": 12, "ts": 1788190749.61256, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.386316299438477, "attn_entropy": 1.4891006857156754, "load_balance": 0.0004723157035186887, "g_entropy": 2.159918260574341, "tau_bias": 0.7969909906387329, "beta": 0.4978165018061797, "train_val_corr": 0.0}
{"t": 13, "ts": 1788190759.9364922, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 36.306175231933594, "attn_entropy": 1.4679257094860076, "load_balance": 0.0004739544878248125, "g_entropy": 2.1600989818573, "tau_bias": 0.8190460205078125, "beta": 0.5043416072924932, "train_val_corr": 0.0}
{"t": 14, "ts": 1788190769.8332746, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.962980270385742, "attn_entropy": 1.464475268125534, "load_balance": 0.0004734020330943167, "g_entropy": 2.1600109577178954, "tau_bias": 0.8211765289306641, "beta": 0.5220527177055677, "train_val_corr": 0.0}
{"t": 15, "ts": 1788190778.9084518, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 32.29684829711914, "attn_entropy": 1.4758265644311905, "load_balance": 0.0004735334136057645, "g_entropy": 2.1601442813873293, "tau_bias": 0.8215038776397705, "beta": 0.4950585129360358, "train_val_corr": 0.0}
{"t": 16, "ts": 1788190788.3735101, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.846327304840088, "attn_entropy": 1.4859458416700364, "load_balance": 0.00047305901534855366, "g_entropy": 2.159901762008667, "tau_bias": 0.832763135433197, "beta": 0.4910122901201248, "train_val_corr": 0.0}
{"t": 17, "ts": 1788190797.6827133, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.099615097045898, "attn_entropy": 1.5096246719360351, "load_balance": 0.0004733966721687466, "g_entropy": 2.159765291213989, "tau_bias": 0.8549951910972595, "beta": 0.5058556025226911, "train_val_corr": 0.0}
{"t": 0, "ts": 1788190961.7838838, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.67063953995239e-05, "advantage": 80.28714752197266, "attn_entropy": 1.3511474013328553, "load_balance": 0.0004755562578793615, "g_entropy": 2.1553136825561525, "tau_bias": 0.6509265899658203, "beta": 0.5608242775003115, "train_val_corr": 0.0}
{"t": 1, "ts": 1788190968.4101572, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725073568428343e-07, "advantage": 52.638301849365234, "attn_entropy": 1.3519277691841125, "load_balance": 0.0004741444252431393, "g_entropy": 2.1576223850250242, "tau_bias": 0.6517000794410706, "beta": 0.5496906166275343, "train_val_corr": 0.0}
{"t": 2, "ts": 1788190977.601202, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.99315643310547, "attn_entropy": 1.3466790974140168, "load_balance": 0.0004759559698868543, "g_entropy": 2.154058504104614, "tau_bias": 0.6490351557731628, "beta": 0.560172013938427, "train_val_corr": 0.0}
{"t": 3, "ts": 1788190986.8685625, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 54.2518424987793, "attn_entropy": 1.3406991958618164, "load_balance": 0.0004734523186925799, "g_entropy": 2.1580838680267336, "tau_bias": 0.654060959815979, "beta": 0.5575732638438543, "train_val_corr": 0.0}
{"t": 4, "ts": 1788190995.9921079, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 46.42334747314453, "attn_entropy": 1.3460229933261871, "load_balance": 0.0004738743242342025, "g_entropy": 2.158155822753906, "tau_bias": 0.665191113948822, "beta": 0.5497982054948807, "train_val_corr": 0.0}
{"t": 5, "ts": 1788191005.599134, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 76.65584564208984, "attn_entropy": 1.3487851321697235, "load_balance": 0.00047352503170259295, "g_entropy": 2.1590274810791015, "tau_bias": 0.6731583476066589, "beta": 0.5808877696593603, "train_val_corr": 0.0}
{"t": 6, "ts": 1788191014.1797101, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.75919723510742, "attn_entropy": 1.3521827518939973, "load_balance": 0.0004752520646434277, "g_entropy": 2.157622766494751, "tau_bias": 0.6818282604217529, "beta": 0.543062187731266, "train_val_corr": 0.0}
{"t": 7, "ts": 1788191035.7920673, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725684635181096e-07, "advantage": 265.5494689941406, "attn_entropy": 1.5031059473752975, "load_balance": 0.0004761442658491433, "g_entropy": 2.1543633937835693, "tau_bias": 0.6877602934837341, "beta": 0.5814952850341797, "train_val_corr": 0.0}
{"t": 8, "ts": 1788191045.212251, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 49.10212707519531, "attn_entropy": 1.3150612592697144, "load_balance": 0.0004749932268168777, "g_entropy": 2.153740739822388, "tau_bias": 0.6911648511886597, "beta": 0.53095892816782, "train_val_corr": 0.0}
{"t": 9, "ts": 1788191054.7059848, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.872671127319336, "attn_entropy": 1.35151726603508, "load_balance": 0.00047834738506935537, "g_entropy": 2.1521800994873046, "tau_bias": 0.7011364698410034, "beta": 0.5137708336114883, "train_val_corr": 0.0}
{"t": 10, "ts": 1788191064.171267, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 209.90025329589844, "attn_entropy": 1.3175441056489945, "load_balance": 0.00047165353898890314, "g_entropy": 2.1589323043823243, "tau_bias": 0.6979400515556335, "beta": 0.5322313755750656, "train_val_corr": 0.0}
{"t": 11, "ts": 1788191073.9091833, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 70.61478424072266, "attn_entropy": 1.3197030812501906, "load_balance": 0.00047560406965203584, "g_entropy": 2.156803274154663, "tau_bias": 0.7040600776672363, "beta": 0.4974934086203575, "train_val_corr": 0.0}
{"t": 12, "ts": 1788191082.6264367, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 53.34919357299805, "attn_entropy": 1.3100007086992265, "load_balance": 0.00047635596129111943, "g_entropy": 2.1545151710510253, "tau_bias": 0.706175684928894, "beta": 0.49785201251506805, "train_val_corr": 0.0}
{"t": 13, "ts": 1788191107.3178203, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 192.78526306152344, "attn_entropy": 1.3156905680894853, "load_balance": 0.0004736549512017518, "g_entropy": 2.156566619873047, "tau_bias": 0.7022916674613953, "beta": 0.5026751110951105, "train_val_corr": 0.0}
{"t": 14, "ts": 1788191116.531838, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 56.16712951660156, "attn_entropy": 1.324126985669136, "load_balance": 0.0004731621884275228, "g_entropy": 2.1578116416931152, "tau_bias": 0.7077902555465698, "beta": 0.48975656429926556, "train_val_corr": 0.0}
{"t": 15, "ts": 1788191125.893501, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -25.076446533203125, "attn_entropy": 1.3185264617204666, "load_balance": 0.00047525627305731175, "g_entropy": 2.155416250228882, "tau_bias": 0.722511351108551, "beta": 0.5013092656930288, "train_val_corr": 0.0}
{"t": 16, "ts": 1788191134.5824075, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -15.7894287109375, "attn_entropy": 1.313843509554863, "load_balance": 0.00047308505163528024, "g_entropy": 2.158518600463867, "tau_bias": 0.7304103374481201, "beta": 0.507084901134173, "train_val_corr": 0.0}
{"t": 17, "ts": 1788191143.376565, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 75.11143493652344, "attn_entropy": 1.3211652100086213, "load_balance": 0.00047718005371280017, "g_entropy": 2.1532440185546875, "tau_bias": 0.7295282483100891, "beta": 0.50057835628589, "train_val_corr": 0.0}
{"t": 18, "ts": 1788191152.0727184, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 15.446611404418945, "attn_entropy": 1.2985500395298004, "load_balance": 0.00047414493164978924, "g_entropy": 2.157845163345337, "tau_bias": 0.7392599582672119, "beta": 0.49165239681800205, "train_val_corr": 0.0}
{"t": 19, "ts": 1788191175.57281, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.306300163269043, "attn_entropy": 1.3256437465548516, "load_balance": 0.0004792233055923134, "g_entropy": 2.154364013671875, "tau_bias": 0.7546268105506897, "beta": 0.507347879310449, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788191183.9144428, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -25.426626205444336, "attn_entropy": 1.3254566878080367, "load_balance": 0.00047193767968565223, "g_entropy": 2.158480644226074, "tau_bias": 0.7604106664657593, "beta": 0.5106916179259618, "train_val_corr": 1.0}
{"t": 21, "ts": 1788191192.9831738, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.36211395263672, "attn_entropy": 1.3223753958940505, "load_balance": 0.0004755445464979857, "g_entropy": 2.156275510787964, "tau_bias": 0.7466249465942383, "beta": 0.5110351319114367, "train_val_corr": 1.0}
{"t": 22, "ts": 1788191201.782078, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.1834602355957, "attn_entropy": 1.312095221877098, "load_balance": 0.0004766503407154232, "g_entropy": 2.1555859088897704, "tau_bias": 0.7403541207313538, "beta": 0.5241915633281072, "train_val_corr": 1.0}
{"t": 23, "ts": 1788191210.5703144, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.48224639892578, "attn_entropy": 1.3205205529928208, "load_balance": 0.00047808611998334527, "g_entropy": 2.155749797821045, "tau_bias": 0.7370033860206604, "beta": 0.5162431870897611, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788191219.1056962, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.53986358642578, "attn_entropy": 1.3045182406902314, "load_balance": 0.00047706118202768267, "g_entropy": 2.1548808574676515, "tau_bias": 0.7335290908813477, "beta": 0.5288137421011925, "train_val_corr": 0.9999999999999998}
{"t": 25, "ts": 1788191241.778229, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 172.13209533691406, "attn_entropy": 1.4803500562906264, "load_balance": 0.0004807867982890457, "g_entropy": 2.1484087467193604, "tau_bias": 0.7350407242774963, "beta": 0.5437633395195007, "train_val_corr": 0.9999999999999998}
{"t": 26, "ts": 1788191250.1658897, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -41.74073028564453, "attn_entropy": 1.3157048523426056, "load_balance": 0.0004769911291077733, "g_entropy": 2.1575958251953127, "tau_bias": 0.7363970279693604, "beta": 0.5353707323471705, "train_val_corr": 1.0}
{"t": 27, "ts": 1788191259.090887, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.3516845703125, "attn_entropy": 1.3392249286174773, "load_balance": 0.0004763724689837545, "g_entropy": 2.1582211017608643, "tau_bias": 0.7372370362281799, "beta": 0.5268775274356207, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788191268.029332, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -41.9948616027832, "attn_entropy": 1.3333421885967254, "load_balance": 0.0004759911505971104, "g_entropy": 2.1586967945098876, "tau_bias": 0.7354714274406433, "beta": 0.5270391801993052, "train_val_corr": 1.0}
{"t": 29, "ts": 1788191277.0334318, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -41.53505325317383, "attn_entropy": 1.3271713078022003, "load_balance": 0.00047801074106246233, "g_entropy": 2.156088399887085, "tau_bias": 0.7358508110046387, "beta": 0.5240898504853249, "train_val_corr": 1.0}
{"t": 30, "ts": 1788191285.775975, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -40.496761322021484, "attn_entropy": 1.325609529018402, "load_balance": 0.00047804900677874687, "g_entropy": 2.1546566009521486, "tau_bias": 0.7325180172920227, "beta": 0.525239979227384, "train_val_corr": 0.9999999999999998}
{"t": 31, "ts": 1788191310.9350197, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.44685745239258, "attn_entropy": 1.3259476900100708, "load_balance": 0.00047618551761843266, "g_entropy": 2.1560480117797853, "tau_bias": 0.7307954430580139, "beta": 0.524636872112751, "train_val_corr": 1.0}
{"t": 32, "ts": 1788191318.2642064, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -34.12226867675781, "attn_entropy": 1.323846560716629, "load_balance": 0.0004757225571665913, "g_entropy": 2.1588810443878175, "tau_bias": 0.7298450469970703, "beta": 0.5418203150232633, "train_val_corr": 1.0}
{"t": 33, "ts": 1788191327.2768927, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.53652572631836, "attn_entropy": 1.3171298503875732, "load_balance": 0.0004734244663268328, "g_entropy": 2.159356451034546, "tau_bias": 0.7262675166130066, "beta": 0.5188914438088735, "train_val_corr": 1.0}
{"t": 34, "ts": 1788191336.273586, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.208309173583984, "attn_entropy": 1.3388242244720459, "load_balance": 0.00047569794696755707, "g_entropy": 2.1571943283081056, "tau_bias": 0.7222710251808167, "beta": 0.5355455527702967, "train_val_corr": 0.9999999999999999}
{"t": 35, "ts": 1788191345.3759384, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.145240783691406, "attn_entropy": 1.328531515598297, "load_balance": 0.0004758451192174107, "g_entropy": 2.1587342262268066, "tau_bias": 0.7211049795150757, "beta": 0.5253021344542503, "train_val_corr": 0.9999999999999999}
{"t": 36, "ts": 1788191354.2835824, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -31.928600311279297, "attn_entropy": 1.317449602484703, "load_balance": 0.0004748356528580189, "g_entropy": 2.1586320877075194, "tau_bias": 0.7210460305213928, "beta": 0.5207086081306139, "train_val_corr": 0.9999999999999999}
{"t": 0, "ts": 1788191432.58923, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.4390410569831147e-06, "advantage": 20.659597396850586, "attn_entropy": 1.3151332408189773, "load_balance": 0.00047510304138995705, "g_entropy": 2.157664966583252, "tau_bias": 0.7187976837158203, "beta": 0.5190132806698481, "train_val_corr": 0.0}
{"t": 1, "ts": 1788191439.9369361, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.081557823885305e-07, "advantage": 221.01998901367188, "attn_entropy": 1.3343413770198822, "load_balance": 0.0004727787512820214, "g_entropy": 2.158558750152588, "tau_bias": 0.7154619693756104, "beta": 0.528774860004584, "train_val_corr": 0.0}
{"t": 2, "ts": 1788191449.3273714, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326736879404e-07, "advantage": 95.33890533447266, "attn_entropy": 1.3093443155288695, "load_balance": 0.00047554735210724174, "g_entropy": 2.1591524124145507, "tau_bias": 0.7128916382789612, "beta": 0.567530964811643, "train_val_corr": 0.0}
{"t": 3, "ts": 1788191458.22349, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 22.832414627075195, "attn_entropy": 1.3307538270950316, "load_balance": 0.0004752476990688592, "g_entropy": 2.158563184738159, "tau_bias": 0.7202548384666443, "beta": 0.5277987023194631, "train_val_corr": 0.0}
{"t": 4, "ts": 1788191467.7383084, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.77531433105469, "attn_entropy": 1.321271723508835, "load_balance": 0.00047595511423423884, "g_entropy": 2.1575121879577637, "tau_bias": 0.7143424153327942, "beta": 0.6637379179398218, "train_val_corr": 0.0}
{"t": 5, "ts": 1788191476.4818401, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 41.54755783081055, "attn_entropy": 1.3196744561195373, "load_balance": 0.00047600630787201224, "g_entropy": 2.156028413772583, "tau_bias": 0.7158395051956177, "beta": 0.5782831832766533, "train_val_corr": 0.0}
{"t": 6, "ts": 1788191485.4748416, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 104.44120788574219, "attn_entropy": 1.3109733879566192, "load_balance": 0.000476967851864174, "g_entropy": 2.157205820083618, "tau_bias": 0.7304596900939941, "beta": 0.5212208280960718, "train_val_corr": 0.0}
{"t": 7, "ts": 1788191508.9776669, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1525.3353271484375, "attn_entropy": 1.6576481342315674, "load_balance": 0.00048230880638584496, "g_entropy": 2.149215650558472, "tau_bias": 0.7341372966766357, "beta": 0.6595984101295471, "train_val_corr": 0.0}
{"t": 8, "ts": 1788191517.9710364, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -57.8937873840332, "attn_entropy": 1.2896866023540496, "load_balance": 0.0004755542962811887, "g_entropy": 2.1590866088867187, "tau_bias": 0.7365947961807251, "beta": 0.5289656991759936, "train_val_corr": 0.0}
{"t": 9, "ts": 1788191527.2669868, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -67.55479431152344, "attn_entropy": 1.3023638427257538, "load_balance": 0.0004738365358207375, "g_entropy": 2.1592109203338623, "tau_bias": 0.7329822182655334, "beta": 0.5252527793248495, "train_val_corr": 0.0}
{"t": 10, "ts": 1788191536.62472, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -62.88569259643555, "attn_entropy": 1.2939707577228545, "load_balance": 0.0004753806279040873, "g_entropy": 2.1576395511627195, "tau_bias": 0.7274279594421387, "beta": 0.5329962794979414, "train_val_corr": 0.0}
{"t": 11, "ts": 1788191545.8984942, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -61.17970275878906, "attn_entropy": 1.292198547720909, "load_balance": 0.00047384879435412585, "g_entropy": 2.1586493015289308, "tau_bias": 0.7234066724777222, "beta": 0.5210781743129095, "train_val_corr": 0.0}
{"t": 12, "ts": 1788191555.2986815, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -63.22133255004883, "attn_entropy": 1.319042545557022, "load_balance": 0.0004723993712104857, "g_entropy": 2.1582553386688232, "tau_bias": 0.7205049395561218, "beta": 0.520583393673102, "train_val_corr": 0.0}
{"t": 13, "ts": 1788191580.5917354, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -63.11842727661133, "attn_entropy": 1.3033084750175477, "load_balance": 0.0004736030416097492, "g_entropy": 2.160091924667358, "tau_bias": 0.7176982760429382, "beta": 0.5238980626066526, "train_val_corr": 0.0}
{"t": 14, "ts": 1788191590.814017, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -60.38515090942383, "attn_entropy": 1.3018266558647156, "load_balance": 0.0004735823778901249, "g_entropy": 2.1595108032226564, "tau_bias": 0.7184072136878967, "beta": 0.5194669291377068, "train_val_corr": 0.0}
{"t": 15, "ts": 1788191600.7082539, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -57.07840347290039, "attn_entropy": 1.302873107790947, "load_balance": 0.00047726817429065703, "g_entropy": 2.1598755359649657, "tau_bias": 0.7182097434997559, "beta": 0.5217977489034334, "train_val_corr": 0.0}
{"t": 16, "ts": 1788191610.6917846, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -54.88044357299805, "attn_entropy": 1.3068834602832795, "load_balance": 0.00047296995180659, "g_entropy": 2.1601535797119142, "tau_bias": 0.7173627018928528, "beta": 0.5148900573452314, "train_val_corr": 0.0}
{"t": 17, "ts": 1788191620.7746716, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -53.796077728271484, "attn_entropy": 1.3030173122882842, "load_balance": 0.00047362374025397005, "g_entropy": 2.160050106048584, "tau_bias": 0.7134758234024048, "beta": 0.5105291903018951, "train_val_corr": 0.0}
{"t": 18, "ts": 1788191645.5111496, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -50.172245025634766, "attn_entropy": 1.314891916513443, "load_balance": 0.0004730180196929723, "g_entropy": 2.1600794315338137, "tau_bias": 0.710090696811676, "beta": 0.5082597558697065, "train_val_corr": 0.0}
{"t": 19, "ts": 1788191654.6671326, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -49.322654724121094, "attn_entropy": 1.308835792541504, "load_balance": 0.0004742457938846201, "g_entropy": 2.1601501941680907, "tau_bias": 0.70819491147995, "beta": 0.510490246117115, "train_val_corr": 1.0}
{"t": 20, "ts": 1788191663.8751645, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -43.495697021484375, "attn_entropy": 1.3095443069934845, "load_balance": 0.0004744844220113009, "g_entropy": 2.160141944885254, "tau_bias": 0.7102585434913635, "beta": 0.5411284938454628, "train_val_corr": 0.9999999999999999}
{"t": 21, "ts": 1788191674.1124794, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -44.85695266723633, "attn_entropy": 1.304306197166443, "load_balance": 0.00047396073932759464, "g_entropy": 2.1601202964782713, "tau_bias": 0.7153213620185852, "beta": 0.5522269085049629, "train_val_corr": 0.9999999999999999}
{"t": 22, "ts": 1788191684.1931152, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.480499267578125, "attn_entropy": 1.297819870710373, "load_balance": 0.00047240822459571064, "g_entropy": 2.1601430892944338, "tau_bias": 0.7220547795295715, "beta": 0.5509944061438242, "train_val_corr": 1.0}
{"t": 23, "ts": 1788191694.264909, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -43.73775863647461, "attn_entropy": 1.2928451359272004, "load_balance": 0.000473043997772038, "g_entropy": 2.1600293636322023, "tau_bias": 0.7235949635505676, "beta": 0.5383264347910881, "train_val_corr": 1.0}
