step,train/reward_mean,train/reward_std,train/policy_kl,train/gen_length_mean,train/grad_norm,train/loss,train/learning_rate,train/beta_adaptive,train/clipped_ratio_frac,train/advantage_mean,train/advantage_std,train/turns_mean,train/r1_mean,train/r2_mean,train/r3_mean,train/r4_mean,train/r5_mean,train/episode_seconds,train/wall_seconds
0,0.23750001192092896,0.03750000149011612,6.653014421463013,512.0,0.018163112923502922,0.26613970994949343,4.997482666353287e-06,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,74.80604290962219,74.80604934692383
1,0.25,0.05000000447034836,12.485302925109863,512.0,4619.2939453125,12.8830898986198,4.989935734988098e-06,1.0,1.0,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,68.50065994262695,150.18300366401672
2,0.22499999403953552,0.02499999850988388,8.762360572814941,512.0,1246.8631591796875,9.100992642343044,4.977374404419838e-06,1.0,1.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,69.8765799999237,220.0717589855194
3,0.22499999403953552,0.02499999850988388,6.899797439575195,512.0,48.56766891479492,6.9075967669487,4.959823971496575e-06,1.0,0.2158203125,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,71.13981032371521,291.248247385025
4,0.23750001192092896,0.03750000149011612,6.589184522628784,512.0,37.53068923950195,6.595208287239075,4.937319780454559e-06,1.0,0.2021484375,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,71.23939824104309,362.50222659111023
5,0.25,0.05000000447034836,6.585293292999268,512.0,31.784725189208984,6.589884877204895,4.909907151739634e-06,1.0,0.0859375,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,72.23862338066101,434.75262475013733
6,0.20000000298023224,0.0,6.475884675979614,512.0,4.087952136993408,6.475884675979614,4.8776412907378845e-06,1.0,0.005859375,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,71.53465509414673,506.30291867256165
7,0.22499999403953552,0.02499999850988388,6.413182258605957,512.0,6.724806308746338,6.413636475801468,4.8405871765993435e-06,1.0,0.0029296875,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,71.68083310127258,578.047988653183
8,0.2750000059604645,0.0,6.420485973358154,512.0,2.4417572021484375,6.420485973358154,4.7988194313786275e-06,1.0,0.0009765625,0.0,0.0,1.0,0.0,0.5,0.5,1.0,0.0,70.74446320533752,648.8044729232788
9,0.20000000298023224,0.0,6.486863851547241,512.0,0.664190948009491,6.486863851547241,4.752422169756048e-06,1.0,0.001953125,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,71.19903922080994,720.0359797477722
10,0.26249998807907104,0.012500002980232239,6.711604118347168,512.0,1.039759635925293,6.7115567326545715,4.701488829641845e-06,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,70.9672462940216,791.0280346870422
11,0.2875000238418579,0.012500002980232239,6.536609411239624,512.0,1.9150428771972656,6.536542177200317,4.646121984004666e-06,1.0,0.001953125,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,70.8780345916748,862.1679155826569
12,0.2750000059604645,0.0,6.548018455505371,512.0,0.16634467244148254,6.548018455505371,4.586433134303257e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.5,1.0,0.0,70.91331028938293,933.0946822166443
13,0.25,0.05000000447034836,6.5434770584106445,512.0,1.1411259174346924,6.543636500835419,4.522542485937369e-06,1.0,0.0,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,70.3309063911438,1003.4358978271484
14,0.22499999403953552,0.02499999850988388,6.8197808265686035,512.0,0.1353868991136551,6.819842755794525,4.454578706170075e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,68.48409962654114,1071.9293241500854
15,0.26249998807907104,0.012500002980232239,6.895052909851074,512.0,0.9044080376625061,6.8951151967048645,4.382678665009028e-06,1.0,0.0029296875,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,67.9190776348114,1139.8634886741638
