{
  "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": 0.6381782293319702,
      "learning_rate": 7.142857142857142e-08,
      "loss": 0.0029,
      "num_tokens": 546936.0,
      "reward": 0.27120537124574184,
      "reward_std": 0.39265505224466324,
      "rewards/accuracy_reward": 0.20647320989519358,
      "rewards/format_reward": 0.06473214365541935,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 471.2355079650879,
      "epoch": 0.029850746268656716,
      "grad_norm": 0.47330012917518616,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0029,
      "num_tokens": 1100635.0,
      "reward": 0.29352679662406445,
      "reward_std": 0.3840556889772415,
      "rewards/accuracy_reward": 0.22767856996506453,
      "rewards/format_reward": 0.06584821548312902,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.05247497558594,
      "epoch": 0.04477611940298507,
      "grad_norm": 0.40551432967185974,
      "learning_rate": 2.1428571428571426e-07,
      "loss": -0.0092,
      "num_tokens": 1698274.0,
      "reward": 0.24888394214212894,
      "reward_std": 0.36480507254600525,
      "rewards/accuracy_reward": 0.20424107275903225,
      "rewards/format_reward": 0.04464285809081048,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 526.8214530944824,
      "epoch": 0.05970149253731343,
      "grad_norm": 0.4093160331249237,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0128,
      "num_tokens": 2299818.0,
      "reward": 0.25781251303851604,
      "reward_std": 0.35233646258711815,
      "rewards/accuracy_reward": 0.1997767873108387,
      "rewards/format_reward": 0.05803571571595967,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 512.1830596923828,
      "epoch": 0.07462686567164178,
      "grad_norm": 0.5151273012161255,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0079,
      "num_tokens": 2896238.0,
      "reward": 0.2667410895228386,
      "reward_std": 0.36962299421429634,
      "rewards/accuracy_reward": 0.18750000279396772,
      "rewards/format_reward": 0.07924107019789517,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 459.7879638671875,
      "epoch": 0.08955223880597014,
      "grad_norm": 0.5247090458869934,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0056,
      "num_tokens": 3432696.0,
      "reward": 0.2645089402794838,
      "reward_std": 0.3864951953291893,
      "rewards/accuracy_reward": 0.21316964365541935,
      "rewards/format_reward": 0.0513392862631008,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.81252670288086,
      "epoch": 0.1044776119402985,
      "grad_norm": 0.7675074338912964,
      "learning_rate": 5e-07,
      "loss": 0.0057,
      "num_tokens": 3997032.0,
      "reward": 0.2901785857975483,
      "reward_std": 0.38608507812023163,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 0.08705357182770967,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 503.35828399658203,
      "epoch": 0.11940298507462686,
      "grad_norm": 0.5886371731758118,
      "learning_rate": 5.714285714285714e-07,
      "loss": -0.0036,
      "num_tokens": 4565689.0,
      "reward": 0.3314732313156128,
      "reward_std": 0.4302351512014866,
      "rewards/accuracy_reward": 0.21205356996506453,
      "rewards/format_reward": 0.11941964458674192,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 454.2031440734863,
      "epoch": 0.13432835820895522,
      "grad_norm": 1.0013484954833984,
      "learning_rate": 6.428571428571429e-07,
      "loss": -0.0081,
      "num_tokens": 5097991.0,
      "reward": 0.3560268022119999,
      "reward_std": 0.4213982783257961,
      "rewards/accuracy_reward": 0.22321428451687098,
      "rewards/format_reward": 0.1328125,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 437.2924270629883,
      "epoch": 0.14925373134328357,
      "grad_norm": 1.295032262802124,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.0078,
      "num_tokens": 5614445.0,
      "reward": 0.4207589514553547,
      "reward_std": 0.49341823533177376,
      "rewards/accuracy_reward": 0.2254464291036129,
      "rewards/format_reward": 0.19531249906867743,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 522.3638687133789,
      "epoch": 0.16417910447761194,
      "grad_norm": 4.296813011169434,
      "learning_rate": 7.857142857142856e-07,
      "loss": 0.0091,
      "num_tokens": 6212539.0,
      "reward": 0.4508928842842579,
      "reward_std": 0.48767242580652237,
      "rewards/accuracy_reward": 0.203125,
      "rewards/format_reward": 0.2477678582072258,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.03127670288086,
      "epoch": 0.1791044776119403,
      "grad_norm": 0.9652361869812012,
      "learning_rate": 8.57142857142857e-07,
      "loss": 0.0202,
      "num_tokens": 6776311.0,
      "reward": 0.525669664144516,
      "reward_std": 0.5405469685792923,
      "rewards/accuracy_reward": 0.2198660708963871,
      "rewards/format_reward": 0.3058035746216774,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 491.166316986084,
      "epoch": 0.19402985074626866,
      "grad_norm": 0.43946775794029236,
      "learning_rate": 9.285714285714285e-07,
      "loss": 0.0234,
      "num_tokens": 7339580.0,
      "reward": 0.7756696790456772,
      "reward_std": 0.5723891109228134,
      "rewards/accuracy_reward": 0.2388392873108387,
      "rewards/format_reward": 0.5368303470313549,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 453.806941986084,
      "epoch": 0.208955223880597,
      "grad_norm": 1.8365284204483032,
      "learning_rate": 1e-06,
      "loss": 0.0215,
      "num_tokens": 7870063.0,
      "reward": 0.8973214775323868,
      "reward_std": 0.5517464429140091,
      "rewards/accuracy_reward": 0.2645089253783226,
      "rewards/format_reward": 0.6328125074505806,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 442.7131881713867,
      "epoch": 0.22388059701492538,
      "grad_norm": 0.4182904362678528,
      "learning_rate": 9.998286624877785e-07,
      "loss": 0.0323,
      "num_tokens": 8381238.0,
      "reward": 0.9241071864962578,
      "reward_std": 0.5212078019976616,
      "rewards/accuracy_reward": 0.2198660708963871,
      "rewards/format_reward": 0.7042410671710968,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 489.45203399658203,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.4200661778450012,
      "learning_rate": 9.99314767377287e-07,
      "loss": 0.0269,
      "num_tokens": 8943155.0,
      "reward": 1.010044701397419,
      "reward_std": 0.5311430767178535,
      "rewards/accuracy_reward": 0.2633928544819355,
      "rewards/format_reward": 0.746651791036129,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 437.2343864440918,
      "epoch": 0.2537313432835821,
      "grad_norm": 0.35031601786613464,
      "learning_rate": 9.98458666866564e-07,
      "loss": -0.0231,
      "num_tokens": 9484653.0,
      "reward": 1.1908482760190964,
      "reward_std": 0.4788440987467766,
      "rewards/accuracy_reward": 0.30691963993012905,
