{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 939,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003194888178913738,
      "grad_norm": 293.204729539606,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -3.59375,
      "logits/rejected": -3.796875,
      "logps/chosen": -332.0,
      "logps/rejected": -151.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03194888178913738,
      "grad_norm": 217.2791429051027,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.734375,
      "logps/chosen": -354.0,
      "logps/rejected": -153.0,
      "loss": 0.6776,
      "rewards/accuracies": 0.3888888955116272,
      "rewards/chosen": 0.01806640625,
      "rewards/margins": 0.035400390625,
      "rewards/rejected": -0.017333984375,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 82.81496720870953,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -3.515625,
      "logits/rejected": -3.625,
      "logps/chosen": -338.0,
      "logps/rejected": -153.0,
      "loss": 0.4387,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 0.53125,
      "rewards/margins": 0.63671875,
      "rewards/rejected": -0.107421875,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 23.086232071557887,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.75,
      "logps/chosen": -346.0,
      "logps/rejected": -160.0,
      "loss": 0.1306,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 1.2890625,
      "rewards/margins": 2.625,
      "rewards/rejected": -1.328125,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 6.811169742780255,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -3.609375,
      "logits/rejected": -3.734375,
      "logps/chosen": -340.0,
      "logps/rejected": -176.0,
      "loss": 0.0516,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.7421875,
      "rewards/margins": 4.5625,
      "rewards/rejected": -2.796875,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 5.778673051920023,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -3.578125,
      "logits/rejected": -3.78125,
      "logps/chosen": -344.0,
      "logps/rejected": -191.0,
      "loss": 0.0142,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.359375,
      "rewards/margins": 6.65625,
      "rewards/rejected": -4.3125,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 0.7376722698835323,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.390625,
      "logps/chosen": -330.0,
      "logps/rejected": -215.0,
      "loss": 0.0168,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.453125,
      "rewards/margins": 9.125,
      "rewards/rejected": -6.65625,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 0.03643652415335125,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -3.140625,
      "logps/chosen": -340.0,
      "logps/rejected": -239.0,
      "loss": 0.0011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.390625,
      "rewards/margins": 11.6875,
      "rewards/rejected": -9.3125,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 0.10868404295869633,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -3.109375,
      "logits/rejected": -3.078125,
      "logps/chosen": -344.0,
      "logps/rejected": -280.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.046875,
      "rewards/margins": 15.1875,
      "rewards/rejected": -13.125,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 0.006992855864033706,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -3.25,
      "logits/rejected": -3.25,
      "logps/chosen": -328.0,
      "logps/rejected": -288.0,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.03125,
      "rewards/margins": 15.25,
      "rewards/rejected": -13.25,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.008609509406233801,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.40625,
      "logps/chosen": -318.0,
      "logps/rejected": -254.0,
      "loss": 0.005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.390625,
      "rewards/margins": 13.3125,
      "rewards/rejected": -9.875,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.2386745976865093,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.578125,
      "logps/chosen": -316.0,
      "logps/rejected": -236.0,
      "loss": 0.0075,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.40625,
      "rewards/margins": 13.1875,
      "rewards/rejected": -8.8125,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.002832864260207118,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.421875,
      "logps/chosen": -334.0,
      "logps/rejected": -270.0,
      "loss": 0.0048,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.34375,
