{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6890756302521008,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 641.9296875,
      "epoch": 0.01680672268907563,
      "grad_norm": 0.1829340010881424,
      "kl": 0.0,
      "learning_rate": 3.333333333333333e-07,
      "loss": -0.0027,
      "num_tokens": 143935.0,
      "reward": 0.015625,
      "reward_std": 0.0289318785071373,
      "rewards/code_format_reward": 0.015625,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 558.2109375,
      "epoch": 0.03361344537815126,
      "grad_norm": 0.24132254719734192,
      "kl": 0.0,
      "learning_rate": 6.666666666666666e-07,
      "loss": 0.0106,
      "num_tokens": 275234.0,
      "reward": 0.015625,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.015625,
      "step": 2
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.5546875,
      "epoch": 0.05042016806722689,
      "grad_norm": 0.2641178071498871,
      "kl": 0.00027370452880859375,
      "learning_rate": 1e-06,
      "loss": 0.0038,
      "num_tokens": 408041.0,
      "reward": 0.015625,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.015625,
      "step": 3
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 580.0703125,
      "epoch": 0.06722689075630252,
      "grad_norm": 0.0005861425888724625,
      "kl": 0.00024080276489257812,
      "learning_rate": 9.997640060704816e-07,
      "loss": 0.0,
      "num_tokens": 541898.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.0,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.8125,
      "epoch": 0.08403361344537816,
      "grad_norm": 0.19232627749443054,
      "kl": 0.00027179718017578125,
      "learning_rate": 9.990562718069702e-07,
      "loss": 0.0049,
      "num_tokens": 679066.0,
      "reward": 0.0234375,
      "reward_std": 0.051028965041041374,
      "rewards/code_format_reward": 0.0234375,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.3046875,
      "epoch": 0.10084033613445378,
      "grad_norm": 0.000793790037278086,
      "kl": 0.0002541542053222656,
      "learning_rate": 9.978775395249762e-07,
      "loss": 0.0,
      "num_tokens": 811689.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.0,
      "step": 6
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.1171875,
      "epoch": 0.11764705882352941,
      "grad_norm": 0.14019447565078735,
      "kl": 0.0003261566162109375,
      "learning_rate": 9.962290455518912e-07,
      "loss": 0.0085,
      "num_tokens": 942160.0,
      "reward": 0.0078125,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.0078125,
      "step": 7
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 602.890625,
      "epoch": 0.13445378151260504,
      "grad_norm": 0.0008551861974410713,
      "kl": 0.00027751922607421875,
      "learning_rate": 9.941125189302508e-07,
      "loss": 0.0,
      "num_tokens": 1081786.0,
      "reward": 0.0,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.0,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 538.1484375,
      "epoch": 0.15126050420168066,
      "grad_norm": 0.6493130326271057,
      "kl": 0.0005221366882324219,
      "learning_rate": 9.915301796042075e-07,
      "loss": 0.0143,
      "num_tokens": 1204973.0,
      "reward": 0.0625,
      "reward_std": 0.16151439771056175,
      "rewards/code_format_reward": 0.0625,
      "step": 9
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.46875,
      "epoch": 0.16806722689075632,
      "grad_norm": 0.4631551206111908,
      "kl": 0.0004162788391113281,
      "learning_rate": 9.884847360911167e-07,
      "loss": -0.0393,
      "num_tokens": 1348945.0,
      "reward": 0.0546875,
      "reward_std": 0.12415501847863197,
      "rewards/code_format_reward": 0.0546875,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 573.3671875,
      "epoch": 0.18487394957983194,
      "grad_norm": 0.46803170442581177,
      "kl": 0.0006732940673828125,
      "learning_rate": 9.84979382640675e-07,
      "loss": -0.0024,
      "num_tokens": 1478632.0,
      "reward": 0.0703125,
      "reward_std": 0.15308689512312412,
      "rewards/code_format_reward": 0.0703125,
      "step": 11
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.7265625,
      "epoch": 0.20168067226890757,
      "grad_norm": 0.40413135290145874,
      "kl": 0.0005345344543457031,
      "learning_rate": 9.81017795884594e-07,
      "loss": 0.0043,
      "num_tokens": 1619165.0,
      "reward": 0.0390625,
      "reward_std": 0.09522314183413982,
      "rewards/code_format_reward": 0.0390625,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 564.96875,
      "epoch": 0.2184873949579832,
      "grad_norm": 0.5944288372993469,
      "kl": 0.0011510848999023438,
      "learning_rate": 9.766041309803217e-07,
