{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9904153354632586,
  "eval_steps": 500,
  "global_step": 468,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.006389776357827476,
      "grad_norm": 11.092770897372258,
      "learning_rate": 1.0638297872340425e-08,
      "logits/chosen": -1.4375,
      "logits/rejected": -1.4765625,
      "logps/chosen": -124.0,
      "logps/rejected": -106.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 11.4814018660731,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -1.6171875,
      "logits/rejected": -1.546875,
      "logps/chosen": -148.0,
      "logps/rejected": -131.0,
      "loss": 0.693,
      "rewards/accuracies": 0.2222222238779068,
      "rewards/chosen": 0.000873565673828125,
      "rewards/margins": -0.0019073486328125,
      "rewards/rejected": 0.002777099609375,
      "step": 10
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 10.808846928393402,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -1.59375,
      "logits/rejected": -1.5,
      "logps/chosen": -148.0,
      "logps/rejected": -119.0,
      "loss": 0.6909,
      "rewards/accuracies": 0.30000001192092896,
      "rewards/chosen": 0.00093841552734375,
      "rewards/margins": 0.00909423828125,
      "rewards/rejected": -0.00811767578125,
      "step": 20
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 11.114832070482858,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -1.5859375,
      "logits/rejected": -1.53125,
      "logps/chosen": -138.0,
      "logps/rejected": -130.0,
      "loss": 0.688,
      "rewards/accuracies": 0.3812499940395355,
      "rewards/chosen": -0.022705078125,
      "rewards/margins": 0.01031494140625,
      "rewards/rejected": -0.032958984375,
      "step": 30
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 10.629612292618301,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -1.5546875,
      "logits/rejected": -1.5234375,
      "logps/chosen": -137.0,
      "logps/rejected": -122.0,
      "loss": 0.6849,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.0517578125,
      "rewards/margins": 0.0220947265625,
      "rewards/rejected": -0.07373046875,
      "step": 40
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 10.309070114270924,
      "learning_rate": 4.96437054631829e-07,
      "logits/chosen": -1.59375,
      "logits/rejected": -1.5703125,
      "logps/chosen": -148.0,
      "logps/rejected": -134.0,
      "loss": 0.6751,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": -0.11767578125,
      "rewards/margins": 0.02880859375,
      "rewards/rejected": -0.146484375,
      "step": 50
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 10.691839409341034,
      "learning_rate": 4.845605700712589e-07,
      "logits/chosen": -1.5390625,
      "logits/rejected": -1.5390625,
      "logps/chosen": -138.0,
      "logps/rejected": -126.0,
      "loss": 0.6763,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.208984375,
      "rewards/margins": 0.0380859375,
      "rewards/rejected": -0.2470703125,
      "step": 60
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 10.976743522383872,
      "learning_rate": 4.7268408551068883e-07,
      "logits/chosen": -1.578125,
      "logits/rejected": -1.5078125,
      "logps/chosen": -146.0,
      "logps/rejected": -130.0,
      "loss": 0.6659,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.158203125,
      "rewards/margins": 0.08154296875,
      "rewards/rejected": -0.2392578125,
      "step": 70
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 9.937714950468724,
      "learning_rate": 4.6080760095011875e-07,
      "logits/chosen": -1.546875,
      "logits/rejected": -1.5390625,
      "logps/chosen": -139.0,
      "logps/rejected": -134.0,
      "loss": 0.6732,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.25390625,
      "rewards/margins": 0.057373046875,
      "rewards/rejected": -0.3125,
      "step": 80
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 10.448001432920702,
      "learning_rate": 4.4893111638954866e-07,
      "logits/chosen": -1.53125,
      "logits/rejected": -1.5078125,
      "logps/chosen": -142.0,
      "logps/rejected": -135.0,
      "loss": 0.659,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.3125,
      "rewards/margins": 0.08935546875,
      "rewards/rejected": -0.40234375,
      "step": 90
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 10.03442318899385,
      "learning_rate": 4.3705463182897863e-07,
      "logits/chosen": -1.59375,
