{
  "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": 92.15871921468805,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": 0.244140625,
      "logits/rejected": -0.169921875,
      "logps/chosen": -101.5,
      "logps/rejected": -260.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03194888178913738,
      "grad_norm": 99.44077976719225,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": 0.01434326171875,
      "logits/rejected": -0.208984375,
      "logps/chosen": -109.5,
      "logps/rejected": -264.0,
      "loss": 0.6936,
      "rewards/accuracies": 0.1805555522441864,
      "rewards/chosen": 0.00347900390625,
      "rewards/margins": -0.0048828125,
      "rewards/rejected": 0.00836181640625,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 85.65852957193766,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -0.02587890625,
      "logits/rejected": -0.2578125,
      "logps/chosen": -130.0,
      "logps/rejected": -268.0,
      "loss": 0.6426,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": 0.0478515625,
      "rewards/margins": 0.12060546875,
      "rewards/rejected": -0.072265625,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 59.91544097187796,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -0.1748046875,
      "logits/rejected": -0.16796875,
      "logps/chosen": -122.0,
      "logps/rejected": -262.0,
      "loss": 0.4771,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.1572265625,
      "rewards/margins": 0.51171875,
      "rewards/rejected": -0.353515625,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 17.593197378045506,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -0.1943359375,
      "logits/rejected": -0.2490234375,
      "logps/chosen": -129.0,
      "logps/rejected": -276.0,
      "loss": 0.236,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3671875,
      "rewards/margins": 1.3984375,
      "rewards/rejected": -1.03125,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 4.727887505913167,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -0.0003910064697265625,
      "logits/rejected": -0.1787109375,
      "logps/chosen": -106.5,
      "logps/rejected": -292.0,
      "loss": 0.0451,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.69921875,
      "rewards/margins": 3.3125,
      "rewards/rejected": -2.609375,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 0.7684129658699129,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": 0.111328125,
      "logits/rejected": -0.177734375,
      "logps/chosen": -111.5,
      "logps/rejected": -316.0,
      "loss": 0.0078,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.84375,
      "rewards/margins": 5.71875,
      "rewards/rejected": -4.875,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 2.689581191618461,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -0.033935546875,
      "logits/rejected": -0.142578125,
      "logps/chosen": -105.0,
      "logps/rejected": -332.0,
      "loss": 0.0017,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.91796875,
      "rewards/margins": 7.71875,
      "rewards/rejected": -6.8125,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 0.03932333441712055,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -0.1708984375,
      "logits/rejected": -0.201171875,
      "logps/chosen": -118.5,
      "logps/rejected": -340.0,
      "loss": 0.0008,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.9140625,
      "rewards/margins": 8.5625,
      "rewards/rejected": -7.625,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 0.19109012307419765,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": 0.02099609375,
      "logits/rejected": -0.1708984375,
      "logps/chosen": -115.0,
      "logps/rejected": -336.0,
      "loss": 0.0003,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.1171875,
      "rewards/margins": 9.0625,
      "rewards/rejected": -7.96875,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.0538559664304332,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": 0.031005859375,
      "logits/rejected": -0.220703125,
      "logps/chosen": -104.5,
      "logps/rejected": -352.0,
      "loss": 0.0005,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.125,
      "rewards/margins": 9.625,
      "rewards/rejected": -8.5625,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.02230319153786611,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -0.0096435546875,
      "logits/rejected": -0.1884765625,
      "logps/chosen": -114.5,
      "logps/rejected": -358.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.9921875,
      "rewards/margins": 9.875,
      "rewards/rejected": -8.9375,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.012850056796005177,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -0.000354766845703125,
      "logits/rejected": -0.208984375,
      "logps/chosen": -107.5,
