{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9499796665311102,
  "eval_steps": 500,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032533550223668155,
      "grad_norm": 103.71580505371094,
      "kl": 0.0,
      "learning_rate": 6.493506493506494e-09,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -158.7671875,
      "logps/rejected": -235.8359375,
      "loss": 0.5,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 10
    },
    {
      "epoch": 0.06506710044733631,
      "grad_norm": 94.47801208496094,
      "kl": 0.05097656324505806,
      "learning_rate": 2.1103896103896103e-08,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -158.2953125,
      "logps/rejected": -232.075,
      "loss": 0.498,
      "rewards/chosen": 0.009334802627563477,
      "rewards/margins": 0.012342309951782227,
      "rewards/rejected": -0.00300750732421875,
      "step": 20
    },
    {
      "epoch": 0.09760065067100447,
      "grad_norm": 143.15447998046875,
      "kl": 0.592968761920929,
      "learning_rate": 3.733766233766234e-08,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -156.840625,
      "logps/rejected": -231.65,
      "loss": 0.471,
      "rewards/chosen": 0.14929332733154296,
      "rewards/margins": 0.1863230228424072,
      "rewards/rejected": -0.037029695510864255,
      "step": 30
    },
    {
      "epoch": 0.13013420089467262,
      "grad_norm": 81.36482238769531,
      "kl": 1.6533203125,
      "learning_rate": 5.3571428571428564e-08,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -151.7375,
      "logps/rejected": -231.0375,
      "loss": 0.3677,
      "rewards/chosen": 0.6113525390625,
      "rewards/margins": 0.9034156799316406,
      "rewards/rejected": -0.29206314086914065,
      "step": 40
    },
    {
      "epoch": 0.16266775111834078,
      "grad_norm": 76.27654266357422,
      "kl": 0.14501953125,
      "learning_rate": 6.98051948051948e-08,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -141.165625,
      "logps/rejected": -251.2015625,
      "loss": 0.1578,
      "rewards/chosen": 1.392755126953125,
      "rewards/margins": 2.920855712890625,
      "rewards/rejected": -1.5281005859375,
      "step": 50
    },
    {
      "epoch": 0.19520130134200894,
      "grad_norm": 3.925260066986084,
      "kl": 0.0025390624068677425,
      "learning_rate": 8.603896103896104e-08,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -127.53515625,
      "logps/rejected": -274.090625,
      "loss": 0.0543,
      "rewards/chosen": 2.5130126953125,
      "rewards/margins": 5.8057373046875,
      "rewards/rejected": -3.292724609375,
      "step": 60
    },
    {
      "epoch": 0.2277348515656771,
      "grad_norm": 0.3158203661441803,
      "kl": 0.0,
      "learning_rate": 1.0227272727272728e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -100.13359375,
      "logps/rejected": -305.453125,
      "loss": 0.0025,
      "rewards/chosen": 4.975732421875,
      "rewards/margins": 10.637548828124999,
      "rewards/rejected": -5.66181640625,
      "step": 70
    },
    {
      "epoch": 0.26026840178934524,
      "grad_norm": 0.01107876282185316,
      "kl": 0.0,
      "learning_rate": 1.1850649350649349e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -75.421875,
      "logps/rejected": -338.096875,
      "loss": 0.0007,
      "rewards/chosen": 6.189892578125,
      "rewards/margins": 13.384521484375,
      "rewards/rejected": -7.19462890625,
      "step": 80
    },
    {
      "epoch": 0.2928019520130134,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 1.3474025974025975e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -56.57421875,
      "logps/rejected": -360.18125,
      "loss": 0.0015,
      "rewards/chosen": 8.42177734375,
      "rewards/margins": 18.254101562499997,
      "rewards/rejected": -9.83232421875,
      "step": 90
    },
    {
      "epoch": 0.32533550223668156,
      "grad_norm": 0.0,
      "kl": 0.15000000596046448,
      "learning_rate": 1.5097402597402597e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.366015625,
      "logps/rejected": -357.228125,
      "loss": 0.0,
      "rewards/chosen": 8.46904296875,
      "rewards/margins": 17.947460937499997,
      "rewards/rejected": -9.47841796875,
      "step": 100
    },
    {
      "epoch": 0.3578690524603497,
      "grad_norm": 0.0,
      "kl": 0.26777344942092896,
      "learning_rate": 1.6720779220779217e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.25625,
      "logps/rejected": -355.371875,
      "loss": 0.0,
      "rewards/chosen": 9.096875,
      "rewards/margins": 19.25791015625,
      "rewards/rejected": -10.16103515625,
      "step": 110
    },
    {
      "epoch": 0.3904026026840179,
