{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0625,
      "grad_norm": 4.144060134887695,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": 3.1186840534210205,
      "logits/rejected": 3.2627792358398438,
      "logps/chosen": -54.367942810058594,
      "logps/rejected": -41.90058517456055,
      "loss": 0.6909,
      "rewards/accuracies": 0.5649999976158142,
      "rewards/chosen": 0.004131872206926346,
      "rewards/margins": 0.004790368489921093,
      "rewards/rejected": -0.0006584958755411208,
      "step": 50
    },
    {
      "epoch": 0.125,
      "grad_norm": 2.6885554790496826,
      "learning_rate": 4.861111111111111e-07,
      "logits/chosen": 3.1031572818756104,
      "logits/rejected": 3.2696139812469482,
      "logps/chosen": -54.12215805053711,
      "logps/rejected": -40.67683029174805,
      "loss": 0.6682,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 0.03350482136011124,
      "rewards/margins": 0.05115475505590439,
      "rewards/rejected": -0.01764993742108345,
      "step": 100
    },
    {
      "epoch": 0.1875,
      "grad_norm": 2.406733751296997,
      "learning_rate": 4.513888888888889e-07,
      "logits/chosen": 3.1083221435546875,
      "logits/rejected": 3.2473440170288086,
      "logps/chosen": -53.839054107666016,
      "logps/rejected": -43.20619583129883,
      "loss": 0.6225,
      "rewards/accuracies": 0.9900000095367432,
      "rewards/chosen": 0.08696311712265015,
      "rewards/margins": 0.1482781618833542,
      "rewards/rejected": -0.06131504476070404,
      "step": 150
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.196671724319458,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": 3.114607810974121,
      "logits/rejected": 3.270479440689087,
      "logps/chosen": -49.034297943115234,
      "logps/rejected": -39.99443817138672,
      "loss": 0.5765,
      "rewards/accuracies": 0.9900000095367432,
      "rewards/chosen": 0.15706641972064972,
      "rewards/margins": 0.2536722421646118,
      "rewards/rejected": -0.09660584479570389,
      "step": 200
    },
    {
      "epoch": 0.3125,
      "grad_norm": 2.1391797065734863,
      "learning_rate": 3.819444444444444e-07,
      "logits/chosen": 3.0528652667999268,
      "logits/rejected": 3.1882102489471436,
      "logps/chosen": -50.195011138916016,
      "logps/rejected": -42.24689865112305,
      "loss": 0.5097,
      "rewards/accuracies": 0.9950000047683716,
      "rewards/chosen": 0.25386032462120056,
      "rewards/margins": 0.4180786609649658,
      "rewards/rejected": -0.16421833634376526,
      "step": 250
    },
    {
      "epoch": 0.375,
      "grad_norm": 2.2767980098724365,
      "learning_rate": 3.472222222222222e-07,
      "logits/chosen": 2.9493257999420166,
      "logits/rejected": 3.099841833114624,
      "logps/chosen": -49.544803619384766,
      "logps/rejected": -41.29837417602539,
      "loss": 0.4578,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2961386442184448,
      "rewards/margins": 0.5633676648139954,
      "rewards/rejected": -0.2672290802001953,
      "step": 300
    },
    {
      "epoch": 0.4375,
      "grad_norm": 1.8134651184082031,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": 2.9306814670562744,
      "logits/rejected": 3.1076226234436035,
      "logps/chosen": -49.26047897338867,
      "logps/rejected": -43.91292953491211,
      "loss": 0.419,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3598053753376007,
      "rewards/margins": 0.677994966506958,
      "rewards/rejected": -0.3181896507740021,
      "step": 350
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.273711919784546,
      "learning_rate": 2.7777777777777776e-07,
      "logits/chosen": 2.9157378673553467,
      "logits/rejected": 3.065505266189575,
      "logps/chosen": -47.57904815673828,
      "logps/rejected": -43.553749084472656,
      "loss": 0.3937,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.33699238300323486,
      "rewards/margins": 0.7744642496109009,
      "rewards/rejected": -0.4374719560146332,
      "step": 400
    },
    {
      "epoch": 0.5625,