      "rewards/margins": 16.375,
      "rewards/rejected": -12.0625,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.9051634037890359,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.3125,
      "logps/chosen": -322.0,
      "logps/rejected": -294.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.921875,
      "rewards/margins": 17.5,
      "rewards/rejected": -14.5,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.0007659043303760018,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.4375,
      "logps/chosen": -300.0,
      "logps/rejected": -264.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.75,
      "rewards/margins": 16.625,
      "rewards/rejected": -11.8125,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 0.003549259881696135,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.421875,
      "logps/chosen": -322.0,
      "logps/rejected": -282.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.625,
      "rewards/margins": 18.25,
      "rewards/rejected": -13.5625,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 0.0003206532972349471,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.328125,
      "logps/chosen": -318.0,
      "logps/rejected": -296.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.90625,
      "rewards/margins": 19.625,
      "rewards/rejected": -14.75,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.0016304269501666504,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.515625,
      "logps/chosen": -302.0,
      "logps/rejected": -298.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.09375,
      "rewards/margins": 20.25,
      "rewards/rejected": -15.1875,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.0001715070042755352,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.3125,
      "logps/chosen": -332.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.75,
      "rewards/margins": 21.375,
      "rewards/rejected": -16.625,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.0004512338192394063,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.421875,
      "logps/chosen": -314.0,
      "logps/rejected": -304.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.71875,
      "rewards/margins": 20.25,
      "rewards/rejected": -15.5625,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 0.061438361455547257,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.390625,
      "logps/chosen": -330.0,
      "logps/rejected": -322.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.1875,
      "rewards/margins": 22.25,
      "rewards/rejected": -17.0,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 0.00034284580567239446,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.390625,
      "logps/chosen": -308.0,
      "logps/rejected": -308.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.8125,
      "rewards/margins": 20.75,
      "rewards/rejected": -16.0,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.00020436480931531622,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.359375,
      "logps/chosen": -320.0,
      "logps/rejected": -308.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.59375,
      "rewards/margins": 20.875,
      "rewards/rejected": -16.25,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 0.02256471664210956,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.40625,
      "logps/chosen": -306.0,
      "logps/rejected": -324.0,
      "loss": 0.0598,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.53125,
      "rewards/margins": 22.125,
      "rewards/rejected": -17.625,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.002374647523069546,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.453125,
      "logps/chosen": -316.0,
      "logps/rejected": -286.0,
      "loss": 0.002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.5625,
      "rewards/margins": 17.25,
      "rewards/rejected": -12.6875,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.029823129672964955,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.4375,
      "logps/chosen": -318.0,
      "logps/rejected": -272.0,
      "loss": 0.0119,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.28125,
      "rewards/margins": 15.75,
      "rewards/rejected": -11.5,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 0.02382260534889851,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.46875,
      "logps/chosen": -316.0,
      "logps/rejected": -284.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.65625,
      "rewards/margins": 17.875,
      "rewards/rejected": -13.25,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 0.00042613742269249753,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.453125,
