{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 478,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020920502092050207,
      "grad_norm": 7.757817121588263,
      "learning_rate": 1.0416666666666666e-08,
      "logits/chosen": -3.21875,
      "logits/rejected": -3.21875,
      "logps/chosen": -420.0,
      "logps/rejected": -456.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.02092050209205021,
      "grad_norm": 7.3309462182136045,
      "learning_rate": 1.0416666666666667e-07,
      "logits/chosen": -3.265625,
      "logits/rejected": -3.265625,
      "logps/chosen": -486.0,
      "logps/rejected": -510.0,
      "loss": 0.6919,
      "rewards/accuracies": 0.1319444477558136,
      "rewards/chosen": -0.000278472900390625,
      "rewards/margins": 0.00167083740234375,
      "rewards/rejected": -0.00194549560546875,
      "step": 10
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 7.629641628995199,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": -3.21875,
      "logits/rejected": -3.21875,
      "logps/chosen": -520.0,
      "logps/rejected": -536.0,
      "loss": 0.6868,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": -0.00144195556640625,
      "rewards/margins": 0.0125732421875,
      "rewards/rejected": -0.0140380859375,
      "step": 20
    },
    {
      "epoch": 0.06276150627615062,
      "grad_norm": 8.637096465661159,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -3.265625,
      "logits/rejected": -3.265625,
      "logps/chosen": -476.0,
      "logps/rejected": -496.0,
      "loss": 0.6566,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.036865234375,
      "rewards/margins": 0.07958984375,
      "rewards/rejected": -0.1162109375,
      "step": 30
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 14.719052878043176,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.234375,
      "logps/chosen": -532.0,
      "logps/rejected": -572.0,
      "loss": 0.5596,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": -0.16015625,
      "rewards/margins": 0.310546875,
      "rewards/rejected": -0.470703125,
      "step": 40
    },
    {
      "epoch": 0.10460251046025104,
      "grad_norm": 16.841624892592055,
      "learning_rate": 4.999733114418725e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.296875,
      "logps/chosen": -572.0,
      "logps/rejected": -652.0,
      "loss": 0.4385,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.70703125,
      "rewards/margins": 0.71875,
      "rewards/rejected": -1.421875,
      "step": 50
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 20.804845236736206,
      "learning_rate": 4.990398100856366e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.34375,
      "logps/chosen": -696.0,
      "logps/rejected": -844.0,
      "loss": 0.3525,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -1.515625,
      "rewards/margins": 1.421875,
      "rewards/rejected": -2.9375,
      "step": 60
    },
    {
      "epoch": 0.14644351464435146,
      "grad_norm": 16.70191345751194,
      "learning_rate": 4.967775735898179e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.390625,
      "logps/chosen": -716.0,
      "logps/rejected": -916.0,
      "loss": 0.3193,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": -2.0,
      "rewards/margins": 1.84375,
      "rewards/rejected": -3.84375,
      "step": 70
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 20.459305442391305,
      "learning_rate": 4.931986719649298e-07,
      "logits/chosen": -3.40625,
      "logits/rejected": -3.359375,
      "logps/chosen": -772.0,
      "logps/rejected": -972.0,
      "loss": 0.3076,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -2.421875,
      "rewards/margins": 1.9765625,
      "rewards/rejected": -4.40625,
      "step": 80
    },
    {
      "epoch": 0.18828451882845187,
      "grad_norm": 17.548957954666694,
      "learning_rate": 4.883222001996351e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.28125,
      "logps/chosen": -744.0,
      "logps/rejected": -1008.0,
      "loss": 0.2369,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -2.421875,
      "rewards/margins": 2.546875,
      "rewards/rejected": -4.96875,
      "step": 90
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 25.991680151043717,
      "learning_rate": 4.821741763807186e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.25,
      "logps/chosen": -748.0,
      "logps/rejected": -1064.0,
      "loss": 0.2037,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -2.609375,
      "rewards/margins": 2.96875,
