{
  "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": 40.63113754109814,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -0.72265625,
      "logits/rejected": -0.40234375,
      "logps/chosen": -296.0,
      "logps/rejected": -97.5,
      "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": 40.83376065165844,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.52734375,
      "logps/chosen": -326.0,
      "logps/rejected": -112.0,
      "loss": 0.6923,
      "rewards/accuracies": 0.1944444477558136,
      "rewards/chosen": -0.00592041015625,
      "rewards/margins": -0.0062255859375,
      "rewards/rejected": 0.0003452301025390625,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 36.83108395610394,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -0.75390625,
      "logits/rejected": -0.6171875,
      "logps/chosen": -360.0,
      "logps/rejected": -116.5,
      "loss": 0.6765,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.00433349609375,
      "rewards/margins": 0.0306396484375,
      "rewards/rejected": -0.03515625,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 33.32858154831051,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -0.8515625,
      "logits/rejected": -0.6015625,
      "logps/chosen": -342.0,
      "logps/rejected": -124.0,
      "loss": 0.6042,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": 0.049072265625,
      "rewards/margins": 0.21484375,
      "rewards/rejected": -0.166015625,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 27.32264489173027,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -0.70703125,
      "logits/rejected": -0.58984375,
      "logps/chosen": -356.0,
      "logps/rejected": -124.0,
      "loss": 0.4774,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.12451171875,
      "rewards/margins": 0.5859375,
      "rewards/rejected": -0.4609375,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 15.975644864718246,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -0.79296875,
      "logits/rejected": -0.5546875,
      "logps/chosen": -286.0,
      "logps/rejected": -126.5,
      "loss": 0.2903,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.05859375,
      "rewards/margins": 1.109375,
      "rewards/rejected": -1.046875,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 6.256507586550031,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.50390625,
      "logps/chosen": -302.0,
      "logps/rejected": -140.0,
      "loss": 0.129,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.056396484375,
      "rewards/margins": 2.328125,
      "rewards/rejected": -2.28125,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 3.6739087198855866,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -0.74609375,
      "logits/rejected": -0.435546875,
      "logps/chosen": -300.0,
      "logps/rejected": -152.0,
      "loss": 0.0324,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.197265625,
      "rewards/margins": 4.375,
      "rewards/rejected": -4.15625,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 5.267432309185282,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -0.76171875,
      "logits/rejected": -0.5,
      "logps/chosen": -296.0,
      "logps/rejected": -174.0,
      "loss": 0.0112,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.341796875,
      "rewards/margins": 5.9375,
      "rewards/rejected": -5.59375,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 2.308803456420075,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -0.60546875,
      "logits/rejected": -0.423828125,
      "logps/chosen": -255.0,
      "logps/rejected": -171.0,
      "loss": 0.006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.462890625,
      "rewards/margins": 6.75,
      "rewards/rejected": -6.28125,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.19555856015239148,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": -0.7421875,
      "logits/rejected": -0.56640625,
      "logps/chosen": -356.0,
      "logps/rejected": -191.0,
      "loss": 0.0026,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.53125,
      "rewards/margins": 7.84375,
      "rewards/rejected": -7.3125,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.11747858932749748,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -0.6640625,
      "logits/rejected": -0.56640625,
      "logps/chosen": -294.0,
      "logps/rejected": -201.0,
      "loss": 0.0062,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.46484375,
      "rewards/margins": 8.25,
      "rewards/rejected": -7.75,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.2954350664468083,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -0.6015625,
      "logits/rejected": -0.373046875,
      "logps/chosen": -240.0,
      "logps/rejected": -189.0,
      "loss": 0.0011,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.380859375,
      "rewards/margins": 8.3125,
