{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 212,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 414.0390625,
      "epoch": 0.009512485136741973,
      "grad_norm": 18.50773734319143,
      "kl": 3.324449062347412e-05,
      "learning_rate": 0.0,
      "loss": 0.0,
      "reward": 1.931640625,
      "reward_std": 0.7540743527933955,
      "rewards/accuracy_reward": 0.501953125,
      "rewards/format_reward": 0.720703125,
      "rewards/influence_reward": 0.318359375,
      "rewards/len_reward": 0.390625,
      "step": 1
    },
    {
      "completion_length": 406.5244140625,
      "epoch": 0.04756242568370987,
      "grad_norm": 12.240397137360413,
      "kl": 0.0038472674787044525,
      "learning_rate": 1.818181818181818e-07,
      "loss": 0.0002,
      "reward": 2.06787109375,
      "reward_std": 0.6699417636264116,
      "rewards/accuracy_reward": 0.552734375,
      "rewards/format_reward": 0.73681640625,
      "rewards/influence_reward": 0.35595703125,
      "rewards/len_reward": 0.42236328125,
      "step": 5
    },
    {
      "completion_length": 408.440625,
      "epoch": 0.09512485136741974,
      "grad_norm": 10.899643199929397,
      "kl": 0.051489830017089844,
      "learning_rate": 4.090909090909091e-07,
      "loss": 0.0021,
      "reward": 2.095703125,
      "reward_std": 0.6716910980641841,
      "rewards/accuracy_reward": 0.564453125,
      "rewards/format_reward": 0.756640625,
      "rewards/influence_reward": 0.368359375,
      "rewards/len_reward": 0.40625,
      "step": 10
    },
    {
      "completion_length": 402.093359375,
      "epoch": 0.1426872770511296,
      "grad_norm": 56.657034116967125,
      "kl": 0.9256591796875,
      "learning_rate": 6.363636363636363e-07,
      "loss": 0.037,
      "reward": 2.055859375,
      "reward_std": 0.6135061264038086,
      "rewards/accuracy_reward": 0.537109375,
      "rewards/format_reward": 0.77890625,
      "rewards/influence_reward": 0.355859375,
      "rewards/len_reward": 0.383984375,
      "step": 15
    },
    {
      "completion_length": 390.37109375,
      "epoch": 0.1902497027348395,
      "grad_norm": 4.269559046091982,
      "kl": 2.8193359375,
      "learning_rate": 8.636363636363636e-07,
      "loss": 0.1128,
      "reward": 2.1671875,
      "reward_std": 0.5897074935957789,
      "rewards/accuracy_reward": 0.56015625,
      "rewards/format_reward": 0.815234375,
      "rewards/influence_reward": 0.384765625,
      "rewards/len_reward": 0.40703125,
      "step": 20
    },
    {
      "completion_length": 355.8265625,
      "epoch": 0.23781212841854935,
      "grad_norm": 5.36647948775656,
      "kl": 2.3447265625,
      "learning_rate": 9.99726628670463e-07,
      "loss": 0.0938,
      "reward": 2.289453125,
      "reward_std": 0.5657233998179436,
      "rewards/accuracy_reward": 0.54140625,
      "rewards/format_reward": 0.87578125,
      "rewards/influence_reward": 0.405859375,
      "rewards/len_reward": 0.46640625,
      "step": 25
    },
    {
      "completion_length": 312.60859375,
      "epoch": 0.2853745541022592,
      "grad_norm": 93.06936011551063,
      "kl": 3.04345703125,
      "learning_rate": 9.966546331768192e-07,
      "loss": 0.1218,
      "reward": 2.478125,
      "reward_std": 0.5446455283090472,
      "rewards/accuracy_reward": 0.5625,
      "rewards/format_reward": 0.953515625,
      "rewards/influence_reward": 0.451171875,
      "rewards/len_reward": 0.5109375,
      "step": 30
    },
    {