      "logits/rejected": -1.5625,
      "logps/chosen": -152.0,
      "logps/rejected": -136.0,
      "loss": 0.6485,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.306640625,
      "rewards/margins": 0.154296875,
      "rewards/rejected": -0.4609375,
      "step": 100
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 10.349608536461915,
      "learning_rate": 4.251781472684085e-07,
      "logits/chosen": -1.5,
      "logits/rejected": -1.484375,
      "logps/chosen": -142.0,
      "logps/rejected": -136.0,
      "loss": 0.6387,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.49609375,
      "rewards/margins": 0.130859375,
      "rewards/rejected": -0.625,
      "step": 110
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 10.539570985663747,
      "learning_rate": 4.1330166270783846e-07,
      "logits/chosen": -1.5078125,
      "logits/rejected": -1.484375,
      "logps/chosen": -140.0,
      "logps/rejected": -132.0,
      "loss": 0.6411,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -0.59375,
      "rewards/margins": 0.1728515625,
      "rewards/rejected": -0.76953125,
      "step": 120
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 9.94604093638921,
      "learning_rate": 4.0142517814726837e-07,
      "logits/chosen": -1.5625,
      "logits/rejected": -1.484375,
      "logps/chosen": -153.0,
      "logps/rejected": -127.5,
      "loss": 0.6196,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.66015625,
      "rewards/margins": 0.337890625,
      "rewards/rejected": -0.99609375,
      "step": 130
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 9.253374940542843,
      "learning_rate": 3.8954869358669834e-07,
      "logits/chosen": -1.5546875,
      "logits/rejected": -1.4765625,
      "logps/chosen": -155.0,
      "logps/rejected": -141.0,
      "loss": 0.6091,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.83984375,
      "rewards/margins": 0.28125,
      "rewards/rejected": -1.1171875,
      "step": 140
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 9.147492605620393,
      "learning_rate": 3.7767220902612825e-07,
      "logits/chosen": -1.5390625,
      "logits/rejected": -1.5078125,
      "logps/chosen": -160.0,
      "logps/rejected": -142.0,
      "loss": 0.6096,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.80859375,
      "rewards/margins": 0.349609375,
      "rewards/rejected": -1.15625,
      "step": 150
    },
    {
      "epoch": 0.9968051118210862,
      "eval_logits/chosen": -1.484375,
      "eval_logits/rejected": -1.4765625,
      "eval_logps/chosen": -145.0,
      "eval_logps/rejected": -134.0,
      "eval_loss": 0.6089843511581421,
      "eval_rewards/accuracies": 0.6607142686843872,
      "eval_rewards/chosen": -0.65234375,
      "eval_rewards/margins": 0.34765625,
      "eval_rewards/rejected": -1.0,
      "eval_runtime": 11.4557,
      "eval_samples_per_second": 17.459,
      "eval_steps_per_second": 0.611,
      "step": 156
    },
    {
      "epoch": 1.0223642172523961,
      "grad_norm": 9.22711379768867,
      "learning_rate": 3.6579572446555817e-07,
      "logits/chosen": -1.4921875,
      "logits/rejected": -1.46875,
      "logps/chosen": -141.0,
      "logps/rejected": -139.0,
      "loss": 0.5811,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.8828125,
      "rewards/margins": 0.34765625,
      "rewards/rejected": -1.2265625,
      "step": 160
    },
    {
      "epoch": 1.0862619808306708,
      "grad_norm": 9.508478185274804,
      "learning_rate": 3.5391923990498813e-07,
      "logits/chosen": -1.5078125,
      "logits/rejected": -1.5,
      "logps/chosen": -145.0,
      "logps/rejected": -134.0,
      "loss": 0.5337,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.71875,
      "rewards/margins": 0.5546875,
      "rewards/rejected": -1.2734375,
      "step": 170
    },
    {
      "epoch": 1.1501597444089458,
      "grad_norm": 9.316918076373987,
      "learning_rate": 3.42042755344418e-07,
      "logits/chosen": -1.5625,
      "logits/rejected": -1.53125,
      "logps/chosen": -157.0,
      "logps/rejected": -146.0,
      "loss": 0.544,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.66015625,
      "rewards/margins": 0.56640625,
      "rewards/rejected": -1.2265625,
      "step": 180
    },
    {
      "epoch": 1.2140575079872205,
      "grad_norm": 8.759483351227418,
      "learning_rate": 3.3016627078384796e-07,
      "logits/chosen": -1.5703125,
      "logits/rejected": -1.5234375,
      "logps/chosen": -152.0,