      "loss": 0.009,
      "num_tokens": 1751681.0,
      "reward": 0.1015625,
      "reward_std": 0.20411096327006817,
      "rewards/code_format_reward": 0.1015625,
      "step": 13
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 509.171875,
      "epoch": 0.23529411764705882,
      "grad_norm": 0.584949254989624,
      "kl": 0.0027894973754882812,
      "learning_rate": 9.717430172528546e-07,
      "loss": 0.0372,
      "num_tokens": 1872271.0,
      "reward": 0.125,
      "reward_std": 0.2398776337504387,
      "rewards/code_format_reward": 0.125,
      "step": 14
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 547.15625,
      "epoch": 0.25210084033613445,
      "grad_norm": 0.6391139030456543,
      "kl": 0.003070831298828125,
      "learning_rate": 9.66439553339217e-07,
      "loss": 0.0615,
      "num_tokens": 2002227.0,
      "reward": 0.296875,
      "reward_std": 0.32535743340849876,
      "rewards/code_format_reward": 0.296875,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 557.8125,
      "epoch": 0.2689075630252101,
      "grad_norm": 0.5913247466087341,
      "kl": 0.00417327880859375,
      "learning_rate": 9.60699301840693e-07,
      "loss": -0.013,
      "num_tokens": 2130827.0,
      "reward": 0.234375,
      "reward_std": 0.2630178965628147,
      "rewards/code_format_reward": 0.234375,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 454.71875,
      "epoch": 0.2857142857142857,
      "grad_norm": 0.6741214990615845,
      "kl": 0.005828857421875,
      "learning_rate": 9.54528283488428e-07,
      "loss": 0.0233,
      "num_tokens": 2246215.0,
      "reward": 0.3125,
      "reward_std": 0.282998226583004,
      "rewards/code_format_reward": 0.3125,
      "step": 17
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 694.3671875,
      "epoch": 0.3025210084033613,
      "grad_norm": 0.2503214478492737,
      "kl": 0.0009298324584960938,
      "learning_rate": 9.479329708285106e-07,
      "loss": 0.0084,
      "num_tokens": 2399230.0,
      "reward": 0.09375,
      "reward_std": 0.0578637570142746,
      "rewards/code_format_reward": 0.09375,
      "step": 18
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 606.7421875,
      "epoch": 0.31932773109243695,
      "grad_norm": 0.3981899917125702,
      "kl": 0.003948211669921875,
      "learning_rate": 9.409202814331679e-07,
      "loss": -0.0357,
      "num_tokens": 2537525.0,
      "reward": 0.265625,
      "reward_std": 0.14806943386793137,
      "rewards/code_format_reward": 0.265625,
      "step": 19
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 442.84375,
      "epoch": 0.33613445378151263,
      "grad_norm": 0.5728533864021301,
      "kl": 0.0042285919189453125,
      "learning_rate": 9.334975706451861e-07,
      "loss": 0.0178,
      "num_tokens": 2653577.0,
      "reward": 0.3671875,
      "reward_std": 0.2314501255750656,
      "rewards/code_format_reward": 0.3671875,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 666.9765625,
      "epoch": 0.35294117647058826,
      "grad_norm": 0.40784379839897156,
      "kl": 0.003398895263671875,
      "learning_rate": 9.256726238631719e-07,
      "loss": 0.0296,
      "num_tokens": 2801526.0,
      "reward": 0.171875,
      "reward_std": 0.12255740165710449,
      "rewards/code_format_reward": 0.171875,
      "step": 21
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 561.1328125,
      "epoch": 0.3697478991596639,
      "grad_norm": 0.3966084122657776,
      "kl": 0.0062713623046875,
      "learning_rate": 9.174536483757448e-07,
      "loss": -0.0061,
      "num_tokens": 2930031.0,
      "reward": 0.3515625,
      "reward_std": 0.11784427240490913,
      "rewards/code_format_reward": 0.3515625,
      "step": 22
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.8125,
      "epoch": 0.3865546218487395,
      "grad_norm": 0.4765741229057312,
      "kl": 0.004345893859863281,
      "learning_rate": 9.088492647532243e-07,
      "loss": 0.0237,
      "num_tokens": 3069271.0,
      "reward": 0.25,
      "reward_std": 0.1065337061882019,
      "rewards/code_format_reward": 0.25,
      "step": 23
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 555.90625,
      "epoch": 0.40336134453781514,
      "grad_norm": 0.42516130208969116,
      "kl": 0.0028820037841796875,
      "learning_rate": 8.998684978058422e-07,
      "loss": -0.0071,
      "num_tokens": 3203811.0,
      "reward": 0.3125,
      "reward_std": 0.1422954723238945,
      "rewards/code_format_reward": 0.3125,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.5703125,
      "epoch": 0.42016806722689076,
      "grad_norm": 0.4370390772819519,
      "kl": 0.008594512939453125,
      "learning_rate": 8.905207671179627e-07,
