{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9850746268656716,
  "eval_steps": 100,
  "global_step": 132,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 468.4821586608887,
      "epoch": 0.014925373134328358,
      "grad_norm": 0.5185047388076782,
      "learning_rate": 7.142857142857142e-08,
      "loss": 0.0022,
      "num_tokens": 546936.0,
      "reward": 0.6997768208384514,
      "reward_std": 1.035411298274994,
      "rewards/format_reward": 0.06473214365541935,
      "rewards/weighted_accuracy_reward": 0.6350446492433548,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 471.2355079650879,
      "epoch": 0.029850746268656716,
      "grad_norm": 0.36792412400245667,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0034,
      "num_tokens": 1100635.0,
      "reward": 0.7187500186264515,
      "reward_std": 0.9904668107628822,
      "rewards/format_reward": 0.06584821548312902,
      "rewards/weighted_accuracy_reward": 0.6529017873108387,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 496.9464569091797,
      "epoch": 0.04477611940298507,
      "grad_norm": 0.5098904967308044,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0123,
      "num_tokens": 1692803.0,
      "reward": 0.729910746216774,
      "reward_std": 1.0777875557541847,
      "rewards/format_reward": 0.06361607217695564,
      "rewards/weighted_accuracy_reward": 0.6662946492433548,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.9453315734863,
      "epoch": 0.05970149253731343,
      "grad_norm": 0.31038016080856323,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0166,
      "num_tokens": 2293562.0,
      "reward": 0.7466518096625805,
      "reward_std": 1.0121628865599632,
      "rewards/format_reward": 0.06250000058207661,
      "rewards/weighted_accuracy_reward": 0.684151791036129,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 496.4643096923828,
      "epoch": 0.07462686567164178,
      "grad_norm": 0.45216551423072815,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0088,
      "num_tokens": 2875898.0,
      "reward": 0.686383955180645,
      "reward_std": 0.912985235452652,
      "rewards/format_reward": 0.08035714249126613,
      "rewards/weighted_accuracy_reward": 0.6060267798602581,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 464.7991256713867,
      "epoch": 0.08955223880597014,
      "grad_norm": 0.5948471426963806,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0162,
      "num_tokens": 3416846.0,
      "reward": 0.6841518133878708,
      "reward_std": 0.9436327293515205,
      "rewards/format_reward": 0.08258928591385484,
      "rewards/weighted_accuracy_reward": 0.6015624962747097,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 472.91185760498047,
      "epoch": 0.1044776119402985,
      "grad_norm": 0.43895602226257324,
      "learning_rate": 5e-07,
      "loss": 0.006,
      "num_tokens": 3969623.0,
      "reward": 0.761160746216774,
      "reward_std": 0.9334519654512405,
      "rewards/format_reward": 0.07812500139698386,
      "rewards/weighted_accuracy_reward": 0.6830357126891613,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 478.17636489868164,
      "epoch": 0.11940298507462686,
      "grad_norm": 0.7910401225090027,
      "learning_rate": 5.714285714285714e-07,
      "loss": 0.036,
      "num_tokens": 4515717.0,
      "reward": 0.7388393171131611,
      "reward_std": 0.9763963147997856,
      "rewards/format_reward": 0.09598214272409678,
      "rewards/weighted_accuracy_reward": 0.6428571492433548,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 473.8560562133789,
      "epoch": 0.13432835820895522,
      "grad_norm": 0.8340532183647156,
      "learning_rate": 6.428571428571429e-07,
      "loss": -0.0025,
      "num_tokens": 5065628.0,
      "reward": 0.8392857536673546,
      "reward_std": 1.0670067071914673,
      "rewards/format_reward": 0.11495535541325808,
      "rewards/weighted_accuracy_reward": 0.7243303544819355,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 467.9285888671875,
      "epoch": 0.14925373134328357,
      "grad_norm": 2.4788100719451904,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.012,
      "num_tokens": 5609532.0,
      "reward": 1.0368304178118706,
      "reward_std": 1.2229549214243889,
      "rewards/format_reward": 0.18526785913854837,
      "rewards/weighted_accuracy_reward": 0.8515624925494194,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 520.4989051818848,
      "epoch": 0.16417910447761194,
      "grad_norm": 0.4840046465396881,
      "learning_rate": 7.857142857142856e-07,
      "loss": -0.0209,
      "num_tokens": 6205955.0,
      "reward": 0.9263393208384514,
      "reward_std": 1.1071469336748123,
      "rewards/format_reward": 0.15736606996506453,
      "rewards/weighted_accuracy_reward": 0.7689732015132904,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 499.55024337768555,
      "epoch": 0.1791044776119403,
      "grad_norm": 2.547011375427246,
      "learning_rate": 8.57142857142857e-07,
      "loss": -0.0148,
      "num_tokens": 6783632.0,
      "reward": 0.9430803880095482,
      "reward_std": 1.2095565646886826,
      "rewards/format_reward": 0.2053571417927742,
      "rewards/weighted_accuracy_reward": 0.7377232164144516,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 505.3348388671875,
      "epoch": 0.19402985074626866,
      "grad_norm": 0.4405680000782013,
      "learning_rate": 9.285714285714285e-07,
      "loss": 0.0163,
      "num_tokens": 7359596.0,
      "reward": 1.1116071790456772,
      "reward_std": 1.0722224935889244,
      "rewards/format_reward": 0.3928571492433548,
      "rewards/weighted_accuracy_reward": 0.7187499925494194,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 432.6082763671875,
      "epoch": 0.208955223880597,
      "grad_norm": 0.47592970728874207,
      "learning_rate": 1e-06,
      "loss": -0.0072,
      "num_tokens": 7871085.0,
      "reward": 1.4955358058214188,
      "reward_std": 1.3228022754192352,
      "rewards/format_reward": 0.5412946492433548,
      "rewards/weighted_accuracy_reward": 0.954241082072258,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 444.87836837768555,
      "epoch": 0.22388059701492538,
      "grad_norm": 0.39034631848335266,
      "learning_rate": 9.998286624877785e-07,
      "loss": -0.0004,
      "num_tokens": 8384200.0,
      "reward": 1.4654018431901932,
      "reward_std": 1.2111344560980797,
      "rewards/format_reward": 0.6350446417927742,
      "rewards/weighted_accuracy_reward": 0.8303571492433548,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.76342391967773,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.7176985144615173,
      "learning_rate": 9.99314767377287e-07,
      "loss": 0.0238,
      "num_tokens": 8962524.0,
      "reward": 1.6573661416769028,
      "reward_std": 1.280972957611084,
      "rewards/format_reward": 0.651785708963871,
      "rewards/weighted_accuracy_reward": 1.005580335855484,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 465.8404235839844,
      "epoch": 0.2537313432835821,
      "grad_norm": 0.35317060351371765,