      "logps/rejected": -146.0,
      "loss": 0.5362,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.74609375,
      "rewards/margins": 0.5,
      "rewards/rejected": -1.25,
      "step": 190
    },
    {
      "epoch": 1.2779552715654952,
      "grad_norm": 9.97195116624129,
      "learning_rate": 3.182897862232779e-07,
      "logits/chosen": -1.515625,
      "logits/rejected": -1.515625,
      "logps/chosen": -151.0,
      "logps/rejected": -141.0,
      "loss": 0.5373,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -0.7890625,
      "rewards/margins": 0.55859375,
      "rewards/rejected": -1.34375,
      "step": 200
    },
    {
      "epoch": 1.34185303514377,
      "grad_norm": 8.576220811968858,
      "learning_rate": 3.0641330166270784e-07,
      "logits/chosen": -1.578125,
      "logits/rejected": -1.5234375,
      "logps/chosen": -151.0,
      "logps/rejected": -140.0,
      "loss": 0.5339,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.83984375,
      "rewards/margins": 0.462890625,
      "rewards/rejected": -1.3046875,
      "step": 210
    },
    {
      "epoch": 1.4057507987220448,
      "grad_norm": 8.688671845017401,
      "learning_rate": 2.9453681710213776e-07,
      "logits/chosen": -1.53125,
      "logits/rejected": -1.53125,
      "logps/chosen": -148.0,
      "logps/rejected": -143.0,
      "loss": 0.5366,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.8671875,
      "rewards/margins": 0.5078125,
      "rewards/rejected": -1.375,
      "step": 220
    },
    {
      "epoch": 1.4696485623003195,
      "grad_norm": 9.030983047385451,
      "learning_rate": 2.8266033254156767e-07,
      "logits/chosen": -1.546875,
      "logits/rejected": -1.515625,
      "logps/chosen": -145.0,
      "logps/rejected": -142.0,
      "loss": 0.5346,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.8046875,
      "rewards/margins": 0.53515625,
      "rewards/rejected": -1.34375,
      "step": 230
    },
    {
      "epoch": 1.5335463258785942,
      "grad_norm": 11.622095941624893,
      "learning_rate": 2.7078384798099764e-07,
      "logits/chosen": -1.484375,
      "logits/rejected": -1.4609375,
      "logps/chosen": -145.0,
      "logps/rejected": -153.0,
      "loss": 0.553,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.80859375,
      "rewards/margins": 0.51171875,
      "rewards/rejected": -1.3203125,
      "step": 240
    },
    {
      "epoch": 1.5974440894568689,
      "grad_norm": 9.38942767820535,
      "learning_rate": 2.589073634204275e-07,
      "logits/chosen": -1.5625,
      "logits/rejected": -1.4765625,
      "logps/chosen": -146.0,
      "logps/rejected": -132.0,
      "loss": 0.54,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.50390625,
      "rewards/margins": 0.578125,
      "rewards/rejected": -1.078125,
      "step": 250
    },
    {
      "epoch": 1.6613418530351438,
      "grad_norm": 9.87005880531835,
      "learning_rate": 2.4703087885985747e-07,
      "logits/chosen": -1.5546875,
      "logits/rejected": -1.5234375,
      "logps/chosen": -160.0,
      "logps/rejected": -148.0,
      "loss": 0.5312,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.55859375,
      "rewards/margins": 0.5546875,
      "rewards/rejected": -1.1171875,
      "step": 260
    },
    {
      "epoch": 1.7252396166134185,
      "grad_norm": 8.251251897175216,
      "learning_rate": 2.351543942992874e-07,
      "logits/chosen": -1.5234375,
      "logits/rejected": -1.4921875,
      "logps/chosen": -146.0,
      "logps/rejected": -143.0,
      "loss": 0.5354,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.609375,
      "rewards/margins": 0.56640625,
      "rewards/rejected": -1.171875,
      "step": 270
    },
    {
      "epoch": 1.7891373801916934,
      "grad_norm": 9.817987165601615,
      "learning_rate": 2.2327790973871732e-07,
      "logits/chosen": -1.53125,
      "logits/rejected": -1.4921875,
      "logps/chosen": -148.0,
      "logps/rejected": -144.0,
      "loss": 0.5347,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.5546875,
      "rewards/margins": 0.58984375,
      "rewards/rejected": -1.1484375,
      "step": 280
    },
    {
      "epoch": 1.8530351437699681,
      "grad_norm": 9.610664273576093,
      "learning_rate": 2.1140142517814726e-07,
      "logits/chosen": -1.5390625,
      "logits/rejected": -1.4921875,
      "logps/chosen": -164.0,
      "logps/rejected": -148.0,
      "loss": 0.5186,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.57421875,