      "rewards/rejected": -5.59375,
      "step": 100
    },
    {
      "epoch": 0.20920502092050208,
      "eval_logits/chosen": -3.34375,
      "eval_logits/rejected": -3.296875,
      "eval_logps/chosen": -732.0,
      "eval_logps/rejected": -1020.0,
      "eval_loss": 0.23008862137794495,
      "eval_rewards/accuracies": 0.8515625,
      "eval_rewards/chosen": -2.328125,
      "eval_rewards/margins": 2.8125,
      "eval_rewards/rejected": -5.15625,
      "eval_runtime": 89.917,
      "eval_samples_per_second": 22.243,
      "eval_steps_per_second": 0.356,
      "step": 100
    },
    {
      "epoch": 0.2301255230125523,
      "grad_norm": 28.455783724679673,
      "learning_rate": 4.747874028753375e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.234375,
      "logps/chosen": -744.0,
      "logps/rejected": -1144.0,
      "loss": 0.2085,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -2.46875,
      "rewards/margins": 3.71875,
      "rewards/rejected": -6.1875,
      "step": 110
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 22.582856541891317,
      "learning_rate": 4.662012913161997e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.25,
      "logps/chosen": -736.0,
      "logps/rejected": -1128.0,
      "loss": 0.1922,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -2.625,
      "rewards/margins": 3.734375,
      "rewards/rejected": -6.34375,
      "step": 120
    },
    {
      "epoch": 0.2719665271966527,
      "grad_norm": 30.22114162417063,
      "learning_rate": 4.5646165232345103e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.25,
      "logps/chosen": -800.0,
      "logps/rejected": -1160.0,
      "loss": 0.1934,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -3.078125,
      "rewards/margins": 3.515625,
      "rewards/rejected": -6.59375,
      "step": 130
    },
    {
      "epoch": 0.2928870292887029,
      "grad_norm": 15.220421215227832,
      "learning_rate": 4.456204510851956e-07,
      "logits/chosen": -3.28125,
      "logits/rejected": -3.21875,
      "logps/chosen": -832.0,
      "logps/rejected": -1256.0,
      "loss": 0.1646,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -3.203125,
      "rewards/margins": 4.21875,
      "rewards/rejected": -7.40625,
      "step": 140
    },
    {
      "epoch": 0.3138075313807531,
      "grad_norm": 23.715990612918446,
      "learning_rate": 4.337355301007335e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.203125,
      "logps/chosen": -880.0,
      "logps/rejected": -1424.0,
      "loss": 0.1569,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -3.84375,
      "rewards/margins": 5.28125,
      "rewards/rejected": -9.125,
      "step": 150
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 15.385270543077525,
      "learning_rate": 4.2087030056579986e-07,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.109375,
      "logps/chosen": -820.0,
      "logps/rejected": -1376.0,
      "loss": 0.147,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -3.578125,
      "rewards/margins": 5.40625,
      "rewards/rejected": -9.0,
      "step": 160
    },
    {
      "epoch": 0.35564853556485354,
      "grad_norm": 22.78878362130454,
      "learning_rate": 4.070934040463998e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -3.03125,
      "logps/chosen": -816.0,
      "logps/rejected": -1288.0,
      "loss": 0.1657,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -3.34375,
      "rewards/margins": 4.625,
      "rewards/rejected": -7.96875,
      "step": 170
    },
    {
      "epoch": 0.37656903765690375,
      "grad_norm": 24.865773370881126,
      "learning_rate": 3.9247834624635404e-07,
      "logits/chosen": -3.125,
      "logits/rejected": -2.984375,
      "logps/chosen": -964.0,
      "logps/rejected": -1560.0,
      "loss": 0.1586,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -4.4375,
      "rewards/margins": 5.90625,
      "rewards/rejected": -10.375,
      "step": 180
    },
    {
      "epoch": 0.39748953974895396,
      "grad_norm": 26.176777590143203,
      "learning_rate": 3.7710310482256523e-07,
      "logits/chosen": -3.15625,
      "logits/rejected": -3.0,
      "logps/chosen": -872.0,
      "logps/rejected": -1512.0,
      "loss": 0.1512,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -4.0,
      "rewards/margins": 6.15625,
      "rewards/rejected": -10.1875,
      "step": 190
    },
    {
      "epoch": 0.41841004184100417,
      "grad_norm": 15.82928504976872,
      "learning_rate": 3.610497133404795e-07,
      "logits/chosen": -3.046875,
      "logits/rejected": -2.90625,