      "completion_length": 299.95390625,
      "epoch": 0.3329369797859691,
      "grad_norm": 4.81826375681457,
      "kl": 2.7115234375,
      "learning_rate": 9.901899829374047e-07,
      "loss": 0.1085,
      "reward": 2.5625,
      "reward_std": 0.5693813040852547,
      "rewards/accuracy_reward": 0.546875,
      "rewards/format_reward": 0.962109375,
      "rewards/influence_reward": 0.44296875,
      "rewards/len_reward": 0.610546875,
      "step": 35
    },
    {
      "completion_length": 276.19296875,
      "epoch": 0.380499405469679,
      "grad_norm": 2.854100596289783,
      "kl": 2.379248046875,
      "learning_rate": 9.803768380684242e-07,
      "loss": 0.0952,
      "reward": 2.51484375,
      "reward_std": 0.5233275255188345,
      "rewards/accuracy_reward": 0.50859375,
      "rewards/format_reward": 0.9640625,
      "rewards/influence_reward": 0.4140625,
      "rewards/len_reward": 0.628125,
      "step": 40
    },
    {
      "completion_length": 281.1796875,
      "epoch": 0.4280618311533888,
      "grad_norm": 3.461375360266424,
      "kl": 2.206005859375,
      "learning_rate": 9.672822322997304e-07,
      "loss": 0.0882,
      "reward": 2.471875,
      "reward_std": 0.5379180932417512,
      "rewards/accuracy_reward": 0.49296875,
      "rewards/format_reward": 0.944140625,
      "rewards/influence_reward": 0.3921875,
      "rewards/len_reward": 0.642578125,
      "step": 45
    },
    {
      "completion_length": 297.3796875,
      "epoch": 0.4756242568370987,
      "grad_norm": 3.3937919439490454,
      "kl": 2.230615234375,
      "learning_rate": 9.509956150664795e-07,
      "loss": 0.0892,
      "reward": 2.546875,
      "reward_std": 0.5505968105047941,
      "rewards/accuracy_reward": 0.53125,
      "rewards/format_reward": 0.923046875,
      "rewards/influence_reward": 0.416015625,
      "rewards/len_reward": 0.6765625,
      "step": 50
    },
    {
      "completion_length": 300.020703125,
      "epoch": 0.5231866825208086,
      "grad_norm": 4.255030544730029,
      "kl": 2.759912109375,
      "learning_rate": 9.316282404787869e-07,
      "loss": 0.1104,
      "reward": 2.500390625,
      "reward_std": 0.5426989603787661,
      "rewards/accuracy_reward": 0.522265625,
      "rewards/format_reward": 0.933984375,
      "rewards/influence_reward": 0.414453125,
      "rewards/len_reward": 0.6296875,
      "step": 55
    },
    {
      "completion_length": 309.028515625,
      "epoch": 0.5707491082045184,
      "grad_norm": 2.734847983664629,
      "kl": 2.937158203125,
      "learning_rate": 9.093124073433462e-07,
      "loss": 0.1175,
      "reward": 2.381640625,
      "reward_std": 0.5930636901408434,
      "rewards/accuracy_reward": 0.46953125,
      "rewards/format_reward": 0.93359375,
      "rewards/influence_reward": 0.3640625,
      "rewards/len_reward": 0.614453125,
      "step": 60
    },
    {
      "completion_length": 305.84296875,
      "epoch": 0.6183115338882283,
      "grad_norm": 4.019506444773187,
      "kl": 3.4853515625,
      "learning_rate": 8.842005554284295e-07,
      "loss": 0.1394,
      "reward": 2.45859375,
      "reward_std": 0.560142171010375,
      "rewards/accuracy_reward": 0.49609375,
      "rewards/format_reward": 0.93671875,
      "rewards/influence_reward": 0.396484375,
      "rewards/len_reward": 0.629296875,
      "step": 65
    },
    {
      "completion_length": 305.07265625,
      "epoch": 0.6658739595719382,
      "grad_norm": 4.116522469679006,