      "loss": 0.0123,
      "num_tokens": 3332956.0,
      "reward": 0.3828125,
      "reward_std": 0.12863079272210598,
      "rewards/code_format_reward": 0.3828125,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.546875,
      "epoch": 0.4369747899159664,
      "grad_norm": 0.6266469955444336,
      "kl": 0.006752490997314453,
      "learning_rate": 8.808158771682401e-07,
      "loss": 0.031,
      "num_tokens": 3451954.0,
      "reward": 0.625,
      "reward_std": 0.1615143995732069,
      "rewards/code_format_reward": 0.625,
      "step": 26
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 623.3125,
      "epoch": 0.453781512605042,
      "grad_norm": 0.3876335620880127,
      "kl": 0.0042819976806640625,
      "learning_rate": 8.707640070460731e-07,
      "loss": -0.0099,
      "num_tokens": 3592458.0,
      "reward": 0.328125,
      "reward_std": 0.10205793380737305,
      "rewards/code_format_reward": 0.328125,
      "step": 27
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 568.046875,
      "epoch": 0.47058823529411764,
      "grad_norm": 0.3156704306602478,
      "kl": 0.006195068359375,
      "learning_rate": 8.60375699775147e-07,
      "loss": 0.0065,
      "num_tokens": 3722080.0,
      "reward": 0.546875,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.546875,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 612.5703125,
      "epoch": 0.48739495798319327,
      "grad_norm": 0.3149147927761078,
      "kl": 0.004520416259765625,
      "learning_rate": 8.496618512552564e-07,
      "loss": 0.0076,
      "num_tokens": 3861281.0,
      "reward": 0.3125,
      "reward_std": 0.0578637570142746,
      "rewards/code_format_reward": 0.3125,
      "step": 29
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 626.0078125,
      "epoch": 0.5042016806722689,
      "grad_norm": 0.41246238350868225,
      "kl": 0.009387016296386719,
      "learning_rate": 8.386336988340129e-07,
      "loss": -0.0028,
      "num_tokens": 4004770.0,
      "reward": 0.34375,
      "reward_std": 0.13258251920342445,
      "rewards/code_format_reward": 0.34375,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.40625,
      "epoch": 0.5210084033613446,
      "grad_norm": 0.2296294867992401,
      "kl": 0.0028858184814453125,
      "learning_rate": 8.273028095204173e-07,
      "loss": 0.0033,
      "num_tokens": 4142014.0,
      "reward": 0.1796875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.1796875,
      "step": 31
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.1328125,
      "epoch": 0.5378151260504201,
      "grad_norm": 0.4845007061958313,
      "kl": 0.0059051513671875,
      "learning_rate": 8.156810678526652e-07,
      "loss": -0.017,
      "num_tokens": 4276015.0,
      "reward": 0.3515625,
      "reward_std": 0.14465449005365372,
      "rewards/code_format_reward": 0.3515625,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 546.265625,
      "epoch": 0.5546218487394958,
      "grad_norm": 0.3567289412021637,
      "kl": 0.005253791809082031,
      "learning_rate": 8.037806634329078e-07,
      "loss": 0.0042,
      "num_tokens": 4407857.0,
      "reward": 0.34375,
      "reward_std": 0.05444391071796417,
      "rewards/code_format_reward": 0.34375,
      "step": 33
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 505.0703125,
      "epoch": 0.5714285714285714,
      "grad_norm": 0.2423199564218521,
      "kl": 0.00742340087890625,
      "learning_rate": 7.916140781420428e-07,
      "loss": 0.0159,
      "num_tokens": 4534298.0,
      "reward": 0.3828125,
      "reward_std": 0.061278700828552246,
      "rewards/code_format_reward": 0.3828125,
      "step": 34
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 544.9296875,
      "epoch": 0.5882352941176471,
      "grad_norm": 0.36438822746276855,
      "kl": 0.011692047119140625,
      "learning_rate": 7.791940730479434e-07,
      "loss": 0.0236,
      "num_tokens": 4665225.0,
      "reward": 0.3359375,
      "reward_std": 0.061278700828552246,
      "rewards/code_format_reward": 0.3359375,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.703125,
      "epoch": 0.6050420168067226,
      "grad_norm": 0.2744235098361969,
      "kl": 0.0067291259765625,
      "learning_rate": 7.665336750208623e-07,
      "loss": -0.0033,
      "num_tokens": 4795883.0,
      "reward": 0.484375,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.484375,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 480.1328125,
      "epoch": 0.6218487394957983,
      "grad_norm": 0.19300399720668793,
      "kl": 0.00769805908203125,
      "learning_rate": 7.536461630700425e-07,
      "loss": -0.0044,
      "num_tokens": 4914988.0,
      "reward": 0.4296875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4296875,