      "logps/rejected": -352.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.1015625,
      "rewards/margins": 10.3125,
      "rewards/rejected": -9.25,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.010788882214532901,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": 0.0230712890625,
      "logits/rejected": -0.083984375,
      "logps/chosen": -99.0,
      "logps/rejected": -360.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.125,
      "rewards/margins": 10.75,
      "rewards/rejected": -9.625,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.007536582701458639,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -0.059326171875,
      "logits/rejected": -0.263671875,
      "logps/chosen": -110.0,
      "logps/rejected": -356.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.140625,
      "rewards/margins": 10.4375,
      "rewards/rejected": -9.25,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 0.006831588583369201,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -0.0712890625,
      "logits/rejected": -0.232421875,
      "logps/chosen": -108.0,
      "logps/rejected": -360.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.171875,
      "rewards/margins": 11.1875,
      "rewards/rejected": -10.0,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 0.011672598505355226,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -0.134765625,
      "logits/rejected": -0.13671875,
      "logps/chosen": -108.0,
      "logps/rejected": -364.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.2578125,
      "rewards/margins": 11.3125,
      "rewards/rejected": -10.0625,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.0072168735171755805,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": 0.08251953125,
      "logits/rejected": -0.2158203125,
      "logps/chosen": -92.5,
      "logps/rejected": -372.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.234375,
      "rewards/margins": 11.8125,
      "rewards/rejected": -10.5625,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.012158762608358928,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -0.004486083984375,
      "logits/rejected": -0.126953125,
      "logps/chosen": -97.5,
      "logps/rejected": -370.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.1328125,
      "rewards/margins": 11.75,
      "rewards/rejected": -10.625,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.005145708105983676,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": 0.0849609375,
      "logits/rejected": -0.228515625,
      "logps/chosen": -101.0,
      "logps/rejected": -366.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.1796875,
      "rewards/margins": 11.5,
      "rewards/rejected": -10.3125,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 0.0043237487176445485,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -0.041015625,
      "logits/rejected": -0.2236328125,
      "logps/chosen": -97.5,
      "logps/rejected": -370.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3125,
      "rewards/margins": 12.125,
      "rewards/rejected": -10.8125,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 0.12874105256796353,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -0.023193359375,
      "logits/rejected": -0.1943359375,
      "logps/chosen": -103.0,
      "logps/rejected": -366.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3515625,
      "rewards/margins": 11.875,
      "rewards/rejected": -10.5,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.007228425803674958,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": 0.10791015625,
      "logits/rejected": -0.140625,
      "logps/chosen": -107.0,
      "logps/rejected": -378.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.25,
      "rewards/margins": 12.3125,
      "rewards/rejected": -11.0625,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 0.003065380853410539,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": 0.03515625,
      "logits/rejected": -0.2001953125,
      "logps/chosen": -129.0,
      "logps/rejected": -374.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.2109375,
      "rewards/margins": 12.125,
      "rewards/rejected": -10.9375,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.0021843027190388093,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -0.015625,
      "logits/rejected": -0.0213623046875,
      "logps/chosen": -98.0,
      "logps/rejected": -384.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.296875,
      "rewards/margins": 12.625,
      "rewards/rejected": -11.375,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.0034189055080181326,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -0.07763671875,
      "logits/rejected": -0.189453125,
      "logps/chosen": -98.0,
      "logps/rejected": -382.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.25,
      "rewards/margins": 12.9375,