      "learning_rate": 9.98458666866564e-07,
      "loss": -0.0088,
      "num_tokens": 9529653.0,
      "reward": 1.9006697088479996,
      "reward_std": 1.2429953515529633,
      "rewards/format_reward": 0.7834821417927742,
      "rewards/weighted_accuracy_reward": 1.1171875,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.01453399658203,
      "epoch": 0.26865671641791045,
      "grad_norm": 0.40738293528556824,
      "learning_rate": 9.972609476841365e-07,
      "loss": 0.0128,
      "num_tokens": 10093242.0,
      "reward": 2.162946492433548,
      "reward_std": 1.393567904829979,
      "rewards/format_reward": 0.854910708963871,
      "rewards/weighted_accuracy_reward": 1.308035708963871,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 531.3359565734863,
      "epoch": 0.2835820895522388,
      "grad_norm": 0.7938212156295776,
      "learning_rate": 9.957224306869053e-07,
      "loss": 0.0227,
      "num_tokens": 10694111.0,
      "reward": 2.3549108654260635,
      "reward_std": 1.3322175592184067,
      "rewards/format_reward": 0.8526785671710968,
      "rewards/weighted_accuracy_reward": 1.5022321492433548,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.4765853881836,
      "epoch": 0.29850746268656714,
      "grad_norm": 0.4721792936325073,
      "learning_rate": 9.938441702975689e-07,
      "loss": 0.0115,
      "num_tokens": 11276090.0,
      "reward": 2.7901786863803864,
      "reward_std": 1.448724240064621,
      "rewards/format_reward": 0.8883928656578064,
      "rewards/weighted_accuracy_reward": 1.9017857164144516,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.2198944091797,
      "epoch": 0.31343283582089554,
      "grad_norm": 0.31892749667167664,
      "learning_rate": 9.916274537819773e-07,
      "loss": 0.0322,
      "num_tokens": 11915063.0,
      "reward": 2.7890626192092896,
      "reward_std": 1.2587374225258827,
      "rewards/format_reward": 0.902901791036129,
      "rewards/weighted_accuracy_reward": 1.8861607313156128,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 516.7757034301758,
      "epoch": 0.3283582089552239,
      "grad_norm": 0.2883875072002411,
      "learning_rate": 9.890738003669027e-07,
      "loss": 0.0306,
      "num_tokens": 12507558.0,
      "reward": 2.7868305146694183,
      "reward_std": 1.135903388261795,
      "rewards/format_reward": 0.9162946417927742,
      "rewards/weighted_accuracy_reward": 1.8705357015132904,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 502.5189971923828,
      "epoch": 0.34328358208955223,
      "grad_norm": 0.4001145660877228,
      "learning_rate": 9.861849601988383e-07,
      "loss": 0.0152,
      "num_tokens": 13088391.0,
      "reward": 2.8191965520381927,
      "reward_std": 0.8589950203895569,
      "rewards/format_reward": 0.8984375,
      "rewards/weighted_accuracy_reward": 1.920758917927742,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.0346221923828,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.20552444458007812,
      "learning_rate": 9.82962913144534e-07,
      "loss": 0.0516,
      "num_tokens": 13746910.0,
      "reward": 3.1841519474983215,
      "reward_std": 1.0728191211819649,
      "rewards/format_reward": 0.926339291036129,
      "rewards/weighted_accuracy_reward": 2.2578124701976776,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 554.3248062133789,
      "epoch": 0.373134328358209,
      "grad_norm": 0.1809108853340149,
      "learning_rate": 9.794098674340966e-07,
      "loss": 0.0144,
      "num_tokens": 14358745.0,
      "reward": 3.2433037161827087,
      "reward_std": 0.7628813087940216,
      "rewards/format_reward": 0.9475446492433548,
      "rewards/weighted_accuracy_reward": 2.295758932828903,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.6506881713867,
      "epoch": 0.3880597014925373,
      "grad_norm": 0.20298637449741364,
      "learning_rate": 9.755282581475767e-07,
      "loss": 0.0378,
      "num_tokens": 15031840.0,
      "reward": 2.978794813156128,
      "reward_std": 0.7949686832726002,
      "rewards/format_reward": 0.9508928582072258,
      "rewards/weighted_accuracy_reward": 2.0279017984867096,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.694221496582,
      "epoch": 0.40298507462686567,
      "grad_norm": 0.16087092459201813,
      "learning_rate": 9.713207455460892e-07,
      "loss": 0.024,
      "num_tokens": 15709294.0,
      "reward": 3.1841519474983215,
      "reward_std": 0.6896311081945896,
      "rewards/format_reward": 0.9520089253783226,
      "rewards/weighted_accuracy_reward": 2.232142850756645,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.9274749755859,
      "epoch": 0.417910447761194,
      "grad_norm": 0.1395013928413391,
      "learning_rate": 9.667902132486008e-07,
      "loss": 0.0218,
      "num_tokens": 16350093.0,
      "reward": 3.0814733505249023,
      "reward_std": 0.7762559447437525,
      "rewards/format_reward": 0.9654017835855484,
      "rewards/weighted_accuracy_reward": 2.116071432828903,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 667.521240234375,
      "epoch": 0.43283582089552236,
      "grad_norm": 0.1379304826259613,
      "learning_rate": 9.619397662556433e-07,
      "loss": 0.0232,
      "num_tokens": 17087432.0,
      "reward": 3.075892984867096,
      "reward_std": 0.8487133756279945,
      "rewards/format_reward": 0.9698660597205162,
      "rewards/weighted_accuracy_reward": 2.1060267984867096,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.9464569091797,
      "epoch": 0.44776119402985076,
      "grad_norm": 0.2469964325428009,
      "learning_rate": 9.567727288213004e-07,
      "loss": 0.0577,
      "num_tokens": 17778168.0,
      "reward": 3.4877233803272247,
      "reward_std": 0.8301229365170002,
      "rewards/format_reward": 0.9698660671710968,
      "rewards/weighted_accuracy_reward": 2.517857149243355,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 635.0401992797852,
      "epoch": 0.4626865671641791,
      "grad_norm": 0.1288292109966278,
      "learning_rate": 9.512926421749303e-07,
      "loss": 0.0367,
      "num_tokens": 18470956.0,
      "reward": 3.154017984867096,
      "reward_std": 0.7471067383885384,
      "rewards/format_reward": 0.9787946417927742,
      "rewards/weighted_accuracy_reward": 2.175223186612129,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.2076110839844,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.11875630170106888,
      "learning_rate": 9.455032620941839e-07,
      "loss": 0.0037,
      "num_tokens": 19133894.0,
      "reward": 3.3839287161827087,
      "reward_std": 0.713517427444458,
      "rewards/format_reward": 0.9765625,
      "rewards/weighted_accuracy_reward": 2.407366096973419,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 672.7835083007812,
      "epoch": 0.4925373134328358,
      "grad_norm": 0.10552665591239929,
      "learning_rate": 9.394085563309826e-07,
      "loss": 0.0331,
      "num_tokens": 19862220.0,
      "reward": 3.217634081840515,
      "reward_std": 0.6950479410588741,
      "rewards/format_reward": 0.9732142835855484,
