{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 46,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10869565217391304,
      "grad_norm": 145.4520263671875,
      "learning_rate": 4.000000000000001e-06,
      "logits/chosen": -7.522208213806152,
      "logits/rejected": -7.187244415283203,
      "logps/chosen": -42.41267776489258,
      "logps/rejected": -228.89913940429688,
      "loss": 5.0182,
      "rewards/accuracies": 0.10000000149011612,
      "rewards/chosen": 3.0174379348754883,
      "rewards/margins": -4.805842399597168,
      "rewards/rejected": 7.823281288146973,
      "step": 5
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 67.39005279541016,
      "learning_rate": 4.883490980137327e-06,
      "logits/chosen": -7.832457542419434,
      "logits/rejected": -7.450900077819824,
      "logps/chosen": -37.844947814941406,
      "logps/rejected": -241.001220703125,
      "loss": 2.3347,
      "rewards/accuracies": 0.36250001192092896,
      "rewards/chosen": 2.7626261711120605,
      "rewards/margins": -1.523572564125061,
      "rewards/rejected": 4.28619909286499,
      "step": 10
    },
    {
      "epoch": 0.32608695652173914,
      "grad_norm": 11.523139953613281,
      "learning_rate": 4.428722949554858e-06,
      "logits/chosen": -8.29098892211914,
      "logits/rejected": -7.737236022949219,
      "logps/chosen": -36.03478240966797,
      "logps/rejected": -299.16802978515625,
      "loss": 0.3164,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 2.3383092880249023,
      "rewards/margins": 3.7309823036193848,
      "rewards/rejected": -1.3926727771759033,
      "step": 15
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 1.9360767602920532,
      "learning_rate": 3.6942995462806574e-06,
      "logits/chosen": -8.896014213562012,
      "logits/rejected": -8.395783424377441,
      "logps/chosen": -44.91072463989258,
      "logps/rejected": -343.45086669921875,
      "loss": 0.0385,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 1.546364426612854,
      "rewards/margins": 9.299250602722168,
      "rewards/rejected": -7.7528862953186035,
      "step": 20
    },
    {
      "epoch": 0.5434782608695652,
      "grad_norm": 0.7242340445518494,
      "learning_rate": 2.786708563496002e-06,
      "logits/chosen": -9.124911308288574,
      "logits/rejected": -8.544282913208008,
      "logps/chosen": -53.40107345581055,
      "logps/rejected": -400.3040466308594,
      "loss": 0.0092,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 1.0653327703475952,
      "rewards/margins": 12.580676078796387,
      "rewards/rejected": -11.515342712402344,
      "step": 25
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.914761483669281,
      "learning_rate": 1.8375462445083464e-06,
      "logits/chosen": -9.442420959472656,
      "logits/rejected": -8.751354217529297,
      "logps/chosen": -58.87113571166992,
      "logps/rejected": -390.2606506347656,
      "loss": 0.035,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.6872259378433228,
      "rewards/margins": 14.051335334777832,
      "rewards/rejected": -13.364108085632324,
      "step": 30
    },
    {
      "epoch": 0.7608695652173914,
      "grad_norm": 2.3408379554748535,
      "learning_rate": 9.844364725834058e-07,
      "logits/chosen": -9.394502639770508,
      "logits/rejected": -8.917421340942383,
      "logps/chosen": -61.983360290527344,
      "logps/rejected": -399.78619384765625,
      "loss": 0.0784,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.25255903601646423,
      "rewards/margins": 15.496400833129883,
      "rewards/rejected": -15.243843078613281,
      "step": 35
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.053074195981025696,
      "learning_rate": 3.510759825319976e-07,
      "logits/chosen": -9.617226600646973,
      "logits/rejected": -8.915563583374023,
      "logps/chosen": -60.818382263183594,
      "logps/rejected": -431.60931396484375,
      "loss": 0.061,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.3885515630245209,
      "rewards/margins": 15.846685409545898,
      "rewards/rejected": -15.45813274383545,
      "step": 40
    },
    {
      "epoch": 0.9782608695652174,
      "grad_norm": 2.3728623390197754,
      "learning_rate": 2.9298940549128962e-08,
      "logits/chosen": -9.523667335510254,
      "logits/rejected": -8.924112319946289,
      "logps/chosen": -63.81079864501953,
      "logps/rejected": -414.37872314453125,
      "loss": 0.0305,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.3749346137046814,
      "rewards/margins": 15.598576545715332,
      "rewards/rejected": -15.223642349243164,
      "step": 45
    }
  ],
  "logging_steps": 5,
  "max_steps": 46,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}