{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9962034927866363,
  "eval_steps": 2000000,
  "global_step": 164,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 803.0022583007812,
      "epoch": 0.006074411541381929,
      "grad_norm": 0.14674668166353436,
      "kl": 0.0,
      "learning_rate": 5.88235294117647e-08,
      "loss": 0.0479,
      "num_tokens": 918402.0,
      "reward": 0.9815848618745804,
      "reward_std": 0.23756355978548527,
      "rewards/accuracy_reward": 0.4933035746216774,
      "rewards/format_reward": 0.9765624925494194,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 734.3027648925781,
      "epoch": 0.030372057706909643,
      "grad_norm": 0.13057922454955487,
      "kl": 5.84721565246582e-05,
      "learning_rate": 2.941176470588235e-07,
      "loss": 0.037,
      "num_tokens": 4258679.0,
      "reward": 1.0901228114962578,
      "reward_std": 0.23264290555380285,
      "rewards/accuracy_reward": 0.5965401763096452,
      "rewards/format_reward": 0.9871651735156775,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 743.0277114868164,
      "epoch": 0.060744115413819286,
      "grad_norm": 0.1343510023679506,
      "kl": 8.401274681091309e-05,
      "learning_rate": 5.88235294117647e-07,
      "loss": 0.0465,
      "num_tokens": 8471139.0,
      "reward": 1.062276841700077,
      "reward_std": 0.23240854553878307,
      "rewards/accuracy_reward": 0.5718749992549419,
      "rewards/format_reward": 0.9808035627007484,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 739.1007049560546,
      "epoch": 0.09111617312072894,
      "grad_norm": 0.738570752534254,
      "kl": 0.00017675161361694336,
      "learning_rate": 8.823529411764705e-07,
      "loss": 0.0358,
      "num_tokens": 12734494.0,
      "reward": 1.0722098737955092,
      "reward_std": 0.2205923892557621,
      "rewards/accuracy_reward": 0.5785714268684388,
      "rewards/format_reward": 0.9872767761349678,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 745.5227966308594,
      "epoch": 0.12148823082763857,
      "grad_norm": 0.30158627332435983,
      "kl": 0.0026874780654907227,
      "learning_rate": 9.989726963751682e-07,
      "loss": 0.037,
      "num_tokens": 16990620.0,
      "reward": 1.0786830827593803,
      "reward_std": 0.23385403044521808,
      "rewards/accuracy_reward": 0.5861607141792774,
      "rewards/format_reward": 0.9850446373224259,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 731.8799407958984,
      "epoch": 0.15186028853454822,
      "grad_norm": 0.10368791276497527,
      "kl": 0.0003843784332275391,
      "learning_rate": 9.927100106776212e-07,
      "loss": 0.0351,
      "num_tokens": 21176106.0,
      "reward": 1.106584869325161,
      "reward_std": 0.21388941686600446,
      "rewards/accuracy_reward": 0.6142857171595096,
      "rewards/format_reward": 0.9845982104539871,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 716.5080688476562,
      "epoch": 0.18223234624145787,
      "grad_norm": 0.24334246130754558,
      "kl": 0.0008969306945800781,
      "learning_rate": 9.808267184205181e-07,
      "loss": 0.0203,
      "num_tokens": 25301046.0,
      "reward": 1.0974330857396126,
      "reward_std": 0.21846173331141472,
      "rewards/accuracy_reward": 0.6022321447730065,
      "rewards/format_reward": 0.9904017791152,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 718.258511352539,
      "epoch": 0.2126044039483675,
      "grad_norm": 0.13060657109261103,
      "kl": 0.0011320114135742188,
      "learning_rate": 9.634583786730108e-07,
      "loss": 0.0247,
      "num_tokens": 29447476.0,
      "reward": 1.116071480512619,
      "reward_std": 0.21333869993686677,
      "rewards/accuracy_reward": 0.620758930593729,
      "rewards/format_reward": 0.9906249955296517,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 734.5969055175781,
      "epoch": 0.24297646165527714,
      "grad_norm": 0.14790799582315342,
      "kl": 0.0016210556030273437,
      "learning_rate": 9.408031213740044e-07,