      "logps/chosen": -314.0,
      "logps/rejected": -282.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.625,
      "rewards/margins": 17.75,
      "rewards/rejected": -13.1875,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 226.14707960640536,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.4375,
      "logps/chosen": -344.0,
      "logps/rejected": -308.0,
      "loss": 0.0071,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.59375,
      "rewards/margins": 17.75,
      "rewards/rejected": -15.1875,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 0.00277159351280118,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.65625,
      "logps/chosen": -298.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.625,
      "rewards/margins": 17.625,
      "rewards/rejected": -13.0,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 0.0006344155132498529,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.65625,
      "logps/chosen": -320.0,
      "logps/rejected": -296.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.71875,
      "rewards/margins": 18.875,
      "rewards/rejected": -14.125,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.004108679708822286,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.578125,
      "logps/chosen": -300.0,
      "logps/rejected": -288.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.15625,
      "rewards/margins": 17.875,
      "rewards/rejected": -13.75,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -3.453125,
      "eval_logits/rejected": -3.421875,
      "eval_logps/chosen": -284.0,
      "eval_logps/rejected": -314.0,
      "eval_loss": 5.706846877728822e-06,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 4.3125,
      "eval_rewards/margins": 21.625,
      "eval_rewards/rejected": -17.25,
      "eval_runtime": 21.7163,
      "eval_samples_per_second": 9.21,
      "eval_steps_per_second": 0.322,
      "step": 313
    },
    {
      "epoch": 1.0223642172523961,
      "grad_norm": 8.473523385083432e-05,
      "learning_rate": 3.662721893491124e-07,
      "logits/chosen": -3.359375,
      "logits/rejected": -3.359375,
      "logps/chosen": -298.0,
      "logps/rejected": -304.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.75,
      "rewards/margins": 20.625,
      "rewards/rejected": -15.875,
      "step": 320
    },
    {
      "epoch": 1.0543130990415335,
      "grad_norm": 6.652483172406472e-05,
      "learning_rate": 3.603550295857988e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.453125,
      "logps/chosen": -288.0,
      "logps/rejected": -310.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.375,
      "rewards/margins": 20.75,
      "rewards/rejected": -16.375,
      "step": 330
    },
    {
      "epoch": 1.0862619808306708,
      "grad_norm": 0.001557792325180698,
      "learning_rate": 3.544378698224852e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.40625,
      "logps/chosen": -332.0,
      "logps/rejected": -298.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.125,
      "rewards/margins": 20.25,
      "rewards/rejected": -15.25,
      "step": 340
    },
    {
      "epoch": 1.1182108626198084,
      "grad_norm": 0.000869024297002781,
      "learning_rate": 3.485207100591716e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.40625,
      "logps/chosen": -288.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.09375,
      "rewards/margins": 20.375,
      "rewards/rejected": -16.25,
      "step": 350
    },
    {
      "epoch": 1.1501597444089458,
      "grad_norm": 0.000502123087378009,
      "learning_rate": 3.42603550295858e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.4375,
      "logps/chosen": -314.0,
      "logps/rejected": -306.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.09375,
      "rewards/margins": 20.625,
      "rewards/rejected": -15.5,
      "step": 360
    },
    {
      "epoch": 1.182108626198083,
      "grad_norm": 0.00015731944634404804,
      "learning_rate": 3.366863905325444e-07,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.53125,
      "logps/chosen": -332.0,
      "logps/rejected": -314.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.1875,
      "rewards/margins": 21.375,
      "rewards/rejected": -16.125,
      "step": 370
    },
    {
      "epoch": 1.2140575079872205,
      "grad_norm": 0.00042280258742680697,
      "learning_rate": 3.3076923076923075e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.5625,
      "logps/chosen": -304.0,
      "logps/rejected": -300.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.6875,