      "grad_norm": 0.0,
      "kl": 0.934863269329071,
      "learning_rate": 1.8344155844155843e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -47.00859375,
      "logps/rejected": -353.48125,
      "loss": 0.0002,
      "rewards/chosen": 8.9310546875,
      "rewards/margins": 17.91005859375,
      "rewards/rejected": -8.97900390625,
      "step": 120
    },
    {
      "epoch": 0.42293615290768605,
      "grad_norm": 0.029346637427806854,
      "kl": 3.103515625,
      "learning_rate": 1.9967532467532466e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -46.458203125,
      "logps/rejected": -350.59375,
      "loss": 0.0,
      "rewards/chosen": 8.2421875,
      "rewards/margins": 17.04013671875,
      "rewards/rejected": -8.79794921875,
      "step": 130
    },
    {
      "epoch": 0.4554697031313542,
      "grad_norm": 0.0,
      "kl": 0.69140625,
      "learning_rate": 2.159090909090909e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -47.403125,
      "logps/rejected": -356.59375,
      "loss": 0.0001,
      "rewards/chosen": 8.01259765625,
      "rewards/margins": 18.42900390625,
      "rewards/rejected": -10.41640625,
      "step": 140
    },
    {
      "epoch": 0.48800325335502237,
      "grad_norm": 0.0,
      "kl": 0.16386719048023224,
      "learning_rate": 2.3214285714285714e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -47.1765625,
      "logps/rejected": -367.81875,
      "loss": 0.0,
      "rewards/chosen": 8.698828125,
      "rewards/margins": 18.972265625,
      "rewards/rejected": -10.2734375,
      "step": 150
    },
    {
      "epoch": 0.5205368035786905,
      "grad_norm": 0.0938071757555008,
      "kl": 3.0655274391174316,
      "learning_rate": 2.483766233766234e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -47.46953125,
      "logps/rejected": -350.990625,
      "loss": 0.0001,
      "rewards/chosen": 8.87724609375,
      "rewards/margins": 17.825390624999997,
      "rewards/rejected": -8.94814453125,
      "step": 160
    },
    {
      "epoch": 0.5530703538023587,
      "grad_norm": 0.0,
      "kl": 0.19960936903953552,
      "learning_rate": 2.6461038961038964e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.409765625,
      "logps/rejected": -372.10625,
      "loss": 0.0,
      "rewards/chosen": 8.11689453125,
      "rewards/margins": 19.501953125,
      "rewards/rejected": -11.38505859375,
      "step": 170
    },
    {
      "epoch": 0.5856039040260268,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 2.8084415584415584e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.136328125,
      "logps/rejected": -387.340625,
      "loss": 0.0,
      "rewards/chosen": 8.95927734375,
      "rewards/margins": 20.40771484375,
      "rewards/rejected": -11.4484375,
      "step": 180
    },
    {
      "epoch": 0.618137454249695,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 2.9707792207792204e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -50.2359375,
      "logps/rejected": -391.128125,
      "loss": 0.0001,
      "rewards/chosen": 8.42666015625,
      "rewards/margins": 20.3193359375,
      "rewards/rejected": -11.89267578125,
      "step": 190
    },
    {
      "epoch": 0.6506710044733631,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 3.133116883116883e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.67421875,
      "logps/rejected": -390.3875,
      "loss": 0.0001,
      "rewards/chosen": 8.5646484375,
      "rewards/margins": 20.190722656250003,
      "rewards/rejected": -11.62607421875,
      "step": 200
    },
    {
      "epoch": 0.6832045546970313,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 3.295454545454545e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -50.19453125,
      "logps/rejected": -393.453125,
      "loss": 0.0,
      "rewards/chosen": 8.9052734375,
      "rewards/margins": 21.5578125,
      "rewards/rejected": -12.6525390625,
      "step": 210
    },
    {
      "epoch": 0.7157381049206994,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 3.457792207792208e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.93125,
      "logps/rejected": -401.3375,
      "loss": 0.0,
      "rewards/chosen": 9.0279296875,
      "rewards/margins": 22.509765625,
      "rewards/rejected": -13.4818359375,
      "step": 220
    },
    {
      "epoch": 0.7482716551443677,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 3.62012987012987e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -50.351171875,
      "logps/rejected": -399.60625,
      "loss": 0.0,
      "rewards/chosen": 8.4427734375,
      "rewards/margins": 22.73662109375,
      "rewards/rejected": -14.29384765625,
      "step": 230