{"t": 24, "ts": 1788191718.7064729, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.0726318359375, "attn_entropy": 1.2929251730442046, "load_balance": 0.00047188103781081736, "g_entropy": 2.1599307537078856, "tau_bias": 0.7194334268569946, "beta": 0.530705489218235, "train_val_corr": 0.9999999999999999}
{"t": 25, "ts": 1788191727.717488, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.2485466003418, "attn_entropy": 1.2976255238056182, "load_balance": 0.0004748501174617559, "g_entropy": 2.1600255966186523, "tau_bias": 0.7166507244110107, "beta": 0.5189858501156172, "train_val_corr": 1.0}
{"t": 26, "ts": 1788191736.905601, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.507999420166016, "attn_entropy": 1.3017673552036286, "load_balance": 0.0004736983042676002, "g_entropy": 2.1600165367126465, "tau_bias": 0.7154030203819275, "beta": 0.523025169968605, "train_val_corr": 1.0}
{"t": 27, "ts": 1788191746.6814904, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.9317512512207, "attn_entropy": 1.2934148788452149, "load_balance": 0.0004730229149572551, "g_entropy": 2.1594719886779785, "tau_bias": 0.713362455368042, "beta": 0.5116128996014595, "train_val_corr": 1.0}
{"t": 28, "ts": 1788191755.635073, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.144901275634766, "attn_entropy": 1.294150996208191, "load_balance": 0.00047449429985135794, "g_entropy": 2.1600807189941404, "tau_bias": 0.710212767124176, "beta": 0.5118365585803986, "train_val_corr": 1.0}
{"t": 29, "ts": 1788191765.9007328, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -34.132720947265625, "attn_entropy": 1.293212080001831, "load_balance": 0.0004748930805362761, "g_entropy": 2.1599632263183595, "tau_bias": 0.70858234167099, "beta": 0.5070852736632029, "train_val_corr": 1.0}
{"t": 30, "ts": 1788191791.6654484, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -33.256778717041016, "attn_entropy": 1.296950250864029, "load_balance": 0.00047338129370473326, "g_entropy": 2.1601417541503904, "tau_bias": 0.7069477438926697, "beta": 0.5070735737681389, "train_val_corr": 0.9999999999999998}
{"t": 31, "ts": 1788191802.267139, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -31.811094284057617, "attn_entropy": 1.2954754889011384, "load_balance": 0.00047605680301785467, "g_entropy": 2.1600422859191895, "tau_bias": 0.7060904502868652, "beta": 0.5062344099084536, "train_val_corr": 1.0}
{"t": 32, "ts": 1788191812.3805857, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.51394271850586, "attn_entropy": 1.301548171043396, "load_balance": 0.0004733407811727375, "g_entropy": 2.160083532333374, "tau_bias": 0.705788254737854, "beta": 0.5050482104221979, "train_val_corr": 1.0}
{"t": 33, "ts": 1788191821.293004, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.644315719604492, "attn_entropy": 1.297587513923645, "load_balance": 0.00047350588720291854, "g_entropy": 2.1599859237670898, "tau_bias": 0.7051270604133606, "beta": 0.5043245007594427, "train_val_corr": 1.0}
{"t": 34, "ts": 1788191830.738794, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.997678756713867, "attn_entropy": 1.302538949251175, "load_balance": 0.0004718289652373642, "g_entropy": 2.1599915981292725, "tau_bias": 0.7041037678718567, "beta": 0.5045509363214175, "train_val_corr": 1.0}
{"t": 35, "ts": 1788191840.0193443, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.798383712768555, "attn_entropy": 1.2987872004508971, "load_balance": 0.00047545555280521513, "g_entropy": 2.1598923206329346, "tau_bias": 0.7027774453163147, "beta": 0.5038483689228693, "train_val_corr": 0.9999999999999999}
{"t": 0, "ts": 1788191979.1815917, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 5.6213743846456055e-06, "advantage": 55.08726119995117, "attn_entropy": 1.5137965261936188, "load_balance": 0.0004746251390315592, "g_entropy": 2.158474636077881, "tau_bias": 0.6509265899658203, "beta": 0.5558276822169622, "train_val_corr": 0.0}
{"t": 1, "ts": 1788191986.1714358, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072615643555764e-07, "advantage": 60.32429885864258, "attn_entropy": 1.9365753650665283, "load_balance": 0.0004755402507726103, "g_entropy": 2.1594616889953615, "tau_bias": 0.6473045349121094, "beta": 0.578659770389398, "train_val_corr": 0.0}
{"t": 2, "ts": 1788191996.602321, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.49761390686035, "attn_entropy": 1.8379575252532958, "load_balance": 0.00047569862217642365, "g_entropy": 2.1578625679016112, "tau_bias": 0.6491557955741882, "beta": 0.5590421011050543, "train_val_corr": 0.0}
{"t": 3, "ts": 1788192006.7688735, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 52.321102142333984, "attn_entropy": 1.744695782661438, "load_balance": 0.00047409418039023876, "g_entropy": 2.159501838684082, "tau_bias": 0.6534193158149719, "beta": 0.5717381263772646, "train_val_corr": 0.0}
{"t": 4, "ts": 1788192016.0907025, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 35.466880798339844, "attn_entropy": 1.6033655405044556, "load_balance": 0.0004738083516713232, "g_entropy": 2.1592835426330566, "tau_bias": 0.6644824743270874, "beta": 0.5688903257250786, "train_val_corr": 0.0}
{"t": 5, "ts": 1788192025.8941448, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 51.81254196166992, "attn_entropy": 1.5739526629447937, "load_balance": 0.0004738339746836573, "g_entropy": 2.159857654571533, "tau_bias": 0.6657261252403259, "beta": 0.5648651818434397, "train_val_corr": 0.0}
{"t": 6, "ts": 1788192035.59394, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 99.6432876586914, "attn_entropy": 1.5635438442230225, "load_balance": 0.00047462457441724835, "g_entropy": 2.159257411956787, "tau_bias": 0.6777182221412659, "beta": 0.522066704928875, "train_val_corr": 0.0}
{"t": 7, "ts": 1788192059.1797032, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072344358339251e-07, "advantage": 215.45748901367188, "attn_entropy": 1.8503149509429933, "load_balance": 0.0004776777874212712, "g_entropy": 2.154471826553345, "tau_bias": 0.6768608093261719, "beta": 0.5874465703964233, "train_val_corr": 0.0}
{"t": 8, "ts": 1788192068.0264266, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.64748001098633, "attn_entropy": 1.8052116870880126, "load_balance": 0.0004751718137413263, "g_entropy": 2.1577607154846192, "tau_bias": 0.6773784756660461, "beta": 0.5229709967970848, "train_val_corr": 0.0}
{"t": 9, "ts": 1788192077.773991, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.967201232910156, "attn_entropy": 1.8922146320343018, "load_balance": 0.0004764471377711743, "g_entropy": 2.1576406002044677, "tau_bias": 0.6874027252197266, "beta": 0.5096148103475571, "train_val_corr": 0.0}
{"t": 10, "ts": 1788192087.5369725, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 95.55701446533203, "attn_entropy": 1.926872730255127, "load_balance": 0.00047577705699950454, "g_entropy": 2.1599441051483153, "tau_bias": 0.6859321594238281, "beta": 0.5252280384302139, "train_val_corr": 0.0}
{"t": 11, "ts": 1788192096.9367661, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 25.0969181060791, "attn_entropy": 1.8286163806915283, "load_balance": 0.0004759201721753925, "g_entropy": 2.159056520462036, "tau_bias": 0.6959159970283508, "beta": 0.4991959258913994, "train_val_corr": 0.0}
{"t": 12, "ts": 1788192106.2042556, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 61.1693000793457, "attn_entropy": 1.850580358505249, "load_balance": 0.00047742120223119854, "g_entropy": 2.1580200672149656, "tau_bias": 0.6994970440864563, "beta": 0.5062166377902031, "train_val_corr": 0.0}
{"t": 13, "ts": 1788192128.6767542, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 47.5068473815918, "attn_entropy": 1.7823785305023194, "load_balance": 0.0004760049108881503, "g_entropy": 2.1587708950042725, "tau_bias": 0.6937175989151001, "beta": 0.5024443839987119, "train_val_corr": 0.0}
{"t": 14, "ts": 1788192136.8095524, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.4298019409179688, "attn_entropy": 1.6963770627975463, "load_balance": 0.0004749999672640115, "g_entropy": 2.1595383167266844, "tau_bias": 0.702537477016449, "beta": 0.49785029515624046, "train_val_corr": 0.0}
{"t": 15, "ts": 1788192145.7117887, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1.7351994514465332, "attn_entropy": 1.771622896194458, "load_balance": 0.0004765777965076268, "g_entropy": 2.1586405754089357, "tau_bias": 0.7136756181716919, "beta": 0.505612793068091, "train_val_corr": 0.0}
{"t": 16, "ts": 1788192154.8775718, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.81105613708496, "attn_entropy": 1.8528468132019043, "load_balance": 0.00047580833779647946, "g_entropy": 2.159486246109009, "tau_bias": 0.7264137268066406, "beta": 0.4954439004262288, "train_val_corr": 0.0}
{"t": 17, "ts": 1788192163.99174, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.290828704833984, "attn_entropy": 1.7314577579498291, "load_balance": 0.0004768848477397114, "g_entropy": 2.1580699920654296, "tau_bias": 0.7411102652549744, "beta": 0.5084469442566236, "train_val_corr": 0.0}
{"t": 18, "ts": 1788192174.2701037, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.95807409286499, "attn_entropy": 1.7724168062210084, "load_balance": 0.0004760462907142937, "g_entropy": 2.1592933177947997, "tau_bias": 0.7458622455596924, "beta": 0.5355451380213102, "train_val_corr": 0.0}
{"t": 19, "ts": 1788192199.3883557, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.441420555114746, "attn_entropy": 1.769619083404541, "load_balance": 0.00047616137890145184, "g_entropy": 2.159033107757568, "tau_bias": 0.7538049817085266, "beta": 0.5119724000493685, "train_val_corr": 1.0}
{"t": 20, "ts": 1788192208.4908342, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.8599460124969482, "attn_entropy": 1.8861191272735596, "load_balance": 0.00047568291192874313, "g_entropy": 2.1595505237579347, "tau_bias": 0.7674970626831055, "beta": 0.518904022872448, "train_val_corr": 1.0}
{"t": 21, "ts": 1788192219.0851486, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.828131198883057, "attn_entropy": 1.7118409633636475, "load_balance": 0.0004756870912387967, "g_entropy": 2.158657360076904, "tau_bias": 0.7649639248847961, "beta": 0.5522857556740443, "train_val_corr": 1.0}
{"t": 22, "ts": 1788192228.2859554, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -14.984050750732422, "attn_entropy": 1.7168451786041259, "load_balance": 0.0004763350996654481, "g_entropy": 2.1582587718963624, "tau_bias": 0.7695327401161194, "beta": 0.5512345631917318, "train_val_corr": 1.0}
{"t": 23, "ts": 1788192238.12325, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -13.245429039001465, "attn_entropy": 1.7175447463989257, "load_balance": 0.00047580656828358767, "g_entropy": 2.158701181411743, "tau_bias": 0.7715408802032471, "beta": 0.5462866698702177, "train_val_corr": 0.9999999999999996}
{"t": 24, "ts": 1788192248.0793173, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.31771183013916, "attn_entropy": 1.7084170818328857, "load_balance": 0.00047569951857440174, "g_entropy": 2.1580219745635985, "tau_bias": 0.7699329257011414, "beta": 0.5601133654514948, "train_val_corr": 1.0}
{"t": 25, "ts": 1788192273.7868073, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 167.77391052246094, "attn_entropy": 1.6141306400299071, "load_balance": 0.0004823929339181632, "g_entropy": 2.147700881958008, "tau_bias": 0.767086386680603, "beta": 0.5342474579811096, "train_val_corr": 1.0}
{"t": 26, "ts": 1788192282.1264145, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.316373825073242, "attn_entropy": 1.7521604299545288, "load_balance": 0.00047494779573753476, "g_entropy": 2.1594425678253173, "tau_bias": 0.7688838243484497, "beta": 0.5448803951342901, "train_val_corr": 1.0}
{"t": 27, "ts": 1788192291.497399, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.793296813964844, "attn_entropy": 1.7389118194580078, "load_balance": 0.00047550839954055844, "g_entropy": 2.1595309734344483, "tau_bias": 0.7678933143615723, "beta": 0.5421381890773773, "train_val_corr": 1.0}
{"t": 28, "ts": 1788192301.1012273, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.80657958984375, "attn_entropy": 1.7249673366546632, "load_balance": 0.00047534558107145133, "g_entropy": 2.159844923019409, "tau_bias": 0.7642956972122192, "beta": 0.5426434775193533, "train_val_corr": 1.0}
{"t": 29, "ts": 1788192310.8156567, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.01604461669922, "attn_entropy": 1.7488488674163818, "load_balance": 0.00047568446607328954, "g_entropy": 2.1585683822631836, "tau_bias": 0.7642895579338074, "beta": 0.5419188166658083, "train_val_corr": 1.0}
{"t": 30, "ts": 1788192320.6264195, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.44978904724121, "attn_entropy": 1.7143171548843383, "load_balance": 0.00047644295846112074, "g_entropy": 2.158155155181885, "tau_bias": 0.7655741572380066, "beta": 0.5371959681312243, "train_val_corr": 1.0}
{"t": 31, "ts": 1788192346.5853405, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.443519592285156, "attn_entropy": 1.7449467658996582, "load_balance": 0.00047588649904355406, "g_entropy": 2.159019422531128, "tau_bias": 0.7636325359344482, "beta": 0.5354726240038872, "train_val_corr": 0.9999999999999999}
{"t": 32, "ts": 1788192355.1819384, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.601503372192383, "attn_entropy": 1.6929487943649293, "load_balance": 0.0004745948361232877, "g_entropy": 2.159850597381592, "tau_bias": 0.754813015460968, "beta": 0.5381007765730222, "train_val_corr": 1.0}
{"t": 33, "ts": 1788192365.0903711, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -16.333127975463867, "attn_entropy": 1.6871184349060058, "load_balance": 0.00047517875791527333, "g_entropy": 2.1599050521850587, "tau_bias": 0.7517573237419128, "beta": 0.5341063837210337, "train_val_corr": 1.0}
{"t": 0, "ts": 1788192408.3711584, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.8621806248120265e-06, "advantage": 27.189247131347656, "attn_entropy": 1.694290542602539, "load_balance": 0.00047504259855486454, "g_entropy": 2.1594614028930663, "tau_bias": 0.7456531524658203, "beta": 0.5307086432973543, "train_val_corr": 0.0}
{"t": 1, "ts": 1788192414.9954746, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0724803562188754e-07, "advantage": 46.16981887817383, "attn_entropy": 1.805910015106201, "load_balance": 0.00047388443490490315, "g_entropy": 2.1594580173492433, "tau_bias": 0.7509832382202148, "beta": 0.5898249248663584, "train_val_corr": 0.0}
{"t": 2, "ts": 1788192424.1046736, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.46323013305664, "attn_entropy": 1.630884861946106, "load_balance": 0.0004742169694509357, "g_entropy": 2.1599101543426515, "tau_bias": 0.7551271915435791, "beta": 0.5591262578964233, "train_val_corr": 0.0}
{"t": 3, "ts": 1788192432.9735086, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 23.073408126831055, "attn_entropy": 1.6215458631515502, "load_balance": 0.00047415627050213517, "g_entropy": 2.159603404998779, "tau_bias": 0.7548097372055054, "beta": 0.5603676239649454, "train_val_corr": 0.0}
{"t": 4, "ts": 1788192441.990043, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 21.40665054321289, "attn_entropy": 1.6985309839248657, "load_balance": 0.00047496271436102687, "g_entropy": 2.1596370697021485, "tau_bias": 0.7528775334358215, "beta": 0.547227015097936, "train_val_corr": 0.0}
{"t": 5, "ts": 1788192451.1947627, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 21.25420570373535, "attn_entropy": 1.6793956518173219, "load_balance": 0.00047494813916273417, "g_entropy": 2.1590322494506835, "tau_bias": 0.7501197457313538, "beta": 0.5493424932161967, "train_val_corr": 0.0}
{"t": 6, "ts": 1788192460.2323103, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 17.561708450317383, "attn_entropy": 1.6400205612182617, "load_balance": 0.00047454110463149843, "g_entropy": 2.159516382217407, "tau_bias": 0.7467533946037292, "beta": 0.5410781453053156, "train_val_corr": 0.0}
{"t": 7, "ts": 1788192483.7819784, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 95.16484832763672, "attn_entropy": 1.462658452987671, "load_balance": 0.000481400202261284, "g_entropy": 2.1489866733551026, "tau_bias": 0.7445595264434814, "beta": 0.5618874430656433, "train_val_corr": 0.0}
{"t": 8, "ts": 1788192492.784217, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 14.667891502380371, "attn_entropy": 1.647813868522644, "load_balance": 0.0004741999611724168, "g_entropy": 2.1598832607269287, "tau_bias": 0.7457114458084106, "beta": 0.5333571607867876, "train_val_corr": 0.0}
{"t": 9, "ts": 1788192501.8944123, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.493561744689941, "attn_entropy": 1.602922224998474, "load_balance": 0.00047395392903126776, "g_entropy": 2.1600021362304687, "tau_bias": 0.7461062669754028, "beta": 0.5328561762968699, "train_val_corr": 0.0}
{"t": 10, "ts": 1788192511.181209, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 11.21000862121582, "attn_entropy": 1.5807446002960206, "load_balance": 0.00047367794904857875, "g_entropy": 2.159491443634033, "tau_bias": 0.7471016049385071, "beta": 0.5267106369137764, "train_val_corr": 0.0}
{"t": 11, "ts": 1788192521.0967114, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.70667552947998, "attn_entropy": 1.6344741582870483, "load_balance": 0.0004741443262901157, "g_entropy": 2.1597596645355224, "tau_bias": 0.7484041452407837, "beta": 0.5258284732699394, "train_val_corr": 0.0}
{"t": 12, "ts": 1788192530.197073, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.727736473083496, "attn_entropy": 1.6375893115997315, "load_balance": 0.00047413842403329907, "g_entropy": 2.1597618579864504, "tau_bias": 0.7466316223144531, "beta": 0.5213510617613792, "train_val_corr": 0.0}
{"t": 13, "ts": 1788192555.0761096, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.818571090698242, "attn_entropy": 1.6181591510772706, "load_balance": 0.00047516177874058486, "g_entropy": 2.1586764812469483, "tau_bias": 0.7438696622848511, "beta": 0.5198462828993797, "train_val_corr": 0.0}
{"t": 14, "ts": 1788192563.898615, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.730005264282227, "attn_entropy": 1.6642889499664306, "load_balance": 0.00047498280182480814, "g_entropy": 2.1595191955566406, "tau_bias": 0.7413464188575745, "beta": 0.5217205882072449, "train_val_corr": 0.0}
{"t": 15, "ts": 1788192573.6754243, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.1940436363220215, "attn_entropy": 1.5670099973678588, "load_balance": 0.0004745416925288737, "g_entropy": 2.159426736831665, "tau_bias": 0.7401598691940308, "beta": 0.5244063039620718, "train_val_corr": 0.0}
{"t": 16, "ts": 1788192583.3383527, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.279629230499268, "attn_entropy": 1.5938910007476808, "load_balance": 0.0004739443771541119, "g_entropy": 2.159975004196167, "tau_bias": 0.7397511005401611, "beta": 0.5143947377800941, "train_val_corr": 0.0}
{"t": 17, "ts": 1788192592.8180447, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.1674580574035645, "attn_entropy": 1.5686415433883667, "load_balance": 0.0004731321765575558, "g_entropy": 2.159443664550781, "tau_bias": 0.7387571930885315, "beta": 0.5162697633107504, "train_val_corr": 0.0}
{"t": 18, "ts": 1788192601.7924483, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.6799635887146, "attn_entropy": 1.5380237579345704, "load_balance": 0.00047305257758125665, "g_entropy": 2.160084915161133, "tau_bias": 0.7367956638336182, "beta": 0.5131167297561964, "train_val_corr": 0.0}
{"t": 19, "ts": 1788192627.5678828, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.842518329620361, "attn_entropy": 1.5662579536437988, "load_balance": 0.00047389077371917665, "g_entropy": 2.1595288276672364, "tau_bias": 0.735284686088562, "beta": 0.511366272966067, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788192636.176953, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.990557670593262, "attn_entropy": 1.553237295150757, "load_balance": 0.0004730782995466143, "g_entropy": 2.159277153015137, "tau_bias": 0.7362818121910095, "beta": 0.5209724555412928, "train_val_corr": 1.0}
{"t": 21, "ts": 1788192645.7842734, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.458471298217773, "attn_entropy": 1.5484247207641602, "load_balance": 0.00047353354748338463, "g_entropy": 2.159897804260254, "tau_bias": 0.7294452786445618, "beta": 0.5558306450645129, "train_val_corr": 1.0}
