{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9846153846153847,
  "eval_steps": 500,
  "global_step": 178,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 239.35076141357422,
      "epoch": 0.011188811188811189,
      "grad_norm": 2.018390655517578,
      "kl": 0.0,
      "learning_rate": 5.555555555555555e-08,
      "loss": 0.0,
      "reward": 1.020408146083355,
      "reward_std": 0.588430143892765,
      "rewards/accuracy_reward": 0.07015305897220969,
      "rewards/format_reward": 0.3928571380674839,
      "rewards/influence_reward": 0.05229591624811292,
      "rewards/len_reward": 0.5051020309329033,
      "step": 1
    },
    {
      "completion_length": 240.83513402938843,
      "epoch": 0.055944055944055944,
      "grad_norm": 1.5813062191009521,
      "kl": 0.00015559792518615723,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.0,
      "reward": 0.9544004816561937,
      "reward_std": 0.6496573686599731,
      "rewards/accuracy_reward": 0.06377550942124799,
      "rewards/format_reward": 0.39317601080983877,
      "rewards/influence_reward": 0.0484693865000736,
      "rewards/len_reward": 0.44897958217188716,
      "step": 5
    },
    {
      "completion_length": 239.87346343994142,
      "epoch": 0.11188811188811189,
      "grad_norm": 1.199715256690979,
      "kl": 0.0015879154205322265,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.0001,
      "reward": 1.2579081252217292,
      "reward_std": 0.6406066298484803,
      "rewards/accuracy_reward": 0.09897958971560002,
      "rewards/format_reward": 0.6145408108830452,
      "rewards/influence_reward": 0.06607142747379839,
      "rewards/len_reward": 0.47831631228327753,
      "step": 10
    },
    {
      "completion_length": 235.00101585388182,
      "epoch": 0.16783216783216784,
      "grad_norm": 0.9491918683052063,
      "kl": 0.012871551513671874,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.0005,
      "reward": 1.9201530247926712,
      "reward_std": 0.5499199964106083,
      "rewards/accuracy_reward": 0.2658163230866194,
      "rewards/format_reward": 0.9533163100481034,
      "rewards/influence_reward": 0.19158162884414195,
      "rewards/len_reward": 0.5094387628138065,
      "step": 15
    },
    {
      "completion_length": 230.20764770507813,
      "epoch": 0.22377622377622378,
      "grad_norm": 1.0375124216079712,
      "kl": 0.034588623046875,
      "learning_rate": 9.996145181203615e-07,
      "loss": 0.0015,
      "reward": 2.358928510546684,
      "reward_std": 0.4863013714551926,
      "rewards/accuracy_reward": 0.46224488839507105,
      "rewards/format_reward": 0.9905612260103226,
      "rewards/influence_reward": 0.3974489726126194,
      "rewards/len_reward": 0.5086734544485807,
      "step": 20
    },
    {
      "completion_length": 221.39234199523926,
      "epoch": 0.27972027972027974,
      "grad_norm": 1.0581080913543701,
      "kl": 0.064349365234375,
      "learning_rate": 9.952846702217885e-07,
      "loss": 0.0027,
      "reward": 2.6903060495853426,
      "reward_std": 0.4831090085208416,
      "rewards/accuracy_reward": 0.5727040730416775,
      "rewards/format_reward": 0.9933673486113548,
      "rewards/influence_reward": 0.516071417927742,
      "rewards/len_reward": 0.6081632524728775,
      "step": 25
    },
    {
      "completion_length": 223.85866737365723,
      "epoch": 0.3356643356643357,
      "grad_norm": 1.002379298210144,
      "kl": 0.0775390625,
      "learning_rate": 9.861849601988383e-07,
      "loss": 0.0032,
      "reward": 2.7147958517074584,
      "reward_std": 0.4902082525193691,
      "rewards/accuracy_reward": 0.552295907586813,
      "rewards/format_reward": 0.9910714283585549,
      "rewards/influence_reward": 0.5109693787992,
      "rewards/len_reward": 0.660459166765213,
      "step": 30
    },
    {
      "completion_length": 242.63366928100587,
      "epoch": 0.3916083916083916,
      "grad_norm": 0.8198888301849365,
      "kl": 0.07734375,