      "rewards/rejected": -7.9375,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.4447368308978073,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": -0.578125,
      "logits/rejected": -0.54296875,
      "logps/chosen": -314.0,
      "logps/rejected": -205.0,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.42578125,
      "rewards/margins": 8.6875,
      "rewards/rejected": -8.25,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.13989545962415664,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -0.6484375,
      "logits/rejected": -0.494140625,
      "logps/chosen": -346.0,
      "logps/rejected": -203.0,
      "loss": 0.0025,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2158203125,
      "rewards/margins": 9.1875,
      "rewards/rejected": -9.0,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 1.005715548136931,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.5859375,
      "logps/chosen": -322.0,
      "logps/rejected": -206.0,
      "loss": 0.0018,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.466796875,
      "rewards/margins": 9.6875,
      "rewards/rejected": -9.1875,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 0.015257918360502878,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -0.76953125,
      "logits/rejected": -0.447265625,
      "logps/chosen": -312.0,
      "logps/rejected": -216.0,
      "loss": 0.0007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.125,
      "rewards/margins": 9.75,
      "rewards/rejected": -9.625,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.14450582552206107,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": -0.76953125,
      "logits/rejected": -0.5625,
      "logps/chosen": -276.0,
      "logps/rejected": -208.0,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.271484375,
      "rewards/margins": 9.9375,
      "rewards/rejected": -9.6875,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.04115746281274911,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.5625,
      "logps/chosen": -332.0,
      "logps/rejected": -217.0,
      "loss": 0.0083,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2470703125,
      "rewards/margins": 9.9375,
      "rewards/rejected": -9.6875,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.027254932142989203,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": -0.578125,
      "logits/rejected": -0.48046875,
      "logps/chosen": -306.0,
      "logps/rejected": -219.0,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1982421875,
      "rewards/margins": 10.375,
      "rewards/rejected": -10.1875,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 0.04076702181699882,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -0.7265625,
      "logits/rejected": -0.478515625,
      "logps/chosen": -348.0,
      "logps/rejected": -226.0,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.125,
      "rewards/margins": 10.6875,
      "rewards/rejected": -10.5,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 0.02362186732429131,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -0.71875,
      "logits/rejected": -0.5,
      "logps/chosen": -316.0,
      "logps/rejected": -220.0,
      "loss": 0.0012,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.11328125,
      "rewards/margins": 10.375,
      "rewards/rejected": -10.5,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.019809535140906925,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": -0.6796875,
      "logits/rejected": -0.57421875,
      "logps/chosen": -356.0,
      "logps/rejected": -228.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.01483154296875,
      "rewards/margins": 10.5625,
      "rewards/rejected": -10.5625,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 1.5883393714345544,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": -0.76953125,
      "logits/rejected": -0.5078125,
      "logps/chosen": -298.0,
      "logps/rejected": -220.0,
      "loss": 0.007,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.0810546875,
      "rewards/margins": 10.6875,
      "rewards/rejected": -10.75,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.020553702943879617,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -0.62890625,
      "logits/rejected": -0.55859375,
      "logps/chosen": -336.0,
      "logps/rejected": -230.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.034423828125,
      "rewards/margins": 10.9375,
      "rewards/rejected": -10.9375,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.18543484770077023,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -0.75,
      "logits/rejected": -0.458984375,
      "logps/chosen": -330.0,
      "logps/rejected": -221.0,
      "loss": 0.001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.2392578125,
      "rewards/margins": 10.4375,
      "rewards/rejected": -10.625,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 0.025956022709723227,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.59765625,