{"t": 22, "ts": 1788192655.1026437, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.049870491027832, "attn_entropy": 1.581041395664215, "load_balance": 0.00047526489943265914, "g_entropy": 2.1594770431518553, "tau_bias": 0.7307599186897278, "beta": 0.5491552253564199, "train_val_corr": 0.9999999999999996}
{"t": 23, "ts": 1788192664.4019747, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.43901252746582, "attn_entropy": 1.578139340877533, "load_balance": 0.00047362307086586954, "g_entropy": 2.15981183052063, "tau_bias": 0.7322118282318115, "beta": 0.5399295091629028, "train_val_corr": 0.9999999999999998}
{"t": 24, "ts": 1788192673.4967325, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.2530510425567627, "attn_entropy": 1.637556266784668, "load_balance": 0.0004737500392366201, "g_entropy": 2.159638833999634, "tau_bias": 0.7302911281585693, "beta": 0.5389902393023173, "train_val_corr": 0.9999999999999996}
{"t": 25, "ts": 1788192696.697372, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 105.5903091430664, "attn_entropy": 1.4965020298957825, "load_balance": 0.00048788918647915127, "g_entropy": 2.1140849113464357, "tau_bias": 0.7297754287719727, "beta": 0.5701241493225098, "train_val_corr": 0.9999999999999998}
{"t": 26, "ts": 1788192704.9223986, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.451451778411865, "attn_entropy": 1.5151653170585633, "load_balance": 0.00047266976325772704, "g_entropy": 2.159657192230225, "tau_bias": 0.7342284321784973, "beta": 0.5296869551142057, "train_val_corr": 1.0}
{"t": 27, "ts": 1788192714.0685246, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.96895694732666, "attn_entropy": 1.5070201396942138, "load_balance": 0.0004733986803330481, "g_entropy": 2.16015887260437, "tau_bias": 0.7383973598480225, "beta": 0.5252759754657745, "train_val_corr": 0.9999999999999997}
{"t": 28, "ts": 1788192723.2303493, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.252096652984619, "attn_entropy": 1.4987539649009705, "load_balance": 0.0004745232814457268, "g_entropy": 2.1588086128234862, "tau_bias": 0.7386090755462646, "beta": 0.5275181159377098, "train_val_corr": 1.0}
{"t": 29, "ts": 1788192732.2724383, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.903544902801514, "attn_entropy": 1.491326665878296, "load_balance": 0.00047294656978920103, "g_entropy": 2.1600019931793213, "tau_bias": 0.7397486567497253, "beta": 0.5207933460672697, "train_val_corr": 1.0}
{"t": 30, "ts": 1788192741.4292517, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.251676559448242, "attn_entropy": 1.517814862728119, "load_balance": 0.00047390510444529356, "g_entropy": 2.159540319442749, "tau_bias": 0.7379783391952515, "beta": 0.5206408053636551, "train_val_corr": 1.0}
{"t": 0, "ts": 1788192849.2183897, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00016496657917741686, "advantage": 233.86448669433594, "attn_entropy": 1.489292985200882, "load_balance": 0.00047873075236566366, "g_entropy": 2.1514075756073, "tau_bias": 0.6509265899658203, "beta": 0.5577196727196375, "train_val_corr": 0.0}
{"t": 1, "ts": 1788192856.3146982, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.073209230957218e-07, "advantage": 135.4520721435547, "attn_entropy": 1.504406327009201, "load_balance": 0.0005040016258135438, "g_entropy": 2.115595817565918, "tau_bias": 0.6485937237739563, "beta": 0.534373015165329, "train_val_corr": 0.0}
{"t": 2, "ts": 1788192865.8234673, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 106.90690612792969, "attn_entropy": 1.4948223650455474, "load_balance": 0.0004752575827296823, "g_entropy": 2.1568811893463136, "tau_bias": 0.6451319456100464, "beta": 0.5370425631602606, "train_val_corr": 0.0}
{"t": 3, "ts": 1788192875.284654, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 90.00999450683594, "attn_entropy": 1.4930770754814149, "load_balance": 0.00048624624032527207, "g_entropy": 2.138133096694946, "tau_bias": 0.6502702236175537, "beta": 0.5410975863536199, "train_val_corr": 0.0}
{"t": 4, "ts": 1788192884.4897997, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 79.22921752929688, "attn_entropy": 1.4895644664764405, "load_balance": 0.00047609547036699953, "g_entropy": 2.1553314685821534, "tau_bias": 0.6526888012886047, "beta": 0.5445701579252878, "train_val_corr": 0.0}
{"t": 5, "ts": 1788192893.6265886, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.597869873046875, "attn_entropy": 1.476958093047142, "load_balance": 0.0004801563394721597, "g_entropy": 2.147358751296997, "tau_bias": 0.6595838069915771, "beta": 0.5430855378508568, "train_val_corr": 0.0}
{"t": 6, "ts": 1788192903.3684685, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.57062911987305, "attn_entropy": 1.4702332824468614, "load_balance": 0.000479798304149881, "g_entropy": 2.148047399520874, "tau_bias": 0.6685022115707397, "beta": 0.543850320080916, "train_val_corr": 0.0}
{"t": 7, "ts": 1788192928.3907442, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 83.86402893066406, "attn_entropy": 1.4705988615751266, "load_balance": 0.0004767290607560426, "g_entropy": 2.151924896240234, "tau_bias": 0.6778479814529419, "beta": 0.5449230074882507, "train_val_corr": 0.0}
{"t": 8, "ts": 1788192937.5102031, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 58.05849838256836, "attn_entropy": 1.487132951617241, "load_balance": 0.00047664226149208845, "g_entropy": 2.1549264907836916, "tau_bias": 0.6788264513015747, "beta": 0.5277107457319895, "train_val_corr": 0.0}
{"t": 9, "ts": 1788192946.7342837, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.9108772277832, "attn_entropy": 1.4734902918338775, "load_balance": 0.0004830178862903267, "g_entropy": 2.1445977687835693, "tau_bias": 0.6880024671554565, "beta": 0.5379047989845276, "train_val_corr": 0.0}
{"t": 10, "ts": 1788192956.221277, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.37331008911133, "attn_entropy": 1.485756081342697, "load_balance": 0.00047561593237333, "g_entropy": 2.156116485595703, "tau_bias": 0.6917641162872314, "beta": 0.5583699494600296, "train_val_corr": 0.0}
{"t": 11, "ts": 1788192965.6041865, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 19.8219051361084, "attn_entropy": 1.488300397992134, "load_balance": 0.00047600738471373914, "g_entropy": 2.155643653869629, "tau_bias": 0.704373300075531, "beta": 0.5196714376409849, "train_val_corr": 0.0}
{"t": 12, "ts": 1788192974.806564, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.291961669921875, "attn_entropy": 1.429060271382332, "load_balance": 0.0004909828654490411, "g_entropy": 2.131763744354248, "tau_bias": 0.7111386060714722, "beta": 0.5366366157929102, "train_val_corr": 0.0}
{"t": 13, "ts": 1788192999.2138379, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 118.08148193359375, "attn_entropy": 1.531869900226593, "load_balance": 0.0004752974899020046, "g_entropy": 2.1547550678253176, "tau_bias": 0.7126057147979736, "beta": 0.529900756975015, "train_val_corr": 0.0}
{"t": 14, "ts": 1788193007.3962996, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.223304748535156, "attn_entropy": 1.4580867111682891, "load_balance": 0.00048351999721489846, "g_entropy": 2.1429885387420655, "tau_bias": 0.7204989790916443, "beta": 0.5127113337318102, "train_val_corr": 0.0}
{"t": 15, "ts": 1788193016.3020005, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.29686564207077026, "attn_entropy": 1.433584713935852, "load_balance": 0.0004787827667314559, "g_entropy": 2.150816774368286, "tau_bias": 0.7216870784759521, "beta": 0.526393656929334, "train_val_corr": 0.0}
{"t": 16, "ts": 1788193025.7831097, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.063535690307617, "attn_entropy": 1.4875813454389573, "load_balance": 0.0004803567077033222, "g_entropy": 2.150985670089722, "tau_bias": 0.7251416444778442, "beta": 0.5204427341620127, "train_val_corr": 0.0}
{"t": 17, "ts": 1788193034.8781154, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.3759169578552246, "attn_entropy": 1.4775080963969232, "load_balance": 0.0004759552713949233, "g_entropy": 2.156014251708984, "tau_bias": 0.7343813180923462, "beta": 0.5155022591352463, "train_val_corr": 0.0}
{"t": 18, "ts": 1788193044.165177, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.894024848937988, "attn_entropy": 1.5333922386169434, "load_balance": 0.0004759374482091516, "g_entropy": 2.157166051864624, "tau_bias": 0.7370041608810425, "beta": 0.5161528637011846, "train_val_corr": 0.0}
{"t": 19, "ts": 1788193067.3968382, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.295143127441406, "attn_entropy": 1.4600316867232324, "load_balance": 0.0004753885034006089, "g_entropy": 2.1552161693573, "tau_bias": 0.7339385747909546, "beta": 0.5131564537684122, "train_val_corr": 1.0}
{"t": 20, "ts": 1788193075.7669337, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.981983184814453, "attn_entropy": 1.4705814749002457, "load_balance": 0.0004749368934426457, "g_entropy": 2.153610849380493, "tau_bias": 0.7318506836891174, "beta": 0.5130152006944021, "train_val_corr": 1.0}
{"t": 21, "ts": 1788193084.9899497, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.952742099761963, "attn_entropy": 1.3807418283075095, "load_balance": 0.0004722240788396448, "g_entropy": 2.154816246032715, "tau_bias": 0.7306451201438904, "beta": 0.541745645304521, "train_val_corr": 0.9999999999999999}
{"t": 22, "ts": 1788193094.1122017, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.607195854187012, "attn_entropy": 1.4029106739908457, "load_balance": 0.00047770116361789405, "g_entropy": 2.1523433208465574, "tau_bias": 0.7285284996032715, "beta": 0.5770759036143621, "train_val_corr": 1.0}
{"t": 23, "ts": 1788193103.7088933, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -14.113017082214355, "attn_entropy": 1.4256901666522026, "load_balance": 0.0004754197492729872, "g_entropy": 2.155956983566284, "tau_bias": 0.7257688045501709, "beta": 0.5752143984039625, "train_val_corr": 1.0}
{"t": 24, "ts": 1788193112.714643, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.482805252075195, "attn_entropy": 1.4184508372098208, "load_balance": 0.000478697958169505, "g_entropy": 2.1504141807556154, "tau_bias": 0.7229959964752197, "beta": 0.5598646725217501, "train_val_corr": 1.0}
{"t": 25, "ts": 1788193137.0860288, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.657544136047363, "attn_entropy": 1.4230731040239335, "load_balance": 0.0004833551414776593, "g_entropy": 2.1419427394866943, "tau_bias": 0.7179423570632935, "beta": 0.5668672919273376, "train_val_corr": 0.9999999999999997}
{"t": 26, "ts": 1788193145.5229895, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -19.55494499206543, "attn_entropy": 1.469193208217621, "load_balance": 0.0004743132391013205, "g_entropy": 2.1570353507995605, "tau_bias": 0.7089336514472961, "beta": 0.5725661093990008, "train_val_corr": 1.0}
{"t": 27, "ts": 1788193154.2155094, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.149471282958984, "attn_entropy": 1.4534897059202194, "load_balance": 0.00047573307529091835, "g_entropy": 2.155047559738159, "tau_bias": 0.7066587209701538, "beta": 0.5640347053607305, "train_val_corr": 1.0}
{"t": 28, "ts": 1788193163.2916203, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.9643497467041, "attn_entropy": 1.4248472712934017, "load_balance": 0.00047483514645136894, "g_entropy": 2.1564094543457033, "tau_bias": 0.7064183950424194, "beta": 0.5528718282779058, "train_val_corr": 1.0}
{"t": 29, "ts": 1788193172.2807221, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.11756134033203, "attn_entropy": 1.4409356124699115, "load_balance": 0.00047185689327307047, "g_entropy": 2.1588532447814943, "tau_bias": 0.7041711807250977, "beta": 0.5626444096366564, "train_val_corr": 0.9999999999999998}
{"t": 30, "ts": 1788193181.9738252, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.89786720275879, "attn_entropy": 1.4820176094770432, "load_balance": 0.0004749827028717846, "g_entropy": 2.1587522506713865, "tau_bias": 0.7013100385665894, "beta": 0.5596639811992645, "train_val_corr": 1.0}
{"t": 31, "ts": 1788193207.2341597, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.56830596923828, "attn_entropy": 1.4493385583162308, "load_balance": 0.0004735155962407589, "g_entropy": 2.1587384223937987, "tau_bias": 0.6972109079360962, "beta": 0.5492019852002462, "train_val_corr": 1.0}
{"t": 32, "ts": 1788193214.9851756, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.716249465942383, "attn_entropy": 1.4522236712276935, "load_balance": 0.0004746820719446987, "g_entropy": 2.156639909744263, "tau_bias": 0.6839556694030762, "beta": 0.5404068231582642, "train_val_corr": 0.9999999999999998}
{"t": 33, "ts": 1788193224.0100036, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.15065574645996, "attn_entropy": 1.4611121721565723, "load_balance": 0.00047611432382836937, "g_entropy": 2.1564822673797606, "tau_bias": 0.6755799651145935, "beta": 0.5450588539242744, "train_val_corr": 0.9999999999999996}
{"t": 34, "ts": 1788193233.0810826, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.584049224853516, "attn_entropy": 1.4608528941869736, "load_balance": 0.00047489738208241763, "g_entropy": 2.1585693836212156, "tau_bias": 0.6719614267349243, "beta": 0.5323830420772234, "train_val_corr": 1.0}
{"t": 35, "ts": 1788193242.3187249, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.090675354003906, "attn_entropy": 1.434397928789258, "load_balance": 0.00047547174617648127, "g_entropy": 2.1556155681610107, "tau_bias": 0.6697440147399902, "beta": 0.5448020994663239, "train_val_corr": 1.0}
{"t": 0, "ts": 1788193275.4737854, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.656435870245332e-06, "advantage": 23.976224899291992, "attn_entropy": 1.4601380176842214, "load_balance": 0.00047438025940209626, "g_entropy": 2.1567759037017824, "tau_bias": 0.6718270778656006, "beta": 0.5399128620823225, "train_val_corr": 0.0}
{"t": 1, "ts": 1788193282.7651165, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0726247385027818e-07, "advantage": 64.60084533691406, "attn_entropy": 1.453570469468832, "load_balance": 0.000482430326519534, "g_entropy": 2.146426773071289, "tau_bias": 0.6609247326850891, "beta": 0.5779191429416338, "train_val_corr": 0.0}
{"t": 2, "ts": 1788193291.7348666, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": 85.27609252929688, "attn_entropy": 1.4418977223336698, "load_balance": 0.00047300157020799813, "g_entropy": 2.159270095825195, "tau_bias": 0.6698293089866638, "beta": 0.5338052585721016, "train_val_corr": 0.0}
{"t": 3, "ts": 1788193301.0258784, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 115.866455078125, "attn_entropy": 1.4258217349648477, "load_balance": 0.00047828759416006504, "g_entropy": 2.152922248840332, "tau_bias": 0.6627852320671082, "beta": 0.5818233961860338, "train_val_corr": 0.0}
{"t": 4, "ts": 1788193310.283464, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 55.56143569946289, "attn_entropy": 1.412215431034565, "load_balance": 0.00047401554766111074, "g_entropy": 2.158366632461548, "tau_bias": 0.6684228777885437, "beta": 0.5259338915348053, "train_val_corr": 0.0}
{"t": 5, "ts": 1788193319.002101, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 40.148380279541016, "attn_entropy": 1.4524424433708192, "load_balance": 0.00047383964410983024, "g_entropy": 2.1554493427276613, "tau_bias": 0.6766847968101501, "beta": 0.5350185011823972, "train_val_corr": 0.0}
{"t": 6, "ts": 1788193327.9308734, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.36295318603516, "attn_entropy": 1.4102187030017377, "load_balance": 0.0004736931237857789, "g_entropy": 2.156524991989136, "tau_bias": 0.6782227158546448, "beta": 0.5461725989977518, "train_val_corr": 0.0}
{"t": 7, "ts": 1788193352.005582, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 68.86883544921875, "attn_entropy": 1.4363857001066207, "load_balance": 0.000475581584032625, "g_entropy": 2.153770685195923, "tau_bias": 0.6887440085411072, "beta": 0.5054866671562195, "train_val_corr": 0.0}
{"t": 8, "ts": 1788193361.1007667, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 15.833781242370605, "attn_entropy": 1.4610347658395768, "load_balance": 0.0004763280507177114, "g_entropy": 2.15785665512085, "tau_bias": 0.6971071362495422, "beta": 0.5254971633354822, "train_val_corr": 0.0}
{"t": 9, "ts": 1788193370.125781, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 105.81881713867188, "attn_entropy": 1.8303028106689454, "load_balance": 0.0004772897344082594, "g_entropy": 2.1565960884094237, "tau_bias": 0.6974976062774658, "beta": 0.5386488934357961, "train_val_corr": 0.0}
{"t": 10, "ts": 1788193379.4884799, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 53.32022476196289, "attn_entropy": 1.6283811450004577, "load_balance": 0.00047375758294947444, "g_entropy": 2.15940899848938, "tau_bias": 0.7158032655715942, "beta": 0.5044478327035904, "train_val_corr": 0.0}
{"t": 11, "ts": 1788193389.491977, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.979759216308594, "attn_entropy": 1.389773005247116, "load_balance": 0.0004733308160211891, "g_entropy": 2.158906126022339, "tau_bias": 0.7244977355003357, "beta": 0.5165184934933981, "train_val_corr": 0.0}
{"t": 12, "ts": 1788193399.5013804, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 146.02171325683594, "attn_entropy": 1.4061006426811218, "load_balance": 0.00048166408087126913, "g_entropy": 2.151128578186035, "tau_bias": 0.7229711413383484, "beta": 0.5113912199934324, "train_val_corr": 0.0}
{"t": 13, "ts": 1788193425.2794366, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.50588035583496, "attn_entropy": 1.4623838424682618, "load_balance": 0.000471229845425114, "g_entropy": 2.1589888095855714, "tau_bias": 0.729915201663971, "beta": 0.48699719086289406, "train_val_corr": 0.0}
{"t": 14, "ts": 1788193434.9975152, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.51024627685547, "attn_entropy": 1.3847083747386932, "load_balance": 0.0004749212821479887, "g_entropy": 2.156910753250122, "tau_bias": 0.7394322752952576, "beta": 0.5058457230528196, "train_val_corr": 0.0}
{"t": 15, "ts": 1788193444.9742343, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.045487403869629, "attn_entropy": 1.3963046222925186, "load_balance": 0.00047312850947491827, "g_entropy": 2.1590830802917482, "tau_bias": 0.747269868850708, "beta": 0.5150434747338295, "train_val_corr": 0.0}
{"t": 16, "ts": 1788193454.4927366, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -1.6205215454101562, "attn_entropy": 1.3845128417015076, "load_balance": 0.00047300459700636564, "g_entropy": 2.1580681800842285, "tau_bias": 0.7656111121177673, "beta": 0.5000121096769968, "train_val_corr": 0.0}
{"t": 17, "ts": 1788193464.1886952, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 48.21352767944336, "attn_entropy": 1.489615535736084, "load_balance": 0.0004739578056614846, "g_entropy": 2.1589680194854735, "tau_bias": 0.7937296628952026, "beta": 0.5059141765038172, "train_val_corr": 0.0}
{"t": 18, "ts": 1788193474.081763, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 21.09170150756836, "attn_entropy": 1.9609086990356446, "load_balance": 0.0004770606174133718, "g_entropy": 2.159427785873413, "tau_bias": 0.7898919582366943, "beta": 0.5310594514012337, "train_val_corr": 0.0}
{"t": 0, "ts": 1788193590.982293, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.00016496657917741686, "advantage": 233.86448669433594, "attn_entropy": 1.489292985200882, "load_balance": 0.00047873075236566366, "g_entropy": 2.1514075756073, "tau_bias": 0.6509265899658203, "beta": 0.5577196727196375, "train_val_corr": 0.0}
{"t": 1, "ts": 1788193598.0945046, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.073209230957218e-07, "advantage": 135.4520721435547, "attn_entropy": 1.504406327009201, "load_balance": 0.0005040016258135438, "g_entropy": 2.115595817565918, "tau_bias": 0.6485937237739563, "beta": 0.534373015165329, "train_val_corr": 0.0}
{"t": 2, "ts": 1788193607.7288315, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 106.90690612792969, "attn_entropy": 1.4948223650455474, "load_balance": 0.0004752575827296823, "g_entropy": 2.1568811893463136, "tau_bias": 0.6451319456100464, "beta": 0.5370425631602606, "train_val_corr": 0.0}