      "learning_rate": 9.72403023233439e-07,
      "loss": 0.0032,
      "reward": 2.704081577062607,
      "reward_std": 0.4683332860469818,
      "rewards/accuracy_reward": 0.5767857022583485,
      "rewards/format_reward": 0.9892857119441032,
      "rewards/influence_reward": 0.5352040722966194,
      "rewards/len_reward": 0.6028061106801033,
      "step": 35
    },
    {
      "completion_length": 219.7584140777588,
      "epoch": 0.44755244755244755,
      "grad_norm": 0.8813366889953613,
      "kl": 0.081610107421875,
      "learning_rate": 9.540715869125407e-07,
      "loss": 0.0033,
      "reward": 2.984948921203613,
      "reward_std": 0.45619520246982576,
      "rewards/accuracy_reward": 0.620918358117342,
      "rewards/format_reward": 0.9966836735606194,
      "rewards/influence_reward": 0.5757652945816517,
      "rewards/len_reward": 0.7915816128253936,
      "step": 40
    },
    {
      "completion_length": 213.20280342102052,
      "epoch": 0.5034965034965035,
      "grad_norm": 0.8298941850662231,
      "kl": 0.09119873046875,
      "learning_rate": 9.313671929888959e-07,
      "loss": 0.0037,
      "reward": 2.999489736557007,
      "reward_std": 0.4441244326531887,
      "rewards/accuracy_reward": 0.6117346778512001,
      "rewards/format_reward": 0.9974489793181419,
      "rewards/influence_reward": 0.5594387598335743,
      "rewards/len_reward": 0.8308673277497292,
      "step": 45
    },
    {
      "completion_length": 239.8193817138672,
      "epoch": 0.5594405594405595,
      "grad_norm": 0.8936271667480469,
      "kl": 0.10096435546875,
      "learning_rate": 9.045084971874737e-07,
      "loss": 0.0041,
      "reward": 2.906377512216568,
      "reward_std": 0.45472528263926504,
      "rewards/accuracy_reward": 0.5836734607815742,
      "rewards/format_reward": 0.9956632673740387,
      "rewards/influence_reward": 0.5321428462862968,
      "rewards/len_reward": 0.7948979437351227,
      "step": 50
    },
    {
      "completion_length": 230.65917854309083,
      "epoch": 0.6153846153846154,
      "grad_norm": 0.7805132865905762,
      "kl": 0.1081298828125,
      "learning_rate": 8.737541634312983e-07,
      "loss": 0.0043,
      "reward": 3.0579080879688263,
      "reward_std": 0.4420431960374117,
      "rewards/accuracy_reward": 0.607397947460413,
      "rewards/format_reward": 0.9959183692932129,
      "rewards/influence_reward": 0.5612244814634323,
      "rewards/len_reward": 0.8933673277497292,
      "step": 55
    },
    {
      "completion_length": 220.4063720703125,
      "epoch": 0.6713286713286714,
      "grad_norm": 1.4386658668518066,
      "kl": 0.1181640625,
      "learning_rate": 8.394003727664709e-07,
      "loss": 0.0047,
      "reward": 3.0977040231227875,
      "reward_std": 0.4241327825933695,
      "rewards/accuracy_reward": 0.6344387613236904,
      "rewards/format_reward": 0.9956632673740387,
      "rewards/influence_reward": 0.5821428462862969,
      "rewards/len_reward": 0.8854591652750969,
      "step": 60
    },
    {
      "completion_length": 228.13443336486816,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.6940521597862244,
      "kl": 0.11019287109375,
      "learning_rate": 8.017779709767857e-07,
      "loss": 0.0044,
      "reward": 2.9755101561546327,
      "reward_std": 0.4135630540549755,
      "rewards/accuracy_reward": 0.5908163137733936,
      "rewards/format_reward": 0.9948979616165161,
      "rewards/influence_reward": 0.5410714164376259,
      "rewards/len_reward": 0.8487244695425034,
      "step": 65
    },
    {
      "completion_length": 217.46045417785643,
      "epoch": 0.7832167832167832,
      "grad_norm": 0.6972203254699707,
      "kl": 0.1288818359375,
      "learning_rate": 7.612492823579744e-07,
      "loss": 0.0052,
      "reward": 3.136479526758194,
      "reward_std": 0.42438863664865495,
      "rewards/accuracy_reward": 0.6517856940627098,
      "rewards/format_reward": 0.9956632673740387,
