{
  "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": 48.569252014160156,
      "learning_rate": 6.249999999999999e-07,
      "logits/chosen": -1.8510873317718506,
      "logits/rejected": -0.29376277327537537,
      "logps/chosen": -214.10960388183594,
      "logps/rejected": -737.373291015625,
      "loss": 0.6961,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": 0.006722441408783197,
      "rewards/margins": 0.026040678843855858,
      "rewards/rejected": -0.019318239763379097,
      "step": 10
    },
    {
      "epoch": 0.128,
      "grad_norm": 14.449111938476562,
      "learning_rate": 9.979871469976195e-07,
      "logits/chosen": -1.8497416973114014,
      "logits/rejected": -0.2751065790653229,
      "logps/chosen": -240.32241821289062,
      "logps/rejected": -844.6730346679688,
      "loss": 0.4739,
      "rewards/accuracies": 0.9468749761581421,
      "rewards/chosen": 0.029739724472165108,
      "rewards/margins": 0.6849702596664429,
      "rewards/rejected": -0.6552305817604065,
      "step": 20
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.7451657056808472,
      "learning_rate": 9.755282581475767e-07,
      "logits/chosen": -2.152054786682129,
      "logits/rejected": -0.8167506456375122,
      "logps/chosen": -241.89797973632812,
      "logps/rejected": -830.0802612304688,
      "loss": 0.1085,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -0.33556511998176575,
      "rewards/margins": 4.663994312286377,
      "rewards/rejected": -4.99955940246582,
      "step": 30
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.9800211191177368,
      "learning_rate": 9.29224396800933e-07,
      "logits/chosen": -2.6119227409362793,
      "logits/rejected": -1.6330392360687256,
      "logps/chosen": -261.77215576171875,
      "logps/rejected": -951.4508666992188,
      "loss": 0.0372,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.5927298069000244,
      "rewards/margins": 14.331648826599121,
      "rewards/rejected": -16.924379348754883,
      "step": 40
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.08055471628904343,
      "learning_rate": 8.613974319136957e-07,
      "logits/chosen": -2.8284103870391846,
      "logits/rejected": -2.016091823577881,
      "logps/chosen": -267.1216735839844,
      "logps/rejected": -1083.85693359375,
      "loss": 0.0246,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -4.838659286499023,
      "rewards/margins": 24.708829879760742,
      "rewards/rejected": -29.547487258911133,
      "step": 50
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.07454250752925873,
      "learning_rate": 7.754484907260512e-07,
      "logits/chosen": -2.8392865657806396,
      "logits/rejected": -2.14508056640625,
      "logps/chosen": -296.3800964355469,
      "logps/rejected": -1137.64892578125,
      "loss": 0.02,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -6.760532379150391,
      "rewards/margins": 29.1066951751709,
      "rewards/rejected": -35.86723327636719,
      "step": 60
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.10226113349199295,
      "learning_rate": 6.756874120406714e-07,
      "logits/chosen": -2.863455295562744,
      "logits/rejected": -2.1374523639678955,
      "logps/chosen": -282.3721008300781,
      "logps/rejected": -1143.27587890625,
      "loss": 0.0304,
      "rewards/accuracies": 0.9874999523162842,
      "rewards/chosen": -6.247722148895264,
      "rewards/margins": 30.036664962768555,
      "rewards/rejected": -36.28438949584961,
      "step": 70
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.06721244752407074,
      "learning_rate": 5.671166329088277e-07,
      "logits/chosen": -2.756829023361206,
      "logits/rejected": -2.045252799987793,
      "logps/chosen": -291.5985412597656,
      "logps/rejected": -1122.9361572265625,
      "loss": 0.019,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -5.791654586791992,
      "rewards/margins": 27.7723331451416,
      "rewards/rejected": -33.563987731933594,
      "step": 80
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.07295263558626175,
      "learning_rate": 4.5518034554828327e-07,
      "logits/chosen": -2.751217842102051,
      "logits/rejected": -2.025172233581543,
      "logps/chosen": -286.55694580078125,
      "logps/rejected": -1081.6651611328125,
      "loss": 0.0353,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -5.1540751457214355,
      "rewards/margins": 25.992502212524414,
      "rewards/rejected": -31.146577835083008,
      "step": 90
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.08157353848218918,
      "learning_rate": 3.454915028125263e-07,
      "logits/chosen": -2.746480703353882,
      "logits/rejected": -1.9662470817565918,
      "logps/chosen": -271.390869140625,
      "logps/rejected": -1062.62109375,
      "loss": 0.0169,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -4.289627552032471,
      "rewards/margins": 24.77456283569336,
      "rewards/rejected": -29.06418800354004,
      "step": 100
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.1165793165564537,
      "learning_rate": 2.4355036129704696e-07,
      "logits/chosen": -2.7295165061950684,
      "logits/rejected": -1.951841950416565,
      "logps/chosen": -265.2632141113281,
      "logps/rejected": -1057.536376953125,
      "loss": 0.0252,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -4.240163803100586,
      "rewards/margins": 23.72610092163086,
      "rewards/rejected": -27.966266632080078,
      "step": 110
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.06965842097997665,
      "learning_rate": 1.5446867550656767e-07,
      "logits/chosen": -2.7134926319122314,
      "logits/rejected": -1.9151828289031982,
      "logps/chosen": -272.508056640625,
      "logps/rejected": -1058.3094482421875,
      "loss": 0.0668,
      "rewards/accuracies": 0.9906249642372131,
      "rewards/chosen": -4.029051303863525,
      "rewards/margins": 23.791399002075195,
      "rewards/rejected": -27.82044792175293,
      "step": 120
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.06551803648471832,
      "learning_rate": 8.271337313934867e-08,
      "logits/chosen": -2.6546225547790527,
      "logits/rejected": -1.8665531873703003,
      "logps/chosen": -285.0191650390625,
      "logps/rejected": -1041.3951416015625,
      "loss": 0.036,
      "rewards/accuracies": 0.9812500476837158,
      "rewards/chosen": -3.9543297290802,
      "rewards/margins": 22.66595458984375,
      "rewards/rejected": -26.620285034179688,
      "step": 130
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.07770426571369171,
      "learning_rate": 3.188256468013139e-08,
      "logits/chosen": -2.7073161602020264,
      "logits/rejected": -1.8695634603500366,
      "logps/chosen": -265.3280029296875,
      "logps/rejected": -1062.6976318359375,
      "loss": 0.0197,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -3.531825304031372,
      "rewards/margins": 23.626256942749023,
      "rewards/rejected": -27.158079147338867,
      "step": 140
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.06992675364017487,
      "learning_rate": 4.5251191160326495e-09,
      "logits/chosen": -2.6707656383514404,
      "logits/rejected": -1.817920446395874,
      "logps/chosen": -286.8282165527344,
      "logps/rejected": -1117.4290771484375,
      "loss": 0.0668,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -3.840498924255371,
      "rewards/margins": 23.849292755126953,
      "rewards/rejected": -27.68979263305664,
      "step": 150
    },
    {
      "epoch": 0.9984,
      "step": 156,
      "total_flos": 1.1115841451898962e+18,
      "train_loss": 0.10875998093531682,
      "train_runtime": 5515.1168,
      "train_samples_per_second": 0.907,
      "train_steps_per_second": 0.028
    }
  ],
  "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
}