{"t": 3, "ts": 1788193617.2822974, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 90.00999450683594, "attn_entropy": 1.4930770754814149, "load_balance": 0.00048624624032527207, "g_entropy": 2.138133096694946, "tau_bias": 0.6502702236175537, "beta": 0.5410975863536199, "train_val_corr": 0.0}
{"t": 4, "ts": 1788193627.406427, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 79.22921752929688, "attn_entropy": 1.4895644664764405, "load_balance": 0.00047609547036699953, "g_entropy": 2.1553314685821534, "tau_bias": 0.6526888012886047, "beta": 0.5445701579252878, "train_val_corr": 0.0}
{"t": 5, "ts": 1788193637.098021, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.597869873046875, "attn_entropy": 1.476958093047142, "load_balance": 0.0004801563394721597, "g_entropy": 2.147358751296997, "tau_bias": 0.6595838069915771, "beta": 0.5430855378508568, "train_val_corr": 0.0}
{"t": 6, "ts": 1788193646.8376012, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 50.57062911987305, "attn_entropy": 1.4702332824468614, "load_balance": 0.000479798304149881, "g_entropy": 2.148047399520874, "tau_bias": 0.6685022115707397, "beta": 0.543850320080916, "train_val_corr": 0.0}
{"t": 7, "ts": 1788193672.7136467, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 83.86402893066406, "attn_entropy": 1.4705988615751266, "load_balance": 0.0004767290607560426, "g_entropy": 2.151924896240234, "tau_bias": 0.6778479814529419, "beta": 0.5449230074882507, "train_val_corr": 0.0}
{"t": 8, "ts": 1788193682.535891, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 58.05849838256836, "attn_entropy": 1.487132951617241, "load_balance": 0.00047664226149208845, "g_entropy": 2.1549264907836916, "tau_bias": 0.6788264513015747, "beta": 0.5277107457319895, "train_val_corr": 0.0}
{"t": 9, "ts": 1788193691.683587, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 39.9108772277832, "attn_entropy": 1.4734902918338775, "load_balance": 0.0004830178862903267, "g_entropy": 2.1445977687835693, "tau_bias": 0.6880024671554565, "beta": 0.5379047989845276, "train_val_corr": 0.0}
{"t": 10, "ts": 1788193700.9979417, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.37331008911133, "attn_entropy": 1.485756081342697, "load_balance": 0.00047561593237333, "g_entropy": 2.156116485595703, "tau_bias": 0.6917641162872314, "beta": 0.5583699494600296, "train_val_corr": 0.0}
{"t": 11, "ts": 1788193710.167038, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 19.8219051361084, "attn_entropy": 1.488300397992134, "load_balance": 0.00047600738471373914, "g_entropy": 2.155643653869629, "tau_bias": 0.704373300075531, "beta": 0.5196714376409849, "train_val_corr": 0.0}
{"t": 12, "ts": 1788193719.5638206, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.291961669921875, "attn_entropy": 1.429060271382332, "load_balance": 0.0004909828654490411, "g_entropy": 2.131763744354248, "tau_bias": 0.7111386060714722, "beta": 0.5366366157929102, "train_val_corr": 0.0}
{"t": 13, "ts": 1788193744.0236504, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 118.08148193359375, "attn_entropy": 1.531869900226593, "load_balance": 0.0004752974899020046, "g_entropy": 2.1547550678253176, "tau_bias": 0.7126057147979736, "beta": 0.529900756975015, "train_val_corr": 0.0}
{"t": 14, "ts": 1788193752.212327, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.223304748535156, "attn_entropy": 1.4580867111682891, "load_balance": 0.00048351999721489846, "g_entropy": 2.1429885387420655, "tau_bias": 0.7204989790916443, "beta": 0.5127113337318102, "train_val_corr": 0.0}
{"t": 15, "ts": 1788193761.0201633, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -0.29686564207077026, "attn_entropy": 1.433584713935852, "load_balance": 0.0004787827667314559, "g_entropy": 2.150816774368286, "tau_bias": 0.7216870784759521, "beta": 0.526393656929334, "train_val_corr": 0.0}
{"t": 16, "ts": 1788193770.0727375, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.063535690307617, "attn_entropy": 1.4875813454389573, "load_balance": 0.0004803567077033222, "g_entropy": 2.150985670089722, "tau_bias": 0.7251416444778442, "beta": 0.5204427341620127, "train_val_corr": 0.0}
{"t": 17, "ts": 1788193780.3285837, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.3759169578552246, "attn_entropy": 1.4775080963969232, "load_balance": 0.0004759552713949233, "g_entropy": 2.156014251708984, "tau_bias": 0.7343813180923462, "beta": 0.5155022591352463, "train_val_corr": 0.0}
{"t": 18, "ts": 1788193790.0753663, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.894024848937988, "attn_entropy": 1.5333922386169434, "load_balance": 0.0004759374482091516, "g_entropy": 2.157166051864624, "tau_bias": 0.7370041608810425, "beta": 0.5161528637011846, "train_val_corr": 0.0}
{"t": 19, "ts": 1788193812.3873262, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.295143127441406, "attn_entropy": 1.4600316867232324, "load_balance": 0.0004753885034006089, "g_entropy": 2.1552161693573, "tau_bias": 0.7339385747909546, "beta": 0.5131564537684122, "train_val_corr": 1.0}
{"t": 20, "ts": 1788193821.2118733, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.981983184814453, "attn_entropy": 1.4705814749002457, "load_balance": 0.0004749368934426457, "g_entropy": 2.153610849380493, "tau_bias": 0.7318506836891174, "beta": 0.5130152006944021, "train_val_corr": 1.0}
{"t": 21, "ts": 1788193831.1836793, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.952742099761963, "attn_entropy": 1.3807418283075095, "load_balance": 0.0004722240788396448, "g_entropy": 2.154816246032715, "tau_bias": 0.7306451201438904, "beta": 0.541745645304521, "train_val_corr": 0.9999999999999999}
{"t": 22, "ts": 1788193841.4108295, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.607195854187012, "attn_entropy": 1.4029106739908457, "load_balance": 0.00047770116361789405, "g_entropy": 2.1523433208465574, "tau_bias": 0.7285284996032715, "beta": 0.5770759036143621, "train_val_corr": 1.0}
{"t": 23, "ts": 1788193851.431503, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -14.113017082214355, "attn_entropy": 1.4256901666522026, "load_balance": 0.0004754197492729872, "g_entropy": 2.155956983566284, "tau_bias": 0.7257688045501709, "beta": 0.5752143984039625, "train_val_corr": 1.0}
{"t": 24, "ts": 1788193861.3924496, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.482805252075195, "attn_entropy": 1.4184508372098208, "load_balance": 0.000478697958169505, "g_entropy": 2.1504141807556154, "tau_bias": 0.7229959964752197, "beta": 0.5598646725217501, "train_val_corr": 1.0}
{"t": 25, "ts": 1788193886.5311606, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.657544136047363, "attn_entropy": 1.4230731040239335, "load_balance": 0.0004833551414776593, "g_entropy": 2.1419427394866943, "tau_bias": 0.7179423570632935, "beta": 0.5668672919273376, "train_val_corr": 0.9999999999999997}
{"t": 26, "ts": 1788193894.3887043, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -19.55494499206543, "attn_entropy": 1.469193208217621, "load_balance": 0.0004743132391013205, "g_entropy": 2.1570353507995605, "tau_bias": 0.7089336514472961, "beta": 0.5725661093990008, "train_val_corr": 1.0}
{"t": 27, "ts": 1788193903.321153, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.149471282958984, "attn_entropy": 1.4534897059202194, "load_balance": 0.00047573307529091835, "g_entropy": 2.155047559738159, "tau_bias": 0.7066587209701538, "beta": 0.5640347053607305, "train_val_corr": 1.0}
{"t": 28, "ts": 1788193912.413654, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.9643497467041, "attn_entropy": 1.4248472712934017, "load_balance": 0.00047483514645136894, "g_entropy": 2.1564094543457033, "tau_bias": 0.7064183950424194, "beta": 0.5528718282779058, "train_val_corr": 1.0}
{"t": 29, "ts": 1788193922.4191475, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.11756134033203, "attn_entropy": 1.4409356124699115, "load_balance": 0.00047185689327307047, "g_entropy": 2.1588532447814943, "tau_bias": 0.7041711807250977, "beta": 0.5626444096366564, "train_val_corr": 0.9999999999999998}
{"t": 30, "ts": 1788193932.8102176, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.89786720275879, "attn_entropy": 1.4820176094770432, "load_balance": 0.0004749827028717846, "g_entropy": 2.1587522506713865, "tau_bias": 0.7013100385665894, "beta": 0.5596639811992645, "train_val_corr": 1.0}
{"t": 31, "ts": 1788193957.18715, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -24.56830596923828, "attn_entropy": 1.4493385583162308, "load_balance": 0.0004735155962407589, "g_entropy": 2.1587384223937987, "tau_bias": 0.6972109079360962, "beta": 0.5492019852002462, "train_val_corr": 1.0}
{"t": 32, "ts": 1788193965.1784503, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.716249465942383, "attn_entropy": 1.4522236712276935, "load_balance": 0.0004746820719446987, "g_entropy": 2.156639909744263, "tau_bias": 0.6839556694030762, "beta": 0.5404068231582642, "train_val_corr": 0.9999999999999998}
{"t": 33, "ts": 1788193974.1181743, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.15065574645996, "attn_entropy": 1.4611121721565723, "load_balance": 0.00047611432382836937, "g_entropy": 2.1564822673797606, "tau_bias": 0.6755799651145935, "beta": 0.5450588539242744, "train_val_corr": 0.9999999999999996}
{"t": 34, "ts": 1788193983.233471, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.584049224853516, "attn_entropy": 1.4608528941869736, "load_balance": 0.00047489738208241763, "g_entropy": 2.1585693836212156, "tau_bias": 0.6719614267349243, "beta": 0.5323830420772234, "train_val_corr": 1.0}
{"t": 0, "ts": 1788194021.4312353, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.5932821497699479e-06, "advantage": 23.574066162109375, "attn_entropy": 1.4596325494349003, "load_balance": 0.0004740151343867183, "g_entropy": 2.157486343383789, "tau_bias": 0.6694812774658203, "beta": 0.5335702275236448, "train_val_corr": 0.0}
{"t": 1, "ts": 1788194028.5663614, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0724515081838035e-07, "advantage": 43.05596160888672, "attn_entropy": 1.4604599103331566, "load_balance": 0.00047992690233513715, "g_entropy": 2.147431516647339, "tau_bias": 0.6698113083839417, "beta": 0.5678501178820928, "train_val_corr": 0.0}
{"t": 2, "ts": 1788194037.5146432, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 22.06309700012207, "attn_entropy": 1.4458961464464664, "load_balance": 0.00047327528591267765, "g_entropy": 2.1592240810394285, "tau_bias": 0.6717329621315002, "beta": 0.5560280059774717, "train_val_corr": 0.0}
{"t": 3, "ts": 1788194046.3149686, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 25.852628707885742, "attn_entropy": 1.4552785366773606, "load_balance": 0.0004790011793375015, "g_entropy": 2.1524893283843993, "tau_bias": 0.6742298007011414, "beta": 0.5598788311084112, "train_val_corr": 0.0}
{"t": 4, "ts": 1788194054.8793602, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 60.43634796142578, "attn_entropy": 1.4649803936481476, "load_balance": 0.00047443113871850076, "g_entropy": 2.1584676265716554, "tau_bias": 0.6759588718414307, "beta": 0.5707544535398483, "train_val_corr": 0.0}
{"t": 5, "ts": 1788194063.583931, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 18.48818588256836, "attn_entropy": 1.4400771513581276, "load_balance": 0.00047271131188608706, "g_entropy": 2.156068468093872, "tau_bias": 0.6737765073776245, "beta": 0.5556010951598486, "train_val_corr": 0.0}
{"t": 6, "ts": 1788194072.6877606, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.58858585357666, "attn_entropy": 1.446292307972908, "load_balance": 0.000474248785758391, "g_entropy": 2.1564708709716798, "tau_bias": 0.6710984706878662, "beta": 0.5504067117969195, "train_val_corr": 0.0}
{"t": 7, "ts": 1788194098.200746, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 20.78583526611328, "attn_entropy": 1.4253369271755219, "load_balance": 0.0004753494518809021, "g_entropy": 2.1521064758300783, "tau_bias": 0.6656938195228577, "beta": 0.5445789098739624, "train_val_corr": 0.0}
{"t": 8, "ts": 1788194108.3203974, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.547415733337402, "attn_entropy": 1.444367841631174, "load_balance": 0.0004749993095174432, "g_entropy": 2.158572483062744, "tau_bias": 0.6624353528022766, "beta": 0.555737574895223, "train_val_corr": 0.0}
{"t": 9, "ts": 1788194118.164642, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.96996784210205, "attn_entropy": 1.4549090206623077, "load_balance": 0.0004751741362269968, "g_entropy": 2.156873083114624, "tau_bias": 0.6612713932991028, "beta": 0.5415557945768038, "train_val_corr": 0.0}
{"t": 10, "ts": 1788194128.5989125, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.759478569030762, "attn_entropy": 1.4751184538006783, "load_balance": 0.0004738975258078426, "g_entropy": 2.1593605995178224, "tau_bias": 0.6574777364730835, "beta": 0.5324633394678434, "train_val_corr": 0.0}
{"t": 11, "ts": 1788194138.7298293, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.26884937286377, "attn_entropy": 1.4438274048268795, "load_balance": 0.0004745150858070701, "g_entropy": 2.1590927600860597, "tau_bias": 0.6565953493118286, "beta": 0.5326097632447878, "train_val_corr": 0.0}
{"t": 12, "ts": 1788194148.0202205, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.991272449493408, "attn_entropy": 1.4598687171936036, "load_balance": 0.0004780864401254803, "g_entropy": 2.152644634246826, "tau_bias": 0.6526188254356384, "beta": 0.5241464152932167, "train_val_corr": 0.0}
{"t": 13, "ts": 1788194171.005638, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.545065879821777, "attn_entropy": 1.4749095141887665, "load_balance": 0.00047305254265666007, "g_entropy": 2.159195566177368, "tau_bias": 0.6476952433586121, "beta": 0.5277427310744921, "train_val_corr": 0.0}
{"t": 14, "ts": 1788194179.0880609, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.31648063659668, "attn_entropy": 1.4361313186585902, "load_balance": 0.0004734818357974291, "g_entropy": 2.157428169250488, "tau_bias": 0.6430888772010803, "beta": 0.5237697983781496, "train_val_corr": 0.0}
{"t": 15, "ts": 1788194187.7817256, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.168776512145996, "attn_entropy": 1.4406652390956878, "load_balance": 0.00047445945092476903, "g_entropy": 2.159170055389404, "tau_bias": 0.6453005075454712, "beta": 0.5235982437928518, "train_val_corr": 0.0}
{"t": 16, "ts": 1788194197.6951187, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.163451194763184, "attn_entropy": 1.4463097803294658, "load_balance": 0.0004753224726300687, "g_entropy": 2.1583879947662354, "tau_bias": 0.6433128714561462, "beta": 0.518310417731603, "train_val_corr": 0.0}
{"t": 17, "ts": 1788194207.5912325, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.352834224700928, "attn_entropy": 1.456115658581257, "load_balance": 0.00047399732284247875, "g_entropy": 2.159459114074707, "tau_bias": 0.6418328285217285, "beta": 0.5195082426071167, "train_val_corr": 0.0}
{"t": 18, "ts": 1788194217.2105217, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.901129722595215, "attn_entropy": 1.463829755783081, "load_balance": 0.0004742951190564781, "g_entropy": 2.15970664024353, "tau_bias": 0.6393529772758484, "beta": 0.5115399758021036, "train_val_corr": 0.0}
{"t": 0, "ts": 1788194365.5866747, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.541656795074232e-05, "advantage": 107.24652862548828, "attn_entropy": 1.5088270902633667, "load_balance": 0.0004780653573106974, "g_entropy": 2.1526050567626953, "tau_bias": 0.6509265899658203, "beta": 0.5584118788441023, "train_val_corr": 0.0}
{"t": 1, "ts": 1788194372.931441, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072697924404565e-07, "advantage": 89.72090148925781, "attn_entropy": 2.03877592086792, "load_balance": 0.0004780834016855806, "g_entropy": 2.156227779388428, "tau_bias": 0.646527886390686, "beta": 0.5712046250700951, "train_val_corr": 0.0}
{"t": 2, "ts": 1788194382.3968048, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 68.68505096435547, "attn_entropy": 1.6358208298683166, "load_balance": 0.0004787362646311522, "g_entropy": 2.1505734443664553, "tau_bias": 0.6498159170150757, "beta": 0.5686033045252165, "train_val_corr": 0.0}
{"t": 3, "ts": 1788194391.4729137, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 73.09050750732422, "attn_entropy": 1.5631667137145997, "load_balance": 0.0004749728774186224, "g_entropy": 2.1569216728210447, "tau_bias": 0.6527172923088074, "beta": 0.5429712931315104, "train_val_corr": 0.0}
{"t": 4, "ts": 1788194400.3872674, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 72.23716735839844, "attn_entropy": 1.6151904463768005, "load_balance": 0.0004749813349917531, "g_entropy": 2.1571628570556642, "tau_bias": 0.6573835611343384, "beta": 0.5722347026069959, "train_val_corr": 0.0}
{"t": 5, "ts": 1788194409.5150623, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 65.53317260742188, "attn_entropy": 1.533086109161377, "load_balance": 0.0004745642829220742, "g_entropy": 2.158458423614502, "tau_bias": 0.6733006238937378, "beta": 0.5313867554068565, "train_val_corr": 0.0}
{"t": 6, "ts": 1788194418.5866613, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 136.11404418945312, "attn_entropy": 1.509199470281601, "load_balance": 0.0004756187787279487, "g_entropy": 2.1563482761383055, "tau_bias": 0.6719450354576111, "beta": 0.559227059284846, "train_val_corr": 0.0}
{"t": 7, "ts": 1788194441.7634487, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723270210964984e-07, "advantage": 175.7641143798828, "attn_entropy": 1.5042088806629181, "load_balance": 0.0004756788606755435, "g_entropy": 2.154452419281006, "tau_bias": 0.6803413033485413, "beta": 0.5269864797592163, "train_val_corr": 0.0}
{"t": 8, "ts": 1788194451.2090821, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.04716491699219, "attn_entropy": 1.5454297959804535, "load_balance": 0.0004784294927958399, "g_entropy": 2.151358127593994, "tau_bias": 0.6817105412483215, "beta": 0.5431402126948038, "train_val_corr": 0.0}
{"t": 9, "ts": 1788194460.9928796, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 52.30317306518555, "attn_entropy": 1.538418424129486, "load_balance": 0.0004805366566870362, "g_entropy": 2.1486629962921144, "tau_bias": 0.6889121532440186, "beta": 0.520272453625997, "train_val_corr": 0.0}
{"t": 10, "ts": 1788194469.9943073, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.933290481567383, "attn_entropy": 1.607443356513977, "load_balance": 0.00047473458107560873, "g_entropy": 2.1584739208221437, "tau_bias": 0.7008935213088989, "beta": 0.5334595938523611, "train_val_corr": 0.0}
{"t": 11, "ts": 1788194478.4037268, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 231.17132568359375, "attn_entropy": 1.8630347728729248, "load_balance": 0.00047701409785076975, "g_entropy": 2.155347204208374, "tau_bias": 0.6985456943511963, "beta": 0.5369742636879286, "train_val_corr": 0.0}
{"t": 12, "ts": 1788194487.2764478, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.683601379394531, "attn_entropy": 1.7435780048370362, "load_balance": 0.00048062226269394157, "g_entropy": 2.15206093788147, "tau_bias": 0.7104957699775696, "beta": 0.5137855360905329, "train_val_corr": 0.0}
{"t": 13, "ts": 1788194510.923387, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 16.932825088500977, "attn_entropy": 1.8686827659606933, "load_balance": 0.00047939049545675516, "g_entropy": 2.155273962020874, "tau_bias": 0.7162169814109802, "beta": 0.5268916835387548, "train_val_corr": 0.0}
{"t": 14, "ts": 1788194519.9921775, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 586.9725952148438, "attn_entropy": 1.94769344329834, "load_balance": 0.0004773543507326394, "g_entropy": 2.157172203063965, "tau_bias": 0.716325581073761, "beta": 0.5271659269928932, "train_val_corr": 0.0}
{"t": 15, "ts": 1788194529.469058, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.58620834350586, "attn_entropy": 1.4907590508461, "load_balance": 0.0004766870813909918, "g_entropy": 2.1543347358703615, "tau_bias": 0.7284765243530273, "beta": 0.49652262901266414, "train_val_corr": 0.0}
{"t": 16, "ts": 1788194538.0705144, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.622438430786133, "attn_entropy": 1.472950428724289, "load_balance": 0.00047396817826665937, "g_entropy": 2.1579267024993896, "tau_bias": 0.751097559928894, "beta": 0.5020412082473437, "train_val_corr": 0.0}