      "step": 37
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 471.4296875,
      "epoch": 0.6386554621848739,
      "grad_norm": 0.01874758116900921,
      "kl": 0.01021575927734375,
      "learning_rate": 7.405450544158706e-07,
      "loss": 0.0001,
      "num_tokens": 5031363.0,
      "reward": 0.5625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5625,
      "step": 38
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.78125,
      "epoch": 0.6554621848739496,
      "grad_norm": 0.056958042085170746,
      "kl": 0.0102996826171875,
      "learning_rate": 7.272440903121791e-07,
      "loss": 0.0001,
      "num_tokens": 5158719.0,
      "reward": 0.375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.375,
      "step": 39
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 587.6953125,
      "epoch": 0.6722689075630253,
      "grad_norm": 0.1463167667388916,
      "kl": 0.006369590759277344,
      "learning_rate": 7.137572216335694e-07,
      "loss": -0.0141,
      "num_tokens": 5294992.0,
      "reward": 0.3671875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.3671875,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 509.125,
      "epoch": 0.6890756302521008,
      "grad_norm": 0.26510393619537354,
      "kl": 0.0092620849609375,
      "learning_rate": 7.000985942428693e-07,
      "loss": -0.0208,
      "num_tokens": 5417744.0,
      "reward": 0.6015625,
      "reward_std": 0.051028965041041374,
      "rewards/code_format_reward": 0.6015625,
      "step": 41
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 408.5625,
      "epoch": 0.7058823529411765,
      "grad_norm": 0.21098148822784424,
      "kl": 0.011627197265625,
      "learning_rate": 6.862825341540778e-07,
      "loss": 0.0141,
      "num_tokens": 5528912.0,
      "reward": 0.6015625,
      "reward_std": 0.03234682232141495,
      "rewards/code_format_reward": 0.6015625,
      "step": 42
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 544.2578125,
      "epoch": 0.7226890756302521,
      "grad_norm": 0.2268461138010025,
      "kl": 0.00897979736328125,
      "learning_rate": 6.723235325063543e-07,
      "loss": -0.0046,
      "num_tokens": 5657177.0,
      "reward": 0.4921875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4921875,
      "step": 43
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 474.453125,
      "epoch": 0.7394957983193278,
      "grad_norm": 0.012474104762077332,
      "kl": 0.00955963134765625,
      "learning_rate": 6.582362303648142e-07,
      "loss": 0.0001,
      "num_tokens": 5780899.0,
      "reward": 0.625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.625,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 563.0078125,
      "epoch": 0.7563025210084033,
      "grad_norm": 0.3377431333065033,
      "kl": 0.00730133056640625,
      "learning_rate": 6.440354033640738e-07,
      "loss": 0.0134,
      "num_tokens": 5912020.0,
      "reward": 0.4765625,
      "reward_std": 0.051028965041041374,
      "rewards/code_format_reward": 0.4765625,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 476.9609375,
      "epoch": 0.773109243697479,
      "grad_norm": 0.0116049125790596,
      "kl": 0.0107574462890625,
      "learning_rate": 6.297359462106502e-07,
      "loss": 0.0001,
      "num_tokens": 6029263.0,
      "reward": 0.625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.625,
      "step": 46
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 536.6484375,
      "epoch": 0.7899159663865546,
      "grad_norm": 0.26681584119796753,
      "kl": 0.006664276123046875,
      "learning_rate": 6.153528570604707e-07,
      "loss": 0.0023,
      "num_tokens": 6159890.0,
      "reward": 0.359375,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.359375,
      "step": 47
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 558.078125,
      "epoch": 0.8067226890756303,
      "grad_norm": 0.3406285345554352,
      "kl": 0.007928848266601562,
      "learning_rate": 6.00901221787878e-07,
      "loss": -0.024,
      "num_tokens": 6292572.0,
      "reward": 0.421875,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.421875,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 543.0625,
      "epoch": 0.8235294117647058,
      "grad_norm": 0.2923266887664795,
      "kl": 0.009540557861328125,
      "learning_rate": 5.86396198162632e-07,
      "loss": -0.011,
      "num_tokens": 6423868.0,
      "reward": 0.421875,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.421875,
      "step": 49
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.78125,
      "epoch": 0.8403361344537815,
      "grad_norm": 0.3251660466194153,
      "kl": 0.00586700439453125,
      "learning_rate": 5.718529999515017e-07,