      "rewards/margins": 19.375,
      "rewards/rejected": -14.625,
      "step": 380
    },
    {
      "epoch": 1.2460063897763578,
      "grad_norm": 0.00042295389832085504,
      "learning_rate": 3.2485207100591716e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.46875,
      "logps/chosen": -322.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.03125,
      "rewards/margins": 21.5,
      "rewards/rejected": -16.5,
      "step": 390
    },
    {
      "epoch": 1.2779552715654952,
      "grad_norm": 4.5067081900441135e-05,
      "learning_rate": 3.189349112426035e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.4375,
      "logps/chosen": -308.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.71875,
      "rewards/margins": 21.375,
      "rewards/rejected": -16.625,
      "step": 400
    },
    {
      "epoch": 1.3099041533546325,
      "grad_norm": 0.0002963548393975308,
      "learning_rate": 3.130177514792899e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.5,
      "logps/chosen": -304.0,
      "logps/rejected": -308.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.8125,
      "rewards/margins": 20.625,
      "rewards/rejected": -15.8125,
      "step": 410
    },
    {
      "epoch": 1.34185303514377,
      "grad_norm": 0.0002097908874806944,
      "learning_rate": 3.071005917159763e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.53125,
      "logps/chosen": -296.0,
      "logps/rejected": -304.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.25,
      "rewards/margins": 20.0,
      "rewards/rejected": -15.75,
      "step": 420
    },
    {
      "epoch": 1.3738019169329074,
      "grad_norm": 0.003056570445261906,
      "learning_rate": 3.011834319526627e-07,
      "logits/chosen": -3.5,
      "logits/rejected": -3.4375,
      "logps/chosen": -324.0,
      "logps/rejected": -310.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.03125,
      "rewards/margins": 21.5,
      "rewards/rejected": -16.5,
      "step": 430
    },
    {
      "epoch": 1.4057507987220448,
      "grad_norm": 0.0011576116160918874,
      "learning_rate": 2.952662721893491e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.453125,
      "logps/chosen": -296.0,
      "logps/rejected": -332.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.78125,
      "rewards/margins": 22.25,
      "rewards/rejected": -17.5,
      "step": 440
    },
    {
      "epoch": 1.4376996805111821,
      "grad_norm": 8.113443408051287e-05,
      "learning_rate": 2.893491124260355e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.34375,
      "logps/chosen": -308.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.5625,
      "rewards/margins": 21.375,
      "rewards/rejected": -16.75,
      "step": 450
    },
    {
      "epoch": 1.4696485623003195,
      "grad_norm": 0.00015803918878388854,
      "learning_rate": 2.834319526627219e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.484375,
      "logps/chosen": -286.0,
      "logps/rejected": -308.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.03125,
      "rewards/margins": 20.125,
      "rewards/rejected": -16.0,
      "step": 460
    },
    {
      "epoch": 1.5015974440894568,
      "grad_norm": 0.0006028423743925809,
      "learning_rate": 2.7751479289940824e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.40625,
      "logps/chosen": -316.0,
      "logps/rejected": -318.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.8125,
      "rewards/margins": 21.625,
      "rewards/rejected": -16.75,
      "step": 470
    },
    {
      "epoch": 1.5335463258785942,
      "grad_norm": 5.114674647424102e-05,
      "learning_rate": 2.715976331360947e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.40625,
      "logps/chosen": -290.0,
      "logps/rejected": -304.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.71875,
      "rewards/margins": 20.75,
      "rewards/rejected": -16.0,
      "step": 480
    },
    {
      "epoch": 1.5654952076677318,
      "grad_norm": 5.075233445509427e-05,
      "learning_rate": 2.6568047337278106e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.453125,
      "logps/chosen": -318.0,
      "logps/rejected": -324.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.5,
      "rewards/margins": 21.25,
      "rewards/rejected": -16.75,
      "step": 490
    },
    {
      "epoch": 1.5974440894568689,
      "grad_norm": 3.736375474749778e-05,
      "learning_rate": 2.5976331360946746e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.375,
      "logps/chosen": -326.0,
      "logps/rejected": -338.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.0,