{"t": 17, "ts": 1788194546.8648663, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 32.07689666748047, "attn_entropy": 1.3497906118631362, "load_balance": 0.00047846404486335815, "g_entropy": 2.1504985809326174, "tau_bias": 0.7559785842895508, "beta": 0.5091952458024025, "train_val_corr": 0.0}
{"t": 18, "ts": 1788194555.2707095, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.704023838043213, "attn_entropy": 1.3439210563898087, "load_balance": 0.0004733814566861838, "g_entropy": 2.1570386409759523, "tau_bias": 0.7603507041931152, "beta": 0.4953181358675162, "train_val_corr": 0.0}
{"t": 19, "ts": 1788194578.0333898, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.25495719909668, "attn_entropy": 1.3679913491010667, "load_balance": 0.0004765985533595085, "g_entropy": 2.153904390335083, "tau_bias": 0.7670421600341797, "beta": 0.4962015338242054, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788194586.2331934, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.781298637390137, "attn_entropy": 1.402373018860817, "load_balance": 0.00047343826736323535, "g_entropy": 2.158175468444824, "tau_bias": 0.7781072854995728, "beta": 0.5035545602440834, "train_val_corr": 1.0}
{"t": 21, "ts": 1788194595.8242276, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.58829689025879, "attn_entropy": 1.3677545428276061, "load_balance": 0.00047486150288023055, "g_entropy": 2.155410146713257, "tau_bias": 0.7776638865470886, "beta": 0.5205506781737009, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788194605.3756216, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -46.50102996826172, "attn_entropy": 1.398021388053894, "load_balance": 0.00047660571290180086, "g_entropy": 2.154541492462158, "tau_bias": 0.7775037884712219, "beta": 0.5097216665744781, "train_val_corr": 1.0}
{"t": 23, "ts": 1788194614.9357314, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.83917236328125, "attn_entropy": 1.3840453386306764, "load_balance": 0.00047706675832159817, "g_entropy": 2.154738664627075, "tau_bias": 0.7739366888999939, "beta": 0.5244797393679619, "train_val_corr": 1.0}
{"t": 24, "ts": 1788194623.7360637, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.23626708984375, "attn_entropy": 1.3669119477272034, "load_balance": 0.00047682768781669437, "g_entropy": 2.153565549850464, "tau_bias": 0.7761063575744629, "beta": 0.52259744455417, "train_val_corr": 1.0}
{"t": 25, "ts": 1788194646.4999194, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.03636169433594, "attn_entropy": 1.3476117163896562, "load_balance": 0.00047982686082832516, "g_entropy": 2.147431993484497, "tau_bias": 0.7789179086685181, "beta": 0.5349597334861755, "train_val_corr": 1.0}
{"t": 26, "ts": 1788194654.79251, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -48.60801315307617, "attn_entropy": 1.4195789337158202, "load_balance": 0.00047398332972079513, "g_entropy": 2.1576677799224853, "tau_bias": 0.7725303769111633, "beta": 0.5320080320040385, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788194663.9767733, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -44.37407684326172, "attn_entropy": 1.4300884038209916, "load_balance": 0.0004736429254990071, "g_entropy": 2.1582190990448, "tau_bias": 0.7672858238220215, "beta": 0.5250358680884043, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788194672.769476, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -46.15907669067383, "attn_entropy": 1.4260217040777206, "load_balance": 0.0004739692143630236, "g_entropy": 2.158786582946777, "tau_bias": 0.7617290019989014, "beta": 0.5237450028459231, "train_val_corr": 1.0}
{"t": 29, "ts": 1788194681.315969, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -43.428741455078125, "attn_entropy": 1.443551704287529, "load_balance": 0.00047591047477908434, "g_entropy": 2.155571222305298, "tau_bias": 0.7596361041069031, "beta": 0.5287181834379832, "train_val_corr": 1.0}
{"t": 30, "ts": 1788194690.2149773, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.606300354003906, "attn_entropy": 1.3982668220996857, "load_balance": 0.0004759089846629649, "g_entropy": 2.154165267944336, "tau_bias": 0.7567456364631653, "beta": 0.5226710935433706, "train_val_corr": 0.9999999999999996}
{"t": 31, "ts": 1788194714.0311654, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.667213439941406, "attn_entropy": 1.423149797320366, "load_balance": 0.0004758568364195526, "g_entropy": 2.156546640396118, "tau_bias": 0.7545661926269531, "beta": 0.5212625364462534, "train_val_corr": 1.0}
{"t": 32, "ts": 1788194721.888741, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.703643798828125, "attn_entropy": 1.384346704185009, "load_balance": 0.00047234864323399963, "g_entropy": 2.1592222690582275, "tau_bias": 0.7530725002288818, "beta": 0.5188408320148786, "train_val_corr": 0.9999999999999993}
{"t": 33, "ts": 1788194731.4363337, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.892982482910156, "attn_entropy": 1.3518488556146622, "load_balance": 0.0004728445434011519, "g_entropy": 2.1594515323638914, "tau_bias": 0.751308262348175, "beta": 0.5222650865713755, "train_val_corr": 1.0}
{"t": 34, "ts": 1788194740.5035484, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -39.038726806640625, "attn_entropy": 1.3794395864009856, "load_balance": 0.00047408321406692264, "g_entropy": 2.1573153018951414, "tau_bias": 0.7478220462799072, "beta": 0.5195876260598501, "train_val_corr": 0.9999999999999998}
{"t": 35, "ts": 1788194749.3327942, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -38.11954879760742, "attn_entropy": 1.3628646373748778, "load_balance": 0.00047315804986283185, "g_entropy": 2.158894157409668, "tau_bias": 0.745747983455658, "beta": 0.5238856673240662, "train_val_corr": 0.9999999999999996}
{"t": 36, "ts": 1788194758.0731826, "step": 360, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.71859359741211, "attn_entropy": 1.3712503388524055, "load_balance": 0.0004723662626929581, "g_entropy": 2.1588478565216063, "tau_bias": 0.7448042035102844, "beta": 0.5172327930728594, "train_val_corr": 1.0}
{"t": 0, "ts": 1788194837.267776, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.8455239114700817e-06, "advantage": 26.846229553222656, "attn_entropy": 1.3836654022336006, "load_balance": 0.00047354287817142905, "g_entropy": 2.158229398727417, "tau_bias": 0.7422351837158203, "beta": 0.5131005868315697, "train_val_corr": 0.0}
{"t": 1, "ts": 1788194844.3041632, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072609959213878e-07, "advantage": 60.76274490356445, "attn_entropy": 1.3927810661494733, "load_balance": 0.00047067334526218474, "g_entropy": 2.1585214138031006, "tau_bias": 0.7483364939689636, "beta": 0.5726271073023478, "train_val_corr": 0.0}
{"t": 2, "ts": 1788194853.7319398, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 27.056724548339844, "attn_entropy": 1.3387424811720847, "load_balance": 0.0004730107961222529, "g_entropy": 2.1593146324157715, "tau_bias": 0.7505127191543579, "beta": 0.5490414102872213, "train_val_corr": 0.0}
{"t": 3, "ts": 1788194863.1055872, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 38.10312271118164, "attn_entropy": 1.3109646048396826, "load_balance": 0.0004728938627522439, "g_entropy": 2.159265375137329, "tau_bias": 0.7513498663902283, "beta": 0.5750054121017456, "train_val_corr": 0.0}
{"t": 4, "ts": 1788194872.2953625, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 28.208959579467773, "attn_entropy": 1.3282313011586666, "load_balance": 0.00047331915702670814, "g_entropy": 2.1587005615234376, "tau_bias": 0.7625691890716553, "beta": 0.5785812363028526, "train_val_corr": 0.0}
{"t": 5, "ts": 1788194881.5062544, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 21.674943923950195, "attn_entropy": 1.3517444416880608, "load_balance": 0.00047517343191429975, "g_entropy": 2.155775260925293, "tau_bias": 0.7735312581062317, "beta": 0.5684969474871954, "train_val_corr": 0.0}
{"t": 6, "ts": 1788194891.3363934, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 17.552053451538086, "attn_entropy": 1.3031622115522623, "load_balance": 0.0004729154228698462, "g_entropy": 2.157785654067993, "tau_bias": 0.7896472215652466, "beta": 0.5609865039587021, "train_val_corr": 0.0}
{"t": 7, "ts": 1788194916.289394, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 62.39039611816406, "attn_entropy": 1.3147990956902504, "load_balance": 0.00048197160940617324, "g_entropy": 2.1463560581207277, "tau_bias": 0.7940338253974915, "beta": 0.5618990659713745, "train_val_corr": 0.0}
{"t": 8, "ts": 1788194925.4038239, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 21.55828285217285, "attn_entropy": 1.3114618442952632, "load_balance": 0.0004715579911135137, "g_entropy": 2.158911609649658, "tau_bias": 0.8003755211830139, "beta": 0.5517746532956759, "train_val_corr": 0.0}
{"t": 9, "ts": 1788194934.5753372, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.210066795349121, "attn_entropy": 1.3066865146160125, "load_balance": 0.00047163369017653165, "g_entropy": 2.1595655918121337, "tau_bias": 0.8038656115531921, "beta": 0.53650979946057, "train_val_corr": 0.0}
{"t": 10, "ts": 1788194944.2653592, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 15.340768814086914, "attn_entropy": 1.3140930846333503, "load_balance": 0.0004730040964204818, "g_entropy": 2.1578542232513427, "tau_bias": 0.8032995462417603, "beta": 0.5383978386720022, "train_val_corr": 0.0}
{"t": 11, "ts": 1788194953.4192836, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.462810516357422, "attn_entropy": 1.3110864877700805, "load_balance": 0.00047228853800334035, "g_entropy": 2.1586824893951415, "tau_bias": 0.804524838924408, "beta": 0.532344621916612, "train_val_corr": 0.0}
{"t": 12, "ts": 1788194962.7010918, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.980320930480957, "attn_entropy": 1.3326971992850303, "load_balance": 0.00047238205443136396, "g_entropy": 2.1587284088134764, "tau_bias": 0.7993857264518738, "beta": 0.5366574575503668, "train_val_corr": 0.0}
{"t": 13, "ts": 1788194985.8858597, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.516908645629883, "attn_entropy": 1.3027584753930568, "load_balance": 0.00047329991939477624, "g_entropy": 2.156495380401611, "tau_bias": 0.7971431016921997, "beta": 0.5286739518245062, "train_val_corr": 0.0}
{"t": 14, "ts": 1788194994.2832005, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.609851837158203, "attn_entropy": 1.3028671093285085, "load_balance": 0.00047285307082347574, "g_entropy": 2.158542013168335, "tau_bias": 0.7959492802619934, "beta": 0.5266600996255875, "train_val_corr": 0.0}
{"t": 15, "ts": 1788195003.4245605, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.370989799499512, "attn_entropy": 1.2937706336379051, "load_balance": 0.00047275457181967796, "g_entropy": 2.1588636875152587, "tau_bias": 0.7970687747001648, "beta": 0.5330915699402491, "train_val_corr": 0.0}
{"t": 16, "ts": 1788195012.4803438, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.20208215713501, "attn_entropy": 1.291519927419722, "load_balance": 0.00047191287740133703, "g_entropy": 2.1597832679748534, "tau_bias": 0.7958289980888367, "beta": 0.5284734989205996, "train_val_corr": 0.0}
{"t": 17, "ts": 1788195021.9787345, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.67823600769043, "attn_entropy": 1.2927681816741825, "load_balance": 0.0004720224183984101, "g_entropy": 2.1583470344543456, "tau_bias": 0.7954632639884949, "beta": 0.5249810243646303, "train_val_corr": 0.0}
{"t": 18, "ts": 1788195031.033695, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.39007568359375, "attn_entropy": 1.2922599976416678, "load_balance": 0.00047147017903625965, "g_entropy": 2.1597671508789062, "tau_bias": 0.7919459342956543, "beta": 0.5259487554430962, "train_val_corr": 0.0}
{"t": 0, "ts": 1788195137.1693118, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.5477211238467135e-05, "advantage": 107.24652862548828, "attn_entropy": 1.5088270902633667, "load_balance": 0.0004780653573106974, "g_entropy": 2.1526050567626953, "tau_bias": 0.6509265899658203, "beta": 0.5584118788441023, "train_val_corr": 0.0}
{"t": 1, "ts": 1788195144.1166234, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0726861293951515e-07, "advantage": 89.72090148925781, "attn_entropy": 2.03877592086792, "load_balance": 0.0004780834016855806, "g_entropy": 2.156227779388428, "tau_bias": 0.646527886390686, "beta": 0.5712046250700951, "train_val_corr": 0.0}
{"t": 2, "ts": 1788195153.7050834, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 68.68505096435547, "attn_entropy": 1.6358208298683166, "load_balance": 0.0004787362646311522, "g_entropy": 2.1505734443664553, "tau_bias": 0.6498159170150757, "beta": 0.5686033045252165, "train_val_corr": 0.0}
{"t": 3, "ts": 1788195162.8788204, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 73.09050750732422, "attn_entropy": 1.5631667137145997, "load_balance": 0.0004749728774186224, "g_entropy": 2.1569216728210447, "tau_bias": 0.6527172923088074, "beta": 0.5429712931315104, "train_val_corr": 0.0}
{"t": 4, "ts": 1788195172.3211956, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 72.23716735839844, "attn_entropy": 1.6151904463768005, "load_balance": 0.0004749813349917531, "g_entropy": 2.1571628570556642, "tau_bias": 0.6573835611343384, "beta": 0.5722347026069959, "train_val_corr": 0.0}
{"t": 5, "ts": 1788195181.9821866, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 65.53317260742188, "attn_entropy": 1.533086109161377, "load_balance": 0.0004745642829220742, "g_entropy": 2.158458423614502, "tau_bias": 0.6733006238937378, "beta": 0.5313867554068565, "train_val_corr": 0.0}
{"t": 6, "ts": 1788195191.8818796, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 136.11404418945312, "attn_entropy": 1.509199470281601, "load_balance": 0.0004756187787279487, "g_entropy": 2.1563482761383055, "tau_bias": 0.6719450354576111, "beta": 0.559227059284846, "train_val_corr": 0.0}
{"t": 7, "ts": 1788195217.1014538, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723268789879512e-07, "advantage": 175.7641143798828, "attn_entropy": 1.5042088806629181, "load_balance": 0.0004756788606755435, "g_entropy": 2.154452419281006, "tau_bias": 0.6803413033485413, "beta": 0.5269864797592163, "train_val_corr": 0.0}
{"t": 8, "ts": 1788195226.910314, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.04716491699219, "attn_entropy": 1.5454297959804535, "load_balance": 0.0004784294927958399, "g_entropy": 2.151358127593994, "tau_bias": 0.6817105412483215, "beta": 0.5431402126948038, "train_val_corr": 0.0}
{"t": 9, "ts": 1788195236.8225622, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 52.30317306518555, "attn_entropy": 1.538418424129486, "load_balance": 0.0004805366566870362, "g_entropy": 2.1486629962921144, "tau_bias": 0.6889121532440186, "beta": 0.520272453625997, "train_val_corr": 0.0}
{"t": 10, "ts": 1788195246.2099643, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.933290481567383, "attn_entropy": 1.607443356513977, "load_balance": 0.00047473458107560873, "g_entropy": 2.1584739208221437, "tau_bias": 0.7008935213088989, "beta": 0.5334595938523611, "train_val_corr": 0.0}
{"t": 11, "ts": 1788195255.9873774, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 231.17132568359375, "attn_entropy": 1.8630347728729248, "load_balance": 0.00047701409785076975, "g_entropy": 2.155347204208374, "tau_bias": 0.6985456943511963, "beta": 0.5369742636879286, "train_val_corr": 0.0}
{"t": 12, "ts": 1788195265.8314562, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.683601379394531, "attn_entropy": 1.7435780048370362, "load_balance": 0.00048062226269394157, "g_entropy": 2.15206093788147, "tau_bias": 0.7104957699775696, "beta": 0.5137855360905329, "train_val_corr": 0.0}
{"t": 13, "ts": 1788195290.5328903, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 16.932825088500977, "attn_entropy": 1.8686827659606933, "load_balance": 0.00047939049545675516, "g_entropy": 2.155273962020874, "tau_bias": 0.7162169814109802, "beta": 0.5268916835387548, "train_val_corr": 0.0}
{"t": 14, "ts": 1788195299.6224077, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 586.9725952148438, "attn_entropy": 1.94769344329834, "load_balance": 0.0004773543507326394, "g_entropy": 2.157172203063965, "tau_bias": 0.716325581073761, "beta": 0.5271659269928932, "train_val_corr": 0.0}
{"t": 15, "ts": 1788195308.5967827, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.58620834350586, "attn_entropy": 1.4907590508461, "load_balance": 0.0004766870813909918, "g_entropy": 2.1543347358703615, "tau_bias": 0.7284765243530273, "beta": 0.49652262901266414, "train_val_corr": 0.0}
{"t": 16, "ts": 1788195317.1251874, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.622438430786133, "attn_entropy": 1.472950428724289, "load_balance": 0.00047396817826665937, "g_entropy": 2.1579267024993896, "tau_bias": 0.751097559928894, "beta": 0.5020412082473437, "train_val_corr": 0.0}
{"t": 17, "ts": 1788195325.9975975, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 32.07689666748047, "attn_entropy": 1.3497906118631362, "load_balance": 0.00047846404486335815, "g_entropy": 2.1504985809326174, "tau_bias": 0.7559785842895508, "beta": 0.5091952458024025, "train_val_corr": 0.0}
{"t": 18, "ts": 1788195334.4794352, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.704023838043213, "attn_entropy": 1.3439210563898087, "load_balance": 0.0004733814566861838, "g_entropy": 2.1570386409759523, "tau_bias": 0.7603507041931152, "beta": 0.4953181358675162, "train_val_corr": 0.0}
{"t": 19, "ts": 1788195357.0251753, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.25495719909668, "attn_entropy": 1.3679913491010667, "load_balance": 0.0004765985533595085, "g_entropy": 2.153904390335083, "tau_bias": 0.7670421600341797, "beta": 0.4962015338242054, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788195365.2699816, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.781298637390137, "attn_entropy": 1.402373018860817, "load_balance": 0.00047343826736323535, "g_entropy": 2.158175468444824, "tau_bias": 0.7781072854995728, "beta": 0.5035545602440834, "train_val_corr": 1.0}
{"t": 21, "ts": 1788195375.3801935, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.58829689025879, "attn_entropy": 1.3677545428276061, "load_balance": 0.00047486150288023055, "g_entropy": 2.155410146713257, "tau_bias": 0.7776638865470886, "beta": 0.5205506781737009, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788195385.2145038, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -46.50102996826172, "attn_entropy": 1.398021388053894, "load_balance": 0.00047660571290180086, "g_entropy": 2.154541492462158, "tau_bias": 0.7775037884712219, "beta": 0.5097216665744781, "train_val_corr": 1.0}
{"t": 23, "ts": 1788195394.6977537, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.83917236328125, "attn_entropy": 1.3840453386306764, "load_balance": 0.00047706675832159817, "g_entropy": 2.154738664627075, "tau_bias": 0.7739366888999939, "beta": 0.5244797393679619, "train_val_corr": 1.0}
{"t": 24, "ts": 1788195404.097176, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.23626708984375, "attn_entropy": 1.3669119477272034, "load_balance": 0.00047682768781669437, "g_entropy": 2.153565549850464, "tau_bias": 0.7761063575744629, "beta": 0.52259744455417, "train_val_corr": 1.0}
{"t": 25, "ts": 1788195427.8850691, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.03636169433594, "attn_entropy": 1.3476117163896562, "load_balance": 0.00047982686082832516, "g_entropy": 2.147431993484497, "tau_bias": 0.7789179086685181, "beta": 0.5349597334861755, "train_val_corr": 1.0}
{"t": 26, "ts": 1788195436.9824433, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -48.60801315307617, "attn_entropy": 1.4195789337158202, "load_balance": 0.00047398332972079513, "g_entropy": 2.1576677799224853, "tau_bias": 0.7725303769111633, "beta": 0.5320080320040385, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788195446.9368136, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -44.37407684326172, "attn_entropy": 1.4300884038209916, "load_balance": 0.0004736429254990071, "g_entropy": 2.1582190990448, "tau_bias": 0.7672858238220215, "beta": 0.5250358680884043, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788195456.4883308, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -46.15907669067383, "attn_entropy": 1.4260217040777206, "load_balance": 0.0004739692143630236, "g_entropy": 2.158786582946777, "tau_bias": 0.7617290019989014, "beta": 0.5237450028459231, "train_val_corr": 1.0}