      "logps/chosen": -896.0,
      "logps/rejected": -1560.0,
      "loss": 0.1326,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -3.859375,
      "rewards/margins": 6.375,
      "rewards/rejected": -10.25,
      "step": 200
    },
    {
      "epoch": 0.41841004184100417,
      "eval_logits/chosen": -3.15625,
      "eval_logits/rejected": -3.046875,
      "eval_logps/chosen": -804.0,
      "eval_logps/rejected": -1304.0,
      "eval_loss": 0.1445154994726181,
      "eval_rewards/accuracies": 0.87890625,
      "eval_rewards/chosen": -3.0625,
      "eval_rewards/margins": 4.90625,
      "eval_rewards/rejected": -7.96875,
      "eval_runtime": 90.1448,
      "eval_samples_per_second": 22.187,
      "eval_steps_per_second": 0.355,
      "step": 200
    },
    {
      "epoch": 0.4393305439330544,
      "grad_norm": 26.675093840651307,
      "learning_rate": 3.4440382358952115e-07,
      "logits/chosen": -3.046875,
      "logits/rejected": -2.875,
      "logps/chosen": -824.0,
      "logps/rejected": -1432.0,
      "loss": 0.1536,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -3.515625,
      "rewards/margins": 6.03125,
      "rewards/rejected": -9.5625,
      "step": 210
    },
    {
      "epoch": 0.4602510460251046,
      "grad_norm": 23.77478679546016,
      "learning_rate": 3.272542485937368e-07,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.875,
      "logps/chosen": -988.0,
      "logps/rejected": -1696.0,
      "loss": 0.1417,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -4.90625,
      "rewards/margins": 6.9375,
      "rewards/rejected": -11.8125,
      "step": 220
    },
    {
      "epoch": 0.4811715481171548,
      "grad_norm": 30.92752961992483,
      "learning_rate": 3.096924887558854e-07,
      "logits/chosen": -3.21875,
      "logits/rejected": -3.0,
      "logps/chosen": -928.0,
      "logps/rejected": -1680.0,
      "loss": 0.1465,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -4.0625,
      "rewards/margins": 7.34375,
      "rewards/rejected": -11.375,
      "step": 230
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 23.019612412596924,
      "learning_rate": 2.9181224366319943e-07,
      "logits/chosen": -3.09375,
      "logits/rejected": -2.90625,
      "logps/chosen": -872.0,
      "logps/rejected": -1584.0,
      "loss": 0.138,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -3.890625,
      "rewards/margins": 6.96875,
      "rewards/rejected": -10.875,
      "step": 240
    },
    {
      "epoch": 0.5230125523012552,
      "grad_norm": 29.646296726971823,
      "learning_rate": 2.7370891215954565e-07,
      "logits/chosen": -3.078125,
      "logits/rejected": -2.859375,
      "logps/chosen": -900.0,
      "logps/rejected": -1680.0,
      "loss": 0.1159,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -4.0625,
      "rewards/margins": 7.8125,
      "rewards/rejected": -11.875,
      "step": 250
    },
    {
      "epoch": 0.5439330543933054,
      "grad_norm": 11.379749179221394,
      "learning_rate": 2.55479083351317e-07,
      "logits/chosen": -3.09375,
      "logits/rejected": -2.890625,
      "logps/chosen": -884.0,
      "logps/rejected": -1696.0,
      "loss": 0.123,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -3.90625,
      "rewards/margins": 7.96875,
      "rewards/rejected": -11.875,
      "step": 260
    },
    {
      "epoch": 0.5648535564853556,
      "grad_norm": 21.73488710471591,
      "learning_rate": 2.3722002126275822e-07,
      "logits/chosen": -3.015625,
      "logits/rejected": -2.8125,
      "logps/chosen": -864.0,
      "logps/rejected": -1608.0,
      "loss": 0.1213,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -3.875,
      "rewards/margins": 7.375,
      "rewards/rejected": -11.25,
      "step": 270
    },
    {
      "epoch": 0.5857740585774058,
      "grad_norm": 15.391997661811661,
      "learning_rate": 2.19029145890313e-07,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.828125,
      "logps/chosen": -1020.0,
      "logps/rejected": -1824.0,
      "loss": 0.1292,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -5.0625,
      "rewards/margins": 7.90625,
      "rewards/rejected": -13.0,
      "step": 280
    },
    {
      "epoch": 0.606694560669456,
      "grad_norm": 25.666971311078395,
      "learning_rate": 2.0100351342479216e-07,
      "logits/chosen": -3.0,
      "logits/rejected": -2.796875,
      "logps/chosen": -932.0,
      "logps/rejected": -1760.0,
      "loss": 0.1308,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -4.375,
      "rewards/margins": 8.25,
      "rewards/rejected": -12.625,
