{
  "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": 0.6388546824455261,
      "learning_rate": 7.142857142857142e-08,
      "loss": 0.0034,
      "num_tokens": 546936.0,
      "reward": 0.27008930034935474,
      "reward_std": 0.3894983232021332,
      "rewards/accuracy_reward": 0.20535713713616133,
      "rewards/format_reward": 0.06473214365541935,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 471.2355079650879,
      "epoch": 0.029850746268656716,
      "grad_norm": 0.4866682291030884,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0046,
      "num_tokens": 1100635.0,
      "reward": 0.29575894586741924,
      "reward_std": 0.38195205107331276,
      "rewards/accuracy_reward": 0.22991071362048388,
      "rewards/format_reward": 0.06584821548312902,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 501.1919822692871,
      "epoch": 0.04477611940298507,
      "grad_norm": 0.5236084461212158,
      "learning_rate": 2.1428571428571426e-07,
      "loss": -0.0034,
      "num_tokens": 1696607.0,
      "reward": 0.2700893022119999,
      "reward_std": 0.40184900537133217,
      "rewards/accuracy_reward": 0.20982142630964518,
      "rewards/format_reward": 0.06026785750873387,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 527.3426513671875,
      "epoch": 0.05970149253731343,
      "grad_norm": 0.4246870279312134,
      "learning_rate": 2.857142857142857e-07,
      "loss": -0.0055,
      "num_tokens": 2298618.0,
      "reward": 0.2533482275903225,
      "reward_std": 0.3664662688970566,
      "rewards/accuracy_reward": 0.18415178544819355,
      "rewards/format_reward": 0.06919642980210483,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 489.1272506713867,
      "epoch": 0.07462686567164178,
      "grad_norm": 0.5821424126625061,
      "learning_rate": 3.5714285714285716e-07,
      "loss": -0.0015,
      "num_tokens": 2874380.0,
      "reward": 0.26339287497103214,
      "reward_std": 0.35569472424685955,
      "rewards/accuracy_reward": 0.18080356810241938,
      "rewards/format_reward": 0.08258928544819355,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 464.8448905944824,
      "epoch": 0.08955223880597014,
      "grad_norm": 0.933703601360321,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0112,
      "num_tokens": 3415369.0,
      "reward": 0.27678573317825794,
      "reward_std": 0.3908194899559021,
      "rewards/accuracy_reward": 0.19531249813735485,
      "rewards/format_reward": 0.0814732143189758,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 476.73328399658203,
      "epoch": 0.1044776119402985,
      "grad_norm": 0.5595778226852417,
      "learning_rate": 5e-07,
      "loss": 0.008,
      "num_tokens": 3971570.0,
      "reward": 0.3303571604192257,
      "reward_std": 0.4130873791873455,
      "rewards/accuracy_reward": 0.22879464458674192,
      "rewards/format_reward": 0.10156250093132257,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.8995780944824,
      "epoch": 0.11940298507462686,
      "grad_norm": 0.9490161538124084,
      "learning_rate": 5.714285714285714e-07,
      "loss": 0.0037,
      "num_tokens": 4524584.0,
      "reward": 0.31919644586741924,
      "reward_std": 0.40963855013251305,
      "rewards/accuracy_reward": 0.2198660746216774,
      "rewards/format_reward": 0.09933035681024194,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 483.68640518188477,
      "epoch": 0.13432835820895522,
      "grad_norm": 0.6994246244430542,
      "learning_rate": 6.428571428571429e-07,
      "loss": 0.0361,
      "num_tokens": 5083303.0,
      "reward": 0.3627232275903225,
      "reward_std": 0.42251865938305855,
      "rewards/accuracy_reward": 0.22321428172290325,
      "rewards/format_reward": 0.13950892817229033,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 436.486629486084,
      "epoch": 0.14925373134328357,
      "grad_norm": 5.183073043823242,
      "learning_rate": 7.142857142857143e-07,
      "loss": -0.0042,
      "num_tokens": 5599035.0,
      "reward": 0.5290178842842579,
      "reward_std": 0.5277893580496311,
      "rewards/accuracy_reward": 0.25669643096625805,
      "rewards/format_reward": 0.2723214216530323,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.1194496154785,
      "epoch": 0.16417910447761194,
      "grad_norm": 1.4668474197387695,
      "learning_rate": 7.857142857142856e-07,
      "loss": 0.0147,
      "num_tokens": 6198702.0,
      "reward": 0.4129464440047741,
      "reward_std": 0.4812769778072834,
      "rewards/accuracy_reward": 0.19642857182770967,
      "rewards/format_reward": 0.2165178582072258,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.21877670288086,
      "epoch": 0.1791044776119403,
      "grad_norm": 0.7429012060165405,
      "learning_rate": 8.57142857142857e-07,
      "loss": -0.0078,
      "num_tokens": 6765330.0,
      "reward": 0.5502232387661934,
      "reward_std": 0.5404816046357155,
      "rewards/accuracy_reward": 0.2343749962747097,
      "rewards/format_reward": 0.31584821082651615,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.33484268188477,
      "epoch": 0.19402985074626866,
      "grad_norm": 1.6211804151535034,
      "learning_rate": 9.285714285714285e-07,
      "loss": 0.0066,
      "num_tokens": 7323374.0,
      "reward": 0.8258928880095482,
      "reward_std": 0.5914651453495026,
      "rewards/accuracy_reward": 0.24665178544819355,
      "rewards/format_reward": 0.5792410671710968,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 431.96653747558594,
      "epoch": 0.208955223880597,
      "grad_norm": 2.787757158279419,
      "learning_rate": 1e-06,
      "loss": 0.0041,
      "num_tokens": 7834288.0,
      "reward": 0.929687537252903,
      "reward_std": 0.5431078746914864,
      "rewards/accuracy_reward": 0.2343749962747097,
      "rewards/format_reward": 0.6953124925494194,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 429.68305587768555,
      "epoch": 0.22388059701492538,
      "grad_norm": 3.2551932334899902,
      "learning_rate": 9.998286624877785e-07,
      "loss": 0.0255,
      "num_tokens": 8333788.0,
      "reward": 0.946428619325161,
      "reward_std": 0.5455193668603897,
      "rewards/accuracy_reward": 0.26674107275903225,
      "rewards/format_reward": 0.6796875074505806,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 478.9319381713867,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.6753084659576416,
      "learning_rate": 9.99314767377287e-07,
      "loss": 0.0273,
      "num_tokens": 8886279.0,
      "reward": 1.0000000447034836,
      "reward_std": 0.5525597706437111,
      "rewards/accuracy_reward": 0.2723214328289032,
      "rewards/format_reward": 0.7276785671710968,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 436.1384086608887,
      "epoch": 0.2537313432835821,
      "grad_norm": 0.4711945354938507,
      "learning_rate": 9.98458666866564e-07,
      "loss": 0.0041,
      "num_tokens": 9426795.0,
      "reward": 1.160714328289032,
      "reward_std": 0.49247992783784866,
      "rewards/accuracy_reward": 0.31026785634458065,