{"t": 29, "ts": 1788195466.035753, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -43.428741455078125, "attn_entropy": 1.443551704287529, "load_balance": 0.00047591047477908434, "g_entropy": 2.155571222305298, "tau_bias": 0.7596361041069031, "beta": 0.5287181834379832, "train_val_corr": 1.0}
{"t": 30, "ts": 1788195475.7645447, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.606300354003906, "attn_entropy": 1.3982668220996857, "load_balance": 0.0004759089846629649, "g_entropy": 2.154165267944336, "tau_bias": 0.7567456364631653, "beta": 0.5226710935433706, "train_val_corr": 0.9999999999999996}
{"t": 31, "ts": 1788195502.2180834, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.667213439941406, "attn_entropy": 1.423149797320366, "load_balance": 0.0004758568364195526, "g_entropy": 2.156546640396118, "tau_bias": 0.7545661926269531, "beta": 0.5212625364462534, "train_val_corr": 1.0}
{"t": 32, "ts": 1788195510.4331408, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.703643798828125, "attn_entropy": 1.384346704185009, "load_balance": 0.00047234864323399963, "g_entropy": 2.1592222690582275, "tau_bias": 0.7530725002288818, "beta": 0.5188408320148786, "train_val_corr": 0.9999999999999993}
{"t": 33, "ts": 1788195519.613575, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.892982482910156, "attn_entropy": 1.3518488556146622, "load_balance": 0.0004728445434011519, "g_entropy": 2.1594515323638914, "tau_bias": 0.751308262348175, "beta": 0.5222650865713755, "train_val_corr": 1.0}
{"t": 0, "ts": 1788195618.6354935, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.5477211238467135e-05, "advantage": 107.24652862548828, "attn_entropy": 1.5088270902633667, "load_balance": 0.0004780653573106974, "g_entropy": 2.1526050567626953, "tau_bias": 0.6509265899658203, "beta": 0.5584118788441023, "train_val_corr": 0.0}
{"t": 1, "ts": 1788195625.285981, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0726861293951515e-07, "advantage": 89.72090148925781, "attn_entropy": 2.03877592086792, "load_balance": 0.0004780834016855806, "g_entropy": 2.156227779388428, "tau_bias": 0.646527886390686, "beta": 0.5712046250700951, "train_val_corr": 0.0}
{"t": 2, "ts": 1788195634.2050755, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 68.68505096435547, "attn_entropy": 1.6358208298683166, "load_balance": 0.0004787362646311522, "g_entropy": 2.1505734443664553, "tau_bias": 0.6498159170150757, "beta": 0.5686033045252165, "train_val_corr": 0.0}
{"t": 3, "ts": 1788195642.932545, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 73.09050750732422, "attn_entropy": 1.5631667137145997, "load_balance": 0.0004749728774186224, "g_entropy": 2.1569216728210447, "tau_bias": 0.6527172923088074, "beta": 0.5429712931315104, "train_val_corr": 0.0}
{"t": 4, "ts": 1788195651.60214, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 72.23716735839844, "attn_entropy": 1.6151904463768005, "load_balance": 0.0004749813349917531, "g_entropy": 2.1571628570556642, "tau_bias": 0.6573835611343384, "beta": 0.5722347026069959, "train_val_corr": 0.0}
{"t": 5, "ts": 1788195660.572962, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 65.53317260742188, "attn_entropy": 1.533086109161377, "load_balance": 0.0004745642829220742, "g_entropy": 2.158458423614502, "tau_bias": 0.6733006238937378, "beta": 0.5313867554068565, "train_val_corr": 0.0}
{"t": 6, "ts": 1788195669.6306646, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 136.11404418945312, "attn_entropy": 1.509199470281601, "load_balance": 0.0004756187787279487, "g_entropy": 2.1563482761383055, "tau_bias": 0.6719450354576111, "beta": 0.559227059284846, "train_val_corr": 0.0}
{"t": 7, "ts": 1788195692.98373, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723268789879512e-07, "advantage": 175.7641143798828, "attn_entropy": 1.5042088806629181, "load_balance": 0.0004756788606755435, "g_entropy": 2.154452419281006, "tau_bias": 0.6803413033485413, "beta": 0.5269864797592163, "train_val_corr": 0.0}
{"t": 8, "ts": 1788195702.321244, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 67.04716491699219, "attn_entropy": 1.5454297959804535, "load_balance": 0.0004784294927958399, "g_entropy": 2.151358127593994, "tau_bias": 0.6817105412483215, "beta": 0.5431402126948038, "train_val_corr": 0.0}
{"t": 9, "ts": 1788195712.1137655, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 52.30317306518555, "attn_entropy": 1.538418424129486, "load_balance": 0.0004805366566870362, "g_entropy": 2.1486629962921144, "tau_bias": 0.6889121532440186, "beta": 0.520272453625997, "train_val_corr": 0.0}
{"t": 10, "ts": 1788195721.9204996, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 29.933290481567383, "attn_entropy": 1.607443356513977, "load_balance": 0.00047473458107560873, "g_entropy": 2.1584739208221437, "tau_bias": 0.7008935213088989, "beta": 0.5334595938523611, "train_val_corr": 0.0}
{"t": 11, "ts": 1788195730.8233905, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 231.17132568359375, "attn_entropy": 1.8630347728729248, "load_balance": 0.00047701409785076975, "g_entropy": 2.155347204208374, "tau_bias": 0.6985456943511963, "beta": 0.5369742636879286, "train_val_corr": 0.0}
{"t": 12, "ts": 1788195739.495562, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.683601379394531, "attn_entropy": 1.7435780048370362, "load_balance": 0.00048062226269394157, "g_entropy": 2.15206093788147, "tau_bias": 0.7104957699775696, "beta": 0.5137855360905329, "train_val_corr": 0.0}
{"t": 13, "ts": 1788195762.6915975, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 16.932825088500977, "attn_entropy": 1.8686827659606933, "load_balance": 0.00047939049545675516, "g_entropy": 2.155273962020874, "tau_bias": 0.7162169814109802, "beta": 0.5268916835387548, "train_val_corr": 0.0}
{"t": 14, "ts": 1788195771.3910127, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 586.9725952148438, "attn_entropy": 1.94769344329834, "load_balance": 0.0004773543507326394, "g_entropy": 2.157172203063965, "tau_bias": 0.716325581073761, "beta": 0.5271659269928932, "train_val_corr": 0.0}
{"t": 15, "ts": 1788195780.8052034, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.58620834350586, "attn_entropy": 1.4907590508461, "load_balance": 0.0004766870813909918, "g_entropy": 2.1543347358703615, "tau_bias": 0.7284765243530273, "beta": 0.49652262901266414, "train_val_corr": 0.0}
{"t": 16, "ts": 1788195790.4678247, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.622438430786133, "attn_entropy": 1.472950428724289, "load_balance": 0.00047396817826665937, "g_entropy": 2.1579267024993896, "tau_bias": 0.751097559928894, "beta": 0.5020412082473437, "train_val_corr": 0.0}
{"t": 17, "ts": 1788195799.5347958, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 32.07689666748047, "attn_entropy": 1.3497906118631362, "load_balance": 0.00047846404486335815, "g_entropy": 2.1504985809326174, "tau_bias": 0.7559785842895508, "beta": 0.5091952458024025, "train_val_corr": 0.0}
{"t": 18, "ts": 1788195808.807476, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.704023838043213, "attn_entropy": 1.3439210563898087, "load_balance": 0.0004733814566861838, "g_entropy": 2.1570386409759523, "tau_bias": 0.7603507041931152, "beta": 0.4953181358675162, "train_val_corr": 0.0}
{"t": 19, "ts": 1788195831.7364333, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.25495719909668, "attn_entropy": 1.3679913491010667, "load_balance": 0.0004765985533595085, "g_entropy": 2.153904390335083, "tau_bias": 0.7670421600341797, "beta": 0.4962015338242054, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788195840.6148117, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.781298637390137, "attn_entropy": 1.402373018860817, "load_balance": 0.00047343826736323535, "g_entropy": 2.158175468444824, "tau_bias": 0.7781072854995728, "beta": 0.5035545602440834, "train_val_corr": 1.0}
{"t": 21, "ts": 1788195850.008701, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.58829689025879, "attn_entropy": 1.3677545428276061, "load_balance": 0.00047486150288023055, "g_entropy": 2.155410146713257, "tau_bias": 0.7776638865470886, "beta": 0.5205506781737009, "train_val_corr": 0.9999999999999998}
{"t": 22, "ts": 1788195859.3010206, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -46.50102996826172, "attn_entropy": 1.398021388053894, "load_balance": 0.00047660571290180086, "g_entropy": 2.154541492462158, "tau_bias": 0.7775037884712219, "beta": 0.5097216665744781, "train_val_corr": 1.0}
{"t": 23, "ts": 1788195868.4878678, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.83917236328125, "attn_entropy": 1.3840453386306764, "load_balance": 0.00047706675832159817, "g_entropy": 2.154738664627075, "tau_bias": 0.7739366888999939, "beta": 0.5244797393679619, "train_val_corr": 1.0}
{"t": 24, "ts": 1788195877.3710284, "step": 240, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.23626708984375, "attn_entropy": 1.3669119477272034, "load_balance": 0.00047682768781669437, "g_entropy": 2.153565549850464, "tau_bias": 0.7761063575744629, "beta": 0.52259744455417, "train_val_corr": 1.0}
{"t": 25, "ts": 1788195900.8827446, "step": 250, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 64.03636169433594, "attn_entropy": 1.3476117163896562, "load_balance": 0.00047982686082832516, "g_entropy": 2.147431993484497, "tau_bias": 0.7789179086685181, "beta": 0.5349597334861755, "train_val_corr": 1.0}
{"t": 26, "ts": 1788195909.308644, "step": 260, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -48.60801315307617, "attn_entropy": 1.4195789337158202, "load_balance": 0.00047398332972079513, "g_entropy": 2.1576677799224853, "tau_bias": 0.7725303769111633, "beta": 0.5320080320040385, "train_val_corr": 0.9999999999999998}
{"t": 27, "ts": 1788195918.795437, "step": 270, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -44.37407684326172, "attn_entropy": 1.4300884038209916, "load_balance": 0.0004736429254990071, "g_entropy": 2.1582190990448, "tau_bias": 0.7672858238220215, "beta": 0.5250358680884043, "train_val_corr": 0.9999999999999998}
{"t": 28, "ts": 1788195927.8879578, "step": 280, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -46.15907669067383, "attn_entropy": 1.4260217040777206, "load_balance": 0.0004739692143630236, "g_entropy": 2.158786582946777, "tau_bias": 0.7617290019989014, "beta": 0.5237450028459231, "train_val_corr": 1.0}
{"t": 29, "ts": 1788195937.1795604, "step": 290, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -43.428741455078125, "attn_entropy": 1.443551704287529, "load_balance": 0.00047591047477908434, "g_entropy": 2.155571222305298, "tau_bias": 0.7596361041069031, "beta": 0.5287181834379832, "train_val_corr": 1.0}
{"t": 30, "ts": 1788195946.6274443, "step": 300, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.606300354003906, "attn_entropy": 1.3982668220996857, "load_balance": 0.0004759089846629649, "g_entropy": 2.154165267944336, "tau_bias": 0.7567456364631653, "beta": 0.5226710935433706, "train_val_corr": 0.9999999999999996}
{"t": 31, "ts": 1788195971.9654298, "step": 310, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -42.667213439941406, "attn_entropy": 1.423149797320366, "load_balance": 0.0004758568364195526, "g_entropy": 2.156546640396118, "tau_bias": 0.7545661926269531, "beta": 0.5212625364462534, "train_val_corr": 1.0}
{"t": 32, "ts": 1788195979.391657, "step": 320, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.703643798828125, "attn_entropy": 1.384346704185009, "load_balance": 0.00047234864323399963, "g_entropy": 2.1592222690582275, "tau_bias": 0.7530725002288818, "beta": 0.5188408320148786, "train_val_corr": 0.9999999999999993}
{"t": 33, "ts": 1788195988.4865885, "step": 330, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.892982482910156, "attn_entropy": 1.3518488556146622, "load_balance": 0.0004728445434011519, "g_entropy": 2.1594515323638914, "tau_bias": 0.751308262348175, "beta": 0.5222650865713755, "train_val_corr": 1.0}
{"t": 34, "ts": 1788195997.6134865, "step": 340, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -39.038726806640625, "attn_entropy": 1.3794395864009856, "load_balance": 0.00047408321406692264, "g_entropy": 2.1573153018951414, "tau_bias": 0.7478220462799072, "beta": 0.5195876260598501, "train_val_corr": 0.9999999999999998}
{"t": 35, "ts": 1788196007.3256934, "step": 350, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -38.11954879760742, "attn_entropy": 1.3628646373748778, "load_balance": 0.00047315804986283185, "g_entropy": 2.158894157409668, "tau_bias": 0.745747983455658, "beta": 0.5238856673240662, "train_val_corr": 0.9999999999999996}
{"t": 0, "ts": 1788196050.9979722, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.8310948917132919e-06, "advantage": 26.778282165527344, "attn_entropy": 1.3821262672543526, "load_balance": 0.000473592261550948, "g_entropy": 2.1582293033599855, "tau_bias": 0.7432117462158203, "beta": 0.5163767511645952, "train_val_corr": 0.0}
{"t": 1, "ts": 1788196057.9228444, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0729385141748935e-07, "advantage": 79.18311309814453, "attn_entropy": 1.4529125839471817, "load_balance": 0.0004717230389360338, "g_entropy": 2.158526802062988, "tau_bias": 0.7493916153907776, "beta": 0.5929005146026611, "train_val_corr": 0.0}
{"t": 2, "ts": 1788196066.9723864, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 37.40401840209961, "attn_entropy": 1.3891335248947143, "load_balance": 0.0004734323418233544, "g_entropy": 2.159298610687256, "tau_bias": 0.7601416110992432, "beta": 0.5849177266160647, "train_val_corr": 0.0}
{"t": 3, "ts": 1788196075.868462, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 34.2718620300293, "attn_entropy": 1.3661392204463483, "load_balance": 0.00047333211405202745, "g_entropy": 2.1592400074005127, "tau_bias": 0.7630330920219421, "beta": 0.5840008780360222, "train_val_corr": 0.0}
{"t": 4, "ts": 1788196084.6977923, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 30.814367294311523, "attn_entropy": 1.420200388133526, "load_balance": 0.0004740404721815139, "g_entropy": 2.1586669921875, "tau_bias": 0.7728768587112427, "beta": 0.581032340725263, "train_val_corr": 0.0}
{"t": 5, "ts": 1788196093.3650813, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 26.188364028930664, "attn_entropy": 1.384943352639675, "load_balance": 0.00047559114173054693, "g_entropy": 2.155769634246826, "tau_bias": 0.774352490901947, "beta": 0.5605333050092062, "train_val_corr": 0.0}
{"t": 6, "ts": 1788196102.4192014, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 17.017276763916016, "attn_entropy": 1.326979923993349, "load_balance": 0.00047350270906463263, "g_entropy": 2.157859373092651, "tau_bias": 0.7738391757011414, "beta": 0.5717272361119589, "train_val_corr": 0.0}
{"t": 7, "ts": 1788196126.217514, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 53.97522735595703, "attn_entropy": 1.319396548718214, "load_balance": 0.0004800056223757565, "g_entropy": 2.1469608306884767, "tau_bias": 0.7717722058296204, "beta": 0.5843812227249146, "train_val_corr": 0.0}
{"t": 8, "ts": 1788196135.00885, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 16.156463623046875, "attn_entropy": 1.3343800406903028, "load_balance": 0.00047191641642712056, "g_entropy": 2.1590033054351805, "tau_bias": 0.7696828246116638, "beta": 0.5639835422237715, "train_val_corr": 0.0}
{"t": 9, "ts": 1788196144.3225193, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.116275787353516, "attn_entropy": 1.3613466918468475, "load_balance": 0.00047267270856536924, "g_entropy": 2.15963454246521, "tau_bias": 0.7657844424247742, "beta": 0.5610587870081266, "train_val_corr": 0.0}
{"t": 10, "ts": 1788196153.8170352, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.727298736572266, "attn_entropy": 1.3605557527393102, "load_balance": 0.00047424097429029646, "g_entropy": 2.158035230636597, "tau_bias": 0.7642542123794556, "beta": 0.5564300417900085, "train_val_corr": 0.0}
{"t": 11, "ts": 1788196163.6004703, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.486845016479492, "attn_entropy": 1.3912285186350346, "load_balance": 0.00047337220748886464, "g_entropy": 2.158825969696045, "tau_bias": 0.7631921172142029, "beta": 0.5533202439546585, "train_val_corr": 0.0}
{"t": 12, "ts": 1788196173.46822, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 7.071537494659424, "attn_entropy": 1.4403756260871887, "load_balance": 0.00047314182738773526, "g_entropy": 2.1588623046875, "tau_bias": 0.7590230703353882, "beta": 0.5460016429424286, "train_val_corr": 0.0}
{"t": 13, "ts": 1788196198.3668458, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.431982040405273, "attn_entropy": 1.3755058992654086, "load_balance": 0.00047440982307307423, "g_entropy": 2.156819152832031, "tau_bias": 0.7566553354263306, "beta": 0.5445439765850703, "train_val_corr": 0.0}
{"t": 14, "ts": 1788196207.020066, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.788814067840576, "attn_entropy": 1.3780044317245483, "load_balance": 0.0004731681779958308, "g_entropy": 2.15868935585022, "tau_bias": 0.7543098330497742, "beta": 0.5456568424900373, "train_val_corr": 0.0}
{"t": 15, "ts": 1788196216.2237718, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 4.93964958190918, "attn_entropy": 1.4137255862355231, "load_balance": 0.0004735040885861963, "g_entropy": 2.158980131149292, "tau_bias": 0.7593077421188354, "beta": 0.5436053921778997, "train_val_corr": 0.0}
{"t": 16, "ts": 1788196225.4936087, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 5.314577102661133, "attn_entropy": 1.3850558873265981, "load_balance": 0.0004728667146991938, "g_entropy": 2.1598334312438965, "tau_bias": 0.7596770524978638, "beta": 0.5380919600526491, "train_val_corr": 0.0}
{"t": 17, "ts": 1788196234.6279578, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.153977632522583, "attn_entropy": 1.3746604472398758, "load_balance": 0.0004725164268165827, "g_entropy": 2.1584869861602782, "tau_bias": 0.7592326402664185, "beta": 0.5383929784099261, "train_val_corr": 0.0}
{"t": 18, "ts": 1788196243.2953262, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.1786632537841797, "attn_entropy": 1.3579531025141478, "load_balance": 0.00047186546144075694, "g_entropy": 2.159819173812866, "tau_bias": 0.7561149597167969, "beta": 0.5432777106761932, "train_val_corr": 0.0}
{"t": 0, "ts": 1788197139.2763722, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 4.104511390323751e-06, "advantage": 2638.746337890625, "attn_entropy": 1.3774088062345982, "load_balance": 0.00047272679512389004, "g_entropy": 2.159939098358154, "tau_bias": 0.7534656524658203, "beta": 0.508636566499869, "train_val_corr": 0.0}
{"t": 1, "ts": 1788197148.209933, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1197254795879417e-07, "advantage": 168.563232421875, "attn_entropy": 1.286908826045692, "load_balance": 0.00047110305749811234, "g_entropy": 2.159309673309326, "tau_bias": 0.7349101305007935, "beta": 0.6316385716199875, "train_val_corr": 0.0}
{"t": 2, "ts": 1788197156.6060455, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723526006349857e-07, "advantage": -48.265869140625, "attn_entropy": 1.2937928345054388, "load_balance": 0.0004726989252958447, "g_entropy": 2.1591532230377197, "tau_bias": 0.7333291172981262, "beta": 0.5296555807193121, "train_val_corr": 0.0}
{"t": 3, "ts": 1788197165.629492, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723273053135927e-07, "advantage": -113.65181732177734, "attn_entropy": 1.2906296174973249, "load_balance": 0.0004713741422165185, "g_entropy": 2.1599358558654784, "tau_bias": 0.7376739978790283, "beta": 0.5251674999793371, "train_val_corr": 0.0}
{"t": 4, "ts": 1788197175.384682, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -10.274995803833008, "attn_entropy": 1.299375089816749, "load_balance": 0.0004703698388766497, "g_entropy": 2.159636449813843, "tau_bias": 0.7329526543617249, "beta": 0.6666464557250341, "train_val_corr": 0.0}
{"t": 5, "ts": 1788197184.8209739, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 19.5880126953125, "attn_entropy": 1.2797943256795405, "load_balance": 0.0004710879293270409, "g_entropy": 2.159078073501587, "tau_bias": 0.729423463344574, "beta": 0.6927658915519714, "train_val_corr": 0.0}