      "kl": 3.28466796875,
      "learning_rate": 8.564642241456986e-07,
      "loss": 0.1314,
      "reward": 2.435546875,
      "reward_std": 0.5443418994545937,
      "rewards/accuracy_reward": 0.48515625,
      "rewards/format_reward": 0.940625,
      "rewards/influence_reward": 0.383203125,
      "rewards/len_reward": 0.6265625,
      "step": 70
    },
    {
      "completion_length": 298.075390625,
      "epoch": 0.713436385255648,
      "grad_norm": 2.9394867850708772,
      "kl": 3.50244140625,
      "learning_rate": 8.262928807620843e-07,
      "loss": 0.1401,
      "reward": 2.416796875,
      "reward_std": 0.5376573745161295,
      "rewards/accuracy_reward": 0.48515625,
      "rewards/format_reward": 0.950390625,
      "rewards/influence_reward": 0.37890625,
      "rewards/len_reward": 0.60234375,
      "step": 75
    },
    {
      "completion_length": 299.733984375,
      "epoch": 0.760998810939358,
      "grad_norm": 3.301846350005546,
      "kl": 3.5205078125,
      "learning_rate": 7.938926261462365e-07,
      "loss": 0.1408,
      "reward": 2.404296875,
      "reward_std": 0.5362825602293014,
      "rewards/accuracy_reward": 0.4703125,
      "rewards/format_reward": 0.94765625,
      "rewards/influence_reward": 0.3671875,
      "rewards/len_reward": 0.619140625,
      "step": 80
    },
    {
      "completion_length": 308.55546875,
      "epoch": 0.8085612366230678,
      "grad_norm": 4.193561063299626,
      "kl": 3.41328125,
      "learning_rate": 7.594847868906076e-07,
      "loss": 0.1365,
      "reward": 2.408203125,
      "reward_std": 0.535981552861631,
      "rewards/accuracy_reward": 0.46796875,
      "rewards/format_reward": 0.945703125,
      "rewards/influence_reward": 0.365625,
      "rewards/len_reward": 0.62890625,
      "step": 85
    },
    {
      "completion_length": 305.84140625,
      "epoch": 0.8561236623067776,
      "grad_norm": 20.941270697353453,
      "kl": 3.77119140625,
      "learning_rate": 7.233044034264033e-07,
      "loss": 0.1509,
      "reward": 2.408203125,
      "reward_std": 0.5046488767489791,
      "rewards/accuracy_reward": 0.475,
      "rewards/format_reward": 0.957421875,
      "rewards/influence_reward": 0.37890625,
      "rewards/len_reward": 0.596875,
      "step": 90
    },
    {
      "completion_length": 318.782421875,
      "epoch": 0.9036860879904876,
      "grad_norm": 5.799431376928617,
      "kl": 3.83603515625,
      "learning_rate": 6.855986244591103e-07,
      "loss": 0.1534,
      "reward": 2.423046875,
      "reward_std": 0.5394395122304558,
      "rewards/accuracy_reward": 0.503515625,
      "rewards/format_reward": 0.943359375,
      "rewards/influence_reward": 0.3859375,
      "rewards/len_reward": 0.590234375,
      "step": 95
    },
    {
      "completion_length": 318.328515625,
      "epoch": 0.9512485136741974,
      "grad_norm": 304.60647459145224,
      "kl": 4.10732421875,
      "learning_rate": 6.466250186922324e-07,
      "loss": 0.1643,
      "reward": 2.353125,
      "reward_std": 0.5590785862877965,
      "rewards/accuracy_reward": 0.461328125,
      "rewards/format_reward": 0.94296875,
      "rewards/influence_reward": 0.359765625,
      "rewards/len_reward": 0.5890625,
      "step": 100
    },
    {
      "completion_length": 326.525390625,
      "epoch": 0.9988109393579072,
      "grad_norm": 5.2538538650037365,
      "kl": 4.06787109375,
      "learning_rate": 6.066498153718734e-07,
      "loss": 0.1627,
      "reward": 2.346484375,