      "grad_norm": 1.6072301864624023,
      "learning_rate": 2.4305555555555555e-07,
      "logits/chosen": 2.9038708209991455,
      "logits/rejected": 3.0539231300354004,
      "logps/chosen": -47.44837951660156,
      "logps/rejected": -45.976158142089844,
      "loss": 0.3513,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.38920170068740845,
      "rewards/margins": 0.9139933586120605,
      "rewards/rejected": -0.5247916579246521,
      "step": 450
    },
    {
      "epoch": 0.625,
      "grad_norm": 2.385540008544922,
      "learning_rate": 2.0833333333333333e-07,
      "logits/chosen": 2.953505754470825,
      "logits/rejected": 3.1365463733673096,
      "logps/chosen": -50.15205001831055,
      "logps/rejected": -49.16044998168945,
      "loss": 0.3337,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3933267295360565,
      "rewards/margins": 0.9976506233215332,
      "rewards/rejected": -0.6043238639831543,
      "step": 500
    },
    {
      "epoch": 0.6875,
      "grad_norm": 2.3325698375701904,
      "learning_rate": 1.736111111111111e-07,
      "logits/chosen": 2.9089577198028564,
      "logits/rejected": 3.068243980407715,
      "logps/chosen": -48.45305633544922,
      "logps/rejected": -48.769771575927734,
      "loss": 0.2935,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.41794300079345703,
      "rewards/margins": 1.1487656831741333,
      "rewards/rejected": -0.7308226227760315,
      "step": 550
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.3242642879486084,
      "learning_rate": 1.3888888888888888e-07,
      "logits/chosen": 2.916025400161743,
      "logits/rejected": 3.075576066970825,
      "logps/chosen": -48.39594268798828,
      "logps/rejected": -47.85335922241211,
      "loss": 0.2915,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.4126347303390503,
      "rewards/margins": 1.1786757707595825,
      "rewards/rejected": -0.7660410404205322,
      "step": 600
    },
    {
      "epoch": 0.8125,
      "grad_norm": 1.8319804668426514,
      "learning_rate": 1.0416666666666667e-07,
      "logits/chosen": 2.8800783157348633,
      "logits/rejected": 3.0461764335632324,
      "logps/chosen": -49.678184509277344,
      "logps/rejected": -50.534385681152344,
      "loss": 0.2617,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3970913290977478,
      "rewards/margins": 1.3191609382629395,
      "rewards/rejected": -0.9220696091651917,
      "step": 650
    },
    {
      "epoch": 0.875,
      "grad_norm": 1.8796759843826294,
      "learning_rate": 6.944444444444444e-08,
      "logits/chosen": 2.8562612533569336,
      "logits/rejected": 3.0434064865112305,
      "logps/chosen": -50.938533782958984,
      "logps/rejected": -51.9145622253418,
      "loss": 0.2487,
      "rewards/accuracies": 0.9950000047683716,
      "rewards/chosen": 0.44071799516677856,
      "rewards/margins": 1.3961044549942017,
      "rewards/rejected": -0.9553865194320679,
      "step": 700
    },
    {
      "epoch": 0.9375,
      "grad_norm": 1.6690614223480225,
      "learning_rate": 3.472222222222222e-08,
      "logits/chosen": 2.904910087585449,
      "logits/rejected": 3.0448944568634033,
      "logps/chosen": -50.198997497558594,
      "logps/rejected": -52.53700637817383,
      "loss": 0.2723,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3550363779067993,
      "rewards/margins": 1.2905099391937256,
      "rewards/rejected": -0.9354735612869263,
      "step": 750
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.8712382316589355,
      "learning_rate": 0.0,
      "logits/chosen": 2.856696128845215,
      "logits/rejected": 3.0441014766693115,
      "logps/chosen": -49.916290283203125,
      "logps/rejected": -52.326080322265625,
      "loss": 0.2438,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.40912488102912903,
      "rewards/margins": 1.3978087902069092,
      "rewards/rejected": -0.9886838793754578,
      "step": 800
    }
  ],
  "logging_steps": 50,
  "max_steps": 800,
  "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": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}