      "rewards/influence_reward": 0.6112244755029679,
      "rewards/len_reward": 0.8778061032295227,
      "step": 70
    },
    {
      "completion_length": 217.32397499084473,
      "epoch": 0.8391608391608392,
      "grad_norm": 0.995740532875061,
      "kl": 0.13599853515625,
      "learning_rate": 7.182046203366709e-07,
      "loss": 0.0055,
      "reward": 3.1479591190814973,
      "reward_std": 0.4072150893509388,
      "rewards/accuracy_reward": 0.6653061121702194,
      "rewards/format_reward": 0.9959183692932129,
      "rewards/influence_reward": 0.6081632517278195,
      "rewards/len_reward": 0.8785714104771614,
      "step": 75
    },
    {
      "completion_length": 204.29948654174805,
      "epoch": 0.8951048951048951,
      "grad_norm": 0.8313683867454529,
      "kl": 0.1267333984375,
      "learning_rate": 6.730585285387465e-07,
      "loss": 0.0051,
      "reward": 3.1954080879688265,
      "reward_std": 0.39486792534589765,
      "rewards/accuracy_reward": 0.6806122347712517,
      "rewards/format_reward": 0.9956632673740387,
      "rewards/influence_reward": 0.6438775353133679,
      "rewards/len_reward": 0.8752550885081292,
      "step": 80
    },
    {
      "completion_length": 203.56530113220214,
      "epoch": 0.951048951048951,
      "grad_norm": 0.7973130345344543,
      "kl": 0.126806640625,
      "learning_rate": 6.262457885075789e-07,
      "loss": 0.0051,
      "reward": 3.2344387233257295,
      "reward_std": 0.3648144776001573,
      "rewards/accuracy_reward": 0.6961734496057034,
      "rewards/format_reward": 0.9956632673740387,
      "rewards/influence_reward": 0.6497448846697808,
      "rewards/len_reward": 0.8928571224212647,
      "step": 85
    },
    {
      "completion_length": 221.41661420549664,
      "epoch": 1.0,
      "grad_norm": 1.3692800998687744,
      "kl": 0.13193359375,
      "learning_rate": 5.782172325201155e-07,
      "loss": 0.0046,
      "reward": 3.196792949948992,
      "reward_std": 0.3790252791983741,
      "rewards/accuracy_reward": 0.676093282018389,
      "rewards/format_reward": 0.9962099126407078,
      "rewards/influence_reward": 0.6265305961881366,
      "rewards/len_reward": 0.897959165913718,
      "step": 90
    },
    {
      "completion_length": 228.48239212036134,
      "epoch": 1.055944055944056,
      "grad_norm": 0.9810852408409119,
      "kl": 0.14306640625,
      "learning_rate": 5.294354018255944e-07,
      "loss": 0.0057,
      "reward": 3.2369897425174714,
      "reward_std": 0.3675705246627331,
      "rewards/accuracy_reward": 0.6959183529019356,
      "rewards/format_reward": 0.9979591846466065,
      "rewards/influence_reward": 0.6489795804023742,
      "rewards/len_reward": 0.8941326335072517,
      "step": 95
    },
    {
      "completion_length": 228.94004592895507,
      "epoch": 1.1118881118881119,
      "grad_norm": 0.814385712146759,
      "kl": 0.14249267578125,
      "learning_rate": 4.803700921204658e-07,
      "loss": 0.0057,
      "reward": 3.299489712715149,
      "reward_std": 0.36674671024084093,
      "rewards/accuracy_reward": 0.7234693765640259,
      "rewards/format_reward": 0.9969387769699096,
      "rewards/influence_reward": 0.6719387613236905,
      "rewards/len_reward": 0.9071428373456001,
      "step": 100
    },
    {
      "completion_length": 227.59565849304198,
      "epoch": 1.167832167832168,
      "grad_norm": 1.3440909385681152,
      "kl": 0.15057373046875,
      "learning_rate": 4.3149382915901606e-07,
      "loss": 0.006,
      "reward": 3.2344387233257295,
      "reward_std": 0.3843592546880245,
      "rewards/accuracy_reward": 0.6931122295558453,
      "rewards/format_reward": 0.9966836735606194,
      "rewards/influence_reward": 0.6403061114251614,
      "rewards/len_reward": 0.9043367177248001,
      "step": 105
    },
    {
      "completion_length": 220.42397651672363,
      "epoch": 1.2237762237762237,
      "grad_norm": 1.2474172115325928,
      "kl": 0.1584716796875,