      "reward_std": 0.5314119735732674,
      "rewards/accuracy_reward": 0.4609375,
      "rewards/format_reward": 0.941796875,
      "rewards/influence_reward": 0.35390625,
      "rewards/len_reward": 0.58984375,
      "step": 105
    },
    {
      "completion_length": 311.8219696969697,
      "epoch": 1.0380499405469679,
      "grad_norm": 4.473570094823953,
      "kl": 3.8953598484848486,
      "learning_rate": 5.659460856710345e-07,
      "loss": 0.1551,
      "reward": 2.414772727272727,
      "reward_std": 0.5575223863124847,
      "rewards/accuracy_reward": 0.5042613636363636,
      "rewards/format_reward": 0.9403409090909091,
      "rewards/influence_reward": 0.4015151515151515,
      "rewards/len_reward": 0.568655303030303,
      "step": 110
    },
    {
      "completion_length": 319.3453125,
      "epoch": 1.0856123662306778,
      "grad_norm": 6.155188589274135,
      "kl": 4.096875,
      "learning_rate": 5.247918773366111e-07,
      "loss": 0.1638,
      "reward": 2.38828125,
      "reward_std": 0.5313196351751686,
      "rewards/accuracy_reward": 0.49140625,
      "rewards/format_reward": 0.947265625,
      "rewards/influence_reward": 0.387109375,
      "rewards/len_reward": 0.5625,
      "step": 115
    },
    {
      "completion_length": 328.0046875,
      "epoch": 1.1331747919143877,
      "grad_norm": 8.779602612585274,
      "kl": 4.08310546875,
      "learning_rate": 4.834683153413459e-07,
      "loss": 0.1633,
      "reward": 2.371484375,
      "reward_std": 0.5432936575263738,
      "rewards/accuracy_reward": 0.48046875,
      "rewards/format_reward": 0.940625,
      "rewards/influence_reward": 0.34453125,
      "rewards/len_reward": 0.605859375,
      "step": 120
    },
    {
      "completion_length": 310.845703125,
      "epoch": 1.1807372175980975,
      "grad_norm": 3.913334859875239,
      "kl": 4.06435546875,
      "learning_rate": 4.4225768151520694e-07,
      "loss": 0.1625,
      "reward": 2.338671875,
      "reward_std": 0.5353632612153888,
      "rewards/accuracy_reward": 0.4703125,
      "rewards/format_reward": 0.94453125,
      "rewards/influence_reward": 0.359375,
      "rewards/len_reward": 0.564453125,
      "step": 125
    },
    {
      "completion_length": 319.834765625,
      "epoch": 1.2282996432818074,
      "grad_norm": 2.821294092421828,
      "kl": 3.9810546875,
      "learning_rate": 4.0144148627425986e-07,
      "loss": 0.1593,
      "reward": 2.343359375,
      "reward_std": 0.5164167508482933,
      "rewards/accuracy_reward": 0.46796875,
      "rewards/format_reward": 0.952734375,
      "rewards/influence_reward": 0.366015625,
      "rewards/len_reward": 0.556640625,
      "step": 130
    },
    {
      "completion_length": 319.56953125,
      "epoch": 1.2758620689655173,
      "grad_norm": 4.533766553167987,
      "kl": 3.80986328125,
      "learning_rate": 3.612985456190778e-07,
      "loss": 0.1524,
      "reward": 2.3515625,
      "reward_std": 0.5141526155173779,
      "rewards/accuracy_reward": 0.46953125,
      "rewards/format_reward": 0.953125,
      "rewards/influence_reward": 0.36171875,
      "rewards/len_reward": 0.5671875,
      "step": 135
    },
    {
      "completion_length": 322.778515625,
      "epoch": 1.323424494649227,
      "grad_norm": 2.6081456311612534,
      "kl": 3.896875,
      "learning_rate": 3.221030765387417e-07,
      "loss": 0.1559,
      "reward": 2.361328125,
      "reward_std": 0.5361660310998559,
      "rewards/accuracy_reward": 0.467578125,