      "rewards/format_reward": 0.8839285671710968,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.06028747558594,
      "epoch": 0.26865671641791045,
      "grad_norm": 0.31605255603790283,
      "learning_rate": 9.972609476841365e-07,
      "loss": 0.031,
      "num_tokens": 10038427.0,
      "reward": 1.2421875596046448,
      "reward_std": 0.4797636419534683,
      "rewards/accuracy_reward": 0.34263392724096775,
      "rewards/format_reward": 0.8995535746216774,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 492.533504486084,
      "epoch": 0.2835820895522388,
      "grad_norm": 0.3311372399330139,
      "learning_rate": 9.957224306869053e-07,
      "loss": 0.0183,
      "num_tokens": 10604529.0,
      "reward": 1.3191964626312256,
      "reward_std": 0.4729856923222542,
      "rewards/accuracy_reward": 0.4151785634458065,
      "rewards/format_reward": 0.9040178507566452,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.89734649658203,
      "epoch": 0.29850746268656714,
      "grad_norm": 0.339732825756073,
      "learning_rate": 9.938441702975689e-07,
      "loss": 0.0107,
      "num_tokens": 11161797.0,
      "reward": 1.3560268729925156,
      "reward_std": 0.47978325933218,
      "rewards/accuracy_reward": 0.4374999962747097,
      "rewards/format_reward": 0.918526791036129,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.010066986084,
      "epoch": 0.31343283582089554,
      "grad_norm": 0.23753176629543304,
      "learning_rate": 9.916274537819773e-07,
      "loss": 0.0283,
      "num_tokens": 11780870.0,
      "reward": 1.4386161267757416,
      "reward_std": 0.46573930606245995,
      "rewards/accuracy_reward": 0.5189732164144516,
      "rewards/format_reward": 0.9196428656578064,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.6372947692871,
      "epoch": 0.3283582089552239,
      "grad_norm": 0.2946343719959259,
      "learning_rate": 9.890738003669027e-07,
      "loss": 0.0361,
      "num_tokens": 12347257.0,
      "reward": 1.491071492433548,
      "reward_std": 0.42494936659932137,
      "rewards/accuracy_reward": 0.5357142835855484,
      "rewards/format_reward": 0.9553571343421936,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 479.8527069091797,
      "epoch": 0.34328358208955223,
      "grad_norm": 0.2775673568248749,
      "learning_rate": 9.861849601988383e-07,
      "loss": 0.013,
      "num_tokens": 12907781.0,
      "reward": 1.545758992433548,
      "reward_std": 0.39185454323887825,
      "rewards/accuracy_reward": 0.6004464291036129,
      "rewards/format_reward": 0.9453125074505806,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.8906555175781,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.2118677943944931,
      "learning_rate": 9.82962913144534e-07,
      "loss": 0.0149,
      "num_tokens": 13555419.0,
      "reward": 1.547991156578064,
      "reward_std": 0.369963688775897,
      "rewards/accuracy_reward": 0.6071428544819355,
      "rewards/format_reward": 0.940848208963871,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.5781478881836,
      "epoch": 0.373134328358209,
      "grad_norm": 0.18608489632606506,
      "learning_rate": 9.794098674340966e-07,
      "loss": 0.011,
      "num_tokens": 14148665.0,
      "reward": 1.6908482909202576,
      "reward_std": 0.2971882149577141,
      "rewards/accuracy_reward": 0.7209821492433548,
      "rewards/format_reward": 0.9698660746216774,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 579.241096496582,
      "epoch": 0.3880597014925373,
      "grad_norm": 0.19908326864242554,
      "learning_rate": 9.755282581475767e-07,
      "loss": 0.0489,
      "num_tokens": 14805265.0,
      "reward": 1.5781250596046448,
      "reward_std": 0.3244300540536642,
      "rewards/accuracy_reward": 0.6238839328289032,
      "rewards/format_reward": 0.9542410746216774,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.6830520629883,
      "epoch": 0.40298507462686567,
      "grad_norm": 0.16485629975795746,
      "learning_rate": 9.713207455460892e-07,
      "loss": 0.0507,
      "num_tokens": 15471061.0,
      "reward": 1.6294643580913544,
      "reward_std": 0.259660467505455,
      "rewards/accuracy_reward": 0.6629464402794838,
      "rewards/format_reward": 0.9665178656578064,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.0223426818848,
      "epoch": 0.417910447761194,
      "grad_norm": 0.1541140377521515,
      "learning_rate": 9.667902132486008e-07,
      "loss": 0.0453,
      "num_tokens": 16104777.0,
      "reward": 1.6495536267757416,
      "reward_std": 0.25614376924932003,
      "rewards/accuracy_reward": 0.684151791036129,
      "rewards/format_reward": 0.9654017835855484,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 663.6596221923828,
      "epoch": 0.43283582089552236,
      "grad_norm": 0.14992094039916992,
      "learning_rate": 9.619397662556433e-07,
      "loss": 0.0387,
      "num_tokens": 16838656.0,
      "reward": 1.617187574505806,
      "reward_std": 0.25877264328300953,
      "rewards/accuracy_reward": 0.6529017873108387,
      "rewards/format_reward": 0.9642857164144516,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.5580749511719,
      "epoch": 0.44776119402985076,
      "grad_norm": 0.1843111515045166,
      "learning_rate": 9.567727288213004e-07,
      "loss": 0.0511,
      "num_tokens": 17528148.0,
      "reward": 1.7154018729925156,
      "reward_std": 0.23728702031075954,
      "rewards/accuracy_reward": 0.7410714328289032,
      "rewards/format_reward": 0.9743303582072258,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.217658996582,
      "epoch": 0.4626865671641791,
      "grad_norm": 0.11407707631587982,
      "learning_rate": 9.512926421749303e-07,
      "loss": 0.0391,
      "num_tokens": 18214823.0,
      "reward": 1.6651786416769028,
      "reward_std": 0.18615373829379678,
      "rewards/accuracy_reward": 0.6841517835855484,
      "rewards/format_reward": 0.9810267835855484,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.1272659301758,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.12041105329990387,
      "learning_rate": 9.455032620941839e-07,
      "loss": 0.0291,
      "num_tokens": 18876793.0,
      "reward": 1.7310268729925156,
      "reward_std": 0.1854203026741743,
      "rewards/accuracy_reward": 0.7500000074505806,
      "rewards/format_reward": 0.9810267835855484,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.4553909301758,
      "epoch": 0.4925373134328358,
      "grad_norm": 0.15564288198947906,
      "learning_rate": 9.394085563309826e-07,
      "loss": 0.0552,
      "num_tokens": 19597657.0,
      "reward": 1.6562500596046448,
      "reward_std": 0.2201189510524273,
      "rewards/accuracy_reward": 0.680803582072258,
      "rewards/format_reward": 0.9754464253783226,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 683.2623062133789,
      "epoch": 0.5074626865671642,
      "grad_norm": 0.11131128668785095,
      "learning_rate": 9.330127018922193e-07,