16,0.25,0.05000000447034836,6.418554306030273,512.0,10.747127532958984,6.418869197368622,4.3069871595684795e-06,1.0,0.0087890625,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,67.56499814987183,1207.4488742351532
17,0.20000000298023224,0.0,6.505070924758911,512.0,0.7914896011352539,6.505070924758911,4.227656622467162e-06,1.0,0.0009765625,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.38874745368958,1274.8516142368317
18,0.20000000298023224,0.0,6.743875026702881,512.0,5.15095853805542,6.743875026702881,4.144846814849282e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.68754363059998,1342.5880963802338
19,0.22499999403953552,0.02499999850988388,6.28533935546875,512.0,3.6141180992126465,6.285608857870102,4.058724504646834e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.42636680603027,1410.031690120697
20,0.2875000238418579,0.012500002980232239,6.567498207092285,512.0,0.5878214240074158,6.567563682794571,3.969463130731183e-06,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,67.85896563529968,1477.923950433731
21,0.23750001192092896,0.03750000149011612,6.713055372238159,512.0,2.0649421215057373,6.713186055421829,3.8772424536302565e-06,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,67.80476975440979,1545.7609391212463
22,0.2750000059604645,0.025000005960464478,6.293972015380859,512.0,0.20943091809749603,6.293927609920502,3.782248193514766e-06,1.0,0.0,0.0,0.9999996423721313,1.0,0.0,0.5,0.5,1.0,0.0,68.17192316055298,1614.1473453044891
23,0.22499999403953552,0.02499999850988388,6.5307416915893555,512.0,0.06552104651927948,6.5307411551475525,3.684671656182497e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,68.1181972026825,1682.2901711463928
24,0.22499999403953552,0.02499999850988388,6.623315095901489,512.0,0.22827006876468658,6.623475939035416,3.5847093477938955e-06,1.0,0.0009765625,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.40779685974121,1749.7287549972534
25,0.20000000298023224,0.0,6.430211305618286,512.0,0.25940921902656555,6.430211305618286,3.4825625791348093e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.48609328269958,1817.2659714221954
26,0.2750000059604645,0.025000005960464478,6.523992538452148,512.0,1.25248122215271,6.524106830358505,3.3784370602033572e-06,1.0,0.001953125,0.0,0.9999996423721313,1.0,0.0,0.5,0.5,1.0,0.0,68.56208443641663,1886.095192193985
27,0.25,0.05000000447034836,6.582975625991821,512.0,0.46089062094688416,6.5830914080142975,3.272542485937369e-06,1.0,0.0,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,68.40218544006348,1954.5390722751617
28,0.25,0.05000000447034836,6.466953277587891,512.0,0.2991197109222412,6.466963291168213,3.165092113916688e-06,1.0,0.0,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,67.63907074928284,2022.2083735466003
29,0.2750000059604645,0.025000005960464478,6.640866756439209,512.0,0.15818898379802704,6.640939027070999,3.056302334890786e-06,1.0,0.0,0.0,0.9999996423721313,1.0,0.0,0.5,0.5,1.0,0.0,68.31145071983337,2090.5504035949707
30,0.22499999403953552,0.02499999850988388,6.804069757461548,512.0,0.08326434344053268,6.803984582424164,2.946392236996592e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.89763021469116,2158.4568123817444
31,0.23750001192092896,0.03750000149011612,6.7474236488342285,512.0,0.04884504899382591,6.747422158718109,2.835583164544139e-06,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,68.01096510887146,2226.490327835083
32,0.2875000238418579,0.012500002980232239,6.557185411453247,512.0,0.14712969958782196,6.557137817144394,2.724098272258584e-06,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,68.27161622047424,2294.8085923194885
33,0.2875000238418579,0.012500002980232239,6.623533487319946,512.0,0.27328166365623474,6.623465955257416,2.6121620758762877e-06,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,68.24250388145447,2363.0834341049194