      "rewards/weighted_accuracy_reward": 2.2444196343421936,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 680.5982360839844,
      "epoch": 0.5074626865671642,
      "grad_norm": 0.1088176816701889,
      "learning_rate": 9.330127018922193e-07,
      "loss": 0.0405,
      "num_tokens": 20602468.0,
      "reward": 3.51450914144516,
      "reward_std": 0.7129604853689671,
      "rewards/format_reward": 0.9866071417927742,
      "rewards/weighted_accuracy_reward": 2.527901813387871,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 654.3750381469727,
      "epoch": 0.5223880597014925,
      "grad_norm": 0.10683911293745041,
      "learning_rate": 9.26320082177046e-07,
      "loss": 0.0316,
      "num_tokens": 21328780.0,
      "reward": 3.409598410129547,
      "reward_std": 0.7418624050915241,
      "rewards/format_reward": 0.9821428507566452,
      "rewards/weighted_accuracy_reward": 2.427455320954323,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 676.232177734375,
      "epoch": 0.5373134328358209,
      "grad_norm": 0.1270638108253479,
      "learning_rate": 9.19335283972712e-07,
      "loss": 0.03,
      "num_tokens": 22078220.0,
      "reward": 3.2477680444717407,
      "reward_std": 0.7615501545369625,
      "rewards/format_reward": 0.9720982164144516,
      "rewards/weighted_accuracy_reward": 2.2756696343421936,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.6629676818848,
      "epoch": 0.5522388059701493,
      "grad_norm": 0.09896102547645569,
      "learning_rate": 9.120630943110077e-07,
      "loss": 0.015,
      "num_tokens": 22778782.0,
      "reward": 3.405134081840515,
      "reward_std": 0.5572724603116512,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.4151785522699356,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.0591697692871,
      "epoch": 0.5671641791044776,
      "grad_norm": 0.11259270459413528,
      "learning_rate": 9.045084971874737e-07,
      "loss": 0.0361,
      "num_tokens": 23515979.0,
      "reward": 3.4151787161827087,
      "reward_std": 0.622731413692236,
      "rewards/format_reward": 0.986607126891613,
      "rewards/weighted_accuracy_reward": 2.4285713881254196,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.693115234375,
      "epoch": 0.582089552238806,
      "grad_norm": 0.11529455333948135,
      "learning_rate": 8.966766701456176e-07,
      "loss": 0.0425,
      "num_tokens": 24222080.0,
      "reward": 3.04241082072258,
      "reward_std": 0.6464412584900856,
      "rewards/format_reward": 0.9821428507566452,
      "rewards/weighted_accuracy_reward": 2.060267835855484,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 732.3973541259766,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.10044367611408234,
      "learning_rate": 8.885729807284854e-07,
      "loss": 0.0353,
      "num_tokens": 25002004.0,
      "reward": 3.623884081840515,
      "reward_std": 0.7944426313042641,
      "rewards/format_reward": 0.9799107164144516,
      "rewards/weighted_accuracy_reward": 2.643973231315613,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.5212326049805,
      "epoch": 0.6119402985074627,
      "grad_norm": 0.15325558185577393,
      "learning_rate": 8.802029828000155e-07,
      "loss": 0.0183,
      "num_tokens": 25733543.0,
      "reward": 3.142857313156128,
      "reward_std": 0.6576039083302021,
      "rewards/format_reward": 0.9832589253783226,
      "rewards/weighted_accuracy_reward": 2.159598186612129,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.9263687133789,
      "epoch": 0.6268656716417911,
      "grad_norm": 0.19637054204940796,
      "learning_rate": 8.71572412738697e-07,
      "loss": 0.0254,
      "num_tokens": 26416853.0,
      "reward": 3.6138394474983215,
      "reward_std": 0.6069203764200211,
      "rewards/format_reward": 0.983258917927742,
      "rewards/weighted_accuracy_reward": 2.630580395460129,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.0602874755859,
      "epoch": 0.6417910447761194,
      "grad_norm": 0.10569639503955841,
      "learning_rate": 8.626871855061437e-07,
      "loss": 0.0086,
      "num_tokens": 27127755.0,
      "reward": 3.3671876788139343,
      "reward_std": 0.6781974174082279,
      "rewards/format_reward": 0.9955357015132904,
      "rewards/weighted_accuracy_reward": 2.371651768684387,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.6172256469727,
      "epoch": 0.6567164179104478,
      "grad_norm": 0.1140342652797699,
      "learning_rate": 8.535533905932737e-07,
      "loss": 0.0125,
      "num_tokens": 27800884.0,
      "reward": 3.2142858803272247,
      "reward_std": 0.6446169354021549,
      "rewards/format_reward": 0.9843749850988388,
      "rewards/weighted_accuracy_reward": 2.229910731315613,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 629.1082992553711,
      "epoch": 0.6716417910447762,
      "grad_norm": 0.08552103489637375,
      "learning_rate": 8.441772878468769e-07,
      "loss": 0.0109,
      "num_tokens": 28490773.0,
      "reward": 3.627232313156128,
      "reward_std": 0.5923671815544367,
      "rewards/format_reward": 0.9933035597205162,
      "rewards/weighted_accuracy_reward": 2.633928567171097,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 724.1573944091797,
      "epoch": 0.6865671641791045,
      "grad_norm": 0.1819816380739212,
      "learning_rate": 8.34565303179429e-07,
      "loss": 0.0455,
      "num_tokens": 29273410.0,
      "reward": 3.3370536863803864,
      "reward_std": 0.8330264016985893,
      "rewards/format_reward": 0.9743303582072258,
      "rewards/weighted_accuracy_reward": 2.362723186612129,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.4085083007812,
      "epoch": 0.7014925373134329,
      "grad_norm": 0.14992384612560272,
      "learning_rate": 8.247240241650917e-07,
      "loss": 0.0176,
      "num_tokens": 29980296.0,
      "reward": 3.369419753551483,
      "reward_std": 0.5236727483570576,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.3816964477300644,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 672.2243728637695,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.24846415221691132,
      "learning_rate": 8.146601955249187e-07,
      "loss": 0.0271,
      "num_tokens": 30709425.0,
      "reward": 3.6026786863803864,
      "reward_std": 0.7045574709773064,
      "rewards/format_reward": 0.9810267835855484,
      "rewards/weighted_accuracy_reward": 2.621651768684387,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 675.1674423217773,
      "epoch": 0.7313432835820896,
      "grad_norm": 0.1271398812532425,
      "learning_rate": 8.043807145043603e-07,
      "loss": 0.0223,
      "num_tokens": 31450775.0,
      "reward": 3.6696430146694183,
      "reward_std": 0.7956224754452705,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.681919664144516,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.8203430175781,
      "epoch": 0.746268656716418,
      "grad_norm": 0.24750620126724243,
      "learning_rate": 7.938926261462365e-07,
      "loss": 0.0433,
      "num_tokens": 32166686.0,
      "reward": 3.260044753551483,
      "reward_std": 0.6283195540308952,