      "rewards/margins": 24.125,
      "rewards/rejected": -19.125,
      "step": 500
    },
    {
      "epoch": 1.6293929712460065,
      "grad_norm": 4.4560481962637565e-05,
      "learning_rate": 2.538461538461538e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.375,
      "logps/chosen": -326.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.9375,
      "rewards/margins": 22.125,
      "rewards/rejected": -17.25,
      "step": 510
    },
    {
      "epoch": 1.6613418530351438,
      "grad_norm": 3.3715749968110866e-05,
      "learning_rate": 2.479289940828402e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.4375,
      "logps/chosen": -312.0,
      "logps/rejected": -322.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.4375,
      "rewards/margins": 21.625,
      "rewards/rejected": -17.125,
      "step": 520
    },
    {
      "epoch": 1.6932907348242812,
      "grad_norm": 1.2624006102344727e-05,
      "learning_rate": 2.420118343195266e-07,
      "logits/chosen": -3.28125,
      "logits/rejected": -3.3125,
      "logps/chosen": -310.0,
      "logps/rejected": -324.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.40625,
      "rewards/margins": 23.5,
      "rewards/rejected": -18.0,
      "step": 530
    },
    {
      "epoch": 1.7252396166134185,
      "grad_norm": 0.004751209486730544,
      "learning_rate": 2.36094674556213e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.390625,
      "logps/chosen": -302.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.8125,
      "rewards/margins": 22.25,
      "rewards/rejected": -17.375,
      "step": 540
    },
    {
      "epoch": 1.7571884984025559,
      "grad_norm": 8.97768658621038e-05,
      "learning_rate": 2.301775147928994e-07,
      "logits/chosen": -3.359375,
      "logits/rejected": -3.40625,
      "logps/chosen": -304.0,
      "logps/rejected": -324.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.6875,
      "rewards/margins": 23.0,
      "rewards/rejected": -18.375,
      "step": 550
    },
    {
      "epoch": 1.7891373801916934,
      "grad_norm": 0.0010722810649243359,
      "learning_rate": 2.242603550295858e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.34375,
      "logps/chosen": -316.0,
      "logps/rejected": -320.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.21875,
      "rewards/margins": 22.25,
      "rewards/rejected": -17.125,
      "step": 560
    },
    {
      "epoch": 1.8210862619808306,
      "grad_norm": 0.00015226817953241815,
      "learning_rate": 2.1834319526627217e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.46875,
      "logps/chosen": -318.0,
      "logps/rejected": -324.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.0,
      "rewards/margins": 22.5,
      "rewards/rejected": -17.5,
      "step": 570
    },
    {
      "epoch": 1.8530351437699681,
      "grad_norm": 8.849357867841071e-05,
      "learning_rate": 2.1242603550295858e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.375,
      "logps/chosen": -316.0,
      "logps/rejected": -328.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.0625,
      "rewards/margins": 22.5,
      "rewards/rejected": -17.375,
      "step": 580
    },
    {
      "epoch": 1.8849840255591053,
      "grad_norm": 3.898268585159925e-05,
      "learning_rate": 2.0650887573964496e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.4375,
      "logps/chosen": -290.0,
      "logps/rejected": -328.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.3125,
      "rewards/margins": 21.625,
      "rewards/rejected": -17.25,
      "step": 590
    },
    {
      "epoch": 1.9169329073482428,
      "grad_norm": 3.925105601437176e-06,
      "learning_rate": 2.0059171597633133e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.421875,
      "logps/chosen": -314.0,
      "logps/rejected": -332.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.53125,
      "rewards/margins": 22.125,
      "rewards/rejected": -17.625,
      "step": 600
    },
    {
      "epoch": 1.9488817891373802,
      "grad_norm": 0.0031742985077368533,
      "learning_rate": 1.9467455621301774e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.296875,
      "logps/chosen": -308.0,
      "logps/rejected": -318.0,
      "loss": 0.0334,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.90625,
      "rewards/margins": 21.375,
      "rewards/rejected": -16.5,
      "step": 610
    },
    {
      "epoch": 1.9808306709265175,
      "grad_norm": 0.001094063345385275,
      "learning_rate": 1.8875739644970412e-07,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.5,