34,0.23750001192092896,0.03750000149011612,6.400196313858032,512.0,0.12724222242832184,6.400171935558319,2.5e-06,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,67.87056040763855,2431.0165536403656
35,0.26249998807907104,0.012500002980232239,6.51344108581543,512.0,0.09748056530952454,6.513444483280182,2.3878379241237136e-06,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,67.37611365318298,2498.4079253673553
36,0.26249998807907104,0.012500002980232239,6.924213171005249,512.0,0.06196940690279007,6.9242377281188965,2.2759017277414165e-06,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,67.29495072364807,2565.7435886859894
37,0.20000000298023224,0.0,6.641408443450928,512.0,0.06769814342260361,6.641408443450928,2.1644168354558623e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.12925672531128,2632.8868074417114
38,0.22499999403953552,0.02499999850988388,6.547806978225708,512.0,0.08400043100118637,6.547786831855774,2.053607763003409e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.3674864768982,2700.2680184841156
39,0.20000000298023224,0.0,6.609237194061279,512.0,0.09566965699195862,6.609237194061279,1.9436976651092143e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.0979676246643,2767.375081539154
40,0.26249998807907104,0.012500002980232239,6.445613145828247,512.0,0.08174767345190048,6.445610165596008,1.8349078860833125e-06,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,68.71477818489075,2836.102038383484
41,0.22499999403953552,0.02499999850988388,6.632544040679932,512.0,0.10063061118125916,6.632496207952499,1.7274575140626318e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.4126353263855,2903.5352952480316
42,0.23149999976158142,0.04350000619888306,6.732983589172363,512.0,0.0416349321603775,6.732965409755707,1.6215629397966432e-06,1.0,0.0,0.0,0.9999997615814209,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,67.19132542610168,2970.7413580417633
43,0.26249998807907104,0.012500002980232239,6.538809061050415,512.0,0.025523146614432335,6.538812667131424,1.5174374208651913e-06,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,67.20357656478882,3037.9879014492035
44,0.26249998807907104,0.012500002980232239,6.773524045944214,512.0,0.01856788992881775,6.773525774478912,1.415290652206105e-06,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,67.48981308937073,3105.510813474655
45,0.20000000298023224,0.0,6.534111738204956,512.0,0.020119013264775276,6.534111738204956,1.3153283438175036e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.11397743225098,3172.6724939346313
46,0.2875000238418579,0.012500002980232239,6.632663726806641,512.0,0.01955726370215416,6.632675409317017,1.217751806485235e-06,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,67.69358539581299,3240.3774976730347
47,0.2750000059604645,0.0,6.409862995147705,512.0,0.026801567524671555,6.409862995147705,1.122757546369744e-06,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.5,1.0,0.0,67.68610858917236,3308.154800891876
48,0.22499999403953552,0.02499999850988388,6.534432649612427,512.0,0.02368723601102829,6.534442186355591,1.0305368692688175e-06,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.47260189056396,3375.671011686325
49,0.23750001192092896,0.03750000149011612,6.547046422958374,512.0,0.01844497211277485,6.547038018703461,9.412754953531664e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,67.45416712760925,3443.137571811676
50,0.20000000298023224,0.0,6.62959098815918,512.0,0.036626186221838,6.62959098815918,8.551531851507186e-07,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,66.7028124332428,3513.0670499801636
51,0.23750001192092896,0.03750000149011612,6.440027475357056,512.0,0.03692486137151718,6.440030634403229,7.723433775328385e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,66.83127331733704,3580.1264748573303
52,0.23750001192092896,0.03750000149011612,6.648930072784424,512.0,0.03147204965353012,6.64895361661911,6.930128404315214e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,67.42068934440613,3647.5602486133575