      "rewards/rejected": -11.6875,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 0.020105190096043807,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -0.1572265625,
      "logits/rejected": -0.2392578125,
      "logps/chosen": -112.0,
      "logps/rejected": -380.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.203125,
      "rewards/margins": 12.6875,
      "rewards/rejected": -11.5,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 0.0015580211267333395,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": 0.126953125,
      "logits/rejected": -0.126953125,
      "logps/chosen": -95.5,
      "logps/rejected": -380.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5078125,
      "rewards/margins": 13.0,
      "rewards/rejected": -11.5,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 0.00127691178022707,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -0.010498046875,
      "logits/rejected": -0.201171875,
      "logps/chosen": -101.0,
      "logps/rejected": -384.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.34375,
      "rewards/margins": 12.875,
      "rewards/rejected": -11.5,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 0.002640021838977143,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -0.01708984375,
      "logits/rejected": -0.150390625,
      "logps/chosen": -109.0,
      "logps/rejected": -378.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3828125,
      "rewards/margins": 12.75,
      "rewards/rejected": -11.375,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 0.0012727790414612228,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": 0.0859375,
      "logits/rejected": -0.0576171875,
      "logps/chosen": -111.5,
      "logps/rejected": -378.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.390625,
      "rewards/margins": 13.0625,
      "rewards/rejected": -11.6875,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.0014064478241843975,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -0.07568359375,
      "logits/rejected": -0.166015625,
      "logps/chosen": -113.5,
      "logps/rejected": -386.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.421875,
      "rewards/margins": 13.375,
      "rewards/rejected": -12.0,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": 0.1689453125,
      "eval_logits/rejected": -0.0250244140625,
      "eval_logps/chosen": -87.0,
      "eval_logps/rejected": -380.0,
      "eval_loss": 1.2890994184999727e-05,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.4453125,
      "eval_rewards/margins": 13.375,
      "eval_rewards/rejected": -11.875,
      "eval_runtime": 13.7816,
      "eval_samples_per_second": 14.512,
      "eval_steps_per_second": 0.508,
      "step": 313
    },
    {
      "epoch": 1.0223642172523961,
      "grad_norm": 0.0017917415724720658,
      "learning_rate": 3.662721893491124e-07,
      "logits/chosen": -0.0654296875,
      "logits/rejected": -0.1328125,
      "logps/chosen": -119.0,
      "logps/rejected": -380.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.34375,
      "rewards/margins": 13.0625,
      "rewards/rejected": -11.75,
      "step": 320
    },
    {
      "epoch": 1.0543130990415335,
      "grad_norm": 0.0020074112783759687,
      "learning_rate": 3.603550295857988e-07,
      "logits/chosen": 0.04541015625,
      "logits/rejected": -0.04248046875,
      "logps/chosen": -102.0,
      "logps/rejected": -380.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.375,
      "rewards/margins": 13.0625,
      "rewards/rejected": -11.6875,
      "step": 330
    },
    {
      "epoch": 1.0862619808306708,
      "grad_norm": 0.0035786925715002543,
      "learning_rate": 3.544378698224852e-07,
      "logits/chosen": 0.0234375,
      "logits/rejected": -0.1484375,
      "logps/chosen": -106.5,
      "logps/rejected": -380.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.359375,
      "rewards/margins": 13.25,
      "rewards/rejected": -11.875,
      "step": 340
    },
    {
      "epoch": 1.1182108626198084,
      "grad_norm": 0.007092466900781082,
      "learning_rate": 3.485207100591716e-07,
      "logits/chosen": 0.0189208984375,
      "logits/rejected": -0.23828125,
      "logps/chosen": -100.0,
      "logps/rejected": -380.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.3984375,
      "rewards/margins": 13.25,
      "rewards/rejected": -11.8125,
      "step": 350
    },
    {
      "epoch": 1.1501597444089458,
      "grad_norm": 0.0025147276597281025,
      "learning_rate": 3.42603550295858e-07,
      "logits/chosen": -0.20703125,
      "logits/rejected": -0.2158203125,
      "logps/chosen": -105.5,
      "logps/rejected": -386.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4140625,
      "rewards/margins": 13.4375,
      "rewards/rejected": -12.0,
      "step": 360
    },
    {
      "epoch": 1.182108626198083,
      "grad_norm": 0.000977769611685395,
      "learning_rate": 3.366863905325444e-07,