    },
    {
      "epoch": 0.7808052053680358,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 3.782467532467532e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -50.937109375,
      "logps/rejected": -398.54375,
      "loss": 0.0,
      "rewards/chosen": 9.748828125,
      "rewards/margins": 22.75693359375,
      "rewards/rejected": -13.00810546875,
      "step": 240
    },
    {
      "epoch": 0.813338755591704,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 3.9448051948051946e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -50.4765625,
      "logps/rejected": -406.259375,
      "loss": 0.0,
      "rewards/chosen": 7.9296875,
      "rewards/margins": 21.64990234375,
      "rewards/rejected": -13.72021484375,
      "step": 250
    },
    {
      "epoch": 0.8458723058153721,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.1071428571428566e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -50.065234375,
      "logps/rejected": -397.978125,
      "loss": 0.0,
      "rewards/chosen": 8.79658203125,
      "rewards/margins": 22.427539062500003,
      "rewards/rejected": -13.63095703125,
      "step": 260
    },
    {
      "epoch": 0.8784058560390403,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.2694805194805197e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.828515625,
      "logps/rejected": -395.05,
      "loss": 0.0,
      "rewards/chosen": 8.353515625,
      "rewards/margins": 19.6958984375,
      "rewards/rejected": -11.3423828125,
      "step": 270
    },
    {
      "epoch": 0.9109394062627084,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.4318181818181817e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.0359375,
      "logps/rejected": -401.634375,
      "loss": 0.0,
      "rewards/chosen": 8.53701171875,
      "rewards/margins": 20.25283203125,
      "rewards/rejected": -11.7158203125,
      "step": 280
    },
    {
      "epoch": 0.9434729564863765,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.5941558441558437e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.019140625,
      "logps/rejected": -393.425,
      "loss": 0.0,
      "rewards/chosen": 8.6474609375,
      "rewards/margins": 20.7501953125,
      "rewards/rejected": -12.102734375,
      "step": 290
    },
    {
      "epoch": 0.9760065067100447,
      "grad_norm": 0.016830716282129288,
      "kl": 0.0,
      "learning_rate": 4.756493506493506e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.725,
      "logps/rejected": -394.653125,
      "loss": 0.0,
      "rewards/chosen": 8.255859375,
      "rewards/margins": 21.61572265625,
      "rewards/rejected": -13.35986328125,
      "step": 300
    },
    {
      "epoch": 1.0065067100447336,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.918831168831168e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.37416666666667,
      "logps/rejected": -396.87,
      "loss": 0.0,
      "rewards/chosen": 7.7615625,
      "rewards/margins": 19.706875,
      "rewards/rejected": -11.9453125,
      "step": 310
    },
    {
      "epoch": 1.0390402602684017,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.999959861406242e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.408203125,
      "logps/rejected": -403.834375,
      "loss": 0.0,
      "rewards/chosen": 9.22265625,
      "rewards/margins": 22.9748046875,
      "rewards/rejected": -13.7521484375,
      "step": 320
    },
    {
      "epoch": 1.07157381049207,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.999638760389452e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.765625,
      "logps/rejected": -396.6625,
      "loss": 0.0,
      "rewards/chosen": 9.08193359375,
      "rewards/margins": 19.86611328125,
      "rewards/rejected": -10.7841796875,
      "step": 330
    },
    {
      "epoch": 1.1041073607157381,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.998996599598879e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.26953125,
      "logps/rejected": -393.775,
      "loss": 0.0,
      "rewards/chosen": 8.9400390625,
      "rewards/margins": 20.629199218750003,
      "rewards/rejected": -11.68916015625,
      "step": 340
    },
    {
      "epoch": 1.1366409109394062,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.998033461515242e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.61484375,
      "logps/rejected": -396.7875,
      "loss": 0.0,
      "rewards/chosen": 9.119140625,
      "rewards/margins": 22.4150390625,
      "rewards/rejected": -13.2958984375,
      "step": 350
    },
    {
      "epoch": 1.1691744611630743,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.996749469846372e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.3609375,