{"t": 6, "ts": 1788197207.8322284, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -59.442604064941406, "attn_entropy": 1.2994714610278606, "load_balance": 0.00047233926015906037, "g_entropy": 2.160013198852539, "tau_bias": 0.7372521758079529, "beta": 0.4977659707268079, "train_val_corr": 0.0}
{"t": 7, "ts": 1788197217.8108628, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -58.911590576171875, "attn_entropy": 1.3129145137965679, "load_balance": 0.0004710166831500828, "g_entropy": 2.1599984645843504, "tau_bias": 0.7431069016456604, "beta": 0.49605684106548625, "train_val_corr": 0.0}
{"t": 8, "ts": 1788197227.766958, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -70.51690673828125, "attn_entropy": 1.3099344935268165, "load_balance": 0.00047195187653414905, "g_entropy": 2.160037088394165, "tau_bias": 0.7556989789009094, "beta": 0.4946834519505501, "train_val_corr": 0.0}
{"t": 9, "ts": 1788197237.437592, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -72.27641296386719, "attn_entropy": 1.2778818037360906, "load_balance": 0.00047107957070693376, "g_entropy": 2.1589313983917235, "tau_bias": 0.7673026919364929, "beta": 0.4999908432364464, "train_val_corr": 0.0}
{"t": 10, "ts": 1788197247.5723834, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -64.34571838378906, "attn_entropy": 1.2691836062818767, "load_balance": 0.00047189241740852594, "g_entropy": 2.159906101226807, "tau_bias": 0.7621312737464905, "beta": 0.5112182050943375, "train_val_corr": 0.0}
{"t": 11, "ts": 1788197257.5148795, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -53.5570068359375, "attn_entropy": 1.27363111525774, "load_balance": 0.00047181888367049397, "g_entropy": 2.16011438369751, "tau_bias": 0.7710855007171631, "beta": 0.5006836255391439, "train_val_corr": 0.0}
{"t": 12, "ts": 1788197283.2004578, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -56.631996154785156, "attn_entropy": 1.265013288334012, "load_balance": 0.00047219346161000433, "g_entropy": 2.1596232891082763, "tau_bias": 0.7813305258750916, "beta": 0.5012209912141165, "train_val_corr": 0.0}
{"t": 13, "ts": 1788197293.3106596, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -45.25475311279297, "attn_entropy": 1.2807740468531847, "load_balance": 0.00047014173469506204, "g_entropy": 2.160054874420166, "tau_bias": 0.7786372303962708, "beta": 0.4824412129819393, "train_val_corr": 0.0}
{"t": 14, "ts": 1788197303.7049925, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -64.34994506835938, "attn_entropy": 1.2885436952114104, "load_balance": 0.0004718240234069526, "g_entropy": 2.160118246078491, "tau_bias": 0.7889055609703064, "beta": 0.46134667346874875, "train_val_corr": 0.0}
{"t": 15, "ts": 1788197313.9118822, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -90.30850219726562, "attn_entropy": 1.282504951953888, "load_balance": 0.0004742981924209744, "g_entropy": 2.15885853767395, "tau_bias": 0.8058553338050842, "beta": 0.49386946732799214, "train_val_corr": 0.0}
{"t": 16, "ts": 1788197324.076989, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -47.65547561645508, "attn_entropy": 1.2961417049169541, "load_balance": 0.0004710727313067764, "g_entropy": 2.160075855255127, "tau_bias": 0.807415783405304, "beta": 0.4930430476864179, "train_val_corr": 0.0}
{"t": 17, "ts": 1788197333.8820608, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -49.775699615478516, "attn_entropy": 1.2809562496840954, "load_balance": 0.0004714580543804914, "g_entropy": 2.1600436210632323, "tau_bias": 0.8109220862388611, "beta": 0.4591047589977582, "train_val_corr": 0.0}
{"t": 0, "ts": 1788197395.1229625, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 9.433430386707187e-05, "advantage": 348.7503356933594, "attn_entropy": 1.282504668086767, "load_balance": 0.000472521799383685, "g_entropy": 2.1598713397979736, "tau_bias": 0.8110828399658203, "beta": 0.47664508720239, "train_val_corr": 0.0}
{"t": 1, "ts": 1788197405.0248806, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0912446530019224e-07, "advantage": 501.1478576660156, "attn_entropy": 1.2293486827984452, "load_balance": 0.0004716860945336521, "g_entropy": 2.159151887893677, "tau_bias": 0.7962626218795776, "beta": 0.46757909034689266, "train_val_corr": 0.0}
{"t": 2, "ts": 1788197414.5280263, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0808522549486952e-07, "advantage": 139.1558380126953, "attn_entropy": 1.2722601694986224, "load_balance": 0.000473541475366801, "g_entropy": 2.1589598178863527, "tau_bias": 0.8001455068588257, "beta": 0.48202675953507423, "train_val_corr": 0.0}
{"t": 3, "ts": 1788197424.4674528, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072596743118993e-07, "advantage": -19.27168083190918, "attn_entropy": 1.2730318823829294, "load_balance": 0.0004714506445452571, "g_entropy": 2.1598426342010497, "tau_bias": 0.8150994777679443, "beta": 0.4965552973250548, "train_val_corr": 0.0}
{"t": 4, "ts": 1788197434.9156785, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0725252625197754e-07, "advantage": -23.673351287841797, "attn_entropy": 1.2509837552905083, "load_balance": 0.0004701354366261512, "g_entropy": 2.1594820976257325, "tau_bias": 0.8161283135414124, "beta": 0.5002806633710861, "train_val_corr": 0.0}
{"t": 5, "ts": 1788197445.1051033, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": 184.87086486816406, "attn_entropy": 1.1672964930534362, "load_balance": 0.0004731380438897759, "g_entropy": 2.1588550567626954, "tau_bias": 0.8044940233230591, "beta": 0.48381384710470837, "train_val_corr": 0.0}
{"t": 6, "ts": 1788197470.7691183, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 56.43259811401367, "attn_entropy": 1.2320567287504673, "load_balance": 0.0004718916432466358, "g_entropy": 2.1599374771118165, "tau_bias": 0.8083890080451965, "beta": 0.44587937369942665, "train_val_corr": 0.0}
{"t": 7, "ts": 1788197480.7119074, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326736879404e-07, "advantage": -35.27780532836914, "attn_entropy": 1.2621728651225568, "load_balance": 0.00047104678233154116, "g_entropy": 2.1599180698394775, "tau_bias": 0.818202018737793, "beta": 0.5037733887632688, "train_val_corr": 0.0}
{"t": 8, "ts": 1788197491.0778825, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 13.65574836730957, "attn_entropy": 1.461172564327717, "load_balance": 0.00047237202525138853, "g_entropy": 2.1599783420562746, "tau_bias": 0.8118674159049988, "beta": 0.524065874516964, "train_val_corr": 0.0}
{"t": 9, "ts": 1788197501.0727947, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 10.443473815917969, "attn_entropy": 1.4361393988132476, "load_balance": 0.00047130275634117426, "g_entropy": 2.158787155151367, "tau_bias": 0.813173234462738, "beta": 0.4634987066189448, "train_val_corr": 0.0}
{"t": 10, "ts": 1788197510.6301453, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.020172119140625, "attn_entropy": 1.4275959998369216, "load_balance": 0.00047644274309277534, "g_entropy": 2.1598252296447753, "tau_bias": 0.8280930519104004, "beta": 0.49747588361303013, "train_val_corr": 0.0}
{"t": 11, "ts": 1788197520.4879825, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.828468322753906, "attn_entropy": 1.3420339465141295, "load_balance": 0.00047805956564843654, "g_entropy": 2.160060262680054, "tau_bias": 0.8369497060775757, "beta": 0.5106357013185819, "train_val_corr": 0.0}
{"t": 12, "ts": 1788197543.8933737, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 69.24685668945312, "attn_entropy": 1.3350422501564025, "load_balance": 0.0004787029465660453, "g_entropy": 2.1595346927642822, "tau_bias": 0.8314118385314941, "beta": 0.5138893450299898, "train_val_corr": 0.0}
{"t": 13, "ts": 1788197553.1168287, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 12.772087097167969, "attn_entropy": 1.279534211754799, "load_balance": 0.0004818175570107996, "g_entropy": 2.1600115299224854, "tau_bias": 0.8257659673690796, "beta": 0.48137378692626953, "train_val_corr": 0.0}
{"t": 14, "ts": 1788197563.1165566, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.358333587646484, "attn_entropy": 1.2763134270906449, "load_balance": 0.00048182035097852347, "g_entropy": 2.160079526901245, "tau_bias": 0.8382874131202698, "beta": 0.49062446132302284, "train_val_corr": 0.0}
{"t": 15, "ts": 1788197572.8024254, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.84510040283203, "attn_entropy": 1.4598352313041687, "load_balance": 0.00048212711699306964, "g_entropy": 2.158698225021362, "tau_bias": 0.8593733906745911, "beta": 0.5032715201377869, "train_val_corr": 0.0}
{"t": 16, "ts": 1788197582.2714584, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 142.227294921875, "attn_entropy": 1.4756388545036316, "load_balance": 0.0004722269542980939, "g_entropy": 2.1600181579589846, "tau_bias": 0.8743664622306824, "beta": 0.5179413283864657, "train_val_corr": 0.0}
{"t": 17, "ts": 1788197591.9294937, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.601607322692871, "attn_entropy": 1.467967301607132, "load_balance": 0.0004730195156298578, "g_entropy": 2.1599740505218508, "tau_bias": 0.8870044350624084, "beta": 0.49235281596581143, "train_val_corr": 0.0}
{"t": 18, "ts": 1788197615.488323, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.69628143310547, "attn_entropy": 1.4379012167453766, "load_balance": 0.0004777257912792265, "g_entropy": 2.1599761486053466, "tau_bias": 0.9026033878326416, "beta": 0.48747305820385617, "train_val_corr": 0.0}
{"t": 19, "ts": 1788197624.471749, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -40.00840759277344, "attn_entropy": 1.229048103074274, "load_balance": 0.0004793355416040868, "g_entropy": 2.1600172996520994, "tau_bias": 0.9093067049980164, "beta": 0.5068641205628713, "train_val_corr": 0.9999999999999998}
{"t": 20, "ts": 1788197633.2174761, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -30.530168533325195, "attn_entropy": 1.2602153879780418, "load_balance": 0.0004792362393345684, "g_entropy": 2.1600495338439942, "tau_bias": 0.9042921662330627, "beta": 0.5110168382525444, "train_val_corr": 1.0}
{"t": 21, "ts": 1788197642.004968, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -39.76776123046875, "attn_entropy": 1.3796361389381218, "load_balance": 0.0004759709816426039, "g_entropy": 2.159761667251587, "tau_bias": 0.9164219498634338, "beta": 0.5020063171784083, "train_val_corr": 1.0}
{"t": 22, "ts": 1788197651.173717, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -41.42706298828125, "attn_entropy": 1.3438738753853614, "load_balance": 0.0004711942397989333, "g_entropy": 2.1600337505340574, "tau_bias": 0.9253600239753723, "beta": 0.507508801917235, "train_val_corr": 0.9999999999999998}
{"t": 23, "ts": 1788197660.600041, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -35.696651458740234, "attn_entropy": 1.298709863424301, "load_balance": 0.00047192347701638935, "g_entropy": 2.1600014686584474, "tau_bias": 0.9238685965538025, "beta": 0.5059515436490377, "train_val_corr": 0.9999999999999999}
{"t": 0, "ts": 1788197809.7153366, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 6.139655306469649e-05, "advantage": 382.0632629394531, "attn_entropy": 1.320936183671165, "load_balance": 0.00047182614798657596, "g_entropy": 2.1597286224365235, "tau_bias": 0.9224109649658203, "beta": 0.5025382762153944, "train_val_corr": 0.0}
{"t": 1, "ts": 1788197818.3771656, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 5.482947358359525e-07, "advantage": 210.12814331054688, "attn_entropy": 1.1976941279834137, "load_balance": 0.00046870543737895785, "g_entropy": 2.1588367462158202, "tau_bias": 0.9077810645103455, "beta": 0.5258843873937925, "train_val_corr": 0.0}
{"t": 2, "ts": 1788197827.3142512, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.1777435676995083e-07, "advantage": 34.08772277832031, "attn_entropy": 1.3199128031730651, "load_balance": 0.0004741265263874084, "g_entropy": 2.1585699558258056, "tau_bias": 0.9190515875816345, "beta": 0.4998909334341685, "train_val_corr": 0.0}
{"t": 3, "ts": 1788197836.3638341, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0800430888812116e-07, "advantage": -9.891522407531738, "attn_entropy": 1.3230350946003455, "load_balance": 0.00047029366251081226, "g_entropy": 2.1597012996673586, "tau_bias": 0.9193277955055237, "beta": 0.5109974220395088, "train_val_corr": 0.0}
{"t": 4, "ts": 1788197845.1232822, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0736095507345453e-07, "advantage": -11.631636619567871, "attn_entropy": 1.3064974721521139, "load_balance": 0.00046915345010347667, "g_entropy": 2.159285306930542, "tau_bias": 0.9122745990753174, "beta": 0.5091996068755785, "train_val_corr": 0.0}
{"t": 5, "ts": 1788197854.4721396, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723391003230063e-07, "advantage": -18.790908813476562, "attn_entropy": 1.2387083060998176, "load_balance": 0.00047476861509494485, "g_entropy": 2.158525085449219, "tau_bias": 0.9057890772819519, "beta": 0.5045262575149536, "train_val_corr": 0.0}
{"t": 6, "ts": 1788197878.123406, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723328475469316e-07, "advantage": -17.69915008544922, "attn_entropy": 1.2524779040870726, "load_balance": 0.0004724011814687401, "g_entropy": 2.159828042984009, "tau_bias": 0.9021627902984619, "beta": 0.5051987394690514, "train_val_corr": 0.0}
{"t": 7, "ts": 1788197887.4088893, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": -13.444483757019043, "attn_entropy": 1.237814858389902, "load_balance": 0.00047166364383883775, "g_entropy": 2.15982346534729, "tau_bias": 0.8982071876525879, "beta": 0.503409797946612, "train_val_corr": 0.0}
{"t": 8, "ts": 1788197896.502768, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -5.480166912078857, "attn_entropy": 1.2436793252825737, "load_balance": 0.000472317315870896, "g_entropy": 2.159885215759277, "tau_bias": 0.8989433646202087, "beta": 0.5082110017538071, "train_val_corr": 0.0}
{"t": 9, "ts": 1788197905.9726346, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 9.329117774963379, "attn_entropy": 1.3028674674800755, "load_balance": 0.0004706720937974751, "g_entropy": 2.1584394454956053, "tau_bias": 0.9108794927597046, "beta": 0.5113037576278051, "train_val_corr": 0.0}
{"t": 10, "ts": 1788197915.486028, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 700.775146484375, "attn_entropy": 1.127944228053093, "load_balance": 0.0004727429070044309, "g_entropy": 2.1597203254699706, "tau_bias": 0.9050332903862, "beta": 0.5219840233524641, "train_val_corr": 0.0}
{"t": 11, "ts": 1788197925.1010432, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 3.542905569076538, "attn_entropy": 1.2671597336491687, "load_balance": 0.0004717979347333312, "g_entropy": 2.1599976062774657, "tau_bias": 0.9069859981536865, "beta": 0.5027426282564799, "train_val_corr": 0.0}
{"t": 12, "ts": 1788197949.4970968, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -25.153034210205078, "attn_entropy": 1.4176754660915534, "load_balance": 0.0004734310496132821, "g_entropy": 2.1593732833862305, "tau_bias": 0.9193263053894043, "beta": 0.5101183777054151, "train_val_corr": 0.0}
{"t": 13, "ts": 1788197958.3703063, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -45.83324432373047, "attn_entropy": 1.3032659264135873, "load_balance": 0.0004769864783156663, "g_entropy": 2.1599140644073485, "tau_bias": 0.9230732321739197, "beta": 0.5313141917188963, "train_val_corr": 0.0}
{"t": 14, "ts": 1788197967.3767195, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -17.828208923339844, "attn_entropy": 1.237376723415673, "load_balance": 0.00047918466734699907, "g_entropy": 2.1600057601928713, "tau_bias": 0.9207709431648254, "beta": 0.5482546041409174, "train_val_corr": 0.0}
{"t": 15, "ts": 1788197976.707345, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.00665283203125, "attn_entropy": 1.1917130008035202, "load_balance": 0.00048255162546411156, "g_entropy": 2.158476686477661, "tau_bias": 0.9237968921661377, "beta": 0.5224831104278564, "train_val_corr": 0.0}
{"t": 16, "ts": 1788197986.5367944, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.088321685791016, "attn_entropy": 1.1984650193946436, "load_balance": 0.0004777951107826084, "g_entropy": 2.1599721908569336, "tau_bias": 0.9236541390419006, "beta": 0.5424510960777601, "train_val_corr": 0.0}
{"t": 17, "ts": 1788197995.9742887, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -37.97018051147461, "attn_entropy": 1.2206587321590632, "load_balance": 0.0004791151091922075, "g_entropy": 2.159926748275757, "tau_bias": 0.9327269792556763, "beta": 0.5295630743106207, "train_val_corr": 0.0}
{"t": 18, "ts": 1788198020.1354089, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.508262634277344, "attn_entropy": 1.2096659291535616, "load_balance": 0.00047765452763997017, "g_entropy": 2.159943962097168, "tau_bias": 0.9351070523262024, "beta": 0.5408517296115557, "train_val_corr": 0.0}
{"t": 19, "ts": 1788198029.8788586, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.27128982543945, "attn_entropy": 1.2069404203721206, "load_balance": 0.0004794704669620842, "g_entropy": 2.1599898815155028, "tau_bias": 0.9399481415748596, "beta": 0.5339716995755831, "train_val_corr": 1.0}
{"t": 20, "ts": 1788198039.6820085, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -34.08359146118164, "attn_entropy": 1.214652762352489, "load_balance": 0.0004797927860636264, "g_entropy": 2.160022258758545, "tau_bias": 0.935590386390686, "beta": 0.5319994861880938, "train_val_corr": 1.0}
{"t": 21, "ts": 1788198049.104224, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -23.37846565246582, "attn_entropy": 1.20963512309321, "load_balance": 0.0004800433409400284, "g_entropy": 2.159733009338379, "tau_bias": 0.9366438388824463, "beta": 0.5405009364088377, "train_val_corr": 1.0}
{"t": 22, "ts": 1788198058.18963, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -1.0339112281799316, "attn_entropy": 1.1920182161033153, "load_balance": 0.00047839703620411456, "g_entropy": 2.160018301010132, "tau_bias": 0.9446942806243896, "beta": 0.4998050257563591, "train_val_corr": 0.9999999999999997}
{"t": 23, "ts": 1788198067.8892927, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -9.385101318359375, "attn_entropy": 1.2043309492757543, "load_balance": 0.00047930332948453724, "g_entropy": 2.1599757194519045, "tau_bias": 0.9521275162696838, "beta": 0.5473087231318156, "train_val_corr": 1.0}
{"t": 0, "ts": 1788198175.9307957, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 1.5015602912171744e-06, "advantage": 667.1997680664062, "attn_entropy": 1.2141851048974133, "load_balance": 0.00047981286770664154, "g_entropy": 2.1596903800964355, "tau_bias": 0.9502429962158203, "beta": 0.5117520391941071, "train_val_corr": 0.0}
{"t": 1, "ts": 1788198185.4640665, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0835521752360364e-07, "advantage": 316.5985412597656, "attn_entropy": 1.1926510766148568, "load_balance": 0.0004716750758234411, "g_entropy": 2.158755922317505, "tau_bias": 0.9329792261123657, "beta": 0.6262685507535934, "train_val_corr": 0.0}
{"t": 2, "ts": 1788198195.3115666, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0724645821701415e-07, "advantage": 142.95358276367188, "attn_entropy": 1.2097139171790332, "load_balance": 0.00047433178406208754, "g_entropy": 2.1585136890411376, "tau_bias": 0.9348617196083069, "beta": 0.5196075340112051, "train_val_corr": 0.0}
{"t": 3, "ts": 1788198204.3778043, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723270210964984e-07, "advantage": -11.034558296203613, "attn_entropy": 1.295223206281662, "load_balance": 0.00047018815530464053, "g_entropy": 2.1596680641174317, "tau_bias": 0.943406343460083, "beta": 0.5217515428860983, "train_val_corr": 0.0}
{"t": 4, "ts": 1788198213.3369927, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": -33.149452209472656, "attn_entropy": 1.2798677483573555, "load_balance": 0.0004698628268670291, "g_entropy": 2.1592201709747316, "tau_bias": 0.9421331882476807, "beta": 0.5195765768488249, "train_val_corr": 0.0}
{"t": 5, "ts": 1788198221.9970703, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -36.06717300415039, "attn_entropy": 1.2683637687281588, "load_balance": 0.0004747908853460103, "g_entropy": 2.158393955230713, "tau_bias": 0.9449976682662964, "beta": 0.5263064205646515, "train_val_corr": 0.0}