      "learning_rate": 3.8327731807204744e-07,
      "loss": 0.0063,
      "reward": 3.296428495645523,
      "reward_std": 0.38065838664770124,
      "rewards/accuracy_reward": 0.7209183543920517,
      "rewards/format_reward": 0.9974489808082581,
      "rewards/influence_reward": 0.6640305913984775,
      "rewards/len_reward": 0.9140305906534195,
      "step": 110
    },
    {
      "completion_length": 228.10331192016602,
      "epoch": 1.2797202797202798,
      "grad_norm": 1.0222375392913818,
      "kl": 0.17257080078125,
      "learning_rate": 3.361849102191533e-07,
      "loss": 0.0069,
      "reward": 3.247704017162323,
      "reward_std": 0.37765960246324537,
      "rewards/accuracy_reward": 0.698979577422142,
      "rewards/format_reward": 0.9948979601264,
      "rewards/influence_reward": 0.6392856992781162,
      "rewards/len_reward": 0.9145407944917678,
      "step": 115
    },
    {
      "completion_length": 212.8178524017334,
      "epoch": 1.3356643356643356,
      "grad_norm": 0.9680814146995544,
      "kl": 0.14952392578125,
      "learning_rate": 2.906701312312861e-07,
      "loss": 0.006,
      "reward": 3.277550941705704,
      "reward_std": 0.36851916685700414,
      "rewards/accuracy_reward": 0.7150510035455226,
      "rewards/format_reward": 0.9966836750507355,
      "rewards/influence_reward": 0.659948968142271,
      "rewards/len_reward": 0.9058673277497291,
      "step": 120
    },
    {
      "completion_length": 218.52907676696776,
      "epoch": 1.3916083916083917,
      "grad_norm": 0.8980757594108582,
      "kl": 0.144287109375,
      "learning_rate": 2.4717131331100774e-07,
      "loss": 0.0058,
      "reward": 3.2676019430160523,
      "reward_std": 0.34619110673666,
      "rewards/accuracy_reward": 0.7015306010842324,
      "rewards/format_reward": 0.9969387769699096,
      "rewards/influence_reward": 0.6543367207050323,
      "rewards/len_reward": 0.9147959008812905,
      "step": 125
    },
    {
      "completion_length": 220.4456588745117,
      "epoch": 1.4475524475524475,
      "grad_norm": 1.6906476020812988,
      "kl": 0.1458251953125,
      "learning_rate": 2.0610737385376348e-07,
      "loss": 0.0058,
      "reward": 3.244897884130478,
      "reward_std": 0.34559424556791785,
      "rewards/accuracy_reward": 0.7017857000231743,
      "rewards/format_reward": 0.9964285731315613,
      "rewards/influence_reward": 0.648724476993084,
      "rewards/len_reward": 0.8979591608047486,
      "step": 130
    },
    {
      "completion_length": 222.28366889953614,
      "epoch": 1.5034965034965035,
      "grad_norm": 2.0042786598205566,
      "kl": 0.15557861328125,
      "learning_rate": 1.6787378104435929e-07,
      "loss": 0.0062,
      "reward": 3.2635203421115877,
      "reward_std": 0.3759432673454285,
      "rewards/accuracy_reward": 0.7086734533309936,
      "rewards/format_reward": 0.9946428582072258,
      "rewards/influence_reward": 0.6635203965008258,
      "rewards/len_reward": 0.8966836482286453,
      "step": 135
    },
    {
      "completion_length": 226.27831115722657,
      "epoch": 1.5594405594405596,
      "grad_norm": 1.0714577436447144,
      "kl": 0.14898681640625,
      "learning_rate": 1.3283874528215733e-07,
      "loss": 0.006,
      "reward": 3.270918291807175,
      "reward_std": 0.3628778774291277,
      "rewards/accuracy_reward": 0.713775496929884,
      "rewards/format_reward": 0.9964285716414452,
      "rewards/influence_reward": 0.6655612140893936,
      "rewards/len_reward": 0.8951530426740646,
      "step": 140
    },
    {
      "completion_length": 216.52473983764648,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.0027109384536743,
      "kl": 0.1531005859375,
      "learning_rate": 1.013396731136465e-07,
      "loss": 0.0061,
      "reward": 3.3109693050384523,
      "reward_std": 0.3781158674508333,
      "rewards/accuracy_reward": 0.7298469215631485,
      "rewards/format_reward": 0.9979591846466065,