      "loss": 0.044,
      "num_tokens": 20340292.0,
      "reward": 1.672991156578064,
      "reward_std": 0.2345678173005581,
      "rewards/accuracy_reward": 0.6941964291036129,
      "rewards/format_reward": 0.978794626891613,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 643.5792694091797,
      "epoch": 0.5223880597014925,
      "grad_norm": 0.10805458575487137,
      "learning_rate": 9.26320082177046e-07,
      "loss": 0.0302,
      "num_tokens": 21056931.0,
      "reward": 1.6930804401636124,
      "reward_std": 0.2090724390000105,
      "rewards/accuracy_reward": 0.7064732126891613,
      "rewards/format_reward": 0.9866071343421936,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 671.2388725280762,
      "epoch": 0.5373134328358209,
      "grad_norm": 0.11671151965856552,
      "learning_rate": 9.19335283972712e-07,
      "loss": 0.0517,
      "num_tokens": 21801897.0,
      "reward": 1.6685268580913544,
      "reward_std": 0.23604967631399632,
      "rewards/accuracy_reward": 0.6908482164144516,
      "rewards/format_reward": 0.9776785671710968,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.1127471923828,
      "epoch": 0.5522388059701493,
      "grad_norm": 0.11113366484642029,
      "learning_rate": 9.120630943110077e-07,
      "loss": 0.0203,
      "num_tokens": 22491214.0,
      "reward": 1.7321429252624512,
      "reward_std": 0.17728407308459282,
      "rewards/accuracy_reward": 0.7477678507566452,
      "rewards/format_reward": 0.984375,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 648.8895416259766,
      "epoch": 0.5671641791044776,
      "grad_norm": 0.09676128625869751,
      "learning_rate": 9.045084971874737e-07,
      "loss": 0.0296,
      "num_tokens": 23218403.0,
      "reward": 1.7366072237491608,
      "reward_std": 0.16557986475527287,
      "rewards/accuracy_reward": 0.745535708963871,
      "rewards/format_reward": 0.9910714104771614,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.256721496582,
      "epoch": 0.582089552238806,
      "grad_norm": 0.10444407165050507,
      "learning_rate": 8.966766701456176e-07,
      "loss": 0.0293,
      "num_tokens": 23901713.0,
      "reward": 1.6908483058214188,
      "reward_std": 0.18806083872914314,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 0.979910708963871,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 713.9565048217773,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.09341216832399368,
      "learning_rate": 8.885729807284854e-07,
      "loss": 0.0245,
      "num_tokens": 24665114.0,
      "reward": 1.6953125596046448,
      "reward_std": 0.1814118754118681,
      "rewards/accuracy_reward": 0.7131696492433548,
      "rewards/format_reward": 0.9821428507566452,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.0513610839844,
      "epoch": 0.6119402985074627,
      "grad_norm": 0.1323300451040268,
      "learning_rate": 8.802029828000155e-07,
      "loss": 0.0425,
      "num_tokens": 25389064.0,
      "reward": 1.6674107909202576,
      "reward_std": 0.2174726091325283,
      "rewards/accuracy_reward": 0.6841517835855484,
      "rewards/format_reward": 0.9832589253783226,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.2377471923828,
      "epoch": 0.6268656716417911,
      "grad_norm": 0.11196275800466537,
      "learning_rate": 8.71572412738697e-07,
      "loss": 0.0262,
      "num_tokens": 26055629.0,
      "reward": 1.8258929252624512,
      "reward_std": 0.15602111723273993,
      "rewards/accuracy_reward": 0.834821417927742,
      "rewards/format_reward": 0.991071417927742,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.4364242553711,
      "epoch": 0.6417910447761194,
      "grad_norm": 0.10694889724254608,
      "learning_rate": 8.626871855061437e-07,
      "loss": 0.0058,
      "num_tokens": 26753428.0,
      "reward": 1.7589286416769028,
      "reward_std": 0.15414638444781303,
      "rewards/accuracy_reward": 0.7645089328289032,
      "rewards/format_reward": 0.9944196343421936,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.2935638427734,
      "epoch": 0.6567164179104478,
      "grad_norm": 0.09472611546516418,
      "learning_rate": 8.535533905932737e-07,
      "loss": 0.0191,
      "num_tokens": 27419099.0,
      "reward": 1.7455357760190964,
      "reward_std": 0.15037838742136955,
      "rewards/accuracy_reward": 0.7544642835855484,
      "rewards/format_reward": 0.9910714253783226,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.0245742797852,
      "epoch": 0.6716417910447762,
      "grad_norm": 0.10363104939460754,
      "learning_rate": 8.441772878468769e-07,
      "loss": 0.034,
      "num_tokens": 28101745.0,
      "reward": 1.743303656578064,
      "reward_std": 0.1812288984656334,
      "rewards/accuracy_reward": 0.756696417927742,
      "rewards/format_reward": 0.9866071343421936,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 713.8482437133789,
      "epoch": 0.6865671641791045,
      "grad_norm": 0.10190044343471527,
      "learning_rate": 8.34565303179429e-07,
      "loss": 0.0563,
      "num_tokens": 28875145.0,
      "reward": 1.6640625894069672,
      "reward_std": 0.2248408328741789,
      "rewards/accuracy_reward": 0.6886160746216774,
      "rewards/format_reward": 0.9754464253783226,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 670.1551666259766,
      "epoch": 0.7014925373134329,
      "grad_norm": 0.11430803686380386,
      "learning_rate": 8.247240241650917e-07,
      "loss": 0.0096,
      "num_tokens": 29591660.0,
      "reward": 1.7377232760190964,
      "reward_std": 0.1653135661035776,
      "rewards/accuracy_reward": 0.7455357052385807,
      "rewards/format_reward": 0.9921874850988388,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 653.9408874511719,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.12349370867013931,
      "learning_rate": 8.146601955249187e-07,
      "loss": 0.0196,
      "num_tokens": 30304407.0,
      "reward": 1.7377232909202576,
      "reward_std": 0.19286326505243778,
      "rewards/accuracy_reward": 0.7511160746216774,
      "rewards/format_reward": 0.9866071343421936,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 667.8393173217773,
      "epoch": 0.7313432835820896,
      "grad_norm": 0.10884106904268265,
      "learning_rate": 8.043807145043603e-07,
      "loss": 0.0286,
      "num_tokens": 31039191.0,
      "reward": 1.7377233058214188,
      "reward_std": 0.1800588108599186,
      "rewards/accuracy_reward": 0.746651791036129,
      "rewards/format_reward": 0.991071417927742,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.0323944091797,
      "epoch": 0.746268656716418,
      "grad_norm": 0.10950858145952225,
      "learning_rate": 7.938926261462365e-07,
      "loss": 0.0221,
      "num_tokens": 31743644.0,
      "reward": 1.7098215073347092,
      "reward_std": 0.15341119468212128,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9910714104771614,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.6294937133789,