      "rewards/format_reward": 0.951953125,
      "rewards/influence_reward": 0.363671875,
      "rewards/len_reward": 0.578125,
      "step": 140
    },
    {
      "completion_length": 315.098828125,
      "epoch": 1.370986920332937,
      "grad_norm": 2.6157397625637695,
      "kl": 4.04404296875,
      "learning_rate": 2.841228238307536e-07,
      "loss": 0.1618,
      "reward": 2.28828125,
      "reward_std": 0.5258904606103897,
      "rewards/accuracy_reward": 0.44140625,
      "rewards/format_reward": 0.95,
      "rewards/influence_reward": 0.3375,
      "rewards/len_reward": 0.559375,
      "step": 145
    },
    {
      "completion_length": 323.634765625,
      "epoch": 1.418549346016647,
      "grad_norm": 5.237449890990668,
      "kl": 3.9771484375,
      "learning_rate": 2.476172311325783e-07,
      "loss": 0.1591,
      "reward": 2.317578125,
      "reward_std": 0.5456763867288827,
      "rewards/accuracy_reward": 0.4640625,
      "rewards/format_reward": 0.94140625,
      "rewards/influence_reward": 0.340625,
      "rewards/len_reward": 0.571484375,
      "step": 150
    },
    {
      "completion_length": 326.392578125,
      "epoch": 1.4661117717003567,
      "grad_norm": 5.719070909048027,
      "kl": 4.08916015625,
      "learning_rate": 2.128356686585282e-07,
      "loss": 0.1636,
      "reward": 2.3203125,
      "reward_std": 0.5764369459822773,
      "rewards/accuracy_reward": 0.47421875,
      "rewards/format_reward": 0.943359375,
      "rewards/influence_reward": 0.36328125,
      "rewards/len_reward": 0.539453125,
      "step": 155
    },
    {
      "completion_length": 336.890625,
      "epoch": 1.5136741973840666,
      "grad_norm": 4.504584223627489,
      "kl": 4.0138671875,
      "learning_rate": 1.8001572974834168e-07,
      "loss": 0.1606,
      "reward": 2.286328125,
      "reward_std": 0.5237330112606287,
      "rewards/accuracy_reward": 0.44375,
      "rewards/format_reward": 0.94375,
      "rewards/influence_reward": 0.336328125,
      "rewards/len_reward": 0.5625,
      "step": 160
    },
    {
      "completion_length": 329.14140625,
      "epoch": 1.5612366230677766,
      "grad_norm": 3.881191755016786,
      "kl": 3.934765625,
      "learning_rate": 1.493816078637557e-07,
      "loss": 0.1574,
      "reward": 2.305078125,
      "reward_std": 0.5420002337545157,
      "rewards/accuracy_reward": 0.460546875,
      "rewards/format_reward": 0.933984375,
      "rewards/influence_reward": 0.340234375,
      "rewards/len_reward": 0.5703125,
      "step": 165
    },
    {
      "completion_length": 326.40625,
      "epoch": 1.6087990487514863,
      "grad_norm": 2.6950604934737026,
      "kl": 3.96904296875,
      "learning_rate": 1.2114256511983274e-07,
      "loss": 0.1588,
      "reward": 2.348046875,
      "reward_std": 0.550262408144772,
      "rewards/accuracy_reward": 0.46796875,
      "rewards/format_reward": 0.9453125,
      "rewards/influence_reward": 0.358203125,
      "rewards/len_reward": 0.5765625,
      "step": 170
    },
    {
      "completion_length": 337.144140625,
      "epoch": 1.6563614744351962,
      "grad_norm": 134.22915926188,
      "kl": 4.2392578125,
      "learning_rate": 9.549150281252632e-08,
      "loss": 0.1695,
      "reward": 2.27265625,
      "reward_std": 0.5525685049593448,
      "rewards/accuracy_reward": 0.44296875,
      "rewards/format_reward": 0.94375,
      "rewards/influence_reward": 0.341796875,
      "rewards/len_reward": 0.544140625,