      "logps/chosen": -322.0,
      "logps/rejected": -237.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.326171875,
      "rewards/margins": 10.9375,
      "rewards/rejected": -11.3125,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 0.768139770920166,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": -0.703125,
      "logits/rejected": -0.57421875,
      "logps/chosen": -304.0,
      "logps/rejected": -228.0,
      "loss": 0.0006,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.0089111328125,
      "rewards/margins": 11.4375,
      "rewards/rejected": -11.4375,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 0.5159058299026532,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -0.59765625,
      "logits/rejected": -0.46484375,
      "logps/chosen": -296.0,
      "logps/rejected": -230.0,
      "loss": 0.0012,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.458984375,
      "rewards/margins": 11.0625,
      "rewards/rejected": -11.5,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 0.0031165260658612332,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -0.76171875,
      "logits/rejected": -0.625,
      "logps/chosen": -338.0,
      "logps/rejected": -234.0,
      "loss": 0.0004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.306640625,
      "rewards/margins": 11.375,
      "rewards/rejected": -11.6875,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 0.013736141734169007,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": -0.69140625,
      "logits/rejected": -0.5078125,
      "logps/chosen": -328.0,
      "logps/rejected": -240.0,
      "loss": 0.0014,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.28125,
      "rewards/margins": 11.75,
      "rewards/rejected": -12.0625,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.031010940779375298,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -0.5390625,
      "logits/rejected": -0.515625,
      "logps/chosen": -300.0,
      "logps/rejected": -237.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.439453125,
      "rewards/margins": 11.75,
      "rewards/rejected": -12.1875,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -0.72265625,
      "eval_logits/rejected": -0.4375,
      "eval_logps/chosen": -296.0,
      "eval_logps/rejected": -233.0,
      "eval_loss": 0.0003721904649864882,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -0.291015625,
      "eval_rewards/margins": 11.625,
      "eval_rewards/rejected": -11.9375,
      "eval_runtime": 19.9934,
      "eval_samples_per_second": 10.003,
      "eval_steps_per_second": 0.35,
      "step": 313
    },
    {
      "epoch": 1.0223642172523961,
      "grad_norm": 0.022677077708026305,
      "learning_rate": 3.662721893491124e-07,
      "logits/chosen": -0.73046875,
      "logits/rejected": -0.44140625,
      "logps/chosen": -278.0,
      "logps/rejected": -240.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.431640625,
      "rewards/margins": 11.75,
      "rewards/rejected": -12.1875,
      "step": 320
    },
    {
      "epoch": 1.0543130990415335,
      "grad_norm": 0.011840880769205434,
      "learning_rate": 3.603550295857988e-07,
      "logits/chosen": -0.72265625,
      "logits/rejected": -0.5546875,
      "logps/chosen": -322.0,
      "logps/rejected": -240.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.380859375,
      "rewards/margins": 12.0,
      "rewards/rejected": -12.375,
      "step": 330
    },
    {
      "epoch": 1.0862619808306708,
      "grad_norm": 0.00391406917206777,
      "learning_rate": 3.544378698224852e-07,
      "logits/chosen": -0.671875,
      "logits/rejected": -0.51953125,
      "logps/chosen": -314.0,
      "logps/rejected": -245.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.359375,
      "rewards/margins": 12.1875,
      "rewards/rejected": -12.5625,
      "step": 340
    },
    {
      "epoch": 1.1182108626198084,
      "grad_norm": 0.0036384946886406336,
      "learning_rate": 3.485207100591716e-07,
      "logits/chosen": -0.7265625,
      "logits/rejected": -0.5546875,
      "logps/chosen": -314.0,
      "logps/rejected": -235.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.263671875,
      "rewards/margins": 12.0,
      "rewards/rejected": -12.25,
      "step": 350
    },
    {
      "epoch": 1.1501597444089458,
      "grad_norm": 0.35575156394070306,
      "learning_rate": 3.42603550295858e-07,
      "logits/chosen": -0.73828125,
      "logits/rejected": -0.546875,
      "logps/chosen": -274.0,
      "logps/rejected": -246.0,
      "loss": 0.0004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.69140625,
      "rewards/margins": 12.0,
      "rewards/rejected": -12.6875,
      "step": 360
    },
    {
      "epoch": 1.182108626198083,
      "grad_norm": 0.4647357188457555,
      "learning_rate": 3.366863905325444e-07,