{"t": 6, "ts": 1788198244.572468, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.78990173339844, "attn_entropy": 1.2683559085708112, "load_balance": 0.0004723908321466297, "g_entropy": 2.15976996421814, "tau_bias": 0.9510927796363831, "beta": 0.5290146718422571, "train_val_corr": 0.0}
{"t": 7, "ts": 1788198253.1910615, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -32.318538665771484, "attn_entropy": 1.2444544469122776, "load_balance": 0.000471495115198195, "g_entropy": 2.1597509384155273, "tau_bias": 0.9567542672157288, "beta": 0.5236607640981674, "train_val_corr": 0.0}
{"t": 8, "ts": 1788198262.112843, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -26.171289443969727, "attn_entropy": 1.2500602908432483, "load_balance": 0.0004719327494967729, "g_entropy": 2.1598093032836916, "tau_bias": 0.9662249088287354, "beta": 0.524953082203865, "train_val_corr": 0.0}
{"t": 9, "ts": 1788198271.1094677, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -29.76529884338379, "attn_entropy": 1.3043057446696336, "load_balance": 0.00047123198164626955, "g_entropy": 2.1582128047943114, "tau_bias": 0.9703240394592285, "beta": 0.5206253503759702, "train_val_corr": 0.0}
{"t": 10, "ts": 1788198280.1768613, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 6.03659200668335, "attn_entropy": 1.2803823949740927, "load_balance": 0.00047183730057440696, "g_entropy": 2.159608840942383, "tau_bias": 0.9859675168991089, "beta": 0.5058176244298617, "train_val_corr": 0.0}
{"t": 11, "ts": 1788198289.4771404, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -15.712390899658203, "attn_entropy": 1.3696451667990892, "load_balance": 0.000475723430281505, "g_entropy": 2.1599207878112794, "tau_bias": 0.98828125, "beta": 0.50716482847929, "train_val_corr": 0.0}
{"t": 12, "ts": 1788198312.5859768, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.458648681640625, "attn_entropy": 1.20333132775307, "load_balance": 0.000480451580369845, "g_entropy": 2.15921893119812, "tau_bias": 0.9966098666191101, "beta": 0.49578995381792385, "train_val_corr": 0.0}
{"t": 13, "ts": 1788198321.672099, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -28.70319175720215, "attn_entropy": 1.2174488071431941, "load_balance": 0.00047757881693542, "g_entropy": 2.159831428527832, "tau_bias": 1.0006835460662842, "beta": 0.5084178869922956, "train_val_corr": 0.0}
{"t": 14, "ts": 1788198330.9084775, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.80393409729004, "attn_entropy": 1.218487692142018, "load_balance": 0.0004799707792699337, "g_entropy": 2.1599318027496337, "tau_bias": 1.0026906728744507, "beta": 0.49917296320199966, "train_val_corr": 0.0}
{"t": 15, "ts": 1788198340.1016884, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.448543548583984, "attn_entropy": 1.2252524891360714, "load_balance": 0.0004843984905164689, "g_entropy": 2.158202886581421, "tau_bias": 1.0009214878082275, "beta": 0.4948017845551173, "train_val_corr": 0.0}
{"t": 16, "ts": 1788198349.2803648, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -19.56268882751465, "attn_entropy": 1.2411006454222187, "load_balance": 0.0004783094569575042, "g_entropy": 2.1598959445953367, "tau_bias": 1.0009722709655762, "beta": 0.4925207483271758, "train_val_corr": 0.0}
{"t": 17, "ts": 1788198358.301049, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.66915512084961, "attn_entropy": 1.3353799850998858, "load_balance": 0.00047144473064690827, "g_entropy": 2.1598528385162354, "tau_bias": 1.0028868913650513, "beta": 0.491023904333512, "train_val_corr": 0.0}
{"t": 18, "ts": 1788198382.3916614, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -21.523653030395508, "attn_entropy": 1.238932371357024, "load_balance": 0.0004776711342856288, "g_entropy": 2.159879541397095, "tau_bias": 0.9983211755752563, "beta": 0.49366508051753044, "train_val_corr": 0.0}
{"t": 19, "ts": 1788198391.810964, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -22.154077529907227, "attn_entropy": 1.2173676969565403, "load_balance": 0.0004797443572897464, "g_entropy": 2.159933614730835, "tau_bias": 1.001594066619873, "beta": 0.4961161936322848, "train_val_corr": 1.0}
{"t": 20, "ts": 1788198401.074673, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.15511703491211, "attn_entropy": 1.227114629970131, "load_balance": 0.00048011147882789376, "g_entropy": 2.159962558746338, "tau_bias": 1.0036389827728271, "beta": 0.5004244148731232, "train_val_corr": 1.0}
{"t": 21, "ts": 1788198410.101672, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -20.449018478393555, "attn_entropy": 1.2191959383056319, "load_balance": 0.0004805484728422016, "g_entropy": 2.1596421718597414, "tau_bias": 1.002302646636963, "beta": 0.498994667083025, "train_val_corr": 1.0}
{"t": 22, "ts": 1788198419.5702596, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -18.571603775024414, "attn_entropy": 1.2172493937371607, "load_balance": 0.0004786593373864889, "g_entropy": 2.1599689960479735, "tau_bias": 0.9999614357948303, "beta": 0.49951491753260296, "train_val_corr": 1.0}
{"t": 23, "ts": 1788198429.5217004, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -19.055042266845703, "attn_entropy": 1.2422140123456162, "load_balance": 0.000479250744683668, "g_entropy": 2.1599331378936766, "tau_bias": 0.9967882037162781, "beta": 0.5019768923521042, "train_val_corr": 1.0}
{"t": 0, "ts": 1788198502.6241076, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.8524111257866025e-05, "advantage": 472.0203857421875, "attn_entropy": 1.2064275745352595, "load_balance": 0.0004798757785465568, "g_entropy": 2.159629249572754, "tau_bias": 0.9941883087158203, "beta": 0.5011125827829043, "train_val_corr": 0.0}
{"t": 1, "ts": 1788198512.0314453, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.099281601886105e-07, "advantage": 46.955352783203125, "attn_entropy": 1.2853257180973527, "load_balance": 0.0004720813070889562, "g_entropy": 2.1585949897766112, "tau_bias": 0.9852002263069153, "beta": 0.5189825048049291, "train_val_corr": 0.0}
{"t": 2, "ts": 1788198521.3154292, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723319948956487e-07, "advantage": -0.35723236203193665, "attn_entropy": 1.2917612554356164, "load_balance": 0.0004743573081213981, "g_entropy": 2.1583064079284666, "tau_bias": 0.9933493137359619, "beta": 0.4961017829676469, "train_val_corr": 0.0}
{"t": 3, "ts": 1788198530.4920673, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 8.482851028442383, "attn_entropy": 1.2410143853953701, "load_balance": 0.00047924171085469427, "g_entropy": 2.159570026397705, "tau_bias": 1.0117980241775513, "beta": 0.5201501846313477, "train_val_corr": 0.0}
{"t": 4, "ts": 1788198539.4184248, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 0.9998496174812317, "attn_entropy": 1.2565785409760577, "load_balance": 0.00047970666200853885, "g_entropy": 2.159051704406738, "tau_bias": 1.017371416091919, "beta": 0.5070437615116438, "train_val_corr": 0.0}
{"t": 5, "ts": 1788198549.1339674, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -12.224342346191406, "attn_entropy": 1.250080967121395, "load_balance": 0.00048389133880846205, "g_entropy": 2.158150339126587, "tau_bias": 1.0136992931365967, "beta": 0.5088276043534279, "train_val_corr": 0.0}
{"t": 6, "ts": 1788198573.000479, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -8.049548149108887, "attn_entropy": 1.2649979116249006, "load_balance": 0.000480303808581084, "g_entropy": 2.159692907333374, "tau_bias": 1.0105606317520142, "beta": 0.5025285879770914, "train_val_corr": 0.0}
{"t": 7, "ts": 1788198582.1796095, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.74118709564209, "attn_entropy": 1.3370676519205673, "load_balance": 0.00047796472208574416, "g_entropy": 2.1596797943115233, "tau_bias": 1.0109578371047974, "beta": 0.5037266587217649, "train_val_corr": 0.0}
{"t": 8, "ts": 1788198591.210956, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -9.068614959716797, "attn_entropy": 1.427145767384159, "load_balance": 0.0004723069199826568, "g_entropy": 2.159755754470825, "tau_bias": 1.0101426839828491, "beta": 0.49945057928562164, "train_val_corr": 0.0}
{"t": 9, "ts": 1788198600.28029, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -6.34265661239624, "attn_entropy": 1.2588502885679511, "load_balance": 0.00047990961465984585, "g_entropy": 2.158038091659546, "tau_bias": 1.0140609741210938, "beta": 0.5005495647589365, "train_val_corr": 0.0}
{"t": 10, "ts": 1788198609.2202096, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -2.848083257675171, "attn_entropy": 1.2726908208657939, "load_balance": 0.0004795783606823534, "g_entropy": 2.159557819366455, "tau_bias": 1.0184073448181152, "beta": 0.49582837894558907, "train_val_corr": 0.0}
{"t": 11, "ts": 1788198618.3083503, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.927951812744141, "attn_entropy": 1.2541212560513597, "load_balance": 0.00047933232854120434, "g_entropy": 2.1598913192749025, "tau_bias": 1.018343448638916, "beta": 0.49904263267914456, "train_val_corr": 0.0}
{"t": 12, "ts": 1788198641.2129066, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -8.266045570373535, "attn_entropy": 1.2394768239812197, "load_balance": 0.0004801047034561634, "g_entropy": 2.1591718673706053, "tau_bias": 1.0184648036956787, "beta": 0.49576952308416367, "train_val_corr": 0.0}
{"t": 13, "ts": 1788198650.4221454, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -10.486668586730957, "attn_entropy": 1.2603933813551342, "load_balance": 0.0004780830407980829, "g_entropy": 2.1598104000091554, "tau_bias": 1.0192995071411133, "beta": 0.4970939780275027, "train_val_corr": 0.0}
{"t": 14, "ts": 1788198659.6874015, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -8.593414306640625, "attn_entropy": 1.3391882421324168, "load_balance": 0.0004783300741109997, "g_entropy": 2.1599181652069093, "tau_bias": 1.0172338485717773, "beta": 0.49050475532809895, "train_val_corr": 0.0}
{"t": 15, "ts": 1788198669.2064312, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -10.168464660644531, "attn_entropy": 1.2925243856336839, "load_balance": 0.00048309554695151745, "g_entropy": 2.1581499576568604, "tau_bias": 1.0157216787338257, "beta": 0.49415888885656994, "train_val_corr": 0.0}
{"t": 16, "ts": 1788198678.6857321, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -8.530773162841797, "attn_entropy": 1.2982280256982834, "load_balance": 0.000478751229820773, "g_entropy": 2.1598814010620115, "tau_bias": 1.0152839422225952, "beta": 0.4920510413746039, "train_val_corr": 0.0}
{"t": 17, "ts": 1788198687.7067883, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -9.42420482635498, "attn_entropy": 1.4307557584606276, "load_balance": 0.0004728361614979804, "g_entropy": 2.1598411560058595, "tau_bias": 1.0142858028411865, "beta": 0.4914189341167609, "train_val_corr": 0.0}
{"t": 18, "ts": 1788198710.5148063, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.522591590881348, "attn_entropy": 1.3003221513964438, "load_balance": 0.00047833565040491524, "g_entropy": 2.159863901138306, "tau_bias": 1.0136349201202393, "beta": 0.4903130816916625, "train_val_corr": 0.0}
{"t": 19, "ts": 1788198719.8928158, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -9.660292625427246, "attn_entropy": 1.3237641336296964, "load_balance": 0.00047894293093122544, "g_entropy": 2.159921884536743, "tau_bias": 1.0132105350494385, "beta": 0.4912393105526765, "train_val_corr": 1.0}
{"t": 20, "ts": 1788198729.592815, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.7969865798950195, "attn_entropy": 1.351787042802896, "load_balance": 0.0004784981021657586, "g_entropy": 2.1599544525146483, "tau_bias": 1.0131016969680786, "beta": 0.49193400144577026, "train_val_corr": 0.9999999999999999}
{"t": 21, "ts": 1788198738.9340103, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -8.339497566223145, "attn_entropy": 1.2815050603678535, "load_balance": 0.00048011335893534125, "g_entropy": 2.159628677368164, "tau_bias": 1.0121710300445557, "beta": 0.4903409332036972, "train_val_corr": 1.0}
{"t": 22, "ts": 1788198748.331188, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.590477466583252, "attn_entropy": 1.2777497293303814, "load_balance": 0.00047840148909017444, "g_entropy": 2.159960889816284, "tau_bias": 1.0094075202941895, "beta": 0.4915248714387417, "train_val_corr": 1.0}
{"t": 23, "ts": 1788198757.82039, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -7.196627616882324, "attn_entropy": 1.4091506482983331, "load_balance": 0.00047481375513598324, "g_entropy": 2.1599230766296387, "tau_bias": 1.0079047679901123, "beta": 0.4911216087639332, "train_val_corr": 1.0}
{"t": 0, "ts": 1788198835.3312411, "step": 0, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 0.0001578065857756883, "advantage": 238.06581115722656, "attn_entropy": 1.2812212945824144, "load_balance": 0.0004795585933607072, "g_entropy": 2.159612607955933, "tau_bias": 1.0054187774658203, "beta": 0.49227356041471165, "train_val_corr": 0.0}
{"t": 1, "ts": 1788198845.5721164, "step": 10, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0747349083194422e-07, "advantage": 11.394448280334473, "attn_entropy": 1.1614617349430407, "load_balance": 0.00046764467842876913, "g_entropy": 2.1585885524749755, "tau_bias": 0.9846935272216797, "beta": 0.5023916438221931, "train_val_corr": 0.0}
{"t": 2, "ts": 1788198861.9361656, "step": 20, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723436477965151e-07, "advantage": 21.65009117126465, "attn_entropy": 1.0554310800358142, "load_balance": 0.000469625648111105, "g_entropy": 2.158340501785278, "tau_bias": 0.9832911491394043, "beta": 0.48687148342529935, "train_val_corr": 0.0}
{"t": 3, "ts": 1788198872.3064315, "step": 30, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0775397047145816e-07, "advantage": -0.5191453695297241, "attn_entropy": 1.2637953284068844, "load_balance": 0.0004721801495179534, "g_entropy": 2.1595871925354, "tau_bias": 0.9985708594322205, "beta": 0.5077130297819773, "train_val_corr": 0.0}
{"t": 4, "ts": 1788198881.3921335, "step": 40, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 1.1808240413665771, "attn_entropy": 1.2555796181359742, "load_balance": 0.00046900229062885046, "g_entropy": 2.1590995311737062, "tau_bias": 0.9983473420143127, "beta": 0.5015321100751559, "train_val_corr": 0.0}
{"t": 5, "ts": 1788198890.4873474, "step": 50, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723362581520632e-07, "advantage": 0.6988407373428345, "attn_entropy": 1.2692433835803418, "load_balance": 0.0004724720085505396, "g_entropy": 2.1581385135650635, "tau_bias": 1.0170496702194214, "beta": 0.5107071697711945, "train_val_corr": 0.0}
{"t": 6, "ts": 1788198914.195818, "step": 60, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -2.9491517543792725, "attn_entropy": 1.26585330980275, "load_balance": 0.00047131492174230517, "g_entropy": 2.1596842288970945, "tau_bias": 1.0180608034133911, "beta": 0.5071801766753197, "train_val_corr": 0.0}
{"t": 7, "ts": 1788198924.0336106, "step": 70, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.072326594770857e-07, "advantage": -4.599849224090576, "attn_entropy": 1.2584989072605475, "load_balance": 0.0004700274963397533, "g_entropy": 2.15967059135437, "tau_bias": 1.0178122520446777, "beta": 0.5111575722694397, "train_val_corr": 0.0}
{"t": 8, "ts": 1788198933.3378115, "step": 80, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.3443820476531982, "attn_entropy": 1.2405314447208764, "load_balance": 0.0004701800586190075, "g_entropy": 2.1597334861755373, "tau_bias": 1.0299491882324219, "beta": 0.4963357349236806, "train_val_corr": 0.0}
{"t": 9, "ts": 1788198942.4330087, "step": 90, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 2.5947439670562744, "attn_entropy": 1.2700792318518177, "load_balance": 0.0004716322582680732, "g_entropy": 2.1579535961151124, "tau_bias": 1.033281922340393, "beta": 0.5004116867979368, "train_val_corr": 0.0}
{"t": 10, "ts": 1788198951.9344852, "step": 100, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": 0.007869639433920383, "attn_entropy": 1.26193957345945, "load_balance": 0.0004704433784354478, "g_entropy": 2.159536838531494, "tau_bias": 1.0315182209014893, "beta": 0.4971039903660615, "train_val_corr": 0.0}
{"t": 11, "ts": 1788198961.3293893, "step": 110, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.345944881439209, "attn_entropy": 1.2661057473977384, "load_balance": 0.00047081116936169564, "g_entropy": 2.1598763465881348, "tau_bias": 1.032990574836731, "beta": 0.4997911565005779, "train_val_corr": 0.0}
{"t": 12, "ts": 1788198984.5361922, "step": 120, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -1.8789234161376953, "attn_entropy": 1.4193555833696128, "load_balance": 0.0004724132653791457, "g_entropy": 2.1591476440429687, "tau_bias": 1.0297192335128784, "beta": 0.5052420794963837, "train_val_corr": 0.0}
{"t": 13, "ts": 1788198994.1970348, "step": 130, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.3824849128723145, "attn_entropy": 1.2519747258970733, "load_balance": 0.000469852815149352, "g_entropy": 2.1597969055175783, "tau_bias": 1.0304738283157349, "beta": 0.5005366032322248, "train_val_corr": 0.0}
{"t": 14, "ts": 1788199003.717262, "step": 140, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.4904727935791016, "attn_entropy": 1.2373472692295462, "load_balance": 0.0004708811698947102, "g_entropy": 2.1599080562591553, "tau_bias": 1.0273330211639404, "beta": 0.5018150260051092, "train_val_corr": 0.0}
{"t": 15, "ts": 1788199012.983443, "step": 150, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.691222667694092, "attn_entropy": 1.2413183690830765, "load_balance": 0.0004715128568932414, "g_entropy": 2.158109426498413, "tau_bias": 1.026123046875, "beta": 0.49766089270512265, "train_val_corr": 0.0}
{"t": 16, "ts": 1788199022.1125193, "step": 160, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -2.472332239151001, "attn_entropy": 1.2334784986302796, "load_balance": 0.00047169410972855983, "g_entropy": 2.1598715782165527, "tau_bias": 1.0246034860610962, "beta": 0.49793854852517444, "train_val_corr": 0.0}
{"t": 17, "ts": 1788199031.3101025, "step": 170, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -4.149685859680176, "attn_entropy": 1.2204695226475102, "load_balance": 0.00046969433315098287, "g_entropy": 2.1598304271698, "tau_bias": 1.0240657329559326, "beta": 0.49682631840308505, "train_val_corr": 0.0}
{"t": 18, "ts": 1788199055.7306168, "step": 180, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -2.6652047634124756, "attn_entropy": 1.224362754993117, "load_balance": 0.00047133222105912866, "g_entropy": 2.1598549365997313, "tau_bias": 1.021152377128601, "beta": 0.4952455709377925, "train_val_corr": 0.0}
{"t": 19, "ts": 1788199065.219984, "step": 190, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.2217390537261963, "attn_entropy": 1.2340524675175095, "load_balance": 0.000470871408469975, "g_entropy": 2.1599148750305175, "tau_bias": 1.0203927755355835, "beta": 0.49504036953051883, "train_val_corr": 1.0}
{"t": 20, "ts": 1788199075.0315113, "step": 200, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -2.8666880130767822, "attn_entropy": 1.226573610477106, "load_balance": 0.0004705613595433533, "g_entropy": 2.15994930267334, "tau_bias": 1.0182541608810425, "beta": 0.49604077885548276, "train_val_corr": 1.0}
{"t": 21, "ts": 1788199084.3771093, "step": 210, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.9243781566619873, "attn_entropy": 1.2426845552250299, "load_balance": 0.00047061003278940915, "g_entropy": 2.1596209049224853, "tau_bias": 1.0172464847564697, "beta": 0.49382580320040387, "train_val_corr": 0.9999999999999997}
{"t": 22, "ts": 1788199093.6275218, "step": 220, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -3.386768341064453, "attn_entropy": 1.2385574819370788, "load_balance": 0.00047010238631628454, "g_entropy": 2.159954881668091, "tau_bias": 1.015773892402649, "beta": 0.49373666445414227, "train_val_corr": 1.0}
{"t": 23, "ts": 1788199103.391652, "step": 230, "action": "ATTN_HYBRID_SPARSE", "entropy_pi": 2.0723264526623097e-07, "advantage": -2.907555341720581, "attn_entropy": 1.1992076875492441, "load_balance": 0.0004708239750470966, "g_entropy": 2.1599164962768556, "tau_bias": 1.0146026611328125, "beta": 0.49340977643926937, "train_val_corr": 1.0}