      "step": 290
    },
    {
      "epoch": 0.6276150627615062,
      "grad_norm": 35.96133294230227,
      "learning_rate": 1.8323929841460178e-07,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.75,
      "logps/chosen": -952.0,
      "logps/rejected": -1840.0,
      "loss": 0.1085,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -4.6875,
      "rewards/margins": 8.8125,
      "rewards/rejected": -13.5,
      "step": 300
    },
    {
      "epoch": 0.6276150627615062,
      "eval_logits/chosen": -3.03125,
      "eval_logits/rejected": -2.828125,
      "eval_logps/chosen": -1032.0,
      "eval_logps/rejected": -1872.0,
      "eval_loss": 0.12868428230285645,
      "eval_rewards/accuracies": 0.890625,
      "eval_rewards/chosen": -5.3125,
      "eval_rewards/margins": 8.375,
      "eval_rewards/rejected": -13.6875,
      "eval_runtime": 89.953,
      "eval_samples_per_second": 22.234,
      "eval_steps_per_second": 0.356,
      "step": 300
    },
    {
      "epoch": 0.6485355648535565,
      "grad_norm": 20.42354045596541,
      "learning_rate": 1.6583128063291573e-07,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.84375,
      "logps/chosen": -944.0,
      "logps/rejected": -1800.0,
      "loss": 0.1185,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -4.5625,
      "rewards/margins": 8.375,
      "rewards/rejected": -12.9375,
      "step": 310
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 12.490212575290906,
      "learning_rate": 1.488723393865766e-07,
      "logits/chosen": -3.0625,
      "logits/rejected": -2.84375,
      "logps/chosen": -1032.0,
      "logps/rejected": -1840.0,
      "loss": 0.1061,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -4.9375,
      "rewards/margins": 8.125,
      "rewards/rejected": -13.0625,
      "step": 320
    },
    {
      "epoch": 0.6903765690376569,
      "grad_norm": 25.455929611409868,
      "learning_rate": 1.3245295796480788e-07,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.828125,
      "logps/chosen": -952.0,
      "logps/rejected": -1800.0,
      "loss": 0.1249,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -4.5,
      "rewards/margins": 8.3125,
      "rewards/rejected": -12.75,
      "step": 330
    },
    {
      "epoch": 0.7112970711297071,
      "grad_norm": 19.337569251631464,
      "learning_rate": 1.1666074087171627e-07,
      "logits/chosen": -3.046875,
      "logits/rejected": -2.828125,
      "logps/chosen": -912.0,
      "logps/rejected": -1744.0,
      "loss": 0.1341,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": -4.0625,
      "rewards/margins": 8.3125,
      "rewards/rejected": -12.375,
      "step": 340
    },
    {
      "epoch": 0.7322175732217573,
      "grad_norm": 26.760973150566464,
      "learning_rate": 1.0157994641835734e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.96875,
      "logps/chosen": -856.0,
      "logps/rejected": -1664.0,
      "loss": 0.1205,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -3.875,
      "rewards/margins": 7.875,
      "rewards/rejected": -11.75,
      "step": 350
    },
    {
      "epoch": 0.7531380753138075,
      "grad_norm": 26.71863682370444,
      "learning_rate": 8.729103716819111e-08,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.859375,
      "logps/chosen": -888.0,
      "logps/rejected": -1632.0,
      "loss": 0.1218,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -3.765625,
      "rewards/margins": 7.375,
      "rewards/rejected": -11.1875,
      "step": 360
    },
    {
      "epoch": 0.7740585774058577,
      "grad_norm": 17.339563343861006,
      "learning_rate": 7.387025063449081e-08,
      "logits/chosen": -3.078125,
      "logits/rejected": -2.875,
      "logps/chosen": -948.0,
      "logps/rejected": -1720.0,
      "loss": 0.1116,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -4.84375,
      "rewards/margins": 7.625,
      "rewards/rejected": -12.5,
      "step": 370
    },
    {
      "epoch": 0.7949790794979079,
      "grad_norm": 37.52855243691648,
      "learning_rate": 6.138919252022435e-08,
      "logits/chosen": -3.109375,
      "logits/rejected": -2.875,
      "logps/chosen": -920.0,
      "logps/rejected": -1824.0,
      "loss": 0.1116,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -4.21875,
      "rewards/margins": 9.0,
      "rewards/rejected": -13.1875,
      "step": 380
    },
    {
      "epoch": 0.8158995815899581,
      "grad_norm": 31.510469937073836,
      "learning_rate": 4.991445467064689e-08,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.90625,