      "logits/chosen": -0.58203125,
      "logits/rejected": -0.5859375,
      "logps/chosen": -376.0,
      "logps/rejected": -251.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.384765625,
      "rewards/margins": 12.25,
      "rewards/rejected": -12.625,
      "step": 370
    },
    {
      "epoch": 1.2140575079872205,
      "grad_norm": 0.006194595470527809,
      "learning_rate": 3.3076923076923075e-07,
      "logits/chosen": -0.6875,
      "logits/rejected": -0.373046875,
      "logps/chosen": -284.0,
      "logps/rejected": -240.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.5234375,
      "rewards/margins": 12.1875,
      "rewards/rejected": -12.6875,
      "step": 380
    },
    {
      "epoch": 1.2460063897763578,
      "grad_norm": 0.08957807843959086,
      "learning_rate": 3.2485207100591716e-07,
      "logits/chosen": -0.7265625,
      "logits/rejected": -0.5234375,
      "logps/chosen": -342.0,
      "logps/rejected": -245.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.5859375,
      "rewards/margins": 12.1875,
      "rewards/rejected": -12.75,
      "step": 390
    },
    {
      "epoch": 1.2779552715654952,
      "grad_norm": 0.0040259036143305595,
      "learning_rate": 3.189349112426035e-07,
      "logits/chosen": -0.65625,
      "logits/rejected": -0.51171875,
      "logps/chosen": -324.0,
      "logps/rejected": -248.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.451171875,
      "rewards/margins": 12.5625,
      "rewards/rejected": -13.0,
      "step": 400
    },
    {
      "epoch": 1.3099041533546325,
      "grad_norm": 0.007242003331847797,
      "learning_rate": 3.130177514792899e-07,
      "logits/chosen": -0.6484375,
      "logits/rejected": -0.44921875,
      "logps/chosen": -318.0,
      "logps/rejected": -248.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.61328125,
      "rewards/margins": 12.875,
      "rewards/rejected": -13.5,
      "step": 410
    },
    {
      "epoch": 1.34185303514377,
      "grad_norm": 0.0064360979279180575,
      "learning_rate": 3.071005917159763e-07,
      "logits/chosen": -0.58203125,
      "logits/rejected": -0.5625,
      "logps/chosen": -296.0,
      "logps/rejected": -249.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.03125,
      "rewards/margins": 12.0625,
      "rewards/rejected": -13.125,
      "step": 420
    },
    {
      "epoch": 1.3738019169329074,
      "grad_norm": 0.00975201883790689,
      "learning_rate": 3.011834319526627e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.51953125,
      "logps/chosen": -314.0,
      "logps/rejected": -252.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.64453125,
      "rewards/margins": 12.6875,
      "rewards/rejected": -13.375,
      "step": 430
    },
    {
      "epoch": 1.4057507987220448,
      "grad_norm": 0.005359836815678541,
      "learning_rate": 2.952662721893491e-07,
      "logits/chosen": -0.65234375,
      "logits/rejected": -0.4609375,
      "logps/chosen": -336.0,
      "logps/rejected": -249.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.69140625,
      "rewards/margins": 12.8125,
      "rewards/rejected": -13.5,
      "step": 440
    },
    {
      "epoch": 1.4376996805111821,
      "grad_norm": 0.07525889423766378,
      "learning_rate": 2.893491124260355e-07,
      "logits/chosen": -0.7109375,
      "logits/rejected": -0.5546875,
      "logps/chosen": -340.0,
      "logps/rejected": -254.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.75,
      "rewards/margins": 12.875,
      "rewards/rejected": -13.625,
      "step": 450
    },
    {
      "epoch": 1.4696485623003195,
      "grad_norm": 0.00193288934052946,
      "learning_rate": 2.834319526627219e-07,
      "logits/chosen": -0.703125,
      "logits/rejected": -0.5078125,
      "logps/chosen": -300.0,
      "logps/rejected": -255.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.8125,
      "rewards/margins": 12.8125,
      "rewards/rejected": -13.6875,
      "step": 460
    },
    {
      "epoch": 1.5015974440894568,
      "grad_norm": 0.004933186728273418,
      "learning_rate": 2.7751479289940824e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.453125,
      "logps/chosen": -294.0,
      "logps/rejected": -252.0,
      "loss": 0.0004,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.796875,
      "rewards/margins": 12.75,
      "rewards/rejected": -13.5625,
      "step": 470
    },
    {
      "epoch": 1.5335463258785942,
      "grad_norm": 0.09834967775177374,
      "learning_rate": 2.715976331360947e-07,
      "logits/chosen": -0.62890625,
      "logits/rejected": -0.478515625,
      "logps/chosen": -360.0,
      "logps/rejected": -247.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.7109375,
      "rewards/margins": 12.5625,
      "rewards/rejected": -13.25,
      "step": 480
    },
    {
      "epoch": 1.5654952076677318,
      "grad_norm": 0.1180240734795196,