      "logits/chosen": -0.0732421875,
      "logits/rejected": -0.1279296875,
      "logps/chosen": -114.5,
      "logps/rejected": -384.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5,
      "rewards/margins": 13.5,
      "rewards/rejected": -12.0,
      "step": 370
    },
    {
      "epoch": 1.2140575079872205,
      "grad_norm": 0.001771417851809341,
      "learning_rate": 3.3076923076923075e-07,
      "logits/chosen": 0.158203125,
      "logits/rejected": -0.15625,
      "logps/chosen": -94.5,
      "logps/rejected": -386.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5625,
      "rewards/margins": 13.9375,
      "rewards/rejected": -12.375,
      "step": 380
    },
    {
      "epoch": 1.2460063897763578,
      "grad_norm": 0.0008419355251056739,
      "learning_rate": 3.2485207100591716e-07,
      "logits/chosen": 0.0859375,
      "logits/rejected": -0.0830078125,
      "logps/chosen": -108.5,
      "logps/rejected": -390.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4375,
      "rewards/margins": 13.5625,
      "rewards/rejected": -12.125,
      "step": 390
    },
    {
      "epoch": 1.2779552715654952,
      "grad_norm": 0.0007460099222667411,
      "learning_rate": 3.189349112426035e-07,
      "logits/chosen": -0.140625,
      "logits/rejected": -0.158203125,
      "logps/chosen": -117.5,
      "logps/rejected": -378.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4375,
      "rewards/margins": 13.25,
      "rewards/rejected": -11.8125,
      "step": 400
    },
    {
      "epoch": 1.3099041533546325,
      "grad_norm": 0.00091340992235047,
      "learning_rate": 3.130177514792899e-07,
      "logits/chosen": -0.1005859375,
      "logits/rejected": -0.208984375,
      "logps/chosen": -110.5,
      "logps/rejected": -392.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.4140625,
      "rewards/margins": 13.75,
      "rewards/rejected": -12.3125,
      "step": 410
    },
    {
      "epoch": 1.34185303514377,
      "grad_norm": 0.00043549652705192343,
      "learning_rate": 3.071005917159763e-07,
      "logits/chosen": -0.00604248046875,
      "logits/rejected": -0.162109375,
      "logps/chosen": -109.5,
      "logps/rejected": -386.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.578125,
      "rewards/margins": 14.0625,
      "rewards/rejected": -12.4375,
      "step": 420
    },
    {
      "epoch": 1.3738019169329074,
      "grad_norm": 0.0007300024465746544,
      "learning_rate": 3.011834319526627e-07,
      "logits/chosen": -0.0269775390625,
      "logits/rejected": -0.1083984375,
      "logps/chosen": -95.5,
      "logps/rejected": -396.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.578125,
      "rewards/margins": 14.375,
      "rewards/rejected": -12.8125,
      "step": 430
    },
    {
      "epoch": 1.4057507987220448,
      "grad_norm": 0.0024316229937478553,
      "learning_rate": 2.952662721893491e-07,
      "logits/chosen": 0.06884765625,
      "logits/rejected": -0.201171875,
      "logps/chosen": -104.0,
      "logps/rejected": -394.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5625,
      "rewards/margins": 14.625,
      "rewards/rejected": -13.0,
      "step": 440
    },
    {
      "epoch": 1.4376996805111821,
      "grad_norm": 0.002150292576631788,
      "learning_rate": 2.893491124260355e-07,
      "logits/chosen": 0.11328125,
      "logits/rejected": -0.173828125,
      "logps/chosen": -104.5,
      "logps/rejected": -392.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.515625,
      "rewards/margins": 14.375,
      "rewards/rejected": -12.875,
      "step": 450
    },
    {
      "epoch": 1.4696485623003195,
      "grad_norm": 0.0013447377195944799,
      "learning_rate": 2.834319526627219e-07,
      "logits/chosen": 0.0301513671875,
      "logits/rejected": -0.1689453125,
      "logps/chosen": -108.0,
      "logps/rejected": -384.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5390625,
      "rewards/margins": 14.1875,
      "rewards/rejected": -12.625,
      "step": 460
    },
    {
      "epoch": 1.5015974440894568,
      "grad_norm": 0.000837506300739043,
      "learning_rate": 2.7751479289940824e-07,
      "logits/chosen": 0.2109375,
      "logits/rejected": -0.11669921875,
      "logps/chosen": -93.5,
      "logps/rejected": -390.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.7109375,
      "rewards/margins": 14.6875,
      "rewards/rejected": -13.0,
      "step": 470
    },
    {
      "epoch": 1.5335463258785942,
      "grad_norm": 0.0007916201596328788,
      "learning_rate": 2.715976331360947e-07,
      "logits/chosen": 0.06787109375,
      "logits/rejected": -0.06103515625,
      "logps/chosen": -96.0,
      "logps/rejected": -392.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6328125,
      "rewards/margins": 14.375,
      "rewards/rejected": -12.75,
      "step": 480
    },
    {
      "epoch": 1.5654952076677318,
      "grad_norm": 0.0006566737497706091,