      "rewards/format_reward": 0.8504464328289032,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 490.244441986084,
      "epoch": 0.26865671641791045,
      "grad_norm": 6.663273811340332,
      "learning_rate": 9.972609476841365e-07,
      "loss": 0.056,
      "num_tokens": 10002238.0,
      "reward": 1.209821492433548,
      "reward_std": 0.480739738792181,
      "rewards/accuracy_reward": 0.32700892724096775,
      "rewards/format_reward": 0.8828125,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 518.3794822692871,
      "epoch": 0.2835820895522388,
      "grad_norm": 0.7554452419281006,
      "learning_rate": 9.957224306869053e-07,
      "loss": 0.0419,
      "num_tokens": 10591498.0,
      "reward": 1.2845982760190964,
      "reward_std": 0.48824567720294,
      "rewards/accuracy_reward": 0.3928571455180645,
      "rewards/format_reward": 0.8917410671710968,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.5770301818848,
      "epoch": 0.29850746268656714,
      "grad_norm": 28.792333602905273,
      "learning_rate": 9.938441702975689e-07,
      "loss": 0.042,
      "num_tokens": 11189695.0,
      "reward": 1.3225446790456772,
      "reward_std": 0.5205758474767208,
      "rewards/accuracy_reward": 0.4386160671710968,
      "rewards/format_reward": 0.8839285671710968,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 533.4453353881836,
      "epoch": 0.31343283582089554,
      "grad_norm": 59.140052795410156,
      "learning_rate": 9.916274537819773e-07,
      "loss": 0.0209,
      "num_tokens": 11792134.0,
      "reward": 1.3314732909202576,
      "reward_std": 0.4882611520588398,
      "rewards/accuracy_reward": 0.4285714253783226,
      "rewards/format_reward": 0.9029017984867096,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 499.65515518188477,
      "epoch": 0.3283582089552239,
      "grad_norm": 0.5003321766853333,
      "learning_rate": 9.890738003669027e-07,
      "loss": 0.0474,
      "num_tokens": 12369289.0,
      "reward": 1.3895089775323868,
      "reward_std": 0.44778596609830856,
      "rewards/accuracy_reward": 0.4687499962747097,
      "rewards/format_reward": 0.920758917927742,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.9509162902832,
      "epoch": 0.34328358208955223,
      "grad_norm": 0.4391040802001953,
      "learning_rate": 9.861849601988383e-07,
      "loss": 0.0479,
      "num_tokens": 12935277.0,
      "reward": 1.4073661267757416,
      "reward_std": 0.4340375326573849,
      "rewards/accuracy_reward": 0.4933035746216774,
      "rewards/format_reward": 0.9140625,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.5480155944824,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.24674437940120697,
      "learning_rate": 9.82962913144534e-07,
      "loss": 0.0413,
      "num_tokens": 13558416.0,
      "reward": 1.4810268431901932,
      "reward_std": 0.44437726214528084,
      "rewards/accuracy_reward": 0.5558035634458065,
      "rewards/format_reward": 0.9252232164144516,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 528.6674461364746,
      "epoch": 0.373134328358209,
      "grad_norm": 0.31784048676490784,
      "learning_rate": 9.794098674340966e-07,
      "loss": -0.004,
      "num_tokens": 14147262.0,
      "reward": 1.577008992433548,
      "reward_std": 0.41776228323578835,
      "rewards/accuracy_reward": 0.6395089216530323,
      "rewards/format_reward": 0.9375,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.8694534301758,
      "epoch": 0.3880597014925373,
      "grad_norm": 0.24819239974021912,
      "learning_rate": 9.755282581475767e-07,
      "loss": 0.0552,
      "num_tokens": 14812489.0,
      "reward": 1.500000074505806,
      "reward_std": 0.40182342007756233,
      "rewards/accuracy_reward": 0.5624999962747097,
      "rewards/format_reward": 0.9375,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 573.0602912902832,
      "epoch": 0.40298507462686567,
      "grad_norm": 0.22896583378314972,
      "learning_rate": 9.713207455460892e-07,
      "loss": 0.0109,
      "num_tokens": 15452639.0,
      "reward": 1.5904018729925156,
      "reward_std": 0.3602062352001667,
      "rewards/accuracy_reward": 0.6305803582072258,
      "rewards/format_reward": 0.9598214253783226,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.6428909301758,
      "epoch": 0.417910447761194,
      "grad_norm": 0.19549427926540375,
      "learning_rate": 9.667902132486008e-07,
      "loss": 0.0429,
      "num_tokens": 16084223.0,
      "reward": 1.6183036714792252,
      "reward_std": 0.29753560572862625,
      "rewards/accuracy_reward": 0.6573660783469677,
      "rewards/format_reward": 0.9609375074505806,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.9185562133789,
      "epoch": 0.43283582089552236,
      "grad_norm": 0.16904062032699585,
      "learning_rate": 9.619397662556433e-07,
      "loss": 0.0436,
      "num_tokens": 16802206.0,
      "reward": 1.6104911267757416,
      "reward_std": 0.2882387526333332,
      "rewards/accuracy_reward": 0.6595982238650322,
      "rewards/format_reward": 0.9508928656578064,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.2879753112793,
      "epoch": 0.44776119402985076,
      "grad_norm": 0.16023597121238708,
      "learning_rate": 9.567727288213004e-07,
      "loss": 0.0488,
      "num_tokens": 17482496.0,
      "reward": 1.696428656578064,
      "reward_std": 0.28165779635310173,
      "rewards/accuracy_reward": 0.7321428507566452,
      "rewards/format_reward": 0.9642857164144516,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.2477951049805,
      "epoch": 0.4626865671641791,
      "grad_norm": 0.11776868999004364,
      "learning_rate": 9.512926421749303e-07,
      "loss": 0.0302,
      "num_tokens": 18146798.0,
      "reward": 1.670758992433548,
      "reward_std": 0.20003954134881496,
      "rewards/accuracy_reward": 0.6875,
      "rewards/format_reward": 0.983258917927742,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 586.9218940734863,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.12123113870620728,
      "learning_rate": 9.455032620941839e-07,
      "loss": 0.0393,
      "num_tokens": 18801416.0,
      "reward": 1.7265625596046448,
      "reward_std": 0.16540441382676363,
      "rewards/accuracy_reward": 0.7388392835855484,
      "rewards/format_reward": 0.987723208963871,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.9687728881836,
      "epoch": 0.4925373134328358,
      "grad_norm": 0.1264243721961975,
      "learning_rate": 9.394085563309826e-07,
      "loss": 0.0452,
      "num_tokens": 19514676.0,
      "reward": 1.6506696939468384,
      "reward_std": 0.2404237724840641,
      "rewards/accuracy_reward": 0.6830357238650322,
      "rewards/format_reward": 0.967633917927742,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 680.2611846923828,
      "epoch": 0.5074626865671642,
      "grad_norm": 0.13258805871009827,
      "learning_rate": 9.330127018922193e-07,
      "loss": 0.0434,
      "num_tokens": 20254622.0,
      "reward": 1.6941965073347092,
      "reward_std": 0.2184577826410532,