      "epoch": 0.7611940298507462,
      "grad_norm": 0.12094567716121674,
      "learning_rate": 7.832031184624164e-07,
      "loss": 0.0344,
      "num_tokens": 32449160.0,
      "reward": 1.7287947237491608,
      "reward_std": 0.184023879468441,
      "rewards/accuracy_reward": 0.7366071343421936,
      "rewards/format_reward": 0.9921874850988388,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.0848617553711,
      "epoch": 0.7761194029850746,
      "grad_norm": 0.10622338950634003,
      "learning_rate": 7.723195175075135e-07,
      "loss": 0.0386,
      "num_tokens": 33168348.0,
      "reward": 1.6964286416769028,
      "reward_std": 0.17664530966430902,
      "rewards/accuracy_reward": 0.717633917927742,
      "rewards/format_reward": 0.9787946417927742,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 671.2433319091797,
      "epoch": 0.7910447761194029,
      "grad_norm": 0.10682238638401031,
      "learning_rate": 7.612492823579744e-07,
      "loss": 0.0376,
      "num_tokens": 33897454.0,
      "reward": 1.671875074505806,
      "reward_std": 0.1950401533395052,
      "rewards/accuracy_reward": 0.6886160746216774,
      "rewards/format_reward": 0.983258917927742,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.1585159301758,
      "epoch": 0.8059701492537313,
      "grad_norm": 0.10302180051803589,
      "learning_rate": 7.5e-07,
      "loss": 0.0104,
      "num_tokens": 34575404.0,
      "reward": 1.7410715073347092,
      "reward_std": 0.16222931072115898,
      "rewards/accuracy_reward": 0.745535708963871,
      "rewards/format_reward": 0.9955357015132904,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.7221260070801,
      "epoch": 0.8208955223880597,
      "grad_norm": 0.10820939391851425,
      "learning_rate": 7.385793801298042e-07,
      "loss": 0.0312,
      "num_tokens": 35275171.0,
      "reward": 1.7243304252624512,
      "reward_std": 0.20950445905327797,
      "rewards/accuracy_reward": 0.7366071417927742,
      "rewards/format_reward": 0.9877232015132904,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.1205596923828,
      "epoch": 0.835820895522388,
      "grad_norm": 0.13289578258991241,
      "learning_rate": 7.269952498697734e-07,
      "loss": 0.0142,
      "num_tokens": 35973503.0,
      "reward": 1.6941965073347092,
      "reward_std": 0.176253211684525,
      "rewards/accuracy_reward": 0.703125,
      "rewards/format_reward": 0.9910714253783226,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.6227951049805,
      "epoch": 0.8507462686567164,
      "grad_norm": 0.13948650658130646,
      "learning_rate": 7.152555484041475e-07,
      "loss": 0.0413,
      "num_tokens": 36657061.0,
      "reward": 1.7879465222358704,
      "reward_std": 0.15797736030071974,
      "rewards/accuracy_reward": 0.8035714253783226,
      "rewards/format_reward": 0.9843749925494194,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 650.3381958007812,
      "epoch": 0.8656716417910447,
      "grad_norm": 0.2045215368270874,
      "learning_rate": 7.033683215379002e-07,
      "loss": 0.0374,
      "num_tokens": 37365732.0,
      "reward": 1.695312574505806,
      "reward_std": 0.17142421007156372,
      "rewards/accuracy_reward": 0.7120535671710968,
      "rewards/format_reward": 0.983258917927742,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.5770378112793,
      "epoch": 0.8805970149253731,
      "grad_norm": 0.13572581112384796,
      "learning_rate": 6.913417161825449e-07,
      "loss": 0.0067,
      "num_tokens": 38065993.0,
      "reward": 1.7890625894069672,
      "reward_std": 0.16594232060015202,
      "rewards/accuracy_reward": 0.7946428656578064,
      "rewards/format_reward": 0.9944196343421936,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.4498100280762,
      "epoch": 0.8955223880597015,
      "grad_norm": 0.118675097823143,
      "learning_rate": 6.7918397477265e-07,
      "loss": 0.0253,
      "num_tokens": 38737236.0,
      "reward": 1.7466518580913544,
      "reward_std": 0.17820186354219913,
      "rewards/accuracy_reward": 0.7522321492433548,
      "rewards/format_reward": 0.9944196343421936,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.1663208007812,
      "epoch": 0.9104477611940298,
      "grad_norm": 0.08891814202070236,
      "learning_rate": 6.669034296168854e-07,
      "loss": 0.0319,
      "num_tokens": 39453897.0,
      "reward": 1.709821492433548,
      "reward_std": 0.14528072997927666,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9910714253783226,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.7120742797852,
      "epoch": 0.9253731343283582,
      "grad_norm": 0.18538488447666168,
      "learning_rate": 6.545084971874736e-07,
      "loss": 0.0169,
      "num_tokens": 40128303.0,
      "reward": 1.7712054252624512,
      "reward_std": 0.16406728280708194,
      "rewards/accuracy_reward": 0.7812499925494194,
      "rewards/format_reward": 0.9899553433060646,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 678.2556076049805,
      "epoch": 0.9402985074626866,
      "grad_norm": 0.1315467208623886,
      "learning_rate": 6.420076723519614e-07,
      "loss": 0.0448,
      "num_tokens": 40865532.0,
      "reward": 1.7254465073347092,
      "reward_std": 0.2296902798116207,
      "rewards/accuracy_reward": 0.7544642984867096,
      "rewards/format_reward": 0.9709821417927742,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.9475708007812,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.1169319748878479,
      "learning_rate": 6.294095225512604e-07,
      "loss": 0.0223,
      "num_tokens": 41525429.0,
      "reward": 1.7522322088479996,
      "reward_std": 0.18113385140895844,
      "rewards/accuracy_reward": 0.7622767761349678,
      "rewards/format_reward": 0.9899553507566452,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.311408996582,
      "epoch": 0.9701492537313433,
      "grad_norm": 0.12513795495033264,
      "learning_rate": 6.167226819279527e-07,
      "loss": 0.0178,
      "num_tokens": 42192916.0,
      "reward": 1.7477679252624512,
      "reward_std": 0.14999555610120296,
      "rewards/accuracy_reward": 0.753348208963871,
      "rewards/format_reward": 0.9944196343421936,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.2395858764648,
      "epoch": 0.9850746268656716,
      "grad_norm": 0.6537142992019653,
      "learning_rate": 6.039558454088795e-07,
      "loss": 0.0283,
      "num_tokens": 42898514.0,
      "reward": 1.733258992433548,
      "reward_std": 0.15077468007802963,
      "rewards/accuracy_reward": 0.7455357164144516,
      "rewards/format_reward": 0.9877232164144516,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.8482437133789,
      "epoch": 1.0149253731343284,
      "grad_norm": 0.09364171326160431,
      "learning_rate": 5.911177627460738e-07,
      "loss": 0.0312,
      "num_tokens": 43570490.0,
      "reward": 1.7712054550647736,
      "reward_std": 0.16040645446628332,