      "learning_rate": 2.6568047337278106e-07,
      "logits/chosen": 0.1728515625,
      "logits/rejected": -0.10546875,
      "logps/chosen": -102.0,
      "logps/rejected": -394.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6015625,
      "rewards/margins": 14.25,
      "rewards/rejected": -12.625,
      "step": 490
    },
    {
      "epoch": 1.5974440894568689,
      "grad_norm": 0.002239282962727829,
      "learning_rate": 2.5976331360946746e-07,
      "logits/chosen": -0.05615234375,
      "logits/rejected": -0.150390625,
      "logps/chosen": -120.0,
      "logps/rejected": -388.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.5703125,
      "rewards/margins": 14.25,
      "rewards/rejected": -12.625,
      "step": 500
    },
    {
      "epoch": 1.6293929712460065,
      "grad_norm": 0.0033182842518145877,
      "learning_rate": 2.538461538461538e-07,
      "logits/chosen": 0.1435546875,
      "logits/rejected": -0.1337890625,
      "logps/chosen": -113.5,
      "logps/rejected": -388.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6328125,
      "rewards/margins": 14.4375,
      "rewards/rejected": -12.8125,
      "step": 510
    },
    {
      "epoch": 1.6613418530351438,
      "grad_norm": 0.0009091464843633191,
      "learning_rate": 2.479289940828402e-07,
      "logits/chosen": 0.1376953125,
      "logits/rejected": -0.033203125,
      "logps/chosen": -91.5,
      "logps/rejected": -398.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.75,
      "rewards/margins": 14.75,
      "rewards/rejected": -12.9375,
      "step": 520
    },
    {
      "epoch": 1.6932907348242812,
      "grad_norm": 0.004616007913180375,
      "learning_rate": 2.420118343195266e-07,
      "logits/chosen": 0.08349609375,
      "logits/rejected": -0.1435546875,
      "logps/chosen": -102.5,
      "logps/rejected": -394.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6796875,
      "rewards/margins": 14.625,
      "rewards/rejected": -12.9375,
      "step": 530
    },
    {
      "epoch": 1.7252396166134185,
      "grad_norm": 0.0018390040873526624,
      "learning_rate": 2.36094674556213e-07,
      "logits/chosen": -0.0057373046875,
      "logits/rejected": -0.201171875,
      "logps/chosen": -106.5,
      "logps/rejected": -396.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6328125,
      "rewards/margins": 14.875,
      "rewards/rejected": -13.1875,
      "step": 540
    },
    {
      "epoch": 1.7571884984025559,
      "grad_norm": 0.0008292543279214604,
      "learning_rate": 2.301775147928994e-07,
      "logits/chosen": 0.003082275390625,
      "logits/rejected": -0.08642578125,
      "logps/chosen": -108.0,
      "logps/rejected": -394.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6640625,
      "rewards/margins": 14.75,
      "rewards/rejected": -13.125,
      "step": 550
    },
    {
      "epoch": 1.7891373801916934,
      "grad_norm": 0.00046900594986925745,
      "learning_rate": 2.242603550295858e-07,
      "logits/chosen": -0.055908203125,
      "logits/rejected": -0.1494140625,
      "logps/chosen": -101.5,
      "logps/rejected": -392.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.625,
      "rewards/margins": 14.625,
      "rewards/rejected": -13.0,
      "step": 560
    },
    {
      "epoch": 1.8210862619808306,
      "grad_norm": 0.0010312073068124495,
      "learning_rate": 2.1834319526627217e-07,
      "logits/chosen": 0.1796875,
      "logits/rejected": -0.1689453125,
      "logps/chosen": -92.5,
      "logps/rejected": -396.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.65625,
      "rewards/margins": 14.6875,
      "rewards/rejected": -13.0,
      "step": 570
    },
    {
      "epoch": 1.8530351437699681,
      "grad_norm": 0.009145821042975717,
      "learning_rate": 2.1242603550295858e-07,
      "logits/chosen": 0.0849609375,
      "logits/rejected": -0.0888671875,
      "logps/chosen": -95.5,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.6796875,
      "rewards/margins": 15.25,
      "rewards/rejected": -13.5625,
      "step": 580
    },
    {
      "epoch": 1.8849840255591053,
      "grad_norm": 0.005520031071378534,
      "learning_rate": 2.0650887573964496e-07,
      "logits/chosen": -0.05810546875,
      "logits/rejected": -0.1640625,
      "logps/chosen": -104.0,
      "logps/rejected": -396.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.703125,
      "rewards/margins": 15.0,
      "rewards/rejected": -13.3125,
      "step": 590
    },
    {
      "epoch": 1.9169329073482428,
      "grad_norm": 0.0025139284722797664,
      "learning_rate": 2.0059171597633133e-07,
      "logits/chosen": 0.11083984375,
      "logits/rejected": -0.08740234375,
      "logps/chosen": -92.5,
      "logps/rejected": -394.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.71875,
      "rewards/margins": 14.6875,
      "rewards/rejected": -13.0,
      "step": 600