      "rewards/accuracy_reward": 0.7087053507566452,
      "rewards/format_reward": 0.9854910746216774,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.7868690490723,
      "epoch": 0.5223880597014925,
      "grad_norm": 0.13021788001060486,
      "learning_rate": 9.26320082177046e-07,
      "loss": 0.0132,
      "num_tokens": 20959799.0,
      "reward": 1.6986608058214188,
      "reward_std": 0.19432559609413147,
      "rewards/accuracy_reward": 0.7154017835855484,
      "rewards/format_reward": 0.9832589253783226,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.8973579406738,
      "epoch": 0.5373134328358209,
      "grad_norm": 0.11859282851219177,
      "learning_rate": 9.19335283972712e-07,
      "loss": 0.0487,
      "num_tokens": 21697291.0,
      "reward": 1.6863840073347092,
      "reward_std": 0.2162930192425847,
      "rewards/accuracy_reward": 0.7087053656578064,
      "rewards/format_reward": 0.9776785597205162,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.2712287902832,
      "epoch": 0.5522388059701493,
      "grad_norm": 0.11509045958518982,
      "learning_rate": 9.120630943110077e-07,
      "loss": 0.0397,
      "num_tokens": 22375998.0,
      "reward": 1.7232143580913544,
      "reward_std": 0.17825941182672977,
      "rewards/accuracy_reward": 0.7310267761349678,
      "rewards/format_reward": 0.9921874925494194,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 645.5435600280762,
      "epoch": 0.5671641791044776,
      "grad_norm": 0.11072342842817307,
      "learning_rate": 9.045084971874737e-07,
      "loss": 0.0259,
      "num_tokens": 23100189.0,
      "reward": 1.734375074505806,
      "reward_std": 0.1874879952520132,
      "rewards/accuracy_reward": 0.7477678582072258,
      "rewards/format_reward": 0.9866071417927742,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.4375267028809,
      "epoch": 0.582089552238806,
      "grad_norm": 0.10572760552167892,
      "learning_rate": 8.966766701456176e-07,
      "loss": 0.0445,
      "num_tokens": 23799789.0,
      "reward": 1.686383992433548,
      "reward_std": 0.1749917520210147,
      "rewards/accuracy_reward": 0.7020089291036129,
      "rewards/format_reward": 0.984375,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 715.3582916259766,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.09404753893613815,
      "learning_rate": 8.885729807284854e-07,
      "loss": 0.0337,
      "num_tokens": 24564446.0,
      "reward": 1.6919643580913544,
      "reward_std": 0.19988763704895973,
      "rewards/accuracy_reward": 0.7109375,
      "rewards/format_reward": 0.9810267835855484,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 637.3683166503906,
      "epoch": 0.6119402985074627,
      "grad_norm": 0.12231509387493134,
      "learning_rate": 8.802029828000155e-07,
      "loss": 0.029,
      "num_tokens": 25276136.0,
      "reward": 1.6841518431901932,
      "reward_std": 0.18472581543028355,
      "rewards/accuracy_reward": 0.6975446417927742,
      "rewards/format_reward": 0.9866071343421936,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 605.458740234375,
      "epoch": 0.6268656716417911,
      "grad_norm": 0.10407140851020813,
      "learning_rate": 8.71572412738697e-07,
      "loss": 0.0281,
      "num_tokens": 25941107.0,
      "reward": 1.8348215073347092,
      "reward_std": 0.14506788924336433,
      "rewards/accuracy_reward": 0.8448660671710968,
      "rewards/format_reward": 0.9899553433060646,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.060302734375,
      "epoch": 0.6417910447761194,
      "grad_norm": 0.11245384812355042,
      "learning_rate": 8.626871855061437e-07,
      "loss": 0.012,
      "num_tokens": 26636777.0,
      "reward": 1.7611607909202576,
      "reward_std": 0.14432393945753574,
      "rewards/accuracy_reward": 0.7667410671710968,
      "rewards/format_reward": 0.9944196343421936,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 598.0089569091797,
      "epoch": 0.6567164179104478,
      "grad_norm": 0.125792995095253,
      "learning_rate": 8.535533905932737e-07,
      "loss": 0.0321,
      "num_tokens": 27299505.0,
      "reward": 1.7377233058214188,
      "reward_std": 0.15178925171494484,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.9877232164144516,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.6283721923828,
      "epoch": 0.6716417910447762,
      "grad_norm": 0.10002588480710983,
      "learning_rate": 8.441772878468769e-07,
      "loss": 0.0195,
      "num_tokens": 27972836.0,
      "reward": 1.7578125894069672,
      "reward_std": 0.14866859279572964,
      "rewards/accuracy_reward": 0.7656250074505806,
      "rewards/format_reward": 0.9921874925494194,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 704.8728103637695,
      "epoch": 0.6865671641791045,
      "grad_norm": 0.10578148812055588,
      "learning_rate": 8.34565303179429e-07,
      "loss": 0.0408,
      "num_tokens": 28738194.0,
      "reward": 1.678571492433548,
      "reward_std": 0.21444039046764374,
      "rewards/accuracy_reward": 0.7020089328289032,
      "rewards/format_reward": 0.9765624925494194,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 656.7991409301758,
      "epoch": 0.7014925373134329,
      "grad_norm": 0.1114787682890892,
      "learning_rate": 8.247240241650917e-07,
      "loss": 0.0164,
      "num_tokens": 29442742.0,
      "reward": 1.7611607909202576,
      "reward_std": 0.15375812584534287,
      "rewards/accuracy_reward": 0.770089291036129,
      "rewards/format_reward": 0.991071417927742,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 653.5190048217773,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.11586251854896545,
      "learning_rate": 8.146601955249187e-07,
      "loss": 0.0291,
      "num_tokens": 30155111.0,
      "reward": 1.7310268580913544,
      "reward_std": 0.18213129229843616,
      "rewards/accuracy_reward": 0.7477678507566452,
      "rewards/format_reward": 0.983258917927742,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.7009201049805,
      "epoch": 0.7313432835820896,
      "grad_norm": 0.11568515747785568,
      "learning_rate": 8.043807145043603e-07,
      "loss": 0.0263,
      "num_tokens": 30888875.0,
      "reward": 1.7109375596046448,
      "reward_std": 0.21294444054365158,
      "rewards/accuracy_reward": 0.7293097600340843,
      "rewards/format_reward": 0.9888392835855484,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.5681076049805,
      "epoch": 0.746268656716418,
      "grad_norm": 0.10673708468675613,
      "learning_rate": 7.938926261462365e-07,
      "loss": 0.0257,
      "num_tokens": 31583952.0,
      "reward": 1.7031250596046448,
      "reward_std": 0.16654676012694836,
      "rewards/accuracy_reward": 0.7098214253783226,
      "rewards/format_reward": 0.9933035671710968,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.3259201049805,
      "epoch": 0.7611940298507462,
      "grad_norm": 0.12650178372859955,
      "learning_rate": 7.832031184624164e-07,
      "loss": 0.0406,
      "num_tokens": 32274860.0,
      "reward": 1.7377233058214188,