      "rewards/format_reward": 0.984375,
      "rewards/weighted_accuracy_reward": 2.275669664144516,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 647.4542694091797,
      "epoch": 0.7611940298507462,
      "grad_norm": 0.4828958809375763,
      "learning_rate": 7.832031184624164e-07,
      "loss": 0.0327,
      "num_tokens": 32884589.0,
      "reward": 3.4520090520381927,
      "reward_std": 0.7472007125616074,
      "rewards/format_reward": 0.9821428433060646,
      "rewards/weighted_accuracy_reward": 2.469866096973419,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.600471496582,
      "epoch": 0.7761194029850746,
      "grad_norm": 0.11773540824651718,
      "learning_rate": 7.723195175075135e-07,
      "loss": 0.0166,
      "num_tokens": 33610511.0,
      "reward": 3.2377233803272247,
      "reward_std": 0.5106364823877811,
      "rewards/format_reward": 0.986607126891613,
      "rewards/weighted_accuracy_reward": 2.2511160522699356,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 657.8962326049805,
      "epoch": 0.7910447761194029,
      "grad_norm": 0.3528241515159607,
      "learning_rate": 7.612492823579744e-07,
      "loss": 0.044,
      "num_tokens": 34327658.0,
      "reward": 3.1495537161827087,
      "reward_std": 0.8184557631611824,
      "rewards/format_reward": 0.989955335855484,
      "rewards/weighted_accuracy_reward": 2.159598186612129,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.9219055175781,
      "epoch": 0.8059701492537313,
      "grad_norm": 0.11597654968500137,
      "learning_rate": 7.5e-07,
      "loss": 0.0177,
      "num_tokens": 35012564.0,
      "reward": 3.069196581840515,
      "reward_std": 0.6249974854290485,
      "rewards/format_reward": 0.991071417927742,
      "rewards/weighted_accuracy_reward": 2.078124985098839,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.5056076049805,
      "epoch": 0.8208955223880597,
      "grad_norm": 0.10310050845146179,
      "learning_rate": 7.385793801298042e-07,
      "loss": 0.0223,
      "num_tokens": 35713929.0,
      "reward": 3.732143074274063,
      "reward_std": 0.6322607882320881,
      "rewards/format_reward": 0.9910714253783226,
      "rewards/weighted_accuracy_reward": 2.741071403026581,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.6663208007812,
      "epoch": 0.835820895522388,
      "grad_norm": 0.2997354567050934,
      "learning_rate": 7.269952498697734e-07,
      "loss": 0.0174,
      "num_tokens": 36419022.0,
      "reward": 3.2734376788139343,
      "reward_std": 0.6772932708263397,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.283482104539871,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 618.4431076049805,
      "epoch": 0.8507462686567164,
      "grad_norm": 0.1617140769958496,
      "learning_rate": 7.152555484041475e-07,
      "loss": 0.0145,
      "num_tokens": 37096147.0,
      "reward": 3.680803745985031,
      "reward_std": 0.5588784627616405,
      "rewards/format_reward": 0.991071417927742,
      "rewards/weighted_accuracy_reward": 2.689732104539871,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.5045051574707,
      "epoch": 0.8656716417910447,
      "grad_norm": 0.09553709626197815,
      "learning_rate": 7.033683215379002e-07,
      "loss": 0.0042,
      "num_tokens": 37796007.0,
      "reward": 3.362723410129547,
      "reward_std": 0.5769157111644745,
      "rewards/format_reward": 0.9933035746216774,
      "rewards/weighted_accuracy_reward": 2.3694196343421936,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.9219017028809,
      "epoch": 0.8805970149253731,
      "grad_norm": 0.11292736232280731,
      "learning_rate": 6.913417161825449e-07,
      "loss": 0.0179,
      "num_tokens": 38494785.0,
      "reward": 3.6517858803272247,
      "reward_std": 0.7219322323799133,
      "rewards/format_reward": 0.9944196343421936,
      "rewards/weighted_accuracy_reward": 2.657366067171097,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 624.2265892028809,
      "epoch": 0.8955223880597015,
      "grad_norm": 0.14145782589912415,
      "learning_rate": 6.7918397477265e-07,
      "loss": 0.0205,
      "num_tokens": 39189124.0,
      "reward": 3.356026917695999,
      "reward_std": 0.5718645751476288,
      "rewards/format_reward": 0.9888392835855484,
      "rewards/weighted_accuracy_reward": 2.367187514901161,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.3337249755859,
      "epoch": 0.9104477611940298,
      "grad_norm": 0.08981050550937653,
      "learning_rate": 6.669034296168854e-07,
      "loss": 0.0265,
      "num_tokens": 39909519.0,
      "reward": 3.1741072833538055,
      "reward_std": 0.5122832171618938,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.186383932828903,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.2890930175781,
      "epoch": 0.9253731343283582,
      "grad_norm": 0.2292553037405014,
      "learning_rate": 6.545084971874736e-07,
      "loss": 0.0214,
      "num_tokens": 40589818.0,
      "reward": 3.665178745985031,
      "reward_std": 0.5537377782166004,
      "rewards/format_reward": 0.9921875,
      "rewards/weighted_accuracy_reward": 2.6729910522699356,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 676.3984680175781,
      "epoch": 0.9402985074626866,
      "grad_norm": 0.09297273308038712,
      "learning_rate": 6.420076723519614e-07,
      "loss": 0.0338,
      "num_tokens": 41325383.0,
      "reward": 3.5714287757873535,
      "reward_std": 0.7664550691843033,
      "rewards/format_reward": 0.9832589253783226,
      "rewards/weighted_accuracy_reward": 2.5881696343421936,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.1931076049805,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.6212418675422668,
      "learning_rate": 6.294095225512604e-07,
      "loss": 0.0002,
      "num_tokens": 41983708.0,
      "reward": 3.470982313156128,
      "reward_std": 0.6227617971599102,
      "rewards/format_reward": 0.9933035597205162,
      "rewards/weighted_accuracy_reward": 2.477678567171097,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.8225631713867,
      "epoch": 0.9701492537313433,
      "grad_norm": 0.1953699290752411,
      "learning_rate": 6.167226819279527e-07,
      "loss": 0.0273,
      "num_tokens": 42660613.0,
      "reward": 3.483259081840515,
      "reward_std": 0.5097917150706053,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.4933036118745804,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.7951507568359,
      "epoch": 0.9850746268656716,
      "grad_norm": 0.3842504918575287,
      "learning_rate": 6.039558454088795e-07,
      "loss": 0.0311,
      "num_tokens": 43376704.0,
      "reward": 3.512276977300644,
      "reward_std": 0.6238038428127766,
      "rewards/format_reward": 0.984375,
      "rewards/weighted_accuracy_reward": 2.527901753783226,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.7198944091797,
      "epoch": 1.0149253731343284,
      "grad_norm": 0.09177397936582565,
      "learning_rate": 5.911177627460738e-07,
      "loss": 0.021,
      "num_tokens": 44062901.0,
      "reward": 3.4553572833538055,