      "loss": 0.0111,
      "num_tokens": 6556552.0,
      "reward": 0.3046875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.3046875,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 581.1640625,
      "epoch": 0.8571428571428571,
      "grad_norm": 0.17117629945278168,
      "kl": 0.007695198059082031,
      "learning_rate": 5.572868809611257e-07,
      "loss": -0.019,
      "num_tokens": 6694565.0,
      "reward": 0.296875,
      "reward_std": 0.0289318785071373,
      "rewards/code_format_reward": 0.296875,
      "step": 51
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 507.4453125,
      "epoch": 0.8739495798319328,
      "grad_norm": 0.2713668644428253,
      "kl": 0.00688934326171875,
      "learning_rate": 5.427131190388743e-07,
      "loss": 0.0161,
      "num_tokens": 6817558.0,
      "reward": 0.4921875,
      "reward_std": 0.0657544732093811,
      "rewards/code_format_reward": 0.4921875,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 524.703125,
      "epoch": 0.8907563025210085,
      "grad_norm": 0.2056041955947876,
      "kl": 0.012969970703125,
      "learning_rate": 5.281470000484985e-07,
      "loss": 0.0066,
      "num_tokens": 6943264.0,
      "reward": 0.4296875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4296875,
      "step": 53
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.6875,
      "epoch": 0.907563025210084,
      "grad_norm": 0.23319439589977264,
      "kl": 0.007537841796875,
      "learning_rate": 5.136038018373682e-07,
      "loss": 0.0105,
      "num_tokens": 7073792.0,
      "reward": 0.4296875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4296875,
      "step": 54
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 529.8125,
      "epoch": 0.9243697478991597,
      "grad_norm": 0.19098737835884094,
      "kl": 0.00901031494140625,
      "learning_rate": 4.990987782121221e-07,
      "loss": -0.0105,
      "num_tokens": 7201304.0,
      "reward": 0.3671875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.3671875,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 526.3515625,
      "epoch": 0.9411764705882353,
      "grad_norm": 0.019878050312399864,
      "kl": 0.00701141357421875,
      "learning_rate": 4.846471429395295e-07,
      "loss": 0.0001,
      "num_tokens": 7331461.0,
      "reward": 0.375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.375,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 486.59375,
      "epoch": 0.957983193277311,
      "grad_norm": 0.010634235106408596,
      "kl": 0.0092620849609375,
      "learning_rate": 4.7026405378934975e-07,
      "loss": 0.0001,
      "num_tokens": 7452657.0,
      "reward": 0.5625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5625,
      "step": 57
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 413.234375,
      "epoch": 0.9747899159663865,
      "grad_norm": 0.210678830742836,
      "kl": 0.011688232421875,
      "learning_rate": 4.5596459663592625e-07,
      "loss": 0.0237,
      "num_tokens": 7560991.0,
      "reward": 0.6171875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.6171875,
      "step": 58
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.8452529907227,
      "epoch": 0.9915966386554622,
      "grad_norm": 0.3087288439273834,
      "kl": 0.00782012939453125,
      "learning_rate": 4.41763769635186e-07,
      "loss": -0.0165,
      "num_tokens": 7690230.0,
      "reward": 0.4140625,
      "reward_std": 0.06629125960171223,
      "rewards/code_format_reward": 0.4140625,
      "step": 59
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 571.84375,
      "epoch": 1.0168067226890756,
      "grad_norm": 0.2766449451446533,
      "kl": 0.009830474853515625,
      "learning_rate": 4.2767646749364574e-07,
      "loss": 0.0012,
      "num_tokens": 7825994.0,
      "reward": 0.359375,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.359375,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 549.734375,
      "epoch": 1.0336134453781514,
      "grad_norm": 0.008657719939947128,
      "kl": 0.00926971435546875,
      "learning_rate": 4.1371746584592227e-07,
      "loss": 0.0001,
      "num_tokens": 7954624.0,
      "reward": 0.4375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.4375,
      "step": 61
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.4296875,
      "epoch": 1.050420168067227,
      "grad_norm": 0.00941454991698265,
      "kl": 0.00757598876953125,
      "learning_rate": 3.999014057571308e-07,
      "loss": 0.0001,
      "num_tokens": 8095783.0,
      "reward": 0.3125,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.3125,
      "step": 62
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.0625,
      "epoch": 1.0672268907563025,