      "reward_std": 0.1876004207879305,
      "rewards/accuracy_reward": 0.75,
      "rewards/format_reward": 0.9877232015132904,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 640.5714530944824,
      "epoch": 0.7761194029850746,
      "grad_norm": 0.10010657459497452,
      "learning_rate": 7.723195175075135e-07,
      "loss": 0.0247,
      "num_tokens": 32977460.0,
      "reward": 1.6852679401636124,
      "reward_std": 0.1428875606507063,
      "rewards/accuracy_reward": 0.7020089253783226,
      "rewards/format_reward": 0.983258917927742,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.3036117553711,
      "epoch": 0.7910447761194029,
      "grad_norm": 0.1098380908370018,
      "learning_rate": 7.612492823579744e-07,
      "loss": 0.0303,
      "num_tokens": 33677052.0,
      "reward": 1.6975446939468384,
      "reward_std": 0.20845805294811726,
      "rewards/accuracy_reward": 0.7053571417927742,
      "rewards/format_reward": 0.9921874850988388,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 597.6797256469727,
      "epoch": 0.8059701492537313,
      "grad_norm": 0.11101842671632767,
      "learning_rate": 7.5e-07,
      "loss": 0.0027,
      "num_tokens": 34347405.0,
      "reward": 1.7656250894069672,
      "reward_std": 0.15746967494487762,
      "rewards/accuracy_reward": 0.770089291036129,
      "rewards/format_reward": 0.995535708963871,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.290210723877,
      "epoch": 0.8208955223880597,
      "grad_norm": 0.10212179273366928,
      "learning_rate": 7.385793801298042e-07,
      "loss": 0.0177,
      "num_tokens": 35035137.0,
      "reward": 1.756696492433548,
      "reward_std": 0.1636977717280388,
      "rewards/accuracy_reward": 0.7683722376823425,
      "rewards/format_reward": 0.995535708963871,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 604.0111885070801,
      "epoch": 0.835820895522388,
      "grad_norm": 0.12705914676189423,
      "learning_rate": 7.269952498697734e-07,
      "loss": 0.0358,
      "num_tokens": 35707387.0,
      "reward": 1.6863840073347092,
      "reward_std": 0.1575933638960123,
      "rewards/accuracy_reward": 0.6908482164144516,
      "rewards/format_reward": 0.995535708963871,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.408504486084,
      "epoch": 0.8507462686567164,
      "grad_norm": 0.11634799093008041,
      "learning_rate": 7.152555484041475e-07,
      "loss": 0.0244,
      "num_tokens": 36362977.0,
      "reward": 1.7946429550647736,
      "reward_std": 0.13385543506592512,
      "rewards/accuracy_reward": 0.7991071417927742,
      "rewards/format_reward": 0.995535708963871,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 620.1696739196777,
      "epoch": 0.8656716417910447,
      "grad_norm": 0.15597358345985413,
      "learning_rate": 7.033683215379002e-07,
      "loss": 0.0285,
      "num_tokens": 37044617.0,
      "reward": 1.7154018878936768,
      "reward_std": 0.1510266363620758,
      "rewards/accuracy_reward": 0.7254464253783226,
      "rewards/format_reward": 0.9899553582072258,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.7701187133789,
      "epoch": 0.8805970149253731,
      "grad_norm": 0.11789924651384354,
      "learning_rate": 6.913417161825449e-07,
      "loss": 0.0313,
      "num_tokens": 37728027.0,
      "reward": 1.7723215073347092,
      "reward_std": 0.1743272840976715,
      "rewards/accuracy_reward": 0.7823660746216774,
      "rewards/format_reward": 0.9899553433060646,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.8102951049805,
      "epoch": 0.8955223880597015,
      "grad_norm": 0.1122424304485321,
      "learning_rate": 6.7918397477265e-07,
      "loss": 0.0096,
      "num_tokens": 38386153.0,
      "reward": 1.7511161416769028,
      "reward_std": 0.1599616724997759,
      "rewards/accuracy_reward": 0.7544642835855484,
      "rewards/format_reward": 0.9966517761349678,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 622.9933395385742,
      "epoch": 0.9104477611940298,
      "grad_norm": 0.10783091187477112,
      "learning_rate": 6.669034296168854e-07,
      "loss": 0.0263,
      "num_tokens": 39086531.0,
      "reward": 1.7232143580913544,
      "reward_std": 0.1590939164161682,
      "rewards/accuracy_reward": 0.7354910671710968,
      "rewards/format_reward": 0.9877232164144516,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 591.811408996582,
      "epoch": 0.9253731343283582,
      "grad_norm": 0.1123439371585846,
      "learning_rate": 6.545084971874736e-07,
      "loss": 0.0166,
      "num_tokens": 39742210.0,
      "reward": 1.8024554401636124,
      "reward_std": 0.14553169254213572,
      "rewards/accuracy_reward": 0.8069196417927742,
      "rewards/format_reward": 0.9955357015132904,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.6227951049805,
      "epoch": 0.9402985074626866,
      "grad_norm": 0.1181529313325882,
      "learning_rate": 6.420076723519614e-07,
      "loss": 0.0333,
      "num_tokens": 40453784.0,
      "reward": 1.751116156578064,
      "reward_std": 0.19144438579678535,
      "rewards/accuracy_reward": 0.7723214253783226,
      "rewards/format_reward": 0.9787946343421936,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.1841850280762,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.11329229176044464,
      "learning_rate": 6.294095225512604e-07,
      "loss": 0.0174,
      "num_tokens": 41096869.0,
      "reward": 1.7622768878936768,
      "reward_std": 0.17155454587191343,
      "rewards/accuracy_reward": 0.7712053582072258,
      "rewards/format_reward": 0.9910714253783226,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 574.6719017028809,
      "epoch": 0.9701492537313433,
      "grad_norm": 0.12106070667505264,
      "learning_rate": 6.167226819279527e-07,
      "loss": 0.0186,
      "num_tokens": 41744967.0,
      "reward": 1.7566965073347092,
      "reward_std": 0.13824818190187216,
      "rewards/accuracy_reward": 0.7633928582072258,
      "rewards/format_reward": 0.9933035671710968,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.8252487182617,
      "epoch": 0.9850746268656716,
      "grad_norm": 0.10446830093860626,
      "learning_rate": 6.039558454088795e-07,
      "loss": 0.0225,
      "num_tokens": 42428760.0,
      "reward": 1.7310268580913544,
      "reward_std": 0.15639658458530903,
      "rewards/accuracy_reward": 0.7399553507566452,
      "rewards/format_reward": 0.9910714253783226,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.428596496582,
      "epoch": 1.0149253731343284,
      "grad_norm": 0.1006438285112381,
      "learning_rate": 5.911177627460738e-07,
      "loss": 0.0184,
      "num_tokens": 43086024.0,
      "reward": 1.7723215073347092,
      "reward_std": 0.17309140413999557,
      "rewards/accuracy_reward": 0.7901785746216774,
      "rewards/format_reward": 0.9821428507566452,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 639.1283798217773,
      "epoch": 1.0298507462686568,
      "grad_norm": 0.10913052409887314,
      "learning_rate": 5.782172325201155e-07,