      "learning_rate": 2.6568047337278106e-07,
      "logits/chosen": -0.69921875,
      "logits/rejected": -0.5078125,
      "logps/chosen": -346.0,
      "logps/rejected": -258.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.015625,
      "rewards/margins": 13.125,
      "rewards/rejected": -14.125,
      "step": 490
    },
    {
      "epoch": 1.5974440894568689,
      "grad_norm": 0.10115647498128895,
      "learning_rate": 2.5976331360946746e-07,
      "logits/chosen": -0.66015625,
      "logits/rejected": -0.54296875,
      "logps/chosen": -372.0,
      "logps/rejected": -256.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.671875,
      "rewards/margins": 12.9375,
      "rewards/rejected": -13.625,
      "step": 500
    },
    {
      "epoch": 1.6293929712460065,
      "grad_norm": 0.005594304401420574,
      "learning_rate": 2.538461538461538e-07,
      "logits/chosen": -0.59375,
      "logits/rejected": -0.52734375,
      "logps/chosen": -312.0,
      "logps/rejected": -258.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.79296875,
      "rewards/margins": 13.4375,
      "rewards/rejected": -14.25,
      "step": 510
    },
    {
      "epoch": 1.6613418530351438,
      "grad_norm": 0.0027735638022890357,
      "learning_rate": 2.479289940828402e-07,
      "logits/chosen": -0.6953125,
      "logits/rejected": -0.5859375,
      "logps/chosen": -292.0,
      "logps/rejected": -250.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1875,
      "rewards/margins": 12.75,
      "rewards/rejected": -13.9375,
      "step": 520
    },
    {
      "epoch": 1.6932907348242812,
      "grad_norm": 0.0005941899982621629,
      "learning_rate": 2.420118343195266e-07,
      "logits/chosen": -0.64453125,
      "logits/rejected": -0.56640625,
      "logps/chosen": -298.0,
      "logps/rejected": -262.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9765625,
      "rewards/margins": 13.3125,
      "rewards/rejected": -14.3125,
      "step": 530
    },
    {
      "epoch": 1.7252396166134185,
      "grad_norm": 0.024572757961485794,
      "learning_rate": 2.36094674556213e-07,
      "logits/chosen": -0.72265625,
      "logits/rejected": -0.51171875,
      "logps/chosen": -366.0,
      "logps/rejected": -256.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.046875,
      "rewards/margins": 12.9375,
      "rewards/rejected": -14.0,
      "step": 540
    },
    {
      "epoch": 1.7571884984025559,
      "grad_norm": 0.0008800867821762263,
      "learning_rate": 2.301775147928994e-07,
      "logits/chosen": -0.6875,
      "logits/rejected": -0.44921875,
      "logps/chosen": -308.0,
      "logps/rejected": -256.0,
      "loss": 0.0009,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.89453125,
      "rewards/margins": 13.0,
      "rewards/rejected": -13.9375,
      "step": 550
    },
    {
      "epoch": 1.7891373801916934,
      "grad_norm": 0.001437566874272442,
      "learning_rate": 2.242603550295858e-07,
      "logits/chosen": -0.76171875,
      "logits/rejected": -0.58984375,
      "logps/chosen": -376.0,
      "logps/rejected": -260.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.78125,
      "rewards/margins": 13.125,
      "rewards/rejected": -13.875,
      "step": 560
    },
    {
      "epoch": 1.8210862619808306,
      "grad_norm": 0.0026793717529034755,
      "learning_rate": 2.1834319526627217e-07,
      "logits/chosen": -0.6953125,
      "logits/rejected": -0.5234375,
      "logps/chosen": -308.0,
      "logps/rejected": -253.0,
      "loss": 0.0014,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.98828125,
      "rewards/margins": 13.1875,
      "rewards/rejected": -14.1875,
      "step": 570
    },
    {
      "epoch": 1.8530351437699681,
      "grad_norm": 0.04424802361114257,
      "learning_rate": 2.1242603550295858e-07,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.56640625,
      "logps/chosen": -330.0,
      "logps/rejected": -262.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1953125,
      "rewards/margins": 13.0,
      "rewards/rejected": -14.1875,
      "step": 580
    },
    {
      "epoch": 1.8849840255591053,
      "grad_norm": 0.0015499356598387029,
      "learning_rate": 2.0650887573964496e-07,
      "logits/chosen": -0.70703125,
      "logits/rejected": -0.5859375,
      "logps/chosen": -380.0,
      "logps/rejected": -262.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.0703125,
      "rewards/margins": 13.25,
      "rewards/rejected": -14.3125,
      "step": 590
    },
    {
      "epoch": 1.9169329073482428,
      "grad_norm": 0.0017185011208288798,
      "learning_rate": 2.0059171597633133e-07,
      "logits/chosen": -0.625,
      "logits/rejected": -0.498046875,
      "logps/chosen": -320.0,
      "logps/rejected": -270.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.015625,
      "rewards/margins": 13.375,
      "rewards/rejected": -14.375,