      "reward_std": 0.4633795730769634,
      "rewards/format_reward": 0.983258917927742,
      "rewards/weighted_accuracy_reward": 2.472098231315613,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 674.7143096923828,
      "epoch": 1.0298507462686568,
      "grad_norm": 0.09129226952791214,
      "learning_rate": 5.782172325201155e-07,
      "loss": 0.0248,
      "num_tokens": 44800397.0,
      "reward": 3.4888394474983215,
      "reward_std": 0.6369249150156975,
      "rewards/format_reward": 0.9810267835855484,
      "rewards/weighted_accuracy_reward": 2.5078125596046448,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.6484680175781,
      "epoch": 1.044776119402985,
      "grad_norm": 0.09588230401277542,
      "learning_rate": 5.652630961100258e-07,
      "loss": 0.0227,
      "num_tokens": 45469290.0,
      "reward": 3.428571581840515,
      "reward_std": 0.6118915155529976,
      "rewards/format_reward": 0.9933035671710968,
      "rewards/weighted_accuracy_reward": 2.4352678656578064,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.7187805175781,
      "epoch": 1.0597014925373134,
      "grad_norm": 0.09058994054794312,
      "learning_rate": 5.522642316338268e-07,
      "loss": 0.0171,
      "num_tokens": 46149166.0,
      "reward": 3.6439733505249023,
      "reward_std": 0.6172333508729935,
      "rewards/format_reward": 0.991071417927742,
      "rewards/weighted_accuracy_reward": 2.652901768684387,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.827033996582,
      "epoch": 1.0746268656716418,
      "grad_norm": 0.15813829004764557,
      "learning_rate": 5.392295478639225e-07,
      "loss": 0.021,
      "num_tokens": 46815331.0,
      "reward": 3.4843751788139343,
      "reward_std": 0.5806327611207962,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.4944196343421936,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 634.513427734375,
      "epoch": 1.0895522388059702,
      "grad_norm": 0.10962940007448196,
      "learning_rate": 5.26167978121472e-07,
      "loss": 0.0185,
      "num_tokens": 47506647.0,
      "reward": 3.4341519474983215,
      "reward_std": 0.7080828212201595,
      "rewards/format_reward": 0.9944196343421936,
      "rewards/weighted_accuracy_reward": 2.439732179045677,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 643.3817253112793,
      "epoch": 1.1044776119402986,
      "grad_norm": 0.1467304229736328,
      "learning_rate": 5.130884741539366e-07,
      "loss": 0.0268,
      "num_tokens": 48215797.0,
      "reward": 3.3772322833538055,
      "reward_std": 0.7876874767243862,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.387276813387871,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.5837440490723,
      "epoch": 1.1194029850746268,
      "grad_norm": 0.331134557723999,
      "learning_rate": 5e-07,
      "loss": 0.0018,
      "num_tokens": 48865360.0,
      "reward": 3.4229912161827087,
      "reward_std": 0.6479158662259579,
      "rewards/format_reward": 0.9944196417927742,
      "rewards/weighted_accuracy_reward": 2.428571403026581,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.5033798217773,
      "epoch": 1.1343283582089552,
      "grad_norm": 0.13394823670387268,
      "learning_rate": 4.869115258460634e-07,
      "loss": 0.0118,
      "num_tokens": 49543027.0,
      "reward": 3.4910715520381927,
      "reward_std": 0.5882082916796207,
      "rewards/format_reward": 0.9910714253783226,
      "rewards/weighted_accuracy_reward": 2.5,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.2634048461914,
      "epoch": 1.1492537313432836,
      "grad_norm": 0.08547962456941605,
      "learning_rate": 4.7383202187852804e-07,
      "loss": 0.01,
      "num_tokens": 50213951.0,
      "reward": 3.5178573429584503,
      "reward_std": 0.5135773122310638,
      "rewards/format_reward": 0.9843749925494194,
      "rewards/weighted_accuracy_reward": 2.53348208963871,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.944221496582,
      "epoch": 1.164179104477612,
      "grad_norm": 0.09146715700626373,
      "learning_rate": 4.6077045213607755e-07,
      "loss": 0.0231,
      "num_tokens": 50820485.0,
      "reward": 3.325892984867096,
      "reward_std": 0.3042021542787552,
      "rewards/format_reward": 0.9977678582072258,
      "rewards/weighted_accuracy_reward": 2.328125,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.6361770629883,
      "epoch": 1.1791044776119404,
      "grad_norm": 0.09851463884115219,
      "learning_rate": 4.477357683661733e-07,
      "loss": 0.0296,
      "num_tokens": 51522199.0,
      "reward": 3.3448662161827087,
      "reward_std": 0.49197270907461643,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.354910731315613,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.2544975280762,
      "epoch": 1.1940298507462686,
      "grad_norm": 0.09435434639453888,
      "learning_rate": 4.347369038899743e-07,
      "loss": 0.0128,
      "num_tokens": 52240451.0,
      "reward": 3.6104912161827087,
      "reward_std": 0.5894973948597908,
      "rewards/format_reward": 0.9977678507566452,
      "rewards/weighted_accuracy_reward": 2.6127232164144516,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 681.8326187133789,
      "epoch": 1.208955223880597,
      "grad_norm": 0.11519414186477661,
      "learning_rate": 4.2178276747988444e-07,
      "loss": 0.0123,
      "num_tokens": 52977789.0,
      "reward": 3.390625149011612,
      "reward_std": 0.6482040733098984,
      "rewards/format_reward": 0.9921874925494194,
      "rewards/weighted_accuracy_reward": 2.398437485098839,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 637.7042770385742,
      "epoch": 1.2238805970149254,
      "grad_norm": 0.1185440942645073,
      "learning_rate": 4.0888223725392624e-07,
      "loss": 0.0315,
      "num_tokens": 53674084.0,
      "reward": 3.4877233505249023,
      "reward_std": 0.5407859422266483,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.497767850756645,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.3571624755859,
      "epoch": 1.2388059701492538,
      "grad_norm": 0.10672775655984879,
      "learning_rate": 3.960441545911204e-07,
      "loss": 0.0106,
      "num_tokens": 54350900.0,
      "reward": 3.640625149011612,
      "reward_std": 0.6587489880621433,
      "rewards/format_reward": 0.9944196417927742,
      "rewards/weighted_accuracy_reward": 2.646205335855484,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.1975708007812,
      "epoch": 1.2537313432835822,
      "grad_norm": 0.30903345346450806,
      "learning_rate": 3.8327731807204744e-07,
      "loss": -0.0004,
      "num_tokens": 55047013.0,
      "reward": 3.271205484867096,
      "reward_std": 0.6298276018351316,
      "rewards/format_reward": 0.9944196343421936,
      "rewards/weighted_accuracy_reward": 2.276785761117935,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.8181076049805,
      "epoch": 1.2686567164179103,
      "grad_norm": 0.11902620643377304,
      "learning_rate": 3.7059047744873955e-07,
      "loss": 0.0013,
      "num_tokens": 55750850.0,