      "logps/chosen": -296.0,
      "logps/rejected": -304.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.34375,
      "rewards/margins": 20.0,
      "rewards/rejected": -15.625,
      "step": 620
    },
    {
      "epoch": 2.0,
      "eval_logits/chosen": -3.546875,
      "eval_logits/rejected": -3.4375,
      "eval_logps/chosen": -288.0,
      "eval_logps/rejected": -310.0,
      "eval_loss": 8.168307431333233e-06,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 3.78125,
      "eval_rewards/margins": 20.5,
      "eval_rewards/rejected": -16.75,
      "eval_runtime": 21.7394,
      "eval_samples_per_second": 9.2,
      "eval_steps_per_second": 0.322,
      "step": 626
    },
    {
      "epoch": 2.012779552715655,
      "grad_norm": 0.0006025863018103904,
      "learning_rate": 1.8284023668639053e-07,
      "logits/chosen": -3.53125,
      "logits/rejected": -3.46875,
      "logps/chosen": -330.0,
      "logps/rejected": -310.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.40625,
      "rewards/margins": 20.375,
      "rewards/rejected": -15.9375,
      "step": 630
    },
    {
      "epoch": 2.0447284345047922,
      "grad_norm": 0.0003708636903725718,
      "learning_rate": 1.7692307692307693e-07,
      "logits/chosen": -3.546875,
      "logits/rejected": -3.515625,
      "logps/chosen": -304.0,
      "logps/rejected": -310.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.125,
      "rewards/margins": 19.75,
      "rewards/rejected": -15.625,
      "step": 640
    },
    {
      "epoch": 2.07667731629393,
      "grad_norm": 0.0010373855132895,
      "learning_rate": 1.710059171597633e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.40625,
      "logps/chosen": -314.0,
      "logps/rejected": -320.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.4375,
      "rewards/margins": 21.5,
      "rewards/rejected": -17.0,
      "step": 650
    },
    {
      "epoch": 2.108626198083067,
      "grad_norm": 5.39936215784096e-05,
      "learning_rate": 1.650887573964497e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.4375,
      "logps/chosen": -342.0,
      "logps/rejected": -326.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.84375,
      "rewards/margins": 22.125,
      "rewards/rejected": -17.375,
      "step": 660
    },
    {
      "epoch": 2.1405750798722045,
      "grad_norm": 0.00019543743585904714,
      "learning_rate": 1.591715976331361e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.453125,
      "logps/chosen": -308.0,
      "logps/rejected": -328.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.09375,
      "rewards/margins": 22.125,
      "rewards/rejected": -18.125,
      "step": 670
    },
    {
      "epoch": 2.1725239616613417,
      "grad_norm": 1.1957658784624645e-05,
      "learning_rate": 1.5325443786982248e-07,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.21875,
      "logps/chosen": -308.0,
      "logps/rejected": -332.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.625,
      "rewards/margins": 22.75,
      "rewards/rejected": -18.125,
      "step": 680
    },
    {
      "epoch": 2.2044728434504792,
      "grad_norm": 0.00013672356144189447,
      "learning_rate": 1.4733727810650885e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.34375,
      "logps/chosen": -294.0,
      "logps/rejected": -336.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 3.9375,
      "rewards/margins": 22.0,
      "rewards/rejected": -18.125,
      "step": 690
    },
    {
      "epoch": 2.236421725239617,
      "grad_norm": 0.0050493316393987265,
      "learning_rate": 1.4142011834319526e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.359375,
      "logps/chosen": -316.0,
      "logps/rejected": -322.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.65625,
      "rewards/margins": 22.25,
      "rewards/rejected": -17.625,
      "step": 700
    },
    {
      "epoch": 2.268370607028754,
      "grad_norm": 0.012682455391037024,
      "learning_rate": 1.3550295857988164e-07,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.34375,
      "logps/chosen": -310.0,
      "logps/rejected": -342.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.46875,
      "rewards/margins": 24.0,
      "rewards/rejected": -19.5,
      "step": 710
    },
    {
      "epoch": 2.3003194888178915,
      "grad_norm": 0.0003196376928219188,
      "learning_rate": 1.2958579881656802e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.390625,
      "logps/chosen": -320.0,
      "logps/rejected": -344.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.0625,
      "rewards/margins": 24.125,
      "rewards/rejected": -19.0,