      "step": 175
    },
    {
      "completion_length": 319.248828125,
      "epoch": 1.7039239001189062,
      "grad_norm": 2.5174984321586007,
      "kl": 4.06650390625,
      "learning_rate": 7.260364370723043e-08,
      "loss": 0.1627,
      "reward": 2.375,
      "reward_std": 0.5062162198126317,
      "rewards/accuracy_reward": 0.480078125,
      "rewards/format_reward": 0.952734375,
      "rewards/influence_reward": 0.3796875,
      "rewards/len_reward": 0.5625,
      "step": 180
    },
    {
      "completion_length": 330.200390625,
      "epoch": 1.7514863258026159,
      "grad_norm": 5.276571460186029,
      "kl": 4.02490234375,
      "learning_rate": 5.263533508961826e-08,
      "loss": 0.161,
      "reward": 2.317578125,
      "reward_std": 0.5432032825425267,
      "rewards/accuracy_reward": 0.462109375,
      "rewards/format_reward": 0.943359375,
      "rewards/influence_reward": 0.343359375,
      "rewards/len_reward": 0.56875,
      "step": 185
    },
    {
      "completion_length": 329.69375,
      "epoch": 1.7990487514863258,
      "grad_norm": 2.749522272839248,
      "kl": 3.97265625,
      "learning_rate": 3.572298075514652e-08,
      "loss": 0.1589,
      "reward": 2.356640625,
      "reward_std": 0.5385594313964248,
      "rewards/accuracy_reward": 0.478125,
      "rewards/format_reward": 0.937890625,
      "rewards/influence_reward": 0.378515625,
      "rewards/len_reward": 0.562109375,
      "step": 190
    },
    {
      "completion_length": 331.41484375,
      "epoch": 1.8466111771700358,
      "grad_norm": 2.3616237811005587,
      "kl": 3.98359375,
      "learning_rate": 2.1982109232821176e-08,
      "loss": 0.1593,
      "reward": 2.29375,
      "reward_std": 0.57948282007128,
      "rewards/accuracy_reward": 0.4765625,
      "rewards/format_reward": 0.93828125,
      "rewards/influence_reward": 0.350390625,
      "rewards/len_reward": 0.528515625,
      "step": 195
    },
    {
      "completion_length": 328.06484375,
      "epoch": 1.8941736028537455,
      "grad_norm": 2.7175467395601,
      "kl": 4.0919921875,
      "learning_rate": 1.1506584608200364e-08,
      "loss": 0.1637,
      "reward": 2.36875,
      "reward_std": 0.5422051591798663,
      "rewards/accuracy_reward": 0.494921875,
      "rewards/format_reward": 0.9421875,
      "rewards/influence_reward": 0.38125,
      "rewards/len_reward": 0.550390625,
      "step": 200
    },
    {
      "completion_length": 331.551171875,
      "epoch": 1.9417360285374554,
      "grad_norm": 2.1666383384741312,
      "kl": 4.04013671875,
      "learning_rate": 4.367965336512403e-09,
      "loss": 0.1616,
      "reward": 2.382421875,
      "reward_std": 0.567980020493269,
      "rewards/accuracy_reward": 0.5,
      "rewards/format_reward": 0.93984375,
      "rewards/influence_reward": 0.37890625,
      "rewards/len_reward": 0.563671875,
      "step": 205
    },
    {
      "completion_length": 339.6515625,
      "epoch": 1.9892984542211654,
      "grad_norm": 3.7196004598100263,
      "kl": 4.05810546875,
      "learning_rate": 6.150154258476314e-10,
      "loss": 0.1623,
      "reward": 2.334375,
      "reward_std": 0.5729038657620549,
      "rewards/accuracy_reward": 0.47890625,
      "rewards/format_reward": 0.949609375,
      "rewards/influence_reward": 0.355859375,
      "rewards/len_reward": 0.55,
      "step": 210
    }
  ],
  "logging_steps": 5,
  "max_steps": 212,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}