      "step": 600
    },
    {
      "epoch": 1.9488817891373802,
      "grad_norm": 0.06027517128383722,
      "learning_rate": 1.9467455621301774e-07,
      "logits/chosen": -0.609375,
      "logits/rejected": -0.443359375,
      "logps/chosen": -324.0,
      "logps/rejected": -254.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.96875,
      "rewards/margins": 13.125,
      "rewards/rejected": -14.125,
      "step": 610
    },
    {
      "epoch": 1.9808306709265175,
      "grad_norm": 0.0034706962600456605,
      "learning_rate": 1.8875739644970412e-07,
      "logits/chosen": -0.75390625,
      "logits/rejected": -0.6015625,
      "logps/chosen": -338.0,
      "logps/rejected": -262.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.15625,
      "rewards/margins": 13.75,
      "rewards/rejected": -14.875,
      "step": 620
    },
    {
      "epoch": 2.0,
      "eval_logits/chosen": -0.71484375,
      "eval_logits/rejected": -0.4375,
      "eval_logps/chosen": -304.0,
      "eval_logps/rejected": -258.0,
      "eval_loss": 0.00012740046076942235,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.1875,
      "eval_rewards/margins": 13.25,
      "eval_rewards/rejected": -14.4375,
      "eval_runtime": 20.0785,
      "eval_samples_per_second": 9.961,
      "eval_steps_per_second": 0.349,
      "step": 626
    },
    {
      "epoch": 2.012779552715655,
      "grad_norm": 0.003092817828863736,
      "learning_rate": 1.8284023668639053e-07,
      "logits/chosen": -0.7421875,
      "logits/rejected": -0.4296875,
      "logps/chosen": -316.0,
      "logps/rejected": -262.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3515625,
      "rewards/margins": 13.3125,
      "rewards/rejected": -14.6875,
      "step": 630
    },
    {
      "epoch": 2.0447284345047922,
      "grad_norm": 0.02088414959200188,
      "learning_rate": 1.7692307692307693e-07,
      "logits/chosen": -0.70703125,
      "logits/rejected": -0.625,
      "logps/chosen": -354.0,
      "logps/rejected": -270.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.296875,
      "rewards/margins": 13.3125,
      "rewards/rejected": -14.5625,
      "step": 640
    },
    {
      "epoch": 2.07667731629393,
      "grad_norm": 0.02978878745088855,
      "learning_rate": 1.710059171597633e-07,
      "logits/chosen": -0.5625,
      "logits/rejected": -0.515625,
      "logps/chosen": -326.0,
      "logps/rejected": -260.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.453125,
      "rewards/margins": 13.375,
      "rewards/rejected": -14.8125,
      "step": 650
    },
    {
      "epoch": 2.108626198083067,
      "grad_norm": 0.014281695666503702,
      "learning_rate": 1.650887573964497e-07,
      "logits/chosen": -0.6640625,
      "logits/rejected": -0.60546875,
      "logps/chosen": -378.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1875,
      "rewards/margins": 13.625,
      "rewards/rejected": -14.75,
      "step": 660
    },
    {
      "epoch": 2.1405750798722045,
      "grad_norm": 0.1075200266687742,
      "learning_rate": 1.591715976331361e-07,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.447265625,
      "logps/chosen": -306.0,
      "logps/rejected": -266.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3515625,
      "rewards/margins": 13.6875,
      "rewards/rejected": -15.0625,
      "step": 670
    },
    {
      "epoch": 2.1725239616613417,
      "grad_norm": 0.013374753124328204,
      "learning_rate": 1.5325443786982248e-07,
      "logits/chosen": -0.81640625,
      "logits/rejected": -0.5234375,
      "logps/chosen": -312.0,
      "logps/rejected": -266.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.484375,
      "rewards/margins": 13.5625,
      "rewards/rejected": -15.0625,
      "step": 680
    },
    {
      "epoch": 2.2044728434504792,
      "grad_norm": 0.05453915761092606,
      "learning_rate": 1.4733727810650885e-07,
      "logits/chosen": -0.6796875,
      "logits/rejected": -0.50390625,
      "logps/chosen": -298.0,
      "logps/rejected": -262.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.515625,
      "rewards/margins": 13.25,
      "rewards/rejected": -14.75,
      "step": 690
    },
    {
      "epoch": 2.236421725239617,
      "grad_norm": 0.004507719187765237,
      "learning_rate": 1.4142011834319526e-07,
      "logits/chosen": -0.65625,
      "logits/rejected": -0.55078125,
      "logps/chosen": -302.0,
      "logps/rejected": -276.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.5859375,
      "rewards/margins": 13.625,
      "rewards/rejected": -15.25,
      "step": 700
    },
    {
      "epoch": 2.268370607028754,
      "grad_norm": 0.0014351004311762242,
      "learning_rate": 1.3550295857988164e-07,
      "logits/chosen": -0.78125,
      "logits/rejected": -0.625,
      "logps/chosen": -348.0,
      "logps/rejected": -278.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3828125,