      "rewards/accuracy_reward": 0.7924107164144516,
      "rewards/format_reward": 0.9787946343421936,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 664.5725784301758,
      "epoch": 1.0298507462686568,
      "grad_norm": 0.09059074521064758,
      "learning_rate": 5.782172325201155e-07,
      "loss": 0.04,
      "num_tokens": 44298899.0,
      "reward": 1.7410715073347092,
      "reward_std": 0.1766328364610672,
      "rewards/accuracy_reward": 0.7555803656578064,
      "rewards/format_reward": 0.9854910671710968,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.5100708007812,
      "epoch": 1.044776119402985,
      "grad_norm": 0.10916374623775482,
      "learning_rate": 5.652630961100258e-07,
      "loss": 0.0065,
      "num_tokens": 44959604.0,
      "reward": 1.7511161714792252,
      "reward_std": 0.14327580528333783,
      "rewards/accuracy_reward": 0.754464291036129,
      "rewards/format_reward": 0.9966517835855484,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.0524826049805,
      "epoch": 1.0597014925373134,
      "grad_norm": 0.11315930634737015,
      "learning_rate": 5.522642316338268e-07,
      "loss": 0.0129,
      "num_tokens": 45635299.0,
      "reward": 1.7700893580913544,
      "reward_std": 0.17250201851129532,
      "rewards/accuracy_reward": 0.776785708963871,
      "rewards/format_reward": 0.9933035671710968,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.2611808776855,
      "epoch": 1.0746268656716418,
      "grad_norm": 0.12196239084005356,
      "learning_rate": 5.392295478639225e-07,
      "loss": 0.0234,
      "num_tokens": 46305437.0,
      "reward": 1.7700893729925156,
      "reward_std": 0.15816278103739023,
      "rewards/accuracy_reward": 0.7756696417927742,
      "rewards/format_reward": 0.9944196417927742,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 642.2678756713867,
      "epoch": 1.0895522388059702,
      "grad_norm": 0.11447075009346008,
      "learning_rate": 5.26167978121472e-07,
      "loss": 0.0379,
      "num_tokens": 47003701.0,
      "reward": 1.75334832072258,
      "reward_std": 0.17948786355555058,
      "rewards/accuracy_reward": 0.7667410746216774,
      "rewards/format_reward": 0.9866071343421936,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.4107437133789,
      "epoch": 1.1044776119402986,
      "grad_norm": 0.11286026239395142,
      "learning_rate": 5.130884741539366e-07,
      "loss": 0.029,
      "num_tokens": 47710189.0,
      "reward": 1.6953125894069672,
      "reward_std": 0.20528794033452868,
      "rewards/accuracy_reward": 0.7042410671710968,
      "rewards/format_reward": 0.991071417927742,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 616.1919898986816,
      "epoch": 1.1194029850746268,
      "grad_norm": 0.11942121386528015,
      "learning_rate": 5e-07,
      "loss": 0.019,
      "num_tokens": 48375529.0,
      "reward": 1.7991072237491608,
      "reward_std": 0.17037044698372483,
      "rewards/accuracy_reward": 0.8058035746216774,
      "rewards/format_reward": 0.9933035671710968,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.8359680175781,
      "epoch": 1.1343283582089552,
      "grad_norm": 0.12714813649654388,
      "learning_rate": 4.869115258460634e-07,
      "loss": 0.0344,
      "num_tokens": 49061558.0,
      "reward": 1.7343750894069672,
      "reward_std": 0.17231870535761118,
      "rewards/accuracy_reward": 0.7488839253783226,
      "rewards/format_reward": 0.9854910597205162,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 593.3370780944824,
      "epoch": 1.1492537313432836,
      "grad_norm": 0.09834135323762894,
      "learning_rate": 4.7383202187852804e-07,
      "loss": 0.0158,
      "num_tokens": 49720004.0,
      "reward": 1.8203125894069672,
      "reward_std": 0.12710504699498415,
      "rewards/accuracy_reward": 0.8325892984867096,
      "rewards/format_reward": 0.987723208963871,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 532.5000267028809,
      "epoch": 1.164179104477612,
      "grad_norm": 0.10094507038593292,
      "learning_rate": 4.6077045213607755e-07,
      "loss": 0.0103,
      "num_tokens": 50322556.0,
      "reward": 1.845982238650322,
      "reward_std": 0.10092606954276562,
      "rewards/accuracy_reward": 0.8459821417927742,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 619.5044860839844,
      "epoch": 1.1791044776119404,
      "grad_norm": 0.09271937608718872,
      "learning_rate": 4.477357683661733e-07,
      "loss": 0.0121,
      "num_tokens": 51004440.0,
      "reward": 1.712053656578064,
      "reward_std": 0.13478131592273712,
      "rewards/accuracy_reward": 0.7243303582072258,
      "rewards/format_reward": 0.987723208963871,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 661.3080673217773,
      "epoch": 1.1940298507462686,
      "grad_norm": 0.1302761286497116,
      "learning_rate": 4.347369038899743e-07,
      "loss": 0.0196,
      "num_tokens": 51728116.0,
      "reward": 1.7321429401636124,
      "reward_std": 0.16369346249848604,
      "rewards/accuracy_reward": 0.7399553507566452,
      "rewards/format_reward": 0.9921875,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 673.4252510070801,
      "epoch": 1.208955223880597,
      "grad_norm": 0.10427302122116089,
      "learning_rate": 4.2178276747988444e-07,
      "loss": 0.0256,
      "num_tokens": 52457921.0,
      "reward": 1.6930804401636124,
      "reward_std": 0.13809803500771523,
      "rewards/accuracy_reward": 0.7008928582072258,
      "rewards/format_reward": 0.9921874925494194,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.3080596923828,
      "epoch": 1.2238805970149254,
      "grad_norm": 0.09508039057254791,
      "learning_rate": 4.0888223725392624e-07,
      "loss": 0.0227,
      "num_tokens": 53149381.0,
      "reward": 1.742187574505806,
      "reward_std": 0.147241884842515,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.9921874850988388,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.2265930175781,
      "epoch": 1.2388059701492538,
      "grad_norm": 0.09547509253025055,
      "learning_rate": 3.960441545911204e-07,
      "loss": 0.0134,
      "num_tokens": 53826080.0,
      "reward": 1.7991072088479996,
      "reward_std": 0.1459453795105219,
      "rewards/accuracy_reward": 0.8002232015132904,
      "rewards/format_reward": 0.9988839253783226,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 634.0826187133789,
      "epoch": 1.2537313432835822,
      "grad_norm": 0.0878840833902359,
      "learning_rate": 3.8327731807204744e-07,
      "loss": 0.0183,
      "num_tokens": 54522986.0,
      "reward": 1.7377232760190964,
      "reward_std": 0.17751624016091228,
      "rewards/accuracy_reward": 0.745535708963871,
      "rewards/format_reward": 0.9921874925494194,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.9698944091797,
      "epoch": 1.2686567164179103,
      "grad_norm": 0.1188497319817543,
      "learning_rate": 3.7059047744873955e-07,
      "loss": 0.0205,