      "grad_norm": 0.2536774277687073,
      "kl": 0.0111541748046875,
      "learning_rate": 3.862427783664306e-07,
      "loss": -0.0146,
      "num_tokens": 8209831.0,
      "reward": 0.609375,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.609375,
      "step": 63
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 585.3203125,
      "epoch": 1.084033613445378,
      "grad_norm": 0.20105506479740143,
      "kl": 0.004669189453125,
      "learning_rate": 3.7275590968782087e-07,
      "loss": 0.0095,
      "num_tokens": 8347160.0,
      "reward": 0.2421875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.2421875,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 493.7578125,
      "epoch": 1.1008403361344539,
      "grad_norm": 0.010381446219980717,
      "kl": 0.00858306884765625,
      "learning_rate": 3.594549455841296e-07,
      "loss": 0.0001,
      "num_tokens": 8471065.0,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5,
      "step": 65
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 487.8125,
      "epoch": 1.1176470588235294,
      "grad_norm": 0.21298491954803467,
      "kl": 0.007595062255859375,
      "learning_rate": 3.4635383692995755e-07,
      "loss": 0.011,
      "num_tokens": 8591161.0,
      "reward": 0.4296875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4296875,
      "step": 66
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 570.90625,
      "epoch": 1.134453781512605,
      "grad_norm": 0.005000706762075424,
      "kl": 0.005382537841796875,
      "learning_rate": 3.3346632497913773e-07,
      "loss": 0.0001,
      "num_tokens": 8723157.0,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.25,
      "step": 67
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 406.5625,
      "epoch": 1.1512605042016806,
      "grad_norm": 0.2167414426803589,
      "kl": 0.011280059814453125,
      "learning_rate": 3.208059269520568e-07,
      "loss": 0.0024,
      "num_tokens": 8834893.0,
      "reward": 0.5546875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.5546875,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 519.734375,
      "epoch": 1.1680672268907564,
      "grad_norm": 0.22041496634483337,
      "kl": 0.011386871337890625,
      "learning_rate": 3.083859218579573e-07,
      "loss": 0.0064,
      "num_tokens": 8962931.0,
      "reward": 0.4765625,
      "reward_std": 0.03234682232141495,
      "rewards/code_format_reward": 0.4765625,
      "step": 69
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 552.875,
      "epoch": 1.184873949579832,
      "grad_norm": 0.008582229726016521,
      "kl": 0.0049285888671875,
      "learning_rate": 2.9621933656709207e-07,
      "loss": 0.0,
      "num_tokens": 9094475.0,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.25,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 505.3203125,
      "epoch": 1.2016806722689075,
      "grad_norm": 0.22665004432201385,
      "kl": 0.00914764404296875,
      "learning_rate": 2.843189321473349e-07,
      "loss": -0.0036,
      "num_tokens": 9220108.0,
      "reward": 0.4921875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4921875,
      "step": 71
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 603.0859375,
      "epoch": 1.2184873949579833,
      "grad_norm": 0.007497387006878853,
      "kl": 0.004810333251953125,
      "learning_rate": 2.7269719047958267e-07,
      "loss": 0.0,
      "num_tokens": 9360135.0,
      "reward": 0.25,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.25,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 546.234375,
      "epoch": 1.2352941176470589,
      "grad_norm": 0.007375677116215229,
      "kl": 0.005222320556640625,
      "learning_rate": 2.613663011659871e-07,
      "loss": 0.0001,
      "num_tokens": 9490949.0,
      "reward": 0.3125,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.3125,
      "step": 73
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 489.0859375,
      "epoch": 1.2521008403361344,
      "grad_norm": 0.24114979803562164,
      "kl": 0.0102996826171875,
      "learning_rate": 2.5033814874474356e-07,
      "loss": 0.0336,
      "num_tokens": 9614784.0,
      "reward": 0.5546875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.5546875,
      "step": 74
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 477.3828125,
      "epoch": 1.26890756302521,
      "grad_norm": 0.013432592153549194,
      "kl": 0.01093292236328125,
      "learning_rate": 2.3962430022485305e-07,
      "loss": 0.0001,
      "num_tokens": 9735377.0,
      "reward": 0.625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.625,
      "step": 75
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 555.890625,
      "epoch": 1.2857142857142856,