      "rewards/margins": 14.3125,
      "rewards/rejected": -15.6875,
      "step": 710
    },
    {
      "epoch": 2.3003194888178915,
      "grad_norm": 0.007661772736529076,
      "learning_rate": 1.2958579881656802e-07,
      "logits/chosen": -0.6328125,
      "logits/rejected": -0.3984375,
      "logps/chosen": -308.0,
      "logps/rejected": -270.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.25,
      "rewards/margins": 14.0,
      "rewards/rejected": -15.25,
      "step": 720
    },
    {
      "epoch": 2.3322683706070286,
      "grad_norm": 0.09440998564444865,
      "learning_rate": 1.2366863905325443e-07,
      "logits/chosen": -0.60546875,
      "logits/rejected": -0.51171875,
      "logps/chosen": -380.0,
      "logps/rejected": -272.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1328125,
      "rewards/margins": 14.3125,
      "rewards/rejected": -15.4375,
      "step": 730
    },
    {
      "epoch": 2.364217252396166,
      "grad_norm": 0.001050873951037513,
      "learning_rate": 1.1775147928994082e-07,
      "logits/chosen": -0.7109375,
      "logits/rejected": -0.6484375,
      "logps/chosen": -336.0,
      "logps/rejected": -274.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3046875,
      "rewards/margins": 13.875,
      "rewards/rejected": -15.1875,
      "step": 740
    },
    {
      "epoch": 2.3961661341853033,
      "grad_norm": 0.0020244185172190623,
      "learning_rate": 1.1183431952662721e-07,
      "logits/chosen": -0.53515625,
      "logits/rejected": -0.439453125,
      "logps/chosen": -282.0,
      "logps/rejected": -272.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3359375,
      "rewards/margins": 14.1875,
      "rewards/rejected": -15.5625,
      "step": 750
    },
    {
      "epoch": 2.428115015974441,
      "grad_norm": 0.0019816508240532195,
      "learning_rate": 1.059171597633136e-07,
      "logits/chosen": -0.66796875,
      "logits/rejected": -0.515625,
      "logps/chosen": -326.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.4609375,
      "rewards/margins": 14.0,
      "rewards/rejected": -15.4375,
      "step": 760
    },
    {
      "epoch": 2.460063897763578,
      "grad_norm": 0.00974446806676642,
      "learning_rate": 1e-07,
      "logits/chosen": -0.6875,
      "logits/rejected": -0.4921875,
      "logps/chosen": -336.0,
      "logps/rejected": -272.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.453125,
      "rewards/margins": 14.0,
      "rewards/rejected": -15.5,
      "step": 770
    },
    {
      "epoch": 2.4920127795527156,
      "grad_norm": 0.0013014264693432427,
      "learning_rate": 9.408284023668639e-08,
      "logits/chosen": -0.6796875,
      "logits/rejected": -0.54296875,
      "logps/chosen": -324.0,
      "logps/rejected": -266.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.640625,
      "rewards/margins": 13.5625,
      "rewards/rejected": -15.1875,
      "step": 780
    },
    {
      "epoch": 2.523961661341853,
      "grad_norm": 0.002562371187665821,
      "learning_rate": 8.816568047337278e-08,
      "logits/chosen": -0.67578125,
      "logits/rejected": -0.5078125,
      "logps/chosen": -354.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.40625,
      "rewards/margins": 13.6875,
      "rewards/rejected": -15.125,
      "step": 790
    },
    {
      "epoch": 2.5559105431309903,
      "grad_norm": 0.000644996434656903,
      "learning_rate": 8.224852071005916e-08,
      "logits/chosen": -0.7265625,
      "logits/rejected": -0.51171875,
      "logps/chosen": -318.0,
      "logps/rejected": -280.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.609375,
      "rewards/margins": 14.5,
      "rewards/rejected": -16.125,
      "step": 800
    },
    {
      "epoch": 2.587859424920128,
      "grad_norm": 0.0008992028582889509,
      "learning_rate": 7.633136094674555e-08,
      "logits/chosen": -0.61328125,
      "logits/rejected": -0.55078125,
      "logps/chosen": -336.0,
      "logps/rejected": -270.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.546875,
      "rewards/margins": 14.375,
      "rewards/rejected": -15.875,
      "step": 810
    },
    {
      "epoch": 2.619808306709265,
      "grad_norm": 0.0012427097539449378,
      "learning_rate": 7.041420118343195e-08,
      "logits/chosen": -0.62890625,
      "logits/rejected": -0.546875,
      "logps/chosen": -354.0,
      "logps/rejected": -278.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.328125,
      "rewards/margins": 14.3125,
      "rewards/rejected": -15.6875,
      "step": 820
    },
    {
      "epoch": 2.6517571884984026,
      "grad_norm": 0.0020240111576199865,
      "learning_rate": 6.449704142011835e-08,
      "logits/chosen": -0.73828125,
      "logits/rejected": -0.53515625,
      "logps/chosen": -308.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3203125,
      "rewards/margins": 14.0,
      "rewards/rejected": -15.375,