      "loss": 0.0307,
      "num_tokens": 33678894.0,
      "reward": 1.0677455827593803,
      "reward_std": 0.2126396529376507,
      "rewards/accuracy_reward": 0.5736607156693936,
      "rewards/format_reward": 0.9881696373224258,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 724.3696716308593,
      "epoch": 0.2733485193621868,
      "grad_norm": 0.1235294567362974,
      "kl": 0.0030605316162109373,
      "learning_rate": 9.131193871579974e-07,
      "loss": 0.0288,
      "num_tokens": 37860574.0,
      "reward": 1.0801339834928512,
      "reward_std": 0.22113933004438877,
      "rewards/accuracy_reward": 0.5854910694062709,
      "rewards/format_reward": 0.9892857074737549,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 719.0857437133789,
      "epoch": 0.30372057706909644,
      "grad_norm": 0.11260442217805001,
      "kl": 0.004908370971679688,
      "learning_rate": 8.807229791845671e-07,
      "loss": 0.0309,
      "num_tokens": 42021414.0,
      "reward": 1.1001116633415222,
      "reward_std": 0.2082567172124982,
      "rewards/accuracy_reward": 0.6053571477532387,
      "rewards/format_reward": 0.9895089223980904,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 710.4136505126953,
      "epoch": 0.3340926347760061,
      "grad_norm": 0.12446746182181773,
      "kl": 0.00711669921875,
      "learning_rate": 8.439834606028593e-07,
      "loss": 0.03,
      "num_tokens": 46149299.0,
      "reward": 1.1018973752856254,
      "reward_std": 0.20198939852416514,
      "rewards/accuracy_reward": 0.6075892850756646,
      "rewards/format_reward": 0.9886160641908646,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 711.5286056518555,
      "epoch": 0.36446469248291574,
      "grad_norm": 0.13140372700302483,
      "kl": 0.01026763916015625,
      "learning_rate": 8.033199387471276e-07,
      "loss": 0.0255,
      "num_tokens": 50248419.0,
      "reward": 1.0939732655882835,
      "reward_std": 0.20504674576222898,
      "rewards/accuracy_reward": 0.6002232126891613,
      "rewards/format_reward": 0.9874999925494194,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 693.0864105224609,
      "epoch": 0.39483675018982534,
      "grad_norm": 5.84593208956697,
      "kl": 0.01532440185546875,
      "learning_rate": 7.591962841552626e-07,
      "loss": 0.0221,
      "num_tokens": 54273302.0,
      "reward": 1.102567011117935,
      "reward_std": 0.19313923437148334,
      "rewards/accuracy_reward": 0.6060267843306064,
      "rewards/format_reward": 0.9930803507566452,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 698.3335159301757,
      "epoch": 0.425208807896735,
      "grad_norm": 0.18797262379226592,
      "kl": 0.0115875244140625,
      "learning_rate": 7.121158389495185e-07,
      "loss": 0.0308,
      "num_tokens": 58317828.0,
      "reward": 1.0886161223053932,
      "reward_std": 0.2050962893292308,
      "rewards/accuracy_reward": 0.5941964283585548,
      "rewards/format_reward": 0.9888392791152001,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 680.4533798217774,
      "epoch": 0.45558086560364464,
      "grad_norm": 0.12462934113614446,
      "kl": 0.01414947509765625,
      "learning_rate": 6.626156749437736e-07,
      "loss": 0.0258,
      "num_tokens": 62259643.0,
      "reward": 1.1068080976605414,
      "reward_std": 0.1903689544647932,
      "rewards/accuracy_reward": 0.6107142798602581,
      "rewards/format_reward": 0.9921874910593033,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 671.9855209350586,
      "epoch": 0.4859529233105543,
      "grad_norm": 0.12209526567257212,
      "kl": 0.01604156494140625,
      "learning_rate": 6.112604669781572e-07,
      "loss": 0.0134,
      "num_tokens": 66175850.0,
      "reward": 1.0722098708152772,
      "reward_std": 0.1989122748374939,
      "rewards/accuracy_reward": 0.5743303574621678,
      "rewards/format_reward": 0.9957589223980904,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 655.3560516357422,