    },
    {
      "epoch": 1.9488817891373802,
      "grad_norm": 0.00024033636078372886,
      "learning_rate": 1.9467455621301774e-07,
      "logits/chosen": 0.1943359375,
      "logits/rejected": -0.000518798828125,
      "logps/chosen": -99.0,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.7421875,
      "rewards/margins": 15.75,
      "rewards/rejected": -14.0,
      "step": 610
    },
    {
      "epoch": 1.9808306709265175,
      "grad_norm": 0.0002764622725855408,
      "learning_rate": 1.8875739644970412e-07,
      "logits/chosen": -0.029296875,
      "logits/rejected": -0.1591796875,
      "logps/chosen": -98.5,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.796875,
      "rewards/margins": 15.0625,
      "rewards/rejected": -13.25,
      "step": 620
    },
    {
      "epoch": 2.0,
      "eval_logits/chosen": 0.2099609375,
      "eval_logits/rejected": 0.0107421875,
      "eval_logps/chosen": -83.5,
      "eval_logps/rejected": -396.0,
      "eval_loss": 3.1499937449552817e-06,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.828125,
      "eval_rewards/margins": 15.375,
      "eval_rewards/rejected": -13.5625,
      "eval_runtime": 13.7265,
      "eval_samples_per_second": 14.57,
      "eval_steps_per_second": 0.51,
      "step": 626
    },
    {
      "epoch": 2.012779552715655,
      "grad_norm": 0.0005327773266414625,
      "learning_rate": 1.8284023668639053e-07,
      "logits/chosen": 0.13671875,
      "logits/rejected": -0.169921875,
      "logps/chosen": -105.0,
      "logps/rejected": -402.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.78125,
      "rewards/margins": 15.5,
      "rewards/rejected": -13.6875,
      "step": 630
    },
    {
      "epoch": 2.0447284345047922,
      "grad_norm": 0.00022759241145421004,
      "learning_rate": 1.7692307692307693e-07,
      "logits/chosen": 0.048095703125,
      "logits/rejected": -0.12060546875,
      "logps/chosen": -99.5,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.84375,
      "rewards/margins": 15.625,
      "rewards/rejected": -13.75,
      "step": 640
    },
    {
      "epoch": 2.07667731629393,
      "grad_norm": 0.001652160843436512,
      "learning_rate": 1.710059171597633e-07,
      "logits/chosen": -0.00069427490234375,
      "logits/rejected": -0.162109375,
      "logps/chosen": -101.5,
      "logps/rejected": -398.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.78125,
      "rewards/margins": 15.625,
      "rewards/rejected": -13.875,
      "step": 650
    },
    {
      "epoch": 2.108626198083067,
      "grad_norm": 0.0016687636602714445,
      "learning_rate": 1.650887573964497e-07,
      "logits/chosen": 0.04638671875,
      "logits/rejected": -0.1396484375,
      "logps/chosen": -95.5,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.796875,
      "rewards/margins": 15.4375,
      "rewards/rejected": -13.625,
      "step": 660
    },
    {
      "epoch": 2.1405750798722045,
      "grad_norm": 0.00042481779313726915,
      "learning_rate": 1.591715976331361e-07,
      "logits/chosen": -0.0157470703125,
      "logits/rejected": -0.146484375,
      "logps/chosen": -96.5,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.796875,
      "rewards/margins": 15.875,
      "rewards/rejected": -14.0625,
      "step": 670
    },
    {
      "epoch": 2.1725239616613417,
      "grad_norm": 0.0003239142196186926,
      "learning_rate": 1.5325443786982248e-07,
      "logits/chosen": 0.1669921875,
      "logits/rejected": -0.11376953125,
      "logps/chosen": -96.5,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.7109375,
      "rewards/margins": 15.8125,
      "rewards/rejected": -14.0625,
      "step": 680
    },
    {
      "epoch": 2.2044728434504792,
      "grad_norm": 0.00015234963642246293,
      "learning_rate": 1.4733727810650885e-07,
      "logits/chosen": -0.0703125,
      "logits/rejected": -0.185546875,
      "logps/chosen": -96.5,
      "logps/rejected": -398.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8828125,
      "rewards/margins": 15.5,
      "rewards/rejected": -13.625,
      "step": 690
    },
    {
      "epoch": 2.236421725239617,
      "grad_norm": 0.0008997814309118761,
      "learning_rate": 1.4142011834319526e-07,
      "logits/chosen": -0.0380859375,
      "logits/rejected": -0.12890625,
      "logps/chosen": -105.5,
      "logps/rejected": -402.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.7734375,
      "rewards/margins": 15.375,
      "rewards/rejected": -13.5625,
      "step": 700
    },
    {
      "epoch": 2.268370607028754,
      "grad_norm": 0.0003177375115825633,
      "learning_rate": 1.3550295857988164e-07,
      "logits/chosen": -0.10498046875,
      "logits/rejected": -0.130859375,
      "logps/chosen": -101.0,
      "logps/rejected": -398.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.875,
      "rewards/margins": 15.75,
      "rewards/rejected": -13.875,
      "step": 710