      "loss": 0.0268,
      "num_tokens": 43791635.0,
      "reward": 1.7321429550647736,
      "reward_std": 0.16498004272580147,
      "rewards/accuracy_reward": 0.7555803656578064,
      "rewards/format_reward": 0.9765625,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.599365234375,
      "epoch": 1.044776119402985,
      "grad_norm": 0.09446422755718231,
      "learning_rate": 5.652630961100258e-07,
      "loss": 0.0132,
      "num_tokens": 44444356.0,
      "reward": 1.757812574505806,
      "reward_std": 0.14226584695279598,
      "rewards/accuracy_reward": 0.7633928507566452,
      "rewards/format_reward": 0.9944196343421936,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.3024826049805,
      "epoch": 1.0597014925373134,
      "grad_norm": 0.15613028407096863,
      "learning_rate": 5.522642316338268e-07,
      "loss": 0.0249,
      "num_tokens": 45096083.0,
      "reward": 1.7767857909202576,
      "reward_std": 0.15461723320186138,
      "rewards/accuracy_reward": 0.7834821492433548,
      "rewards/format_reward": 0.9933035597205162,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.2790412902832,
      "epoch": 1.0746268656716418,
      "grad_norm": 0.1187715157866478,
      "learning_rate": 5.392295478639225e-07,
      "loss": 0.0358,
      "num_tokens": 45740253.0,
      "reward": 1.7812500894069672,
      "reward_std": 0.147844135761261,
      "rewards/accuracy_reward": 0.7935267835855484,
      "rewards/format_reward": 0.987723208963871,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 605.3872985839844,
      "epoch": 1.0895522388059702,
      "grad_norm": 0.12618952989578247,
      "learning_rate": 5.26167978121472e-07,
      "loss": 0.015,
      "num_tokens": 46405472.0,
      "reward": 1.7745536416769028,
      "reward_std": 0.17927894555032253,
      "rewards/accuracy_reward": 0.7801339253783226,
      "rewards/format_reward": 0.994419626891613,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.8515968322754,
      "epoch": 1.1044776119402986,
      "grad_norm": 0.12401104718446732,
      "learning_rate": 5.130884741539366e-07,
      "loss": 0.024,
      "num_tokens": 47088163.0,
      "reward": 1.7075893431901932,
      "reward_std": 0.20213122107088566,
      "rewards/accuracy_reward": 0.7131696417927742,
      "rewards/format_reward": 0.9944196343421936,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.5457916259766,
      "epoch": 1.1194029850746268,
      "grad_norm": 0.14666379988193512,
      "learning_rate": 5e-07,
      "loss": 0.0017,
      "num_tokens": 47723356.0,
      "reward": 1.782366156578064,
      "reward_std": 0.15405129920691252,
      "rewards/accuracy_reward": 0.7901785597205162,
      "rewards/format_reward": 0.9921875,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.9419898986816,
      "epoch": 1.1343283582089552,
      "grad_norm": 0.12369884550571442,
      "learning_rate": 4.869115258460634e-07,
      "loss": 0.0281,
      "num_tokens": 48385288.0,
      "reward": 1.7633929252624512,
      "reward_std": 0.16522505693137646,
      "rewards/accuracy_reward": 0.7712053656578064,
      "rewards/format_reward": 0.9921874925494194,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.2332763671875,
      "epoch": 1.1492537313432836,
      "grad_norm": 0.1021357998251915,
      "learning_rate": 4.7383202187852804e-07,
      "loss": 0.0142,
      "num_tokens": 49023929.0,
      "reward": 1.8303572237491608,
      "reward_std": 0.12529135216027498,
      "rewards/accuracy_reward": 0.8404017761349678,
      "rewards/format_reward": 0.9899553582072258,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 519.0881881713867,
      "epoch": 1.164179104477612,
      "grad_norm": 0.09221172332763672,
      "learning_rate": 4.6077045213607755e-07,
      "loss": 0.0251,
      "num_tokens": 49614464.0,
      "reward": 1.8314733058214188,
      "reward_std": 0.1043315613642335,
      "rewards/accuracy_reward": 0.8415178507566452,
      "rewards/format_reward": 0.9899553433060646,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.8861885070801,
      "epoch": 1.1791044776119404,
      "grad_norm": 0.10308758169412613,
      "learning_rate": 4.477357683661733e-07,
      "loss": 0.0347,
      "num_tokens": 50289522.0,
      "reward": 1.7142857909202576,
      "reward_std": 0.1305637713521719,
      "rewards/accuracy_reward": 0.7287946417927742,
      "rewards/format_reward": 0.9854910746216774,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.0011520385742,
      "epoch": 1.1940298507462686,
      "grad_norm": 0.10439736396074295,
      "learning_rate": 4.347369038899743e-07,
      "loss": 0.0094,
      "num_tokens": 50981563.0,
      "reward": 1.7589286416769028,
      "reward_std": 0.1464823465794325,
      "rewards/accuracy_reward": 0.761160708963871,
      "rewards/format_reward": 0.9977678507566452,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 662.6283645629883,
      "epoch": 1.208955223880597,
      "grad_norm": 0.11504602432250977,
      "learning_rate": 4.2178276747988444e-07,
      "loss": 0.0304,
      "num_tokens": 51701694.0,
      "reward": 1.6930804252624512,
      "reward_std": 0.1656303908675909,
      "rewards/accuracy_reward": 0.7042410746216774,
      "rewards/format_reward": 0.9888392761349678,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.9006958007812,
      "epoch": 1.2238805970149254,
      "grad_norm": 0.1070084497332573,
      "learning_rate": 4.0888223725392624e-07,
      "loss": 0.0111,
      "num_tokens": 52373973.0,
      "reward": 1.7544643580913544,
      "reward_std": 0.11468092259019613,
      "rewards/accuracy_reward": 0.7566964328289032,
      "rewards/format_reward": 0.9977678507566452,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 576.9643096923828,
      "epoch": 1.2388059701492538,
      "grad_norm": 0.12054042518138885,
      "learning_rate": 3.960441545911204e-07,
      "loss": 0.0152,
      "num_tokens": 53028037.0,
      "reward": 1.7879464775323868,
      "reward_std": 0.15595215652137995,
      "rewards/accuracy_reward": 0.7935267761349678,
      "rewards/format_reward": 0.9944196343421936,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 605.6752471923828,
      "epoch": 1.2537313432835822,
      "grad_norm": 0.10490835458040237,
      "learning_rate": 3.8327731807204744e-07,
      "loss": 0.0112,
      "num_tokens": 53699490.0,
      "reward": 1.7388393580913544,
      "reward_std": 0.17471368238329887,
      "rewards/accuracy_reward": 0.7444196343421936,
      "rewards/format_reward": 0.9944196343421936,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.0904235839844,
      "epoch": 1.2686567164179103,
      "grad_norm": 0.12984751164913177,
      "learning_rate": 3.7059047744873955e-07,
      "loss": 0.0298,
      "num_tokens": 54387443.0,
      "reward": 1.7399554550647736,
      "reward_std": 0.17055520229041576,
      "rewards/accuracy_reward": 0.7466517761349678,
      "rewards/format_reward": 0.9933035671710968,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.9888687133789,
      "epoch": 1.2835820895522387,