      "epoch": 0.5163249810174639,
      "grad_norm": 0.12987027750323757,
      "kl": 0.017596435546875,
      "learning_rate": 5.586360513712009e-07,
      "loss": 0.0202,
      "num_tokens": 70039477.0,
      "reward": 1.1373884499073028,
      "reward_std": 0.17809431692585348,
      "rewards/accuracy_reward": 0.6401785723865032,
      "rewards/format_reward": 0.9944196373224259,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 679.969448852539,
      "epoch": 0.5466970387243736,
      "grad_norm": 0.11569644648116158,
      "kl": 0.01629638671875,
      "learning_rate": 5.053427429716866e-07,
      "loss": 0.0261,
      "num_tokens": 74010748.0,
      "reward": 1.1206473752856254,
      "reward_std": 0.16938802655786275,
      "rewards/accuracy_reward": 0.62518887296319,
      "rewards/format_reward": 0.9930803492665291,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 670.8172203063965,
      "epoch": 0.5770690964312832,
      "grad_norm": 0.14653189798827554,
      "kl": 0.01604461669921875,
      "learning_rate": 4.519884870461591e-07,
      "loss": 0.0062,
      "num_tokens": 77924425.0,
      "reward": 1.1319196939468383,
      "reward_std": 0.18193732015788555,
      "rewards/accuracy_reward": 0.6334821462631226,
      "rewards/format_reward": 0.9968749970197678,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 689.6049453735352,
      "epoch": 0.6074411541381929,
      "grad_norm": 0.1373443920915848,
      "kl": 0.0151153564453125,
      "learning_rate": 3.991819241221835e-07,
      "loss": 0.0202,
      "num_tokens": 81937855.0,
      "reward": 1.1222098782658576,
      "reward_std": 0.1860942555591464,
      "rewards/accuracy_reward": 0.625,
      "rewards/format_reward": 0.9944196373224259,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 675.8683364868164,
      "epoch": 0.6378132118451025,
      "grad_norm": 0.14762350547769187,
      "kl": 0.015521240234375,
      "learning_rate": 3.4752544690038643e-07,
      "loss": 0.0151,
      "num_tokens": 85899921.0,
      "reward": 1.128125049173832,
      "reward_std": 0.1950968151912093,
      "rewards/accuracy_reward": 0.6305803559720516,
      "rewards/format_reward": 0.9950892791152001,
      "step": 105
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 696.7239181518555,
      "epoch": 0.6681852695520122,
      "grad_norm": 0.12897190097851505,
      "kl": 0.0144561767578125,
      "learning_rate": 2.976083284388031e-07,
      "loss": 0.0225,
      "num_tokens": 89967132.0,
      "reward": 1.0918527334928512,
      "reward_std": 0.1823650782927871,
      "rewards/accuracy_reward": 0.5953125059604645,
      "rewards/format_reward": 0.9930803492665291,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 719.4877540588379,
      "epoch": 0.6985573272589218,
      "grad_norm": 0.12020301457071197,
      "kl": 0.014898681640625,
      "learning_rate": 2.500000000000001e-07,
      "loss": 0.0153,
      "num_tokens": 94129389.0,
      "reward": 1.1255580872297286,
      "reward_std": 0.19005396589636803,
      "rewards/accuracy_reward": 0.6283482149243355,
      "rewards/format_reward": 0.9944196343421936,
      "step": 115
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 681.8125282287598,
      "epoch": 0.7289293849658315,
      "grad_norm": 0.1344831996031667,
      "kl": 0.01582489013671875,
      "learning_rate": 2.0524355524417015e-07,
      "loss": 0.0192,
      "num_tokens": 98103221.0,
      "reward": 1.1333705812692643,
      "reward_std": 0.20433492437005044,
      "rewards/accuracy_reward": 0.6354910746216774,
      "rewards/format_reward": 0.9957589209079742,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 681.395115661621,
      "epoch": 0.7593014426727411,
      "grad_norm": 0.13865803911159388,
      "kl": 0.0163787841796875,
      "learning_rate": 1.6384955486934154e-07,
      "loss": 0.0122,
      "num_tokens": 102096279.0,
      "reward": 1.1371652349829673,
      "reward_std": 0.18053851332515478,
      "rewards/accuracy_reward": 0.6404017873108387,