      "step": 830
    },
    {
      "epoch": 2.68370607028754,
      "grad_norm": 0.001503960612923789,
      "learning_rate": 5.857988165680473e-08,
      "logits/chosen": -0.640625,
      "logits/rejected": -0.5,
      "logps/chosen": -288.0,
      "logps/rejected": -276.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.2890625,
      "rewards/margins": 14.375,
      "rewards/rejected": -15.6875,
      "step": 840
    },
    {
      "epoch": 2.7156549520766773,
      "grad_norm": 0.0010888686135269838,
      "learning_rate": 5.266272189349112e-08,
      "logits/chosen": -0.703125,
      "logits/rejected": -0.546875,
      "logps/chosen": -326.0,
      "logps/rejected": -274.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.625,
      "rewards/margins": 13.875,
      "rewards/rejected": -15.5,
      "step": 850
    },
    {
      "epoch": 2.747603833865815,
      "grad_norm": 0.003491886927346326,
      "learning_rate": 4.674556213017751e-08,
      "logits/chosen": -0.58984375,
      "logits/rejected": -0.48046875,
      "logps/chosen": -348.0,
      "logps/rejected": -270.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.34375,
      "rewards/margins": 14.125,
      "rewards/rejected": -15.4375,
      "step": 860
    },
    {
      "epoch": 2.779552715654952,
      "grad_norm": 0.0003618045449194657,
      "learning_rate": 4.082840236686391e-08,
      "logits/chosen": -0.74609375,
      "logits/rejected": -0.53125,
      "logps/chosen": -352.0,
      "logps/rejected": -276.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.4609375,
      "rewards/margins": 14.25,
      "rewards/rejected": -15.6875,
      "step": 870
    },
    {
      "epoch": 2.8115015974440896,
      "grad_norm": 0.0020187053364208985,
      "learning_rate": 3.4911242603550294e-08,
      "logits/chosen": -0.69921875,
      "logits/rejected": -0.50390625,
      "logps/chosen": -258.0,
      "logps/rejected": -272.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.5078125,
      "rewards/margins": 14.1875,
      "rewards/rejected": -15.6875,
      "step": 880
    },
    {
      "epoch": 2.8434504792332267,
      "grad_norm": 0.0010734719479623244,
      "learning_rate": 2.8994082840236687e-08,
      "logits/chosen": -0.578125,
      "logits/rejected": -0.51171875,
      "logps/chosen": -382.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.34375,
      "rewards/margins": 13.8125,
      "rewards/rejected": -15.125,
      "step": 890
    },
    {
      "epoch": 2.8753993610223643,
      "grad_norm": 0.0007849148592248786,
      "learning_rate": 2.3076923076923076e-08,
      "logits/chosen": -0.734375,
      "logits/rejected": -0.47265625,
      "logps/chosen": -278.0,
      "logps/rejected": -272.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.6796875,
      "rewards/margins": 14.1875,
      "rewards/rejected": -15.875,
      "step": 900
    },
    {
      "epoch": 2.9073482428115014,
      "grad_norm": 0.0009071422276518199,
      "learning_rate": 1.7159763313609465e-08,
      "logits/chosen": -0.546875,
      "logits/rejected": -0.32421875,
      "logps/chosen": -324.0,
      "logps/rejected": -262.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.5859375,
      "rewards/margins": 14.0625,
      "rewards/rejected": -15.625,
      "step": 910
    },
    {
      "epoch": 2.939297124600639,
      "grad_norm": 0.02336710919043686,
      "learning_rate": 1.1242603550295858e-08,
      "logits/chosen": -0.765625,
      "logits/rejected": -0.427734375,
      "logps/chosen": -348.0,
      "logps/rejected": -268.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.390625,
      "rewards/margins": 14.1875,
      "rewards/rejected": -15.625,
      "step": 920
    },
    {
      "epoch": 2.9712460063897765,
      "grad_norm": 0.03716794354422435,
      "learning_rate": 5.325443786982248e-09,
      "logits/chosen": -0.7421875,
      "logits/rejected": -0.578125,
      "logps/chosen": -326.0,
      "logps/rejected": -272.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.515625,
      "rewards/margins": 14.0625,
      "rewards/rejected": -15.5625,
      "step": 930
    },
    {
      "epoch": 3.0,
      "eval_logits/chosen": -0.71875,
      "eval_logits/rejected": -0.44921875,
      "eval_logps/chosen": -306.0,
      "eval_logps/rejected": -268.0,
      "eval_loss": 7.638990791747347e-05,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.3515625,
      "eval_rewards/margins": 13.9375,
      "eval_rewards/rejected": -15.3125,
      "eval_runtime": 20.8014,
      "eval_samples_per_second": 9.615,
      "eval_steps_per_second": 0.337,
      "step": 939
    },
    {
      "epoch": 3.0,
      "step": 939,
      "total_flos": 0.0,
      "train_loss": 0.03158865887803126,
      "train_runtime": 7648.7703,
      "train_samples_per_second": 3.922,
      "train_steps_per_second": 0.123
    }
  ],
  "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
}