      "grad_norm": 0.013460615649819374,
      "kl": 0.0073699951171875,
      "learning_rate": 2.2923599295392694e-07,
      "loss": 0.0001,
      "num_tokens": 9867139.0,
      "reward": 0.4375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.4375,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 409.7734375,
      "epoch": 1.3025210084033614,
      "grad_norm": 0.3048118054866791,
      "kl": 0.0102081298828125,
      "learning_rate": 2.1918412283175994e-07,
      "loss": -0.003,
      "num_tokens": 9980222.0,
      "reward": 0.671875,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.671875,
      "step": 77
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.71875,
      "epoch": 1.319327731092437,
      "grad_norm": 0.2550560235977173,
      "kl": 0.007965087890625,
      "learning_rate": 2.0947923288203713e-07,
      "loss": 0.0955,
      "num_tokens": 10107666.0,
      "reward": 0.484375,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.484375,
      "step": 78
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 517.8125,
      "epoch": 1.3361344537815127,
      "grad_norm": 0.2195437252521515,
      "kl": 0.00948333740234375,
      "learning_rate": 2.0013150219415793e-07,
      "loss": -0.0028,
      "num_tokens": 10232114.0,
      "reward": 0.46875,
      "reward_std": 0.033407654613256454,
      "rewards/code_format_reward": 0.46875,
      "step": 79
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.390625,
      "epoch": 1.3529411764705883,
      "grad_norm": 0.008532202802598476,
      "kl": 0.006832122802734375,
      "learning_rate": 1.9115073524677572e-07,
      "loss": 0.0001,
      "num_tokens": 10357436.0,
      "reward": 0.4375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.4375,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.96875,
      "epoch": 1.3697478991596639,
      "grad_norm": 0.011241392232477665,
      "kl": 0.00920867919921875,
      "learning_rate": 1.8254635162425503e-07,
      "loss": 0.0001,
      "num_tokens": 10481120.0,
      "reward": 0.625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.625,
      "step": 81
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 485.390625,
      "epoch": 1.3865546218487395,
      "grad_norm": 0.2583635151386261,
      "kl": 0.01049041748046875,
      "learning_rate": 1.7432737613682807e-07,
      "loss": 0.007,
      "num_tokens": 10603258.0,
      "reward": 0.5546875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.5546875,
      "step": 82
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.484375,
      "epoch": 1.403361344537815,
      "grad_norm": 0.007445089519023895,
      "kl": 0.00675201416015625,
      "learning_rate": 1.6650242935481388e-07,
      "loss": 0.0001,
      "num_tokens": 10736448.0,
      "reward": 0.375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.375,
      "step": 83
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 565.8515625,
      "epoch": 1.4201680672268908,
      "grad_norm": 0.012575228698551655,
      "kl": 0.006989479064941406,
      "learning_rate": 1.59079718566832e-07,
      "loss": 0.0001,
      "num_tokens": 10867749.0,
      "reward": 0.375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.375,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.5,
      "epoch": 1.4369747899159664,
      "grad_norm": 0.22415108978748322,
      "kl": 0.0146026611328125,
      "learning_rate": 1.5206702917148945e-07,
      "loss": 0.0055,
      "num_tokens": 10988109.0,
      "reward": 0.6171875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.6171875,
      "step": 85
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 425.671875,
      "epoch": 1.453781512605042,
      "grad_norm": 0.012631588615477085,
      "kl": 0.00989532470703125,
      "learning_rate": 1.4547171651157214e-07,
      "loss": 0.0001,
      "num_tokens": 11100171.0,
      "reward": 0.5625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5625,
      "step": 86
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 484.8125,
      "epoch": 1.4705882352941178,
      "grad_norm": 0.19023600220680237,
      "kl": 0.01287078857421875,
      "learning_rate": 1.3930069815930697e-07,
      "loss": -0.0073,
      "num_tokens": 11218611.0,
      "reward": 0.5546875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.5546875,
      "step": 87
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 542.421875,
      "epoch": 1.4873949579831933,
      "grad_norm": 0.18787921965122223,
      "kl": 0.005619049072265625,
      "learning_rate": 1.3356044666078315e-07,
      "loss": -0.0035,
      "num_tokens": 11349593.0,
      "reward": 0.2421875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.2421875,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 414.28125,