      "rewards/format_reward": 0.9935267820954323,
      "step": 125
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 702.8656555175781,
      "epoch": 0.7896735003796507,
      "grad_norm": 0.10840464189916307,
      "kl": 0.01602935791015625,
      "learning_rate": 1.262902023724824e-07,
      "loss": 0.0162,
      "num_tokens": 106168389.0,
      "reward": 1.1045759484171866,
      "reward_std": 0.18122291592881085,
      "rewards/accuracy_reward": 0.6062500029802322,
      "rewards/format_reward": 0.9966517820954323,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 696.1946792602539,
      "epoch": 0.8200455580865603,
      "grad_norm": 0.1591189816361499,
      "kl": 0.01680145263671875,
      "learning_rate": 9.299395737170757e-08,
      "loss": 0.0167,
      "num_tokens": 110246405.0,
      "reward": 1.0947545170783997,
      "reward_std": 0.20664523243904115,
      "rewards/accuracy_reward": 0.5973214291036129,
      "rewards/format_reward": 0.9948660656809807,
      "step": 135
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 698.2578453063965,
      "epoch": 0.85041761579347,
      "grad_norm": 0.13052581661637774,
      "kl": 0.0175323486328125,
      "learning_rate": 6.43406479383053e-08,
      "loss": 0.0206,
      "num_tokens": 114326168.0,
      "reward": 1.1349330857396125,
      "reward_std": 0.19575350042432546,
      "rewards/accuracy_reward": 0.6372767858207226,
      "rewards/format_reward": 0.9953124925494194,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 678.3460144042969,
      "epoch": 0.8807896735003796,
      "grad_norm": 0.12342327662368928,
      "kl": 0.01726837158203125,
      "learning_rate": 4.065713769482082e-08,
      "loss": 0.0217,
      "num_tokens": 118268150.0,
      "reward": 1.131250050663948,
      "reward_std": 0.19892821311950684,
      "rewards/accuracy_reward": 0.6337053582072258,
      "rewards/format_reward": 0.9950892806053162,
      "step": 145
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 699.1832885742188,
      "epoch": 0.9111617312072893,
      "grad_norm": 0.11347140306400243,
      "kl": 0.01734161376953125,
      "learning_rate": 2.2213597106929605e-08,
      "loss": 0.0206,
      "num_tokens": 122323331.0,
      "reward": 1.123325940966606,
      "reward_std": 0.19308128226548432,
      "rewards/accuracy_reward": 0.625669640302658,
      "rewards/format_reward": 0.9953124925494194,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 699.7196716308594,
      "epoch": 0.9415337889141989,
      "grad_norm": 0.1475830222889839,
      "kl": 0.01719207763671875,
      "learning_rate": 9.22042150446728e-09,
      "loss": 0.0185,
      "num_tokens": 126366451.0,
      "reward": 1.154241117835045,
      "reward_std": 0.1731583815999329,
      "rewards/accuracy_reward": 0.6560267820954323,
      "rewards/format_reward": 0.9964285656809807,
      "step": 155
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 695.9060577392578,
      "epoch": 0.9719058466211086,
      "grad_norm": 0.13967302854618574,
      "kl": 0.01683197021484375,
      "learning_rate": 1.8258309893965374e-09,
      "loss": 0.0191,
      "num_tokens": 130430318.0,
      "reward": 1.1508929118514062,
      "reward_std": 0.20208097249269485,
      "rewards/accuracy_reward": 0.6546875029802323,
      "rewards/format_reward": 0.9924107119441032,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.7267169952393,
      "epoch": 0.9962034927866363,
      "kl": 0.017988204956054688,
      "num_tokens": 133627034.0,
      "reward": 1.118443138897419,
      "reward_std": 0.1834622365422547,
      "rewards/accuracy_reward": 0.6226820051670074,
      "rewards/format_reward": 0.9946986511349678,
      "step": 164,
      "total_flos": 0.0,
      "train_loss": 0.024119453254814554,
      "train_runtime": 33084.5686,
      "train_samples_per_second": 0.557,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 5,
  "max_steps": 164,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}