      "rewards/influence_reward": 0.6729591734707355,
      "rewards/len_reward": 0.9102040633559227,
      "step": 145
    },
    {
      "completion_length": 214.229333114624,
      "epoch": 1.6713286713286712,
      "grad_norm": 1.2131189107894897,
      "kl": 0.16943359375,
      "learning_rate": 7.36799178229539e-08,
      "loss": 0.0068,
      "reward": 3.2859693229198457,
      "reward_std": 0.35086961574852465,
      "rewards/accuracy_reward": 0.7020408011972904,
      "rewards/format_reward": 0.9961734697222709,
      "rewards/influence_reward": 0.6668367192149163,
      "rewards/len_reward": 0.920918345451355,
      "step": 150
    },
    {
      "completion_length": 218.42678260803223,
      "epoch": 1.7272727272727273,
      "grad_norm": 28.282617568969727,
      "kl": 0.187109375,
      "learning_rate": 5.012585797388935e-08,
      "loss": 0.0075,
      "reward": 3.2318876922130584,
      "reward_std": 0.35199854988604784,
      "rewards/accuracy_reward": 0.6829081475734711,
      "rewards/format_reward": 0.9956632673740387,
      "rewards/influence_reward": 0.6392856985330582,
      "rewards/len_reward": 0.9140305936336517,
      "step": 155
    },
    {
      "completion_length": 215.96785354614258,
      "epoch": 1.7832167832167833,
      "grad_norm": 1.0345922708511353,
      "kl": 0.17052001953125,
      "learning_rate": 3.0904332038757974e-08,
      "loss": 0.0068,
      "reward": 3.295152986049652,
      "reward_std": 0.3517200522124767,
      "rewards/accuracy_reward": 0.7298469223082066,
      "rewards/format_reward": 0.9943877562880516,
      "rewards/influence_reward": 0.6688775345683098,
      "rewards/len_reward": 0.9020407944917679,
      "step": 160
    },
    {
      "completion_length": 220.99591369628905,
      "epoch": 1.8391608391608392,
      "grad_norm": 0.9470728039741516,
      "kl": 0.16759033203125,
      "learning_rate": 1.6200453819870118e-08,
      "loss": 0.0067,
      "reward": 3.3061223804950712,
      "reward_std": 0.3646129764616489,
      "rewards/accuracy_reward": 0.7349489614367485,
      "rewards/format_reward": 0.9943877577781677,
      "rewards/influence_reward": 0.6849489629268646,
      "rewards/len_reward": 0.8918367147445678,
      "step": 165
    },
    {
      "completion_length": 212.3209140777588,
      "epoch": 1.895104895104895,
      "grad_norm": 1.04275381565094,
      "kl": 0.53802490234375,
      "learning_rate": 6.15582970243117e-09,
      "loss": 0.0216,
      "reward": 3.295663195848465,
      "reward_std": 0.35336949974298476,
      "rewards/accuracy_reward": 0.7135203927755356,
      "rewards/format_reward": 0.9966836735606194,
      "rewards/influence_reward": 0.6688775405287742,
      "rewards/len_reward": 0.9165816113352776,
      "step": 170
    },
    {
      "completion_length": 214.2193832397461,
      "epoch": 1.951048951048951,
      "grad_norm": 1.4784653186798096,
      "kl": 0.1680908203125,
      "learning_rate": 8.671949076420881e-10,
      "loss": 0.0067,
      "reward": 3.292857068777084,
      "reward_std": 0.3482844814658165,
      "rewards/accuracy_reward": 0.7214285537600518,
      "rewards/format_reward": 0.9946428582072258,
      "rewards/influence_reward": 0.6732142709195614,
      "rewards/len_reward": 0.9035714089870452,
      "step": 175
    },
    {
      "completion_length": 220.2002493540446,
      "epoch": 1.9846153846153847,
      "kl": 0.19038899739583334,
      "reward": 3.2593536575635276,
      "reward_std": 0.3590250660975774,
      "rewards/accuracy_reward": 0.7032312775651614,
      "rewards/format_reward": 0.9940476194024086,
      "rewards/influence_reward": 0.6611394360661507,
      "rewards/len_reward": 0.9009353543321291,
      "step": 178,
      "total_flos": 0.0,
      "train_loss": 0.005316848002538006,
      "train_runtime": 54287.5063,
      "train_samples_per_second": 0.368,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 178,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}