      "logps/rejected": -396.115625,
      "loss": 0.0,
      "rewards/chosen": 8.59248046875,
      "rewards/margins": 22.016796875,
      "rewards/rejected": -13.42431640625,
      "step": 360
    },
    {
      "epoch": 1.2017080113867427,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.995144789511329e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.6953125,
      "logps/rejected": -401.90625,
      "loss": 0.0,
      "rewards/chosen": 8.97890625,
      "rewards/margins": 22.325878906249997,
      "rewards/rejected": -13.34697265625,
      "step": 370
    },
    {
      "epoch": 1.2342415616104108,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.993219626619219e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.890625,
      "logps/rejected": -401.4625,
      "loss": 0.0,
      "rewards/chosen": 8.95244140625,
      "rewards/margins": 22.37314453125,
      "rewards/rejected": -13.420703125,
      "step": 380
    },
    {
      "epoch": 1.2667751118340789,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.990974228442717e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.7734375,
      "logps/rejected": -403.28125,
      "loss": 0.0,
      "rewards/chosen": 8.423828125,
      "rewards/margins": 20.5177734375,
      "rewards/rejected": -12.0939453125,
      "step": 390
    },
    {
      "epoch": 1.299308662057747,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.988408883386307e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.671875,
      "logps/rejected": -399.890625,
      "loss": 0.0,
      "rewards/chosen": 8.93525390625,
      "rewards/margins": 21.9240234375,
      "rewards/rejected": -12.98876953125,
      "step": 400
    },
    {
      "epoch": 1.331842212281415,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.985523920949242e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.679296875,
      "logps/rejected": -402.096875,
      "loss": 0.0,
      "rewards/chosen": 8.42080078125,
      "rewards/margins": 20.993359374999997,
      "rewards/rejected": -12.57255859375,
      "step": 410
    },
    {
      "epoch": 1.3643757625050834,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.982319711683221e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.55546875,
      "logps/rejected": -395.29375,
      "loss": 0.0,
      "rewards/chosen": 9.683203125,
      "rewards/margins": 22.695703125,
      "rewards/rejected": -13.0125,
      "step": 420
    },
    {
      "epoch": 1.3969093127287515,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.978796667144791e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.65859375,
      "logps/rejected": -403.265625,
      "loss": 0.0,
      "rewards/chosen": 8.24130859375,
      "rewards/margins": 21.69130859375,
      "rewards/rejected": -13.45,
      "step": 430
    },
    {
      "epoch": 1.4294428629524196,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.974955239842493e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.275390625,
      "logps/rejected": -410.9125,
      "loss": 0.0,
      "rewards/chosen": 8.18759765625,
      "rewards/margins": 21.62451171875,
      "rewards/rejected": -13.4369140625,
      "step": 440
    },
    {
      "epoch": 1.461976413176088,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.970795923178733e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.339453125,
      "logps/rejected": -405.471875,
      "loss": 0.0,
      "rewards/chosen": 7.85322265625,
      "rewards/margins": 21.57744140625,
      "rewards/rejected": -13.72421875,
      "step": 450
    },
    {
      "epoch": 1.494509963399756,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.966319251386412e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.5921875,
      "logps/rejected": -401.665625,
      "loss": 0.0,
      "rewards/chosen": 8.530078125,
      "rewards/margins": 21.6490234375,
      "rewards/rejected": -13.1189453125,
      "step": 460
    },
    {
      "epoch": 1.5270435136234242,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.961525799460308e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.540625,
      "logps/rejected": -408.703125,
      "loss": 0.0,
      "rewards/chosen": 8.6064453125,
      "rewards/margins": 22.0228515625,
      "rewards/rejected": -13.41640625,
      "step": 470
    },
    {
      "epoch": 1.5595770638470923,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.956416183083221e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.923046875,
      "logps/rejected": -407.4,
      "loss": 0.0,
      "rewards/chosen": 8.303515625,
      "rewards/margins": 23.2646484375,
      "rewards/rejected": -14.9611328125,
      "step": 480
    },