      "num_tokens": 55224271.0,
      "reward": 1.7555804401636124,
      "reward_std": 0.16282380744814873,
      "rewards/accuracy_reward": 0.761160708963871,
      "rewards/format_reward": 0.9944196417927742,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.5402069091797,
      "epoch": 1.2835820895522387,
      "grad_norm": 0.0854121670126915,
      "learning_rate": 3.5799232764803867e-07,
      "loss": 0.0124,
      "num_tokens": 55889819.0,
      "reward": 1.8024554699659348,
      "reward_std": 0.12497595604509115,
      "rewards/accuracy_reward": 0.8035714328289032,
      "rewards/format_reward": 0.9988839253783226,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.6161003112793,
      "epoch": 1.2985074626865671,
      "grad_norm": 0.1087036058306694,
      "learning_rate": 3.454915028125263e-07,
      "loss": 0.0278,
      "num_tokens": 56559627.0,
      "reward": 1.710937574505806,
      "reward_std": 0.15050204377621412,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9921874925494194,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.1116409301758,
      "epoch": 1.3134328358208955,
      "grad_norm": 0.09039624035358429,
      "learning_rate": 3.330965703831146e-07,
      "loss": 0.0385,
      "num_tokens": 57282351.0,
      "reward": 1.7176340222358704,
      "reward_std": 0.16426212899386883,
      "rewards/accuracy_reward": 0.7321428582072258,
      "rewards/format_reward": 0.9854910671710968,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.8861999511719,
      "epoch": 1.328358208955224,
      "grad_norm": 0.09018764644861221,
      "learning_rate": 3.2081602522734985e-07,
      "loss": 0.0131,
      "num_tokens": 57939417.0,
      "reward": 1.85491082072258,
      "reward_std": 0.11039695050567389,
      "rewards/accuracy_reward": 0.8560267835855484,
      "rewards/format_reward": 0.9988839253783226,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 578.7924346923828,
      "epoch": 1.3432835820895521,
      "grad_norm": 0.12793904542922974,
      "learning_rate": 3.086582838174551e-07,
      "loss": 0.0431,
      "num_tokens": 58583903.0,
      "reward": 1.77678582072258,
      "reward_std": 0.1774543970823288,
      "rewards/accuracy_reward": 0.7834821417927742,
      "rewards/format_reward": 0.9933035597205162,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 635.9888610839844,
      "epoch": 1.3582089552238805,
      "grad_norm": 0.10590506345033646,
      "learning_rate": 2.9663167846209996e-07,
      "loss": 0.0308,
      "num_tokens": 59286877.0,
      "reward": 1.705357238650322,
      "reward_std": 0.16925656888633966,
      "rewards/accuracy_reward": 0.71875,
      "rewards/format_reward": 0.9866071417927742,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 559.2232398986816,
      "epoch": 1.373134328358209,
      "grad_norm": 0.11253120750188828,
      "learning_rate": 2.847444515958523e-07,
      "loss": 0.0138,
      "num_tokens": 59920069.0,
      "reward": 1.8337054401636124,
      "reward_std": 0.1393702062778175,
      "rewards/accuracy_reward": 0.8370535671710968,
      "rewards/format_reward": 0.9966517761349678,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.5044937133789,
      "epoch": 1.3880597014925373,
      "grad_norm": 0.10160024464130402,
      "learning_rate": 2.730047501302266e-07,
      "loss": 0.0299,
      "num_tokens": 60608377.0,
      "reward": 1.7879465222358704,
      "reward_std": 0.1563611626625061,
      "rewards/accuracy_reward": 0.8024553507566452,
      "rewards/format_reward": 0.9854910671710968,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.7399826049805,
      "epoch": 1.4029850746268657,
      "grad_norm": 0.09554547071456909,
      "learning_rate": 2.6142061987019574e-07,
      "loss": 0.0153,
      "num_tokens": 61288944.0,
      "reward": 1.783482238650322,
      "reward_std": 0.1566515825688839,
      "rewards/accuracy_reward": 0.7890625,
      "rewards/format_reward": 0.9944196417927742,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.5256996154785,
      "epoch": 1.417910447761194,
      "grad_norm": 0.08792009949684143,
      "learning_rate": 2.500000000000001e-07,
      "loss": 0.0273,
      "num_tokens": 61952575.0,
      "reward": 1.7310268878936768,
      "reward_std": 0.14925330318510532,
      "rewards/accuracy_reward": 0.7377232164144516,
      "rewards/format_reward": 0.9933035671710968,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.5502471923828,
      "epoch": 1.4328358208955223,
      "grad_norm": 0.12849754095077515,
      "learning_rate": 2.387507176420256e-07,
      "loss": 0.0301,
      "num_tokens": 62647396.0,
      "reward": 1.7087054401636124,
      "reward_std": 0.1948286723345518,
      "rewards/accuracy_reward": 0.7310267835855484,
      "rewards/format_reward": 0.9776785597205162,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.8973579406738,
      "epoch": 1.4477611940298507,
      "grad_norm": 0.30540189146995544,
      "learning_rate": 2.2768048249248644e-07,
      "loss": 0.0235,
      "num_tokens": 63355344.0,
      "reward": 1.7477679252624512,
      "reward_std": 0.16450455971062183,
      "rewards/accuracy_reward": 0.7533482164144516,
      "rewards/format_reward": 0.9944196343421936,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.8024864196777,
      "epoch": 1.462686567164179,
      "grad_norm": 0.09905046969652176,
      "learning_rate": 2.167968815375837e-07,
      "loss": 0.0151,
      "num_tokens": 64012431.0,
      "reward": 1.7756697088479996,
      "reward_std": 0.13652249239385128,
      "rewards/accuracy_reward": 0.78125,
      "rewards/format_reward": 0.9944196417927742,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 658.9587326049805,
      "epoch": 1.4776119402985075,
      "grad_norm": 0.10181548446416855,
      "learning_rate": 2.0610737385376348e-07,
      "loss": 0.0381,
      "num_tokens": 64744754.0,
      "reward": 1.699776828289032,
      "reward_std": 0.19275180995464325,
      "rewards/accuracy_reward": 0.7098214253783226,
      "rewards/format_reward": 0.9899553507566452,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 672.4922180175781,
      "epoch": 1.4925373134328357,
      "grad_norm": 0.10780028253793716,
      "learning_rate": 1.9561928549563966e-07,
      "loss": 0.0107,
      "num_tokens": 65485395.0,
      "reward": 1.633928656578064,
      "reward_std": 0.1676093488931656,
      "rewards/accuracy_reward": 0.6350446492433548,
      "rewards/format_reward": 0.9988839253783226,
      "step": 99
    },
    {
      "epoch": 1.5074626865671643,
      "grad_norm": 0.09826894104480743,
      "learning_rate": 1.8533980447508135e-07,
      "loss": 0.0277,
      "step": 100
    },
    {
      "epoch": 1.5074626865671643,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 608.394744617313,
      "eval_loss": 0.02166515402495861,
      "eval_num_tokens": 66141067.0,
      "eval_reward": 1.7128642008291277,
      "eval_reward_std": 0.17956038917201525,