      "reward": 3.3716518878936768,
      "reward_std": 0.6535772457718849,
      "rewards/format_reward": 0.9966517835855484,
      "rewards/weighted_accuracy_reward": 2.374999985098839,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 614.3605155944824,
      "epoch": 1.2835820895522387,
      "grad_norm": 0.08017314225435257,
      "learning_rate": 3.5799232764803867e-07,
      "loss": 0.0147,
      "num_tokens": 56419821.0,
      "reward": 3.7165180146694183,
      "reward_std": 0.48502405919134617,
      "rewards/format_reward": 0.9910714104771614,
      "rewards/weighted_accuracy_reward": 2.725446403026581,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.4442291259766,
      "epoch": 1.2985074626865671,
      "grad_norm": 0.10848066955804825,
      "learning_rate": 3.454915028125263e-07,
      "loss": 0.0274,
      "num_tokens": 57105603.0,
      "reward": 3.363839417695999,
      "reward_std": 0.5629270784556866,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.376116082072258,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 663.1919860839844,
      "epoch": 1.3134328358208955,
      "grad_norm": 0.09586381912231445,
      "learning_rate": 3.330965703831146e-07,
      "loss": 0.0312,
      "num_tokens": 57831983.0,
      "reward": 3.292410910129547,
      "reward_std": 0.6855169534683228,
      "rewards/format_reward": 0.9854910671710968,
      "rewards/weighted_accuracy_reward": 2.306919649243355,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.4564895629883,
      "epoch": 1.328358208955224,
      "grad_norm": 0.09536378085613251,
      "learning_rate": 3.2081602522734985e-07,
      "loss": 0.0108,
      "num_tokens": 58503000.0,
      "reward": 3.876116245985031,
      "reward_std": 0.5105698518455029,
      "rewards/format_reward": 0.9921874850988388,
      "rewards/weighted_accuracy_reward": 2.8839285373687744,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.9163131713867,
      "epoch": 1.3432835820895521,
      "grad_norm": 0.0982845276594162,
      "learning_rate": 3.086582838174551e-07,
      "loss": 0.0152,
      "num_tokens": 59141325.0,
      "reward": 3.4933037757873535,
      "reward_std": 0.6200135685503483,
      "rewards/format_reward": 0.9977678507566452,
      "rewards/weighted_accuracy_reward": 2.4955357015132904,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.834846496582,
      "epoch": 1.3582089552238805,
      "grad_norm": 0.10763094574213028,
      "learning_rate": 2.9663167846209996e-07,
      "loss": 0.0184,
      "num_tokens": 59848641.0,
      "reward": 3.3325894474983215,
      "reward_std": 0.5141867641359568,
      "rewards/format_reward": 0.9899553433060646,
      "rewards/weighted_accuracy_reward": 2.342633903026581,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.4598426818848,
      "epoch": 1.373134328358209,
      "grad_norm": 0.129378542304039,
      "learning_rate": 2.847444515958523e-07,
      "loss": 0.0064,
      "num_tokens": 60487421.0,
      "reward": 3.6238841116428375,
      "reward_std": 0.5190795473754406,
      "rewards/format_reward": 0.9977678582072258,
      "rewards/weighted_accuracy_reward": 2.626116067171097,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 637.1183319091797,
      "epoch": 1.3880597014925373,
      "grad_norm": 0.10381485521793365,
      "learning_rate": 2.730047501302266e-07,
      "loss": 0.0291,
      "num_tokens": 61184343.0,
      "reward": 3.595982313156128,
      "reward_std": 0.6280332431197166,
      "rewards/format_reward": 0.9810267835855484,
      "rewards/weighted_accuracy_reward": 2.6149553656578064,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.6730117797852,
      "epoch": 1.4029850746268657,
      "grad_norm": 0.09659765660762787,
      "learning_rate": 2.6142061987019574e-07,
      "loss": 0.0242,
      "num_tokens": 61876498.0,
      "reward": 3.5133930146694183,
      "reward_std": 0.6254775524139404,
      "rewards/format_reward": 0.9910714253783226,
      "rewards/weighted_accuracy_reward": 2.522321417927742,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 604.9219055175781,
      "epoch": 1.417910447761194,
      "grad_norm": 0.08201196044683456,
      "learning_rate": 2.500000000000001e-07,
      "loss": 0.035,
      "num_tokens": 62561092.0,
      "reward": 3.3928572833538055,
      "reward_std": 0.5540718920528889,
      "rewards/format_reward": 0.9843749925494194,
      "rewards/weighted_accuracy_reward": 2.4084821343421936,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.8661003112793,
      "epoch": 1.4328358208955223,
      "grad_norm": 0.11465168744325638,
      "learning_rate": 2.387507176420256e-07,
      "loss": 0.0276,
      "num_tokens": 63275012.0,
      "reward": 3.4040180146694183,
      "reward_std": 0.5879912413656712,
      "rewards/format_reward": 0.9799107164144516,
      "rewards/weighted_accuracy_reward": 2.4241071343421936,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.3303871154785,
      "epoch": 1.4477611940298507,
      "grad_norm": 0.10745745152235031,
      "learning_rate": 2.2768048249248644e-07,
      "loss": 0.0254,
      "num_tokens": 63999476.0,
      "reward": 3.5502233505249023,
      "reward_std": 0.5421314351260662,
      "rewards/format_reward": 0.9910714253783226,
      "rewards/weighted_accuracy_reward": 2.559151768684387,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 584.6071662902832,
      "epoch": 1.462686567164179,
      "grad_norm": 0.2066802978515625,
      "learning_rate": 2.167968815375837e-07,
      "loss": 0.0095,
      "num_tokens": 64659076.0,
      "reward": 3.3404019474983215,
      "reward_std": 0.5494298487901688,
      "rewards/format_reward": 0.9910714253783226,
      "rewards/weighted_accuracy_reward": 2.349330335855484,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.5993576049805,
      "epoch": 1.4776119402985075,
      "grad_norm": 0.08914008736610413,
      "learning_rate": 2.0610737385376348e-07,
      "loss": 0.0147,
      "num_tokens": 65394661.0,
      "reward": 3.401785910129547,
      "reward_std": 0.6547242775559425,
      "rewards/format_reward": 0.9933035671710968,
      "rewards/weighted_accuracy_reward": 2.40848208963871,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 704.4453506469727,
      "epoch": 1.4925373134328357,
      "grad_norm": 0.09854733198881149,
      "learning_rate": 1.9561928549563966e-07,
      "loss": 0.0102,
      "num_tokens": 66163932.0,
      "reward": 3.2098215520381927,
      "reward_std": 0.7065087556838989,
      "rewards/format_reward": 0.9955357015132904,
      "rewards/weighted_accuracy_reward": 2.214285731315613,
      "step": 99
    },
    {
      "epoch": 1.5074626865671643,
      "grad_norm": 0.13317689299583435,
      "learning_rate": 1.8533980447508135e-07,
      "loss": 0.0244,
      "step": 100
    },
    {
      "epoch": 1.5074626865671643,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 621.4327540051338,
      "eval_loss": 0.019861649721860886,
      "eval_num_tokens": 66830283.0,
      "eval_reward": 3.260200669645597,
      "eval_reward_std": 0.6498415311746424,
      "eval_rewards/format_reward": 0.9916949257171354,