      "step": 720
    },
    {
      "epoch": 2.3322683706070286,
      "grad_norm": 0.0018609559316608846,
      "learning_rate": 1.2366863905325443e-07,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.40625,
      "logps/chosen": -308.0,
      "logps/rejected": -346.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.46875,
      "rewards/margins": 23.625,
      "rewards/rejected": -19.125,
      "step": 730
    },
    {
      "epoch": 2.364217252396166,
      "grad_norm": 0.00014085216796591186,
      "learning_rate": 1.1775147928994082e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.46875,
      "logps/chosen": -322.0,
      "logps/rejected": -336.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.75,
      "rewards/margins": 23.625,
      "rewards/rejected": -18.875,
      "step": 740
    },
    {
      "epoch": 2.3961661341853033,
      "grad_norm": 4.269109270342446e-05,
      "learning_rate": 1.1183431952662721e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.359375,
      "logps/chosen": -302.0,
      "logps/rejected": -336.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 5.0,
      "rewards/margins": 23.75,
      "rewards/rejected": -18.625,
      "step": 750
    },
    {
      "epoch": 2.428115015974441,
      "grad_norm": 0.0001767088421848351,
      "learning_rate": 1.059171597633136e-07,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.421875,
      "logps/chosen": -306.0,
      "logps/rejected": -338.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.125,
      "rewards/margins": 23.0,
      "rewards/rejected": -18.875,
      "step": 760
    },
    {
      "epoch": 2.460063897763578,
      "grad_norm": 0.00023091107677415993,
      "learning_rate": 1e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.390625,
      "logps/chosen": -304.0,
      "logps/rejected": -342.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.34375,
      "rewards/margins": 24.0,
      "rewards/rejected": -19.75,
      "step": 770
    },
    {
      "epoch": 2.4920127795527156,
      "grad_norm": 0.0001142577200873669,
      "learning_rate": 9.408284023668639e-08,
      "logits/chosen": -3.359375,
      "logits/rejected": -3.4375,
      "logps/chosen": -314.0,
      "logps/rejected": -348.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.5625,
      "rewards/margins": 24.25,
      "rewards/rejected": -19.625,
      "step": 780
    },
    {
      "epoch": 2.523961661341853,
      "grad_norm": 1.2907685571594175e-05,
      "learning_rate": 8.816568047337278e-08,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.28125,
      "logps/chosen": -344.0,
      "logps/rejected": -334.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.875,
      "rewards/margins": 24.5,
      "rewards/rejected": -19.625,
      "step": 790
    },
    {
      "epoch": 2.5559105431309903,
      "grad_norm": 1.9061508094334936e-05,
      "learning_rate": 8.224852071005916e-08,
      "logits/chosen": -3.4375,
      "logits/rejected": -3.34375,
      "logps/chosen": -294.0,
      "logps/rejected": -340.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.25,
      "rewards/margins": 23.625,
      "rewards/rejected": -19.375,
      "step": 800
    },
    {
      "epoch": 2.587859424920128,
      "grad_norm": 0.001107918941822443,
      "learning_rate": 7.633136094674555e-08,
      "logits/chosen": -3.375,
      "logits/rejected": -3.328125,
      "logps/chosen": -312.0,
      "logps/rejected": -356.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.96875,
      "rewards/margins": 25.625,
      "rewards/rejected": -20.75,
      "step": 810
    },
    {
      "epoch": 2.619808306709265,
      "grad_norm": 0.00012242964249948313,
      "learning_rate": 7.041420118343195e-08,
      "logits/chosen": -3.453125,
      "logits/rejected": -3.296875,
      "logps/chosen": -332.0,
      "logps/rejected": -356.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.90625,
      "rewards/margins": 25.75,
      "rewards/rejected": -20.875,
      "step": 820
    },
    {
      "epoch": 2.6517571884984026,
      "grad_norm": 0.000314369306849524,
      "learning_rate": 6.449704142011835e-08,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.46875,
      "logps/chosen": -330.0,
      "logps/rejected": -342.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.28125,
      "rewards/margins": 23.5,
      "rewards/rejected": -19.25,
      "step": 830
    },
    {
      "epoch": 2.68370607028754,
      "grad_norm": 3.995134657031263e-05,
      "learning_rate": 5.857988165680473e-08,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.296875,
      "logps/chosen": -308.0,