      "eval_rewards/accuracy_reward": 0.7203461689323021,
      "eval_rewards/format_reward": 0.99251795181349,
      "eval_runtime": 7500.3981,
      "eval_samples_per_second": 0.667,
      "eval_steps_per_second": 0.006,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.5853996276855,
      "epoch": 1.5223880597014925,
      "grad_norm": 0.09803326427936554,
      "learning_rate": 1.7527597583490823e-07,
      "loss": 0.0144,
      "num_tokens": 66825204.0,
      "reward": 1.758370615541935,
      "reward_std": 0.14415120193734765,
      "rewards/accuracy_reward": 0.7678571417927742,
      "rewards/format_reward": 0.9905133880674839,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 589.2846298217773,
      "epoch": 1.537313432835821,
      "grad_norm": 0.09236069023609161,
      "learning_rate": 1.6543469682057104e-07,
      "loss": 0.023,
      "num_tokens": 67489387.0,
      "reward": 1.7488840073347092,
      "reward_std": 0.12999783549457788,
      "rewards/accuracy_reward": 0.7578124925494194,
      "rewards/format_reward": 0.9910714253783226,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 631.8917617797852,
      "epoch": 1.5522388059701493,
      "grad_norm": 0.09945357590913773,
      "learning_rate": 1.5582271215312293e-07,
      "loss": 0.0348,
      "num_tokens": 68182122.0,
      "reward": 1.7488840073347092,
      "reward_std": 0.14755096472799778,
      "rewards/accuracy_reward": 0.7589285746216774,
      "rewards/format_reward": 0.9899553433060646,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.780158996582,
      "epoch": 1.5671641791044775,
      "grad_norm": 0.1044122576713562,
      "learning_rate": 1.4644660940672627e-07,
      "loss": 0.0259,
      "num_tokens": 68843053.0,
      "reward": 1.765625074505806,
      "reward_std": 0.15429440513253212,
      "rewards/accuracy_reward": 0.7834821417927742,
      "rewards/format_reward": 0.9821428582072258,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.1797103881836,
      "epoch": 1.582089552238806,
      "grad_norm": 0.11087001115083694,
      "learning_rate": 1.3731281449385628e-07,
      "loss": 0.0144,
      "num_tokens": 69536934.0,
      "reward": 1.7779018729925156,
      "reward_std": 0.1481210682541132,
      "rewards/accuracy_reward": 0.7823660671710968,
      "rewards/format_reward": 0.995535708963871,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 660.2511444091797,
      "epoch": 1.5970149253731343,
      "grad_norm": 0.09636900573968887,
      "learning_rate": 1.284275872613028e-07,
      "loss": 0.0152,
      "num_tokens": 70259407.0,
      "reward": 1.7109375894069672,
      "reward_std": 0.17005854472517967,
      "rewards/accuracy_reward": 0.7220982164144516,
      "rewards/format_reward": 0.9888392761349678,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 674.9152069091797,
      "epoch": 1.6119402985074627,
      "grad_norm": 0.10703866928815842,
      "learning_rate": 1.1979701719998454e-07,
      "loss": 0.0272,
      "num_tokens": 70998891.0,
      "reward": 1.7343750894069672,
      "reward_std": 0.18828168138861656,
      "rewards/accuracy_reward": 0.7444196343421936,
      "rewards/format_reward": 0.9899553433060646,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 604.8225708007812,
      "epoch": 1.626865671641791,
      "grad_norm": 0.107986681163311,
      "learning_rate": 1.1142701927151454e-07,
      "loss": 0.0242,
      "num_tokens": 71660740.0,
      "reward": 1.805803656578064,
      "reward_std": 0.1751266662031412,
      "rewards/accuracy_reward": 0.8125000074505806,
      "rewards/format_reward": 0.9933035522699356,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.0335083007812,
      "epoch": 1.6417910447761193,
      "grad_norm": 0.09717655181884766,
      "learning_rate": 1.0332332985438247e-07,
      "loss": 0.0221,
      "num_tokens": 72367578.0,
      "reward": 1.7700893580913544,
      "reward_std": 0.16510999109596014,
      "rewards/accuracy_reward": 0.7779017835855484,
      "rewards/format_reward": 0.9921874850988388,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 596.9654197692871,
      "epoch": 1.6567164179104479,
      "grad_norm": 0.09363789856433868,
      "learning_rate": 9.549150281252632e-08,
      "loss": 0.0057,
      "num_tokens": 73028371.0,
      "reward": 1.7689733058214188,
      "reward_std": 0.1158353891223669,
      "rewards/accuracy_reward": 0.7745535671710968,
      "rewards/format_reward": 0.9944196417927742,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.8337478637695,
      "epoch": 1.671641791044776,
      "grad_norm": 0.10138875991106033,
      "learning_rate": 8.793690568899215e-08,
      "loss": 0.0239,
      "num_tokens": 73731510.0,
      "reward": 1.7578125894069672,
      "reward_std": 0.15553954057395458,
      "rewards/accuracy_reward": 0.762276791036129,
      "rewards/format_reward": 0.995535708963871,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.131721496582,
      "epoch": 1.6865671641791045,
      "grad_norm": 0.08721350878477097,
      "learning_rate": 8.066471602728803e-08,
      "loss": 0.0261,
      "num_tokens": 74432588.0,
      "reward": 1.7589286416769028,
      "reward_std": 0.14256859384477139,
      "rewards/accuracy_reward": 0.7712053507566452,
      "rewards/format_reward": 0.9877232164144516,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.5703277587891,
      "epoch": 1.7014925373134329,
      "grad_norm": 0.10357489436864853,
      "learning_rate": 7.36799178229539e-08,
      "loss": 0.0318,
      "num_tokens": 75104779.0,
      "reward": 1.7220982760190964,
      "reward_std": 0.16208719182759523,
      "rewards/accuracy_reward": 0.7377232164144516,
      "rewards/format_reward": 0.984375,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.2377395629883,
      "epoch": 1.716417910447761,
      "grad_norm": 0.09729992598295212,
      "learning_rate": 6.698729810778064e-08,
      "loss": 0.015,
      "num_tokens": 75758304.0,
      "reward": 1.7935268580913544,
      "reward_std": 0.1367027312517166,
      "rewards/accuracy_reward": 0.7957589328289032,
      "rewards/format_reward": 0.9977678582072258,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.0134124755859,
      "epoch": 1.7313432835820897,
      "grad_norm": 0.10120779275894165,
      "learning_rate": 6.059144366901736e-08,
      "loss": 0.0228,
      "num_tokens": 76463884.0,
      "reward": 1.7354911714792252,
      "reward_std": 0.14026323426514864,
      "rewards/accuracy_reward": 0.7488839253783226,
      "rewards/format_reward": 0.9866071343421936,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.9464569091797,
      "epoch": 1.7462686567164178,
      "grad_norm": 0.08987674117088318,
      "learning_rate": 5.44967379058161e-08,
      "loss": 0.0337,
      "num_tokens": 77102308.0,
      "reward": 1.7555804550647736,
      "reward_std": 0.1271761180832982,
      "rewards/accuracy_reward": 0.762276791036129,