53,0.22499999403953552,0.02499999850988388,6.440209150314331,512.0,0.025097964331507683,6.44022274017334,6.17321334990973e-07,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,67.45666480064392,3715.028964281082
54,0.2875000238418579,0.012500002980232239,6.854901313781738,512.0,0.020589470863342285,6.85493266582489,5.454212938299256e-07,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,67.54383993148804,3782.5855252742767
55,0.20000000298023224,0.0,6.5117833614349365,512.0,0.0219567883759737,6.5117833614349365,4.774575140626317e-07,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,67.64683890342712,3850.260645389557
56,0.23750001192092896,0.03750000149011612,6.539593458175659,512.0,0.014322347939014435,6.539586782455444,4.1356686569674344e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,67.68236684799194,3917.9918751716614
57,0.23750001192092896,0.03750000149011612,6.527055740356445,512.0,0.017170514911413193,6.52706852555275,3.538780159953348e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,68.62618064880371,3986.664388656616
58,0.20000000298023224,0.0,6.508911609649658,512.0,0.02966112270951271,6.508911609649658,2.98511170358155e-07,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,70.75544619560242,4057.4292657375336
59,0.23750001192092896,0.03750000149011612,6.586055755615234,512.0,0.014161341823637486,6.586097449064255,2.4757783024395244e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,69.78725004196167,4127.236268281937
60,0.2750000059604645,0.0,6.768088102340698,512.0,0.016272814944386482,6.768088102340698,2.0118056862137358e-07,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.5,1.0,0.0,70.25813555717468,4197.508693695068
61,0.23750001192092896,0.03750000149011612,6.573902130126953,512.0,0.03289478272199631,6.573881685733795,1.59412823400657e-07,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,68.99141192436218,4266.518006324768
62,0.22499999403953552,0.02499999850988388,6.334805011749268,512.0,0.013292730785906315,6.334813416004181,1.223587092621162e-07,1.0,0.0,2.980232238769531e-07,0.9999996423721313,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,68.75240087509155,4335.3025279045105
63,0.2875000238418579,0.012500002980232239,6.530715465545654,512.0,0.013113042339682579,6.53072914481163,9.00928482603669e-08,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,68.75915765762329,4404.095731258392
64,0.25,0.05000000447034836,6.7753520011901855,512.0,0.013271667994558811,6.775352835655212,6.268021954544095e-08,1.0,0.0,1.4901161193847656e-07,0.9999997615814209,1.0,0.0,0.5,0.3333333333333333,1.0,0.0,69.44468283653259,4473.551738023758
65,0.20000000298023224,0.04999999701976776,6.370165586471558,512.0,0.013597301207482815,6.370205938816071,4.017602850342584e-08,1.0,0.0,0.0,0.9999997615814209,1.0,0.0,0.5,0.16666666666666666,1.0,0.0,69.26078653335571,4542.876961946487
66,0.20000000298023224,0.0,6.605443239212036,512.0,0.012532848864793777,6.605443239212036,2.262559558016325e-08,1.0,0.0,0.0,0.0,1.0,0.0,0.5,0.0,1.0,0.0,69.03273606300354,4611.958700418472
67,0.26249998807907104,0.012500002980232239,6.8500494956970215,512.0,0.012855819426476955,6.850044518709183,1.006426501190233e-08,1.0,0.0,1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.41666666666666663,1.0,0.0,68.72005248069763,4680.709627389908
68,0.23750001192092896,0.03750000149011612,6.721463203430176,512.0,0.012744296342134476,6.721463710069656,2.5173336467135266e-09,1.0,0.0,-1.7881393432617188e-07,0.9999997019767761,1.0,0.0,0.5,0.25,1.0,0.0,68.38904666900635,4749.138332128525
69,0.2875000238418579,0.012500002980232239,6.368884086608887,512.0,0.015068421140313148,6.368874192237854,0.0,1.0,0.0,-1.1920928955078125e-06,0.9999991655349731,1.0,0.0,0.5,0.5833333333333333,1.0,0.0,68.82629942893982,4817.991739988327