    {
      "epoch": 1.5921106140707604,
      "grad_norm": 0.0061869011260569096,
      "kl": 0.0,
      "learning_rate": 4.950991058546892e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.971484375,
      "logps/rejected": -404.6,
      "loss": 0.0,
      "rewards/chosen": 8.97802734375,
      "rewards/margins": 21.27333984375,
      "rewards/rejected": -12.2953125,
      "step": 490
    },
    {
      "epoch": 1.6246441642944287,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.945251122667714e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.584765625,
      "logps/rejected": -411.175,
      "loss": 0.0,
      "rewards/chosen": 8.23486328125,
      "rewards/margins": 21.76025390625,
      "rewards/rejected": -13.525390625,
      "step": 500
    },
    {
      "epoch": 1.6571777145180968,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.93919711269722e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.616015625,
      "logps/rejected": -403.053125,
      "loss": 0.0,
      "rewards/chosen": 8.54775390625,
      "rewards/margins": 21.637109375,
      "rewards/rejected": -13.08935546875,
      "step": 510
    },
    {
      "epoch": 1.689711264741765,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.932829806227398e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.595703125,
      "logps/rejected": -401.678125,
      "loss": 0.0,
      "rewards/chosen": 9.25576171875,
      "rewards/margins": 22.1841796875,
      "rewards/rejected": -12.92841796875,
      "step": 520
    },
    {
      "epoch": 1.7222448149654332,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.926150021090812e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.584765625,
      "logps/rejected": -408.21875,
      "loss": 0.0,
      "rewards/chosen": 9.191796875,
      "rewards/margins": 23.53251953125,
      "rewards/rejected": -14.34072265625,
      "step": 530
    },
    {
      "epoch": 1.7547783651891011,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.919158615255555e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.476953125,
      "logps/rejected": -402.609375,
      "loss": 0.0,
      "rewards/chosen": 8.7083984375,
      "rewards/margins": 22.474316406249997,
      "rewards/rejected": -13.76591796875,
      "step": 540
    },
    {
      "epoch": 1.7873119154127695,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.911856486715056e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.780078125,
      "logps/rejected": -407.21875,
      "loss": 0.0,
      "rewards/chosen": 9.62373046875,
      "rewards/margins": 23.37060546875,
      "rewards/rejected": -13.746875,
      "step": 550
    },
    {
      "epoch": 1.8198454656364376,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.904244573372733e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.086328125,
      "logps/rejected": -407.178125,
      "loss": 0.0,
      "rewards/chosen": 8.03115234375,
      "rewards/margins": 22.18642578125,
      "rewards/rejected": -14.1552734375,
      "step": 560
    },
    {
      "epoch": 1.8523790158601057,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.896323852921527e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -49.037890625,
      "logps/rejected": -406.178125,
      "loss": 0.0,
      "rewards/chosen": 8.99189453125,
      "rewards/margins": 22.690234375000003,
      "rewards/rejected": -13.69833984375,
      "step": 570
    },
    {
      "epoch": 1.884912566083774,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.888095342718329e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.119140625,
      "logps/rejected": -405.759375,
      "loss": 0.0,
      "rewards/chosen": 8.451171875,
      "rewards/margins": 21.19541015625,
      "rewards/rejected": -12.74423828125,
      "step": 580
    },
    {
      "epoch": 1.9174461163074419,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.879560099653306e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.169921875,
      "logps/rejected": -406.915625,
      "loss": 0.0,
      "rewards/chosen": 8.71015625,
      "rewards/margins": 20.17109375,
      "rewards/rejected": -11.4609375,
      "step": 590
    },
    {
      "epoch": 1.9499796665311102,
      "grad_norm": 0.0,
      "kl": 0.0,
      "learning_rate": 4.870719220014149e-07,
      "logits/chosen": NaN,
      "logits/rejected": NaN,
      "logps/chosen": -48.301953125,
      "logps/rejected": -401.065625,
      "loss": 0.0,
      "rewards/chosen": 8.30107421875,
      "rewards/margins": 21.400097656249997,
      "rewards/rejected": -13.0990234375,
      "step": 600
    }
  ],
  "logging_steps": 10,
  "max_steps": 3080,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}