      "rewards/format_reward": 0.9933035597205162,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.6161041259766,
      "epoch": 1.7611940298507462,
      "grad_norm": 0.11021538078784943,
      "learning_rate": 4.870735782506979e-08,
      "loss": 0.0105,
      "num_tokens": 77835332.0,
      "reward": 1.7142857909202576,
      "reward_std": 0.18193747848272324,
      "rewards/accuracy_reward": 0.7165178582072258,
      "rewards/format_reward": 0.9977678507566452,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.1942176818848,
      "epoch": 1.7761194029850746,
      "grad_norm": 0.12281131744384766,
      "learning_rate": 4.322727117869951e-08,
      "loss": 0.0103,
      "num_tokens": 78535354.0,
      "reward": 1.7522322237491608,
      "reward_std": 0.15158831980079412,
      "rewards/accuracy_reward": 0.7555803507566452,
      "rewards/format_reward": 0.9966517835855484,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 620.7723617553711,
      "epoch": 1.7910447761194028,
      "grad_norm": 0.09940478950738907,
      "learning_rate": 3.806023374435663e-08,
      "loss": 0.0351,
      "num_tokens": 79216974.0,
      "reward": 1.7254465073347092,
      "reward_std": 0.16859996505081654,
      "rewards/accuracy_reward": 0.7366071492433548,
      "rewards/format_reward": 0.9888392835855484,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.8460083007812,
      "epoch": 1.8059701492537314,
      "grad_norm": 0.09752647578716278,
      "learning_rate": 3.3209786751399184e-08,
      "loss": 0.0242,
      "num_tokens": 79931012.0,
      "reward": 1.7723215073347092,
      "reward_std": 0.17135184817016125,
      "rewards/accuracy_reward": 0.7845982164144516,
      "rewards/format_reward": 0.9877232164144516,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 605.100471496582,
      "epoch": 1.8208955223880596,
      "grad_norm": 0.09651315957307816,
      "learning_rate": 2.8679254453910785e-08,
      "loss": 0.025,
      "num_tokens": 80603486.0,
      "reward": 1.7366072237491608,
      "reward_std": 0.14862325228750706,
      "rewards/accuracy_reward": 0.7410714328289032,
      "rewards/format_reward": 0.9955357164144516,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 651.4765853881836,
      "epoch": 1.835820895522388,
      "grad_norm": 0.09396825730800629,
      "learning_rate": 2.4471741852423233e-08,
      "loss": 0.0172,
      "num_tokens": 81334001.0,
      "reward": 1.7354911714792252,
      "reward_std": 0.16738921124488115,
      "rewards/accuracy_reward": 0.7511160746216774,
      "rewards/format_reward": 0.9843749850988388,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.3326187133789,
      "epoch": 1.8507462686567164,
      "grad_norm": 0.1153474748134613,
      "learning_rate": 2.0590132565903473e-08,
      "loss": 0.0411,
      "num_tokens": 82038987.0,
      "reward": 1.7734375894069672,
      "reward_std": 0.2178222257643938,
      "rewards/accuracy_reward": 0.7868303507566452,
      "rewards/format_reward": 0.9866071417927742,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.5000228881836,
      "epoch": 1.8656716417910446,
      "grad_norm": 0.10678599029779434,
      "learning_rate": 1.7037086855465898e-08,
      "loss": 0.0201,
      "num_tokens": 82700603.0,
      "reward": 1.7354911714792252,
      "reward_std": 0.1632389174774289,
      "rewards/accuracy_reward": 0.7421874925494194,
      "rewards/format_reward": 0.9933035597205162,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.1205596923828,
      "epoch": 1.8805970149253732,
      "grad_norm": 0.08962120860815048,
      "learning_rate": 1.3815039801161722e-08,
      "loss": 0.0185,
      "num_tokens": 83404503.0,
      "reward": 1.750000074505806,
      "reward_std": 0.15682824235409498,
      "rewards/accuracy_reward": 0.7566964253783226,
      "rewards/format_reward": 0.9933035671710968,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.2109603881836,
      "epoch": 1.8955223880597014,
      "grad_norm": 0.10807648301124573,
      "learning_rate": 1.0926199633097154e-08,
      "loss": 0.0042,
      "num_tokens": 84037108.0,
      "reward": 1.8046875894069672,
      "reward_std": 0.13849271647632122,
      "rewards/accuracy_reward": 0.8080357164144516,
      "rewards/format_reward": 0.9966517835855484,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.560302734375,
      "epoch": 1.9104477611940298,
      "grad_norm": 0.10242209583520889,
      "learning_rate": 8.372546218022746e-09,
      "loss": 0.0202,
      "num_tokens": 84741714.0,
      "reward": 1.7120536714792252,
      "reward_std": 0.158757739700377,
      "rewards/accuracy_reward": 0.7187500074505806,
      "rewards/format_reward": 0.9933035671710968,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.8248062133789,
      "epoch": 1.9253731343283582,
      "grad_norm": 0.09526298940181732,
      "learning_rate": 6.15582970243117e-09,
      "loss": 0.0091,
      "num_tokens": 85407733.0,
      "reward": 1.8069197237491608,
      "reward_std": 0.11288107186555862,
      "rewards/accuracy_reward": 0.8069196492433548,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.9765853881836,
      "epoch": 1.9402985074626866,
      "grad_norm": 0.09915687888860703,
      "learning_rate": 4.277569313094809e-09,
      "loss": 0.0311,
      "num_tokens": 86108960.0,
      "reward": 1.7622768729925156,
      "reward_std": 0.1561026033014059,
      "rewards/accuracy_reward": 0.7756696343421936,
      "rewards/format_reward": 0.9866071343421936,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 646.7611846923828,
      "epoch": 1.955223880597015,
      "grad_norm": 0.09854050725698471,
      "learning_rate": 2.739052315863355e-09,
      "loss": 0.007,
      "num_tokens": 86830202.0,
      "reward": 1.6919643580913544,
      "reward_std": 0.14856339246034622,
      "rewards/accuracy_reward": 0.6953125074505806,
      "rewards/format_reward": 0.9966517835855484,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.3638725280762,
      "epoch": 1.9701492537313432,
      "grad_norm": 0.11233126372098923,
      "learning_rate": 1.541333133436018e-09,
      "loss": -0.002,
      "num_tokens": 87492608.0,
      "reward": 1.7845982909202576,
      "reward_std": 0.14027062244713306,
      "rewards/accuracy_reward": 0.7868303582072258,
      "rewards/format_reward": 0.9977678507566452,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.9294013977051,
      "epoch": 1.9850746268656716,
      "grad_norm": 0.10069043189287186,
      "learning_rate": 6.852326227130833e-10,
      "loss": 0.0303,
      "num_tokens": 88193234.0,
      "reward": 1.7500000894069672,
      "reward_std": 0.14365093689411879,
      "rewards/accuracy_reward": 0.7555803582072258,
      "rewards/format_reward": 0.9944196343421936,
      "step": 132
    },
    {
      "epoch": 1.9850746268656716,
      "step": 132,
      "total_flos": 0.0,
      "train_loss": 0.023034477807496758,
      "train_runtime": 35029.9588,
      "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
}