      "eval_rewards/weighted_accuracy_reward": 2.268505583928284,
      "eval_runtime": 7542.0097,
      "eval_samples_per_second": 0.663,
      "eval_steps_per_second": 0.006,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.0842876434326,
      "epoch": 1.5223880597014925,
      "grad_norm": 0.14756238460540771,
      "learning_rate": 1.7527597583490823e-07,
      "loss": 0.0162,
      "num_tokens": 67535099.0,
      "reward": 3.4536832123994827,
      "reward_std": 0.5698292423039675,
      "rewards/format_reward": 0.9921874925494194,
      "rewards/weighted_accuracy_reward": 2.4614954963326454,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.2366333007812,
      "epoch": 1.537313432835821,
      "grad_norm": 0.09353147447109222,
      "learning_rate": 1.6543469682057104e-07,
      "loss": 0.0171,
      "num_tokens": 68216263.0,
      "reward": 3.502232253551483,
      "reward_std": 0.46381750889122486,
      "rewards/format_reward": 0.9899553507566452,
      "rewards/weighted_accuracy_reward": 2.512276753783226,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 653.2835083007812,
      "epoch": 1.5522388059701493,
      "grad_norm": 0.10310540348291397,
      "learning_rate": 1.5582271215312293e-07,
      "loss": 0.0336,
      "num_tokens": 68928165.0,
      "reward": 3.600446581840515,
      "reward_std": 0.657109696418047,
      "rewards/format_reward": 0.984375,
      "rewards/weighted_accuracy_reward": 2.6160714626312256,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.0993537902832,
      "epoch": 1.5671641791044775,
      "grad_norm": 0.08973117172718048,
      "learning_rate": 1.4644660940672627e-07,
      "loss": 0.0264,
      "num_tokens": 69600134.0,
      "reward": 3.590401917695999,
      "reward_std": 0.4744480177760124,
      "rewards/format_reward": 0.9843749925494194,
      "rewards/weighted_accuracy_reward": 2.606026828289032,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.853816986084,
      "epoch": 1.582089552238806,
      "grad_norm": 0.113929882645607,
      "learning_rate": 1.3731281449385628e-07,
      "loss": 0.022,
      "num_tokens": 70304475.0,
      "reward": 3.375000149011612,
      "reward_std": 0.5547266900539398,
      "rewards/format_reward": 0.9933035671710968,
      "rewards/weighted_accuracy_reward": 2.381696432828903,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.9710159301758,
      "epoch": 1.5970149253731343,
      "grad_norm": 0.08515556901693344,
      "learning_rate": 1.284275872613028e-07,
      "loss": 0.0187,
      "num_tokens": 71044617.0,
      "reward": 3.436384081840515,
      "reward_std": 0.5531730111688375,
      "rewards/format_reward": 0.9944196343421936,
      "rewards/weighted_accuracy_reward": 2.4419642984867096,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.2087249755859,
      "epoch": 1.6119402985074627,
      "grad_norm": 0.17523446679115295,
      "learning_rate": 1.1979701719998454e-07,
      "loss": 0.035,
      "num_tokens": 71787948.0,
      "reward": 3.4341519474983215,
      "reward_std": 0.7938684299588203,
      "rewards/format_reward": 0.9854910671710968,
      "rewards/weighted_accuracy_reward": 2.4486607164144516,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.8147583007812,
      "epoch": 1.626865671641791,
      "grad_norm": 0.10510087013244629,
      "learning_rate": 1.1142701927151454e-07,
      "loss": 0.0584,
      "num_tokens": 72474878.0,
      "reward": 3.508928745985031,
      "reward_std": 0.7126019038259983,
      "rewards/format_reward": 0.9821428433060646,
      "rewards/weighted_accuracy_reward": 2.526785686612129,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.9955749511719,
      "epoch": 1.6417910447761193,
      "grad_norm": 0.0960945188999176,
      "learning_rate": 1.0332332985438247e-07,
      "loss": 0.0049,
      "num_tokens": 73193330.0,
      "reward": 3.4676341116428375,
      "reward_std": 0.5328357554972172,
      "rewards/format_reward": 0.9888392761349678,
      "rewards/weighted_accuracy_reward": 2.4787946343421936,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.6908721923828,
      "epoch": 1.6567164179104479,
      "grad_norm": 0.08923044800758362,
      "learning_rate": 9.549150281252632e-08,
      "loss": 0.0199,
      "num_tokens": 73869109.0,
      "reward": 3.454241245985031,
      "reward_std": 0.5548066087067127,
      "rewards/format_reward": 0.9921875,
      "rewards/weighted_accuracy_reward": 2.462053582072258,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.4821624755859,
      "epoch": 1.671641791044776,
      "grad_norm": 0.09829309582710266,
      "learning_rate": 8.793690568899215e-08,
      "loss": 0.017,
      "num_tokens": 74586269.0,
      "reward": 3.4151787161827087,
      "reward_std": 0.5523700322955847,
      "rewards/format_reward": 0.9921874925494194,
      "rewards/weighted_accuracy_reward": 2.422991067171097,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 665.600471496582,
      "epoch": 1.6865671641791045,
      "grad_norm": 0.07953444123268127,
      "learning_rate": 8.066471602728803e-08,
      "loss": 0.0106,
      "num_tokens": 75300311.0,
      "reward": 3.603794813156128,
      "reward_std": 0.45538773387670517,
      "rewards/format_reward": 0.9888392761349678,
      "rewards/weighted_accuracy_reward": 2.6149553805589676,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.7455673217773,
      "epoch": 1.7014925373134329,
      "grad_norm": 0.35571610927581787,
      "learning_rate": 7.36799178229539e-08,
      "loss": 0.0293,
      "num_tokens": 75994163.0,
      "reward": 3.3214287161827087,
      "reward_std": 0.7195953875780106,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.333705335855484,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.6495666503906,
      "epoch": 1.716417910447761,
      "grad_norm": 0.10555040091276169,
      "learning_rate": 6.698729810778064e-08,
      "loss": 0.0029,
      "num_tokens": 76662393.0,
      "reward": 3.4866072833538055,
      "reward_std": 0.4477080851793289,
      "rewards/format_reward": 0.9977678507566452,
      "rewards/weighted_accuracy_reward": 2.4888392984867096,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 657.5870819091797,
      "epoch": 1.7313432835820897,
      "grad_norm": 0.09108369797468185,
      "learning_rate": 6.059144366901736e-08,
      "loss": 0.0201,
      "num_tokens": 77380135.0,
      "reward": 3.297991245985031,
      "reward_std": 0.5759816411882639,
      "rewards/format_reward": 0.9866071343421936,
      "rewards/weighted_accuracy_reward": 2.311383917927742,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.610523223877,
      "epoch": 1.7462686567164178,
      "grad_norm": 0.09139251708984375,
      "learning_rate": 5.44967379058161e-08,
      "loss": 0.0248,
      "num_tokens": 78033490.0,
      "reward": 3.414062649011612,
      "reward_std": 0.34683902747929096,
      "rewards/format_reward": 0.9921874925494194,
      "rewards/weighted_accuracy_reward": 2.4218749701976776,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.6774826049805,