      "rewards/margins": 0.59375,
      "rewards/rejected": -1.1640625,
      "step": 290
    },
    {
      "epoch": 1.9169329073482428,
      "grad_norm": 9.80773406850179,
      "learning_rate": 1.9952494061757718e-07,
      "logits/chosen": -1.53125,
      "logits/rejected": -1.5078125,
      "logps/chosen": -151.0,
      "logps/rejected": -148.0,
      "loss": 0.5476,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.75,
      "rewards/margins": 0.439453125,
      "rewards/rejected": -1.1875,
      "step": 300
    },
    {
      "epoch": 1.9808306709265175,
      "grad_norm": 9.05899810741374,
      "learning_rate": 1.876484560570071e-07,
      "logits/chosen": -1.578125,
      "logits/rejected": -1.53125,
      "logps/chosen": -156.0,
      "logps/rejected": -144.0,
      "loss": 0.5245,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.71875,
      "rewards/margins": 0.5703125,
      "rewards/rejected": -1.2890625,
      "step": 310
    },
    {
      "epoch": 2.0,
      "eval_logits/chosen": -1.5,
      "eval_logits/rejected": -1.4921875,
      "eval_logps/chosen": -144.0,
      "eval_logps/rejected": -134.0,
      "eval_loss": 0.5994531512260437,
      "eval_rewards/accuracies": 0.6964285969734192,
      "eval_rewards/chosen": -0.546875,
      "eval_rewards/margins": 0.458984375,
      "eval_rewards/rejected": -1.0078125,
      "eval_runtime": 11.4667,
      "eval_samples_per_second": 17.442,
      "eval_steps_per_second": 0.61,
      "step": 313
    },
    {
      "epoch": 2.0447284345047922,
      "grad_norm": 8.064019736193583,
      "learning_rate": 1.7577197149643706e-07,
      "logits/chosen": -1.546875,
      "logits/rejected": -1.5,
      "logps/chosen": -155.0,
      "logps/rejected": -145.0,
      "loss": 0.4849,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.84765625,
      "rewards/margins": 0.671875,
      "rewards/rejected": -1.515625,
      "step": 320
    },
    {
      "epoch": 2.108626198083067,
      "grad_norm": 7.485198225503147,
      "learning_rate": 1.6389548693586697e-07,
      "logits/chosen": -1.5625,
      "logits/rejected": -1.5078125,
      "logps/chosen": -144.0,
      "logps/rejected": -136.0,
      "loss": 0.495,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -0.8671875,
      "rewards/margins": 0.60546875,
      "rewards/rejected": -1.4765625,
      "step": 330
    },
    {
      "epoch": 2.1725239616613417,
      "grad_norm": 9.966020126485736,
      "learning_rate": 1.520190023752969e-07,
      "logits/chosen": -1.546875,
      "logits/rejected": -1.5,
      "logps/chosen": -149.0,
      "logps/rejected": -137.0,
      "loss": 0.4747,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -0.83203125,
      "rewards/margins": 0.7109375,
      "rewards/rejected": -1.546875,
      "step": 340
    },
    {
      "epoch": 2.236421725239617,
      "grad_norm": 9.297651299143556,
      "learning_rate": 1.4014251781472683e-07,
      "logits/chosen": -1.5625,
      "logits/rejected": -1.53125,
      "logps/chosen": -152.0,
      "logps/rejected": -144.0,
      "loss": 0.4823,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -0.8203125,
      "rewards/margins": 0.640625,
      "rewards/rejected": -1.4609375,
      "step": 350
    },
    {
      "epoch": 2.3003194888178915,
      "grad_norm": 9.172329731503968,
      "learning_rate": 1.2826603325415677e-07,
      "logits/chosen": -1.53125,
      "logits/rejected": -1.5078125,
      "logps/chosen": -146.0,
      "logps/rejected": -137.0,
      "loss": 0.4743,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.8046875,
      "rewards/margins": 0.74609375,
      "rewards/rejected": -1.546875,
      "step": 360
    },
    {
      "epoch": 2.364217252396166,
      "grad_norm": 9.394249924534705,
      "learning_rate": 1.163895486935867e-07,
      "logits/chosen": -1.5859375,
      "logits/rejected": -1.5390625,
      "logps/chosen": -164.0,
      "logps/rejected": -151.0,
      "loss": 0.4695,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.5703125,
      "rewards/margins": 0.69921875,
      "rewards/rejected": -1.2734375,
      "step": 370
    },
    {
      "epoch": 2.428115015974441,
      "grad_norm": 8.43211550191306,
      "learning_rate": 1.0451306413301662e-07,
      "logits/chosen": -1.546875,
      "logits/rejected": -1.515625,
      "logps/chosen": -141.0,
      "logps/rejected": -136.0,
      "loss": 0.4747,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": -0.8515625,
      "rewards/margins": 0.69140625,