      "grad_norm": 0.09954959899187088,
      "learning_rate": 3.5799232764803867e-07,
      "loss": 0.0307,
      "num_tokens": 55045329.0,
      "reward": 1.782366156578064,
      "reward_std": 0.14454064890742302,
      "rewards/accuracy_reward": 0.7935267835855484,
      "rewards/format_reward": 0.9888392761349678,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 604.8571624755859,
      "epoch": 1.2985074626865671,
      "grad_norm": 0.09568862617015839,
      "learning_rate": 3.454915028125263e-07,
      "loss": 0.0255,
      "num_tokens": 55709977.0,
      "reward": 1.7232143729925156,
      "reward_std": 0.14090445078909397,
      "rewards/accuracy_reward": 0.7321428582072258,
      "rewards/format_reward": 0.9910714328289032,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.7187881469727,
      "epoch": 1.3134328358208955,
      "grad_norm": 0.09479253739118576,
      "learning_rate": 3.330965703831146e-07,
      "loss": 0.0362,
      "num_tokens": 56409949.0,
      "reward": 1.7209822088479996,
      "reward_std": 0.1328377453610301,
      "rewards/accuracy_reward": 0.7462225183844566,
      "rewards/format_reward": 0.9821428582072258,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.6261444091797,
      "epoch": 1.328358208955224,
      "grad_norm": 0.0859806090593338,
      "learning_rate": 3.2081602522734985e-07,
      "loss": 0.0102,
      "num_tokens": 57056030.0,
      "reward": 1.8861607909202576,
      "reward_std": 0.08803954161703587,
      "rewards/accuracy_reward": 0.8872767761349678,
      "rewards/format_reward": 0.9988839253783226,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 561.8058242797852,
      "epoch": 1.3432835820895521,
      "grad_norm": 0.11504408717155457,
      "learning_rate": 3.086582838174551e-07,
      "loss": 0.0164,
      "num_tokens": 57685296.0,
      "reward": 1.7801340073347092,
      "reward_std": 0.1496917586773634,
      "rewards/accuracy_reward": 0.7834821343421936,
      "rewards/format_reward": 0.9966517761349678,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 604.1942291259766,
      "epoch": 1.3582089552238805,
      "grad_norm": 0.11696956306695938,
      "learning_rate": 2.9663167846209996e-07,
      "loss": 0.0174,
      "num_tokens": 58359782.0,
      "reward": 1.7410715073347092,
      "reward_std": 0.13343470823019743,
      "rewards/accuracy_reward": 0.746651791036129,
      "rewards/format_reward": 0.9944196343421936,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.7087287902832,
      "epoch": 1.373134328358209,
      "grad_norm": 0.10284429043531418,
      "learning_rate": 2.847444515958523e-07,
      "loss": 0.0248,
      "num_tokens": 58984449.0,
      "reward": 1.829241156578064,
      "reward_std": 0.14068395644426346,
      "rewards/accuracy_reward": 0.8348214253783226,
      "rewards/format_reward": 0.9944196417927742,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.9464569091797,
      "epoch": 1.3880597014925373,
      "grad_norm": 0.09784922748804092,
      "learning_rate": 2.730047501302266e-07,
      "loss": 0.0335,
      "num_tokens": 59657025.0,
      "reward": 1.7834822237491608,
      "reward_std": 0.14826951827853918,
      "rewards/accuracy_reward": 0.7979910671710968,
      "rewards/format_reward": 0.9854910597205162,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.3013610839844,
      "epoch": 1.4029850746268657,
      "grad_norm": 0.1128157526254654,
      "learning_rate": 2.6142061987019574e-07,
      "loss": 0.0299,
      "num_tokens": 60321071.0,
      "reward": 1.7957590073347092,
      "reward_std": 0.15062199719250202,
      "rewards/accuracy_reward": 0.8024553582072258,
      "rewards/format_reward": 0.9933035597205162,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.7411003112793,
      "epoch": 1.417910447761194,
      "grad_norm": 0.11607693880796432,
      "learning_rate": 2.500000000000001e-07,
      "loss": 0.0219,
      "num_tokens": 60970559.0,
      "reward": 1.7712054252624512,
      "reward_std": 0.13864925876259804,
      "rewards/accuracy_reward": 0.7779017835855484,
      "rewards/format_reward": 0.9933035597205162,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.8002548217773,
      "epoch": 1.4328358208955223,
      "grad_norm": 0.1278841495513916,
      "learning_rate": 2.387507176420256e-07,
      "loss": 0.0224,
      "num_tokens": 61642308.0,
      "reward": 1.7198661416769028,
      "reward_std": 0.1696074465289712,
      "rewards/accuracy_reward": 0.7332589328289032,
      "rewards/format_reward": 0.9866071417927742,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.9654312133789,
      "epoch": 1.4477611940298507,
      "grad_norm": 0.1310645341873169,
      "learning_rate": 2.2768048249248644e-07,
      "loss": 0.0202,
      "num_tokens": 62334189.0,
      "reward": 1.7812501043081284,
      "reward_std": 0.14759367797523737,
      "rewards/accuracy_reward": 0.7879464253783226,
      "rewards/format_reward": 0.9933035597205162,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 560.3035926818848,
      "epoch": 1.462686567164179,
      "grad_norm": 0.10888738930225372,
      "learning_rate": 2.167968815375837e-07,
      "loss": 0.0074,
      "num_tokens": 62972013.0,
      "reward": 1.7946429252624512,
      "reward_std": 0.12166838068515062,
      "rewards/accuracy_reward": 0.800223208963871,
      "rewards/format_reward": 0.9944196417927742,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 638.4877548217773,
      "epoch": 1.4776119402985075,
      "grad_norm": 0.12343649566173553,
      "learning_rate": 2.0610737385376348e-07,
      "loss": 0.0268,
      "num_tokens": 63685994.0,
      "reward": 1.7287947237491608,
      "reward_std": 0.1671175602823496,
      "rewards/accuracy_reward": 0.7366071492433548,
      "rewards/format_reward": 0.9921874776482582,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 659.0502548217773,
      "epoch": 1.4925373134328357,
      "grad_norm": 0.11799073964357376,
      "learning_rate": 1.9561928549563966e-07,
      "loss": 0.0153,
      "num_tokens": 64414591.0,
      "reward": 1.6685268580913544,
      "reward_std": 0.16868568398058414,
      "rewards/accuracy_reward": 0.6718750037252903,
      "rewards/format_reward": 0.9966517835855484,
      "step": 99
    },
    {
      "epoch": 1.5074626865671643,
      "grad_norm": 0.09030181914567947,
      "learning_rate": 1.8533980447508135e-07,
      "loss": 0.0049,
      "step": 100
    },
    {
      "epoch": 1.5074626865671643,
      "eval_clip_ratio": 0.0,
      "eval_completion_length": 592.329727769564,
      "eval_loss": 0.021697601303458214,
      "eval_num_tokens": 65050155.0,
      "eval_reward": 1.715956783161483,
      "eval_reward_std": 0.17178419383253132,
      "eval_rewards/accuracy_reward": 0.7222416201949785,
      "eval_rewards/format_reward": 0.9937150783045998,
      "eval_runtime": 7214.1655,
      "eval_samples_per_second": 0.693,
      "eval_steps_per_second": 0.006,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 592.2946720123291,
      "epoch": 1.5223880597014925,