      "epoch": 1.504201680672269,
      "grad_norm": 0.2461971491575241,
      "kl": 0.01172637939453125,
      "learning_rate": 1.2825698274714542e-07,
      "loss": 0.0123,
      "num_tokens": 11459157.0,
      "reward": 0.7421875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.7421875,
      "step": 89
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 535.8671875,
      "epoch": 1.5210084033613445,
      "grad_norm": 0.009749805554747581,
      "kl": 0.0081024169921875,
      "learning_rate": 1.233958690196783e-07,
      "loss": 0.0001,
      "num_tokens": 11587884.0,
      "reward": 0.4375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.4375,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 599.015625,
      "epoch": 1.53781512605042,
      "grad_norm": 0.0069138724356889725,
      "kl": 0.00567626953125,
      "learning_rate": 1.1898220411540583e-07,
      "loss": 0.0001,
      "num_tokens": 11723398.0,
      "reward": 0.375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.375,
      "step": 91
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 516.984375,
      "epoch": 1.5546218487394958,
      "grad_norm": 0.26793172955513,
      "kl": 0.00763702392578125,
      "learning_rate": 1.1502061735932499e-07,
      "loss": -0.0202,
      "num_tokens": 11850164.0,
      "reward": 0.359375,
      "reward_std": 0.04419417306780815,
      "rewards/code_format_reward": 0.359375,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 506.890625,
      "epoch": 1.5714285714285714,
      "grad_norm": 0.010911713354289532,
      "kl": 0.00727081298828125,
      "learning_rate": 1.115152639088833e-07,
      "loss": 0.0001,
      "num_tokens": 11975358.0,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5,
      "step": 93
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 569.6796875,
      "epoch": 1.5882352941176472,
      "grad_norm": 0.00520605593919754,
      "kl": 0.0041522979736328125,
      "learning_rate": 1.0846982039579242e-07,
      "loss": 0.0,
      "num_tokens": 12111517.0,
      "reward": 0.3125,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.3125,
      "step": 94
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 391.6796875,
      "epoch": 1.6050420168067228,
      "grad_norm": 0.014589727856218815,
      "kl": 0.0121002197265625,
      "learning_rate": 1.0588748106974918e-07,
      "loss": 0.0001,
      "num_tokens": 12219404.0,
      "reward": 0.625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.625,
      "step": 95
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 583.4453125,
      "epoch": 1.6218487394957983,
      "grad_norm": 0.010287419892847538,
      "kl": 0.0075836181640625,
      "learning_rate": 1.0377095444810871e-07,
      "loss": 0.0001,
      "num_tokens": 12354245.0,
      "reward": 0.4375,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.4375,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 551.484375,
      "epoch": 1.638655462184874,
      "grad_norm": 0.21819841861724854,
      "kl": 0.00997161865234375,
      "learning_rate": 1.0212246047502372e-07,
      "loss": -0.0046,
      "num_tokens": 12482835.0,
      "reward": 0.4296875,
      "reward_std": 0.022097086533904076,
      "rewards/code_format_reward": 0.4296875,
      "step": 97
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 486.953125,
      "epoch": 1.6554621848739495,
      "grad_norm": 0.011982027441263199,
      "kl": 0.0090484619140625,
      "learning_rate": 1.0094372819302977e-07,
      "loss": 0.0001,
      "num_tokens": 12605909.0,
      "reward": 0.5625,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5625,
      "step": 98
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 516.625,
      "epoch": 1.6722689075630253,
      "grad_norm": 0.008852253668010235,
      "kl": 0.00608062744140625,
      "learning_rate": 1.0023599392951829e-07,
      "loss": 0.0001,
      "num_tokens": 12732141.0,
      "reward": 0.3125,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.3125,
      "step": 99
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 505.9921875,
      "epoch": 1.6890756302521008,
      "grad_norm": 0.016563883051276207,
      "kl": 0.0088958740234375,
      "learning_rate": 1e-07,
      "loss": 0.0001,
      "num_tokens": 12854748.0,
      "reward": 0.5,
      "reward_std": 0.0,
      "rewards/code_format_reward": 0.5,
      "step": 100
    },
    {
      "epoch": 1.6890756302521008,
      "step": 100,
      "total_flos": 0.0,
      "train_loss": 0.003220523014695118,
      "train_runtime": 15510.0274,
      "train_samples_per_second": 0.825,
      "train_steps_per_second": 0.006
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}