      "rewards/rejected": -1.5390625,
      "step": 380
    },
    {
      "epoch": 2.4920127795527156,
      "grad_norm": 9.162321870225993,
      "learning_rate": 9.263657957244655e-08,
      "logits/chosen": -1.5234375,
      "logits/rejected": -1.4921875,
      "logps/chosen": -148.0,
      "logps/rejected": -144.0,
      "loss": 0.4763,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -0.8671875,
      "rewards/margins": 0.6640625,
      "rewards/rejected": -1.53125,
      "step": 390
    },
    {
      "epoch": 2.5559105431309903,
      "grad_norm": 9.156556482449458,
      "learning_rate": 8.076009501187649e-08,
      "logits/chosen": -1.5703125,
      "logits/rejected": -1.5234375,
      "logps/chosen": -153.0,
      "logps/rejected": -139.0,
      "loss": 0.468,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -0.79296875,
      "rewards/margins": 0.66796875,
      "rewards/rejected": -1.4609375,
      "step": 400
    },
    {
      "epoch": 2.619808306709265,
      "grad_norm": 8.694584668338557,
      "learning_rate": 6.88836104513064e-08,
      "logits/chosen": -1.5625,
      "logits/rejected": -1.5,
      "logps/chosen": -155.0,
      "logps/rejected": -143.0,
      "loss": 0.4845,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.79296875,
      "rewards/margins": 0.67578125,
      "rewards/rejected": -1.46875,
      "step": 410
    },
    {
      "epoch": 2.68370607028754,
      "grad_norm": 8.76105196917074,
      "learning_rate": 5.700712589073634e-08,
      "logits/chosen": -1.5390625,
      "logits/rejected": -1.5,
      "logps/chosen": -151.0,
      "logps/rejected": -144.0,
      "loss": 0.4679,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -0.80078125,
      "rewards/margins": 0.70703125,
      "rewards/rejected": -1.5078125,
      "step": 420
    },
    {
      "epoch": 2.747603833865815,
      "grad_norm": 10.082882017826297,
      "learning_rate": 4.5130641330166267e-08,
      "logits/chosen": -1.5078125,
      "logits/rejected": -1.46875,
      "logps/chosen": -145.0,
      "logps/rejected": -142.0,
      "loss": 0.4617,
      "rewards/accuracies": 0.856249988079071,
      "rewards/chosen": -0.765625,
      "rewards/margins": 0.78125,
      "rewards/rejected": -1.546875,
      "step": 430
    },
    {
      "epoch": 2.8115015974440896,
      "grad_norm": 8.539769794548567,
      "learning_rate": 3.32541567695962e-08,
      "logits/chosen": -1.53125,
      "logits/rejected": -1.4765625,
      "logps/chosen": -142.0,
      "logps/rejected": -136.0,
      "loss": 0.4761,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -0.8203125,
      "rewards/margins": 0.80859375,
      "rewards/rejected": -1.625,
      "step": 440
    },
    {
      "epoch": 2.8753993610223643,
      "grad_norm": 9.882517602717023,
      "learning_rate": 2.1377672209026125e-08,
      "logits/chosen": -1.5859375,
      "logits/rejected": -1.5390625,
      "logps/chosen": -149.0,
      "logps/rejected": -148.0,
      "loss": 0.4793,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -0.75390625,
      "rewards/margins": 0.66796875,
      "rewards/rejected": -1.421875,
      "step": 450
    },
    {
      "epoch": 2.939297124600639,
      "grad_norm": 8.662355843337348,
      "learning_rate": 9.501187648456057e-09,
      "logits/chosen": -1.5546875,
      "logits/rejected": -1.5,
      "logps/chosen": -153.0,
      "logps/rejected": -143.0,
      "loss": 0.4931,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.87890625,
      "rewards/margins": 0.6953125,
      "rewards/rejected": -1.578125,
      "step": 460
    },
    {
      "epoch": 2.9904153354632586,
      "eval_logits/chosen": -1.5,
      "eval_logits/rejected": -1.4921875,
      "eval_logps/chosen": -145.0,
      "eval_logps/rejected": -135.0,
      "eval_loss": 0.5950781106948853,
      "eval_rewards/accuracies": 0.6785714030265808,
      "eval_rewards/chosen": -0.59765625,
      "eval_rewards/margins": 0.498046875,
      "eval_rewards/rejected": -1.1015625,
      "eval_runtime": 14.1407,
      "eval_samples_per_second": 14.144,
      "eval_steps_per_second": 0.495,
      "step": 468
    },
    {
      "epoch": 2.9904153354632586,
      "step": 468,
      "total_flos": 0.0,
      "train_loss": 0.5567819970285791,
      "train_runtime": 4306.1528,
      "train_samples_per_second": 6.965,
      "train_steps_per_second": 0.109
    }
  ],
  "logging_steps": 10,
  "max_steps": 468,
  "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
}