      "grad_norm": 0.12870648503303528,
      "learning_rate": 1.7527597583490823e-07,
      "loss": 0.0078,
      "num_tokens": 65721623.0,
      "reward": 1.771205447614193,
      "reward_std": 0.13381139980629086,
      "rewards/accuracy_reward": 0.7804344072937965,
      "rewards/format_reward": 0.9944196417927742,
      "step": 101
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.9241256713867,
      "epoch": 1.537313432835821,
      "grad_norm": 0.1074695810675621,
      "learning_rate": 1.6543469682057104e-07,
      "loss": 0.0277,
      "num_tokens": 66366667.0,
      "reward": 1.7712054401636124,
      "reward_std": 0.1381669770926237,
      "rewards/accuracy_reward": 0.7790178582072258,
      "rewards/format_reward": 0.9921874925494194,
      "step": 102
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.3058319091797,
      "epoch": 1.5522388059701493,
      "grad_norm": 0.11733277142047882,
      "learning_rate": 1.5582271215312293e-07,
      "loss": 0.016,
      "num_tokens": 67046333.0,
      "reward": 1.7388393580913544,
      "reward_std": 0.13305745087563992,
      "rewards/accuracy_reward": 0.7444196492433548,
      "rewards/format_reward": 0.9944196343421936,
      "step": 103
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.180835723877,
      "epoch": 1.5671641791044775,
      "grad_norm": 0.12533283233642578,
      "learning_rate": 1.4644660940672627e-07,
      "loss": 0.0358,
      "num_tokens": 67691495.0,
      "reward": 1.7756697237491608,
      "reward_std": 0.12638314347714186,
      "rewards/accuracy_reward": 0.7857142761349678,
      "rewards/format_reward": 0.9899553507566452,
      "step": 104
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.5212249755859,
      "epoch": 1.582089552238806,
      "grad_norm": 0.09835629165172577,
      "learning_rate": 1.3731281449385628e-07,
      "loss": 0.0094,
      "num_tokens": 68369554.0,
      "reward": 1.781250074505806,
      "reward_std": 0.1372467102482915,
      "rewards/accuracy_reward": 0.7868303582072258,
      "rewards/format_reward": 0.9944196417927742,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 630.0357513427734,
      "epoch": 1.5970149253731343,
      "grad_norm": 0.11403670161962509,
      "learning_rate": 1.284275872613028e-07,
      "loss": 0.0133,
      "num_tokens": 69064954.0,
      "reward": 1.7243304252624512,
      "reward_std": 0.16303362511098385,
      "rewards/accuracy_reward": 0.7265624925494194,
      "rewards/format_reward": 0.9977678507566452,
      "step": 106
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 657.9710083007812,
      "epoch": 1.6119402985074627,
      "grad_norm": 0.1182536780834198,
      "learning_rate": 1.1979701719998454e-07,
      "loss": 0.0307,
      "num_tokens": 69789256.0,
      "reward": 1.7455357909202576,
      "reward_std": 0.168695330619812,
      "rewards/accuracy_reward": 0.7555803656578064,
      "rewards/format_reward": 0.9899553507566452,
      "step": 107
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 588.330379486084,
      "epoch": 1.626865671641791,
      "grad_norm": 0.11358051002025604,
      "learning_rate": 1.1142701927151454e-07,
      "loss": 0.0274,
      "num_tokens": 70436328.0,
      "reward": 1.805803656578064,
      "reward_std": 0.1442921068519354,
      "rewards/accuracy_reward": 0.8147321492433548,
      "rewards/format_reward": 0.9910714253783226,
      "step": 108
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 617.8203506469727,
      "epoch": 1.6417910447761193,
      "grad_norm": 0.13703642785549164,
      "learning_rate": 1.0332332985438247e-07,
      "loss": 0.0042,
      "num_tokens": 71124159.0,
      "reward": 1.7767858058214188,
      "reward_std": 0.14657451026141644,
      "rewards/accuracy_reward": 0.7790178656578064,
      "rewards/format_reward": 0.9977678507566452,
      "step": 109
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.6573905944824,
      "epoch": 1.6567164179104479,
      "grad_norm": 0.0975952073931694,
      "learning_rate": 9.549150281252632e-08,
      "loss": 0.0203,
      "num_tokens": 71770340.0,
      "reward": 1.7689732909202576,
      "reward_std": 0.12042983993887901,
      "rewards/accuracy_reward": 0.7745535671710968,
      "rewards/format_reward": 0.9944196343421936,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.9743576049805,
      "epoch": 1.671641791044776,
      "grad_norm": 0.10074656456708908,
      "learning_rate": 8.793690568899215e-08,
      "loss": 0.0126,
      "num_tokens": 72464645.0,
      "reward": 1.7578125894069672,
      "reward_std": 0.15030533727258444,
      "rewards/accuracy_reward": 0.7633928507566452,
      "rewards/format_reward": 0.9944196417927742,
      "step": 111
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 634.4721145629883,
      "epoch": 1.6865671641791045,
      "grad_norm": 0.09204529970884323,
      "learning_rate": 8.066471602728803e-08,
      "loss": 0.0321,
      "num_tokens": 73150796.0,
      "reward": 1.7801340222358704,
      "reward_std": 0.12358444835990667,
      "rewards/accuracy_reward": 0.7901785746216774,
      "rewards/format_reward": 0.9899553507566452,
      "step": 112
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 586.6027069091797,
      "epoch": 1.7014925373134329,
      "grad_norm": 0.10714556276798248,
      "learning_rate": 7.36799178229539e-08,
      "loss": 0.029,
      "num_tokens": 73804200.0,
      "reward": 1.7455357760190964,
      "reward_std": 0.1615738943219185,
      "rewards/accuracy_reward": 0.7566964328289032,
      "rewards/format_reward": 0.9888392761349678,
      "step": 113
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 567.3649749755859,
      "epoch": 1.716417910447761,
      "grad_norm": 0.11726563423871994,
      "learning_rate": 6.698729810778064e-08,
      "loss": 0.0008,
      "num_tokens": 74444399.0,
      "reward": 1.8069197237491608,
      "reward_std": 0.12931114435195923,
      "rewards/accuracy_reward": 0.8102678582072258,
      "rewards/format_reward": 0.9966517761349678,
      "step": 114
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 625.3214569091797,
      "epoch": 1.7313432835820897,
      "grad_norm": 0.1247309222817421,
      "learning_rate": 6.059144366901736e-08,
      "loss": 0.0196,
      "num_tokens": 75133231.0,
      "reward": 1.73214291036129,
      "reward_std": 0.14865244133397937,
      "rewards/accuracy_reward": 0.7433035671710968,
      "rewards/format_reward": 0.9888392761349678,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 562.8538246154785,
      "epoch": 1.7462686567164178,
      "grad_norm": 0.09697449207305908,
      "learning_rate": 5.44967379058161e-08,
      "loss": 0.0212,
      "num_tokens": 75758132.0,
      "reward": 1.7455357909202576,
      "reward_std": 0.1268181591294706,
      "rewards/accuracy_reward": 0.753348208963871,
      "rewards/format_reward": 0.9921875,
      "step": 116
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 628.1741333007812,
      "epoch": 1.7611940298507462,
      "grad_norm": 0.11467966437339783,
