{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.064,
      "grad_norm": 46.88199996948242,
      "learning_rate": 6.249999999999999e-07,
      "logits/chosen": -1.8514013290405273,
      "logits/rejected": -0.29395362734794617,
      "logps/chosen": -214.1394500732422,
      "logps/rejected": -737.39794921875,
      "loss": 0.722,
      "rewards/accuracies": 0.4906249940395355,
      "rewards/chosen": 0.0037395646795630455,
      "rewards/margins": 0.025518950074911118,
      "rewards/rejected": -0.021779386326670647,
      "step": 10
    },
    {
      "epoch": 0.128,
      "grad_norm": 13.165843963623047,
      "learning_rate": 9.979871469976195e-07,
      "logits/chosen": -1.8478971719741821,
      "logits/rejected": -0.2753606140613556,
      "logps/chosen": -240.26084899902344,
      "logps/rejected": -844.6398315429688,
      "loss": 0.5035,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 0.035897765308618546,
      "rewards/margins": 0.6878108978271484,
      "rewards/rejected": -0.6519131660461426,
      "step": 20
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.7543407082557678,
      "learning_rate": 9.755282581475767e-07,
      "logits/chosen": -2.1524031162261963,
      "logits/rejected": -0.8206841945648193,
      "logps/chosen": -241.90380859375,
      "logps/rejected": -830.2996215820312,
      "loss": 0.1368,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -0.3361475467681885,
      "rewards/margins": 4.685339450836182,
      "rewards/rejected": -5.021487236022949,
      "step": 30
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.9449922442436218,
      "learning_rate": 9.29224396800933e-07,
      "logits/chosen": -2.6018669605255127,
      "logits/rejected": -1.616600513458252,
      "logps/chosen": -260.47198486328125,
      "logps/rejected": -947.6165161132812,
      "loss": 0.0662,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.462709665298462,
      "rewards/margins": 14.078239440917969,
      "rewards/rejected": -16.54094886779785,
      "step": 40
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.21557988226413727,
      "learning_rate": 8.613974319136957e-07,
      "logits/chosen": -2.7778477668762207,
      "logits/rejected": -1.9211829900741577,
      "logps/chosen": -257.9867248535156,
      "logps/rejected": -1050.994384765625,
      "loss": 0.0562,
      "rewards/accuracies": 0.9906250238418579,
      "rewards/chosen": -3.9251632690429688,
      "rewards/margins": 22.336076736450195,
      "rewards/rejected": -26.26123809814453,
      "step": 50
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.16911636292934418,
      "learning_rate": 7.754484907260512e-07,
      "logits/chosen": -2.7103047370910645,
      "logits/rejected": -1.904673457145691,
      "logps/chosen": -272.1762390136719,
      "logps/rejected": -1050.43212890625,
      "loss": 0.051,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.340147018432617,
      "rewards/margins": 22.80539321899414,
      "rewards/rejected": -27.14554214477539,
      "step": 60
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.3055303394794464,
      "learning_rate": 6.756874120406714e-07,
      "logits/chosen": -2.6818180084228516,
      "logits/rejected": -1.7974838018417358,
      "logps/chosen": -252.34637451171875,
      "logps/rejected": -1027.539306640625,
      "loss": 0.0596,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -3.2451460361480713,
      "rewards/margins": 21.46558952331543,
      "rewards/rejected": -24.710735321044922,
      "step": 70
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.15121270716190338,
      "learning_rate": 5.671166329088277e-07,
      "logits/chosen": -2.543553590774536,
      "logits/rejected": -1.6689860820770264,
      "logps/chosen": -260.95526123046875,
      "logps/rejected": -1009.7645874023438,
      "loss": 0.0476,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.7273285388946533,
      "rewards/margins": 19.519502639770508,
      "rewards/rejected": -22.246829986572266,
      "step": 80
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.17506758868694305,
      "learning_rate": 4.5518034554828327e-07,
      "logits/chosen": -2.5249202251434326,
      "logits/rejected": -1.6314860582351685,
      "logps/chosen": -258.52435302734375,
      "logps/rejected": -976.0228271484375,
      "loss": 0.0575,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -2.350813150405884,
      "rewards/margins": 18.23154067993164,
      "rewards/rejected": -20.582355499267578,
      "step": 90
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.20912963151931763,
      "learning_rate": 3.454915028125263e-07,
      "logits/chosen": -2.5144877433776855,
      "logits/rejected": -1.5639787912368774,
      "logps/chosen": -247.0535430908203,
      "logps/rejected": -963.437744140625,
      "loss": 0.0454,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.8558976650238037,
      "rewards/margins": 17.28997230529785,
      "rewards/rejected": -19.145870208740234,
      "step": 100
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.2550797462463379,
      "learning_rate": 2.4355036129704696e-07,
      "logits/chosen": -2.4958443641662598,
      "logits/rejected": -1.5513837337493896,
      "logps/chosen": -241.15231323242188,
      "logps/rejected": -962.02587890625,
      "loss": 0.0481,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.829073190689087,
      "rewards/margins": 16.586135864257812,
      "rewards/rejected": -18.41520881652832,
      "step": 110
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.15323293209075928,
      "learning_rate": 1.5446867550656767e-07,
      "logits/chosen": -2.4677257537841797,
      "logits/rejected": -1.5029109716415405,
      "logps/chosen": -249.34861755371094,
      "logps/rejected": -962.841552734375,
      "loss": 0.0838,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -1.7131061553955078,
      "rewards/margins": 16.56055450439453,
      "rewards/rejected": -18.27366065979004,
      "step": 120
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.14431753754615784,
      "learning_rate": 8.271337313934867e-08,
      "logits/chosen": -2.411092519760132,
      "logits/rejected": -1.4570066928863525,
      "logps/chosen": -262.32379150390625,
      "logps/rejected": -950.74560546875,
      "loss": 0.0652,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -1.68479585647583,
      "rewards/margins": 15.87053394317627,
      "rewards/rejected": -17.555328369140625,
      "step": 130
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.17921504378318787,
      "learning_rate": 3.188256468013139e-08,
      "logits/chosen": -2.4623465538024902,
      "logits/rejected": -1.4546464681625366,
      "logps/chosen": -243.91656494140625,
      "logps/rejected": -970.7305297851562,
      "loss": 0.0472,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -1.3906798362731934,
      "rewards/margins": 16.57068634033203,
      "rewards/rejected": -17.961366653442383,
      "step": 140
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.1529853641986847,
      "learning_rate": 4.5251191160326495e-09,
      "logits/chosen": -2.4238433837890625,
      "logits/rejected": -1.397405982017517,
      "logps/chosen": -264.49542236328125,
      "logps/rejected": -1022.7637329101562,
      "loss": 0.0801,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -1.6072216033935547,
      "rewards/margins": 16.616043090820312,
      "rewards/rejected": -18.2232666015625,
      "step": 150
    },
    {
      "epoch": 0.9984,
      "step": 156,
      "total_flos": 1.1115841451898962e+18,
      "train_loss": 0.1350558667610853,
      "train_runtime": 5886.4808,
      "train_samples_per_second": 0.849,
      "train_steps_per_second": 0.027
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 1.1115841451898962e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}