    },
    {
      "epoch": 2.3003194888178915,
      "grad_norm": 0.0007482430969870445,
      "learning_rate": 1.2958579881656802e-07,
      "logits/chosen": 0.0927734375,
      "logits/rejected": -0.1328125,
      "logps/chosen": -106.5,
      "logps/rejected": -406.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.765625,
      "rewards/margins": 15.6875,
      "rewards/rejected": -13.9375,
      "step": 720
    },
    {
      "epoch": 2.3322683706070286,
      "grad_norm": 0.00010184283390916899,
      "learning_rate": 1.2366863905325443e-07,
      "logits/chosen": 0.1552734375,
      "logits/rejected": -0.05908203125,
      "logps/chosen": -95.5,
      "logps/rejected": -412.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.921875,
      "rewards/margins": 16.5,
      "rewards/rejected": -14.625,
      "step": 730
    },
    {
      "epoch": 2.364217252396166,
      "grad_norm": 0.0008696203319851818,
      "learning_rate": 1.1775147928994082e-07,
      "logits/chosen": -0.0260009765625,
      "logits/rejected": -0.23828125,
      "logps/chosen": -104.5,
      "logps/rejected": -398.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8359375,
      "rewards/margins": 15.6875,
      "rewards/rejected": -13.8125,
      "step": 740
    },
    {
      "epoch": 2.3961661341853033,
      "grad_norm": 0.0005906389744401823,
      "learning_rate": 1.1183431952662721e-07,
      "logits/chosen": -0.019287109375,
      "logits/rejected": -0.1416015625,
      "logps/chosen": -108.0,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.7890625,
      "rewards/margins": 15.8125,
      "rewards/rejected": -14.0625,
      "step": 750
    },
    {
      "epoch": 2.428115015974441,
      "grad_norm": 0.0009197680517853092,
      "learning_rate": 1.059171597633136e-07,
      "logits/chosen": 0.029052734375,
      "logits/rejected": -0.193359375,
      "logps/chosen": -97.0,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8046875,
      "rewards/margins": 15.8125,
      "rewards/rejected": -14.0,
      "step": 760
    },
    {
      "epoch": 2.460063897763578,
      "grad_norm": 0.01155545960955193,
      "learning_rate": 1e-07,
      "logits/chosen": 0.1630859375,
      "logits/rejected": -0.1318359375,
      "logps/chosen": -82.0,
      "logps/rejected": -412.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.859375,
      "rewards/margins": 16.25,
      "rewards/rejected": -14.4375,
      "step": 770
    },
    {
      "epoch": 2.4920127795527156,
      "grad_norm": 0.0001793365966728007,
      "learning_rate": 9.408284023668639e-08,
      "logits/chosen": 0.140625,
      "logits/rejected": -0.1416015625,
      "logps/chosen": -99.0,
      "logps/rejected": -408.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.765625,
      "rewards/margins": 16.375,
      "rewards/rejected": -14.6875,
      "step": 780
    },
    {
      "epoch": 2.523961661341853,
      "grad_norm": 0.0035154170380901225,
      "learning_rate": 8.816568047337278e-08,
      "logits/chosen": 0.02880859375,
      "logits/rejected": -0.076171875,
      "logps/chosen": -95.0,
      "logps/rejected": -402.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.015625,
      "rewards/margins": 15.6875,
      "rewards/rejected": -13.6875,
      "step": 790
    },
    {
      "epoch": 2.5559105431309903,
      "grad_norm": 0.0006999365033696467,
      "learning_rate": 8.224852071005916e-08,
      "logits/chosen": -0.287109375,
      "logits/rejected": -0.1630859375,
      "logps/chosen": -111.0,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8828125,
      "rewards/margins": 15.875,
      "rewards/rejected": -14.0,
      "step": 800
    },
    {
      "epoch": 2.587859424920128,
      "grad_norm": 0.00043385978100053777,
      "learning_rate": 7.633136094674555e-08,
      "logits/chosen": 0.0234375,
      "logits/rejected": -0.1416015625,
      "logps/chosen": -94.0,
      "logps/rejected": -408.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.828125,
      "rewards/margins": 16.5,
      "rewards/rejected": -14.6875,
      "step": 810
    },
    {
      "epoch": 2.619808306709265,
      "grad_norm": 0.00028654137725038975,
      "learning_rate": 7.041420118343195e-08,
      "logits/chosen": -0.0263671875,
      "logits/rejected": -0.171875,
      "logps/chosen": -102.0,
      "logps/rejected": -408.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.984375,
      "rewards/margins": 16.375,
      "rewards/rejected": -14.4375,
      "step": 820
    },
    {
      "epoch": 2.6517571884984026,
      "grad_norm": 0.010220233121677299,
      "learning_rate": 6.449704142011835e-08,
      "logits/chosen": -0.038330078125,
      "logits/rejected": -0.04638671875,
      "logps/chosen": -103.5,
      "logps/rejected": -410.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9140625,
      "rewards/margins": 16.5,
      "rewards/rejected": -14.625,
      "step": 830