      "learning_rate": 4.870735782506979e-08,
      "loss": 0.0212,
      "num_tokens": 76476424.0,
      "reward": 1.7254465222358704,
      "reward_std": 0.18577638221904635,
      "rewards/accuracy_reward": 0.7299107164144516,
      "rewards/format_reward": 0.9955357015132904,
      "step": 117
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 637.7288208007812,
      "epoch": 1.7761194029850746,
      "grad_norm": 0.12015992403030396,
      "learning_rate": 4.322727117869951e-08,
      "loss": 0.0363,
      "num_tokens": 77174237.0,
      "reward": 1.7243304252624512,
      "reward_std": 0.16919867880642414,
      "rewards/accuracy_reward": 0.737723208963871,
      "rewards/format_reward": 0.986607126891613,
      "step": 118
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 608.5301666259766,
      "epoch": 1.7910447761194028,
      "grad_norm": 0.1321968138217926,
      "learning_rate": 3.806023374435663e-08,
      "loss": 0.0319,
      "num_tokens": 77844888.0,
      "reward": 1.742187574505806,
      "reward_std": 0.1826935103163123,
      "rewards/accuracy_reward": 0.7500000111758709,
      "rewards/format_reward": 0.9921874925494194,
      "step": 119
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 615.6049194335938,
      "epoch": 1.8059701492537314,
      "grad_norm": 0.12096427381038666,
      "learning_rate": 3.3209786751399184e-08,
      "loss": 0.0184,
      "num_tokens": 78535414.0,
      "reward": 1.7957590073347092,
      "reward_std": 0.1389077603816986,
      "rewards/accuracy_reward": 0.8035714253783226,
      "rewards/format_reward": 0.9921874925494194,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 601.6015853881836,
      "epoch": 1.8208955223880596,
      "grad_norm": 0.10550054907798767,
      "learning_rate": 2.8679254453910785e-08,
      "loss": 0.0298,
      "num_tokens": 79204753.0,
      "reward": 1.7220982909202576,
      "reward_std": 0.1378917433321476,
      "rewards/accuracy_reward": 0.7287946417927742,
      "rewards/format_reward": 0.9933035597205162,
      "step": 121
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 647.8415374755859,
      "epoch": 1.835820895522388,
      "grad_norm": 0.10303810238838196,
      "learning_rate": 2.4471741852423233e-08,
      "loss": 0.0258,
      "num_tokens": 79932011.0,
      "reward": 1.7421875894069672,
      "reward_std": 0.17180508747696877,
      "rewards/accuracy_reward": 0.7566964253783226,
      "rewards/format_reward": 0.9854910597205162,
      "step": 122
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 624.6116333007812,
      "epoch": 1.8507462686567164,
      "grad_norm": 0.1059068813920021,
      "learning_rate": 2.0590132565903473e-08,
      "loss": 0.0385,
      "num_tokens": 80624703.0,
      "reward": 1.7924107760190964,
      "reward_std": 0.1603873297572136,
      "rewards/accuracy_reward": 0.8080357164144516,
      "rewards/format_reward": 0.984375,
      "step": 123
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 577.8995819091797,
      "epoch": 1.8656716417910446,
      "grad_norm": 0.12102499604225159,
      "learning_rate": 1.7037086855465898e-08,
      "loss": 0.0069,
      "num_tokens": 81266965.0,
      "reward": 1.7555804401636124,
      "reward_std": 0.15357153490185738,
      "rewards/accuracy_reward": 0.7555803582072258,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.3482322692871,
      "epoch": 1.8805970149253732,
      "grad_norm": 0.1061650738120079,
      "learning_rate": 1.3815039801161722e-08,
      "loss": 0.0114,
      "num_tokens": 81960317.0,
      "reward": 1.758928656578064,
      "reward_std": 0.1489480398595333,
      "rewards/accuracy_reward": 0.7622767835855484,
      "rewards/format_reward": 0.9966517835855484,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 553.6853065490723,
      "epoch": 1.8955223880597014,
      "grad_norm": 0.11372170597314835,
      "learning_rate": 1.0926199633097154e-08,
      "loss": 0.0107,
      "num_tokens": 82582595.0,
      "reward": 1.805803656578064,
      "reward_std": 0.12326187361031771,
      "rewards/accuracy_reward": 0.809151791036129,
      "rewards/format_reward": 0.9966517835855484,
      "step": 126
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 621.6317367553711,
      "epoch": 1.9104477611940298,
      "grad_norm": 0.10079541057348251,
      "learning_rate": 8.372546218022746e-09,
      "loss": 0.027,
      "num_tokens": 83276513.0,
      "reward": 1.7187500894069672,
      "reward_std": 0.14888622052967548,
      "rewards/accuracy_reward": 0.7287946343421936,
      "rewards/format_reward": 0.9899553582072258,
      "step": 127
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.2857360839844,
      "epoch": 1.9253731343283582,
      "grad_norm": 0.11577489972114563,
      "learning_rate": 6.15582970243117e-09,
      "loss": 0.0138,
      "num_tokens": 83929505.0,
      "reward": 1.7991072237491608,
      "reward_std": 0.11922843009233475,
      "rewards/accuracy_reward": 0.800223208963871,
      "rewards/format_reward": 0.9988839253783226,
      "step": 128
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 632.0848541259766,
      "epoch": 1.9402985074626866,
      "grad_norm": 0.10834372043609619,
      "learning_rate": 4.277569313094809e-09,
      "loss": 0.0197,
      "num_tokens": 84619181.0,
      "reward": 1.7712054401636124,
      "reward_std": 0.15788256376981735,
      "rewards/accuracy_reward": 0.785714291036129,
      "rewards/format_reward": 0.9854910597205162,
      "step": 129
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.475471496582,
      "epoch": 1.955223880597015,
      "grad_norm": 0.10320267826318741,
      "learning_rate": 2.739052315863355e-09,
      "loss": 0.0212,
      "num_tokens": 85331207.0,
      "reward": 1.7131697237491608,
      "reward_std": 0.1358252428472042,
      "rewards/accuracy_reward": 0.7187500074505806,
      "rewards/format_reward": 0.9944196343421936,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 582.7500343322754,
      "epoch": 1.9701492537313432,
      "grad_norm": 0.12011269479990005,
      "learning_rate": 1.541333133436018e-09,
      "loss": 0.0075,
      "num_tokens": 85980519.0,
      "reward": 1.8113840222358704,
      "reward_std": 0.12723324913531542,
      "rewards/accuracy_reward": 0.8136160597205162,
      "rewards/format_reward": 0.9977678582072258,
      "step": 131
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 611.2233848571777,
      "epoch": 1.9850746268656716,
      "grad_norm": 0.09882749617099762,
      "learning_rate": 6.852326227130833e-10,
      "loss": 0.0123,
      "num_tokens": 86666974.0,
      "reward": 1.7555804550647736,
      "reward_std": 0.13798785768449306,
      "rewards/accuracy_reward": 0.7611607238650322,
      "rewards/format_reward": 0.9944196343421936,
      "step": 132
    },
    {
      "epoch": 1.9850746268656716,
      "step": 132,
      "total_flos": 0.0,
      "train_loss": 0.0225595414226656,
      "train_runtime": 34371.2658,
      "train_samples_per_second": 0.436,
      "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
}