      "logps/chosen": -952.0,
      "logps/rejected": -1888.0,
      "loss": 0.1118,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -4.71875,
      "rewards/margins": 9.1875,
      "rewards/rejected": -13.9375,
      "step": 390
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 20.130351984433958,
      "learning_rate": 3.9507259776993954e-08,
      "logits/chosen": -3.015625,
      "logits/rejected": -2.765625,
      "logps/chosen": -1012.0,
      "logps/rejected": -1864.0,
      "loss": 0.1209,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -5.375,
      "rewards/margins": 8.375,
      "rewards/rejected": -13.75,
      "step": 400
    },
    {
      "epoch": 0.8368200836820083,
      "eval_logits/chosen": -3.0625,
      "eval_logits/rejected": -2.828125,
      "eval_logps/chosen": -1040.0,
      "eval_logps/rejected": -1944.0,
      "eval_loss": 0.11481054872274399,
      "eval_rewards/accuracies": 0.89453125,
      "eval_rewards/chosen": -5.375,
      "eval_rewards/margins": 9.0,
      "eval_rewards/rejected": -14.375,
      "eval_runtime": 89.9041,
      "eval_samples_per_second": 22.246,
      "eval_steps_per_second": 0.356,
      "step": 400
    },
    {
      "epoch": 0.8577405857740585,
      "grad_norm": 28.288073652571832,
      "learning_rate": 3.022313472693447e-08,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.90625,
      "logps/chosen": -1004.0,
      "logps/rejected": -1856.0,
      "loss": 0.1248,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -5.15625,
      "rewards/margins": 8.4375,
      "rewards/rejected": -13.5625,
      "step": 410
    },
    {
      "epoch": 0.8786610878661087,
      "grad_norm": 22.266390291165507,
      "learning_rate": 2.2111614344599684e-08,
      "logits/chosen": -3.015625,
      "logits/rejected": -2.765625,
      "logps/chosen": -1004.0,
      "logps/rejected": -1968.0,
      "loss": 0.1045,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -5.15625,
      "rewards/margins": 9.4375,
      "rewards/rejected": -14.5625,
      "step": 420
    },
    {
      "epoch": 0.899581589958159,
      "grad_norm": 19.883477029807672,
      "learning_rate": 1.521597710086439e-08,
      "logits/chosen": -3.125,
      "logits/rejected": -2.875,
      "logps/chosen": -1008.0,
      "logps/rejected": -1976.0,
      "loss": 0.1156,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -5.15625,
      "rewards/margins": 9.4375,
      "rewards/rejected": -14.625,
      "step": 430
    },
    {
      "epoch": 0.9205020920502092,
      "grad_norm": 19.251903060475673,
      "learning_rate": 9.57301420397924e-09,
      "logits/chosen": -3.109375,
      "logits/rejected": -2.859375,
      "logps/chosen": -992.0,
      "logps/rejected": -1960.0,
      "loss": 0.1283,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -4.9375,
      "rewards/margins": 9.625,
      "rewards/rejected": -14.5625,
      "step": 440
    },
    {
      "epoch": 0.9414225941422594,
      "grad_norm": 25.369937691827406,
      "learning_rate": 5.212833302556258e-09,
      "logits/chosen": -3.03125,
      "logits/rejected": -2.796875,
      "logps/chosen": -940.0,
      "logps/rejected": -1912.0,
      "loss": 0.1019,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": -4.6875,
      "rewards/margins": 9.4375,
      "rewards/rejected": -14.125,
      "step": 450
    },
    {
      "epoch": 0.9623430962343096,
      "grad_norm": 21.145873817748413,
      "learning_rate": 2.158697848236607e-09,
      "logits/chosen": -3.046875,
      "logits/rejected": -2.8125,
      "logps/chosen": -1020.0,
      "logps/rejected": -1960.0,
      "loss": 0.1128,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -4.90625,
      "rewards/margins": 9.3125,
      "rewards/rejected": -14.25,
      "step": 460
    },
    {
      "epoch": 0.9832635983263598,
      "grad_norm": 27.719184177090515,
      "learning_rate": 4.269029751107489e-10,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.875,
      "logps/chosen": -900.0,
      "logps/rejected": -1872.0,
      "loss": 0.1302,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -4.40625,
      "rewards/margins": 9.5625,
      "rewards/rejected": -13.9375,
      "step": 470
    },
    {
      "epoch": 1.0,
      "step": 478,
      "total_flos": 0.0,
      "train_loss": 0.19883761919691972,
      "train_runtime": 7237.2429,
      "train_samples_per_second": 8.447,
      "train_steps_per_second": 0.066
    }
  ],
  "logging_steps": 10,
  "max_steps": 478,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "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
}