    },
    {
      "epoch": 2.68370607028754,
      "grad_norm": 0.012838197327440185,
      "learning_rate": 5.857988165680473e-08,
      "logits/chosen": 0.0185546875,
      "logits/rejected": -0.130859375,
      "logps/chosen": -95.0,
      "logps/rejected": -412.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 2.03125,
      "rewards/margins": 16.625,
      "rewards/rejected": -14.625,
      "step": 840
    },
    {
      "epoch": 2.7156549520766773,
      "grad_norm": 0.00029967980239553816,
      "learning_rate": 5.266272189349112e-08,
      "logits/chosen": 0.08447265625,
      "logits/rejected": -0.1015625,
      "logps/chosen": -102.0,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9296875,
      "rewards/margins": 15.9375,
      "rewards/rejected": -14.0,
      "step": 850
    },
    {
      "epoch": 2.747603833865815,
      "grad_norm": 0.0002762678724235396,
      "learning_rate": 4.674556213017751e-08,
      "logits/chosen": 0.048828125,
      "logits/rejected": -0.1591796875,
      "logps/chosen": -100.0,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.890625,
      "rewards/margins": 16.0,
      "rewards/rejected": -14.0625,
      "step": 860
    },
    {
      "epoch": 2.779552715654952,
      "grad_norm": 0.00012800387502098561,
      "learning_rate": 4.082840236686391e-08,
      "logits/chosen": 0.16796875,
      "logits/rejected": -0.0634765625,
      "logps/chosen": -108.5,
      "logps/rejected": -418.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8984375,
      "rewards/margins": 16.625,
      "rewards/rejected": -14.75,
      "step": 870
    },
    {
      "epoch": 2.8115015974440896,
      "grad_norm": 0.005831428455641523,
      "learning_rate": 3.4911242603550294e-08,
      "logits/chosen": 0.0576171875,
      "logits/rejected": -0.08837890625,
      "logps/chosen": -87.5,
      "logps/rejected": -410.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9375,
      "rewards/margins": 16.375,
      "rewards/rejected": -14.375,
      "step": 880
    },
    {
      "epoch": 2.8434504792332267,
      "grad_norm": 0.00012405404139800464,
      "learning_rate": 2.8994082840236687e-08,
      "logits/chosen": -0.045166015625,
      "logits/rejected": -0.041259765625,
      "logps/chosen": -108.0,
      "logps/rejected": -404.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9765625,
      "rewards/margins": 16.125,
      "rewards/rejected": -14.125,
      "step": 890
    },
    {
      "epoch": 2.8753993610223643,
      "grad_norm": 0.00014955962263616494,
      "learning_rate": 2.3076923076923076e-08,
      "logits/chosen": -0.05517578125,
      "logits/rejected": -0.171875,
      "logps/chosen": -101.5,
      "logps/rejected": -400.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.9296875,
      "rewards/margins": 15.875,
      "rewards/rejected": -13.9375,
      "step": 900
    },
    {
      "epoch": 2.9073482428115014,
      "grad_norm": 9.318676138250297e-05,
      "learning_rate": 1.7159763313609465e-08,
      "logits/chosen": 0.00396728515625,
      "logits/rejected": -0.0693359375,
      "logps/chosen": -114.5,
      "logps/rejected": -412.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8125,
      "rewards/margins": 16.375,
      "rewards/rejected": -14.5625,
      "step": 910
    },
    {
      "epoch": 2.939297124600639,
      "grad_norm": 0.009008026329567341,
      "learning_rate": 1.1242603550295858e-08,
      "logits/chosen": 0.1337890625,
      "logits/rejected": -0.123046875,
      "logps/chosen": -97.0,
      "logps/rejected": -406.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.890625,
      "rewards/margins": 16.125,
      "rewards/rejected": -14.25,
      "step": 920
    },
    {
      "epoch": 2.9712460063897765,
      "grad_norm": 0.00016004830074262195,
      "learning_rate": 5.325443786982248e-09,
      "logits/chosen": -0.08251953125,
      "logits/rejected": -0.16015625,
      "logps/chosen": -99.0,
      "logps/rejected": -406.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.8828125,
      "rewards/margins": 16.25,
      "rewards/rejected": -14.3125,
      "step": 930
    },
    {
      "epoch": 3.0,
      "eval_logits/chosen": 0.193359375,
      "eval_logits/rejected": 0.004974365234375,
      "eval_logps/chosen": -82.0,
      "eval_logps/rejected": -406.0,
      "eval_loss": 1.4230236047296785e-06,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 1.9609375,
      "eval_rewards/margins": 16.375,
      "eval_rewards/rejected": -14.375,
      "eval_runtime": 16.4141,
      "eval_samples_per_second": 12.185,
      "eval_steps_per_second": 0.426,
      "step": 939
    },
    {
      "epoch": 3.0,
      "step": 939,
      "total_flos": 0.0,
      "train_loss": 0.022443719598399206,
      "train_runtime": 5691.11,
      "train_samples_per_second": 5.271,
      "train_steps_per_second": 0.165
    }
  ],
  "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
}