      "epoch": 1.7611940298507462,
      "grad_norm": 0.12718737125396729,
      "learning_rate": 4.870735782506979e-08,
      "loss": 0.0102,
      "num_tokens": 78784489.0,
      "reward": 3.27678582072258,
      "reward_std": 0.7959724003449082,
      "rewards/format_reward": 0.995535708963871,
      "rewards/weighted_accuracy_reward": 2.281250014901161,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.4810562133789,
      "epoch": 1.7761194029850746,
      "grad_norm": 0.11492962390184402,
      "learning_rate": 4.322727117869951e-08,
      "loss": 0.0204,
      "num_tokens": 79504480.0,
      "reward": 3.4542412161827087,
      "reward_std": 0.695434695109725,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.466517850756645,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 637.9006958007812,
      "epoch": 1.7910447761194028,
      "grad_norm": 0.1276756078004837,
      "learning_rate": 3.806023374435663e-08,
      "loss": 0.0298,
      "num_tokens": 80201447.0,
      "reward": 3.447544813156128,
      "reward_std": 0.5874622464179993,
      "rewards/format_reward": 0.991071417927742,
      "rewards/weighted_accuracy_reward": 2.456473171710968,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.5703430175781,
      "epoch": 1.8059701492537314,
      "grad_norm": 0.110611692070961,
      "learning_rate": 3.3209786751399184e-08,
      "loss": 0.0234,
      "num_tokens": 80916134.0,
      "reward": 3.529018074274063,
      "reward_std": 0.5716630481183529,
      "rewards/format_reward": 0.9921874925494194,
      "rewards/weighted_accuracy_reward": 2.5368303656578064,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.752254486084,
      "epoch": 1.8208955223880596,
      "grad_norm": 0.1067574992775917,
      "learning_rate": 2.8679254453910785e-08,
      "loss": 0.0174,
      "num_tokens": 81620552.0,
      "reward": 3.264509081840515,
      "reward_std": 0.5513999536633492,
      "rewards/format_reward": 0.9977678507566452,
      "rewards/weighted_accuracy_reward": 2.266741082072258,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 670.5993499755859,
      "epoch": 1.835820895522388,
      "grad_norm": 0.12840881943702698,
      "learning_rate": 2.4471741852423233e-08,
      "loss": 0.0317,
      "num_tokens": 82368201.0,
      "reward": 3.504464477300644,
      "reward_std": 0.6936709322035313,
      "rewards/format_reward": 0.9799107015132904,
      "rewards/weighted_accuracy_reward": 2.524553507566452,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 661.0960083007812,
      "epoch": 1.8507462686567164,
      "grad_norm": 0.09926487505435944,
      "learning_rate": 2.0590132565903473e-08,
      "loss": 0.0341,
      "num_tokens": 83093583.0,
      "reward": 3.514509081840515,
      "reward_std": 0.7100039795041084,
      "rewards/format_reward": 0.9810267835855484,
      "rewards/weighted_accuracy_reward": 2.533482149243355,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.849365234375,
      "epoch": 1.8656716417910446,
      "grad_norm": 0.17390291392803192,
      "learning_rate": 1.7037086855465898e-08,
      "loss": 0.012,
      "num_tokens": 83764472.0,
      "reward": 3.3571430444717407,
      "reward_std": 0.6937255449593067,
      "rewards/format_reward": 0.9988839253783226,
      "rewards/weighted_accuracy_reward": 2.358258917927742,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.6027069091797,
      "epoch": 1.8805970149253732,
      "grad_norm": 0.11428766697645187,
      "learning_rate": 1.3815039801161722e-08,
      "loss": 0.011,
      "num_tokens": 84486724.0,
      "reward": 3.5513394474983215,
      "reward_std": 0.675942525267601,
      "rewards/format_reward": 0.9944196417927742,
      "rewards/weighted_accuracy_reward": 2.556919664144516,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.9576225280762,
      "epoch": 1.8955223880597014,
      "grad_norm": 0.10255751013755798,
      "learning_rate": 1.0926199633097154e-08,
      "loss": 0.0228,
      "num_tokens": 85133438.0,
      "reward": 3.5234377086162567,
      "reward_std": 0.6280252300202847,
      "rewards/format_reward": 0.9944196343421936,
      "rewards/weighted_accuracy_reward": 2.5290178656578064,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.1819458007812,
      "epoch": 1.9104477611940298,
      "grad_norm": 0.11163829267024994,
      "learning_rate": 8.372546218022746e-09,
      "loss": 0.0287,
      "num_tokens": 85861001.0,
      "reward": 3.127232253551483,
      "reward_std": 0.6194400154054165,
      "rewards/format_reward": 0.987723208963871,
      "rewards/weighted_accuracy_reward": 2.1395089477300644,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.6774826049805,
      "epoch": 1.9253731343283582,
      "grad_norm": 0.09224532544612885,
      "learning_rate": 6.15582970243117e-09,
      "loss": 0.0158,
      "num_tokens": 86536744.0,
      "reward": 3.6886162161827087,
      "reward_std": 0.43035920709371567,
      "rewards/format_reward": 0.9921875,
      "rewards/weighted_accuracy_reward": 2.696428567171097,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.1183319091797,
      "epoch": 1.9402985074626866,
      "grad_norm": 0.11262743175029755,
      "learning_rate": 4.277569313094809e-09,
      "loss": 0.0264,
      "num_tokens": 87253330.0,
      "reward": 3.5011161863803864,
      "reward_std": 0.6001675426959991,
      "rewards/format_reward": 0.9866071343421936,
      "rewards/weighted_accuracy_reward": 2.514508992433548,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.9408721923828,
      "epoch": 1.955223880597015,
      "grad_norm": 0.11845642328262329,
      "learning_rate": 2.739052315863355e-09,
      "loss": 0.0081,
      "num_tokens": 87990861.0,
      "reward": 3.352678805589676,
      "reward_std": 0.5374037250876427,
      "rewards/format_reward": 0.9977678507566452,
      "rewards/weighted_accuracy_reward": 2.3549107015132904,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.1004829406738,
      "epoch": 1.9701492537313432,
      "grad_norm": 0.13290314376354218,
      "learning_rate": 1.541333133436018e-09,
      "loss": 0.0136,
      "num_tokens": 88666471.0,
      "reward": 3.5290180146694183,
      "reward_std": 0.6786738075315952,
      "rewards/format_reward": 0.9966517761349678,
      "rewards/weighted_accuracy_reward": 2.5323661118745804,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.555549621582,
      "epoch": 1.9850746268656716,
      "grad_norm": 0.08534468710422516,
      "learning_rate": 6.852326227130833e-10,
      "loss": 0.0023,
      "num_tokens": 89371858.0,
      "reward": 3.5000001788139343,
      "reward_std": 0.5822820998728275,
      "rewards/format_reward": 0.9988839253783226,
      "rewards/weighted_accuracy_reward": 2.501116082072258,
      "step": 132
    },
    {
      "epoch": 1.9850746268656716,
      "step": 132,
      "total_flos": 0.0,
      "train_loss": 0.019874975508476684,
      "train_runtime": 35069.7154,
      "train_samples_per_second": 0.428,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 1,
  "max_steps": 134,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}