      "logps/rejected": -344.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.375,
      "rewards/margins": 24.0,
      "rewards/rejected": -19.625,
      "step": 840
    },
    {
      "epoch": 2.7156549520766773,
      "grad_norm": 5.2231188693328664e-05,
      "learning_rate": 5.266272189349112e-08,
      "logits/chosen": -3.359375,
      "logits/rejected": -3.296875,
      "logps/chosen": -320.0,
      "logps/rejected": -350.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.46875,
      "rewards/margins": 24.875,
      "rewards/rejected": -20.5,
      "step": 850
    },
    {
      "epoch": 2.747603833865815,
      "grad_norm": 8.639376019043504e-05,
      "learning_rate": 4.674556213017751e-08,
      "logits/chosen": -3.375,
      "logits/rejected": -3.375,
      "logps/chosen": -318.0,
      "logps/rejected": -336.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.8125,
      "rewards/margins": 23.875,
      "rewards/rejected": -19.0,
      "step": 860
    },
    {
      "epoch": 2.779552715654952,
      "grad_norm": 1.3383019872983316e-06,
      "learning_rate": 4.082840236686391e-08,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.40625,
      "logps/chosen": -312.0,
      "logps/rejected": -348.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.09375,
      "rewards/margins": 24.0,
      "rewards/rejected": -19.875,
      "step": 870
    },
    {
      "epoch": 2.8115015974440896,
      "grad_norm": 2.150109438478383e-05,
      "learning_rate": 3.4911242603550294e-08,
      "logits/chosen": -3.359375,
      "logits/rejected": -3.390625,
      "logps/chosen": -290.0,
      "logps/rejected": -336.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.1875,
      "rewards/margins": 23.25,
      "rewards/rejected": -19.0,
      "step": 880
    },
    {
      "epoch": 2.8434504792332267,
      "grad_norm": 7.993011042618744e-06,
      "learning_rate": 2.8994082840236687e-08,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.34375,
      "logps/chosen": -312.0,
      "logps/rejected": -362.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.46875,
      "rewards/margins": 25.25,
      "rewards/rejected": -20.75,
      "step": 890
    },
    {
      "epoch": 2.8753993610223643,
      "grad_norm": 0.00022999878510051856,
      "learning_rate": 2.3076923076923076e-08,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.375,
      "logps/chosen": -312.0,
      "logps/rejected": -356.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.28125,
      "rewards/margins": 25.125,
      "rewards/rejected": -20.75,
      "step": 900
    },
    {
      "epoch": 2.9073482428115014,
      "grad_norm": 3.0026149327471167e-06,
      "learning_rate": 1.7159763313609465e-08,
      "logits/chosen": -3.375,
      "logits/rejected": -3.390625,
      "logps/chosen": -304.0,
      "logps/rejected": -354.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.59375,
      "rewards/margins": 24.25,
      "rewards/rejected": -19.625,
      "step": 910
    },
    {
      "epoch": 2.939297124600639,
      "grad_norm": 3.2133287645940767e-06,
      "learning_rate": 1.1242603550295858e-08,
      "logits/chosen": -3.359375,
      "logits/rejected": -3.40625,
      "logps/chosen": -304.0,
      "logps/rejected": -348.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.4375,
      "rewards/margins": 24.25,
      "rewards/rejected": -19.875,
      "step": 920
    },
    {
      "epoch": 2.9712460063897765,
      "grad_norm": 0.0001442658837879914,
      "learning_rate": 5.325443786982248e-09,
      "logits/chosen": -3.390625,
      "logits/rejected": -3.296875,
      "logps/chosen": -320.0,
      "logps/rejected": -330.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 4.875,
      "rewards/margins": 24.125,
      "rewards/rejected": -19.25,
      "step": 930
    },
    {
      "epoch": 3.0,
      "eval_logits/chosen": -3.453125,
      "eval_logits/rejected": -3.328125,
      "eval_logps/chosen": -286.0,
      "eval_logps/rejected": -356.0,
      "eval_loss": 3.605603069445351e-06,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 4.09375,
      "eval_rewards/margins": 25.375,
      "eval_rewards/rejected": -21.375,
      "eval_runtime": 24.9625,
      "eval_samples_per_second": 8.012,
      "eval_steps_per_second": 0.28,
      "step": 939
    },
    {
      "epoch": 3.0,
      "step": 939,
      "total_flos": 0.0,
      "train_loss": 0.015602122592439932,
      "train_runtime": 8087.8179,
      "train_samples_per_second": 3.709,
      "train_steps_per_second": 0.116
    }
  ],
  "logging_steps": 10,
  "max_steps": 939,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}