{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.29832401871681213,
      "learning_rate": 3.461538461538462e-05,
      "logits/chosen": -1.1990242004394531,
      "logits/rejected": -1.1565947532653809,
      "logps/chosen": -34.47224044799805,
      "logps/rejected": -57.595638275146484,
      "loss": 0.4812039852142334,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -1.6355812549591064,
      "rewards/margins": 1.8687772750854492,
      "rewards/rejected": -3.5043587684631348,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.31452351808547974,
      "learning_rate": 4.967625656594782e-05,
      "logits/chosen": -1.1797922849655151,
      "logits/rejected": -1.1334644556045532,
      "logps/chosen": -30.18818473815918,
      "logps/rejected": -54.1025505065918,
      "loss": 0.38157622814178466,
      "rewards/accuracies": 0.8062500357627869,
      "rewards/chosen": -1.2342875003814697,
      "rewards/margins": 1.9759091138839722,
      "rewards/rejected": -3.2101962566375732,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.2767532467842102,
      "learning_rate": 4.77281074950681e-05,
      "logits/chosen": -1.218713641166687,
      "logits/rejected": -1.1803112030029297,
      "logps/chosen": -31.594274520874023,
      "logps/rejected": -58.6033935546875,
      "loss": 0.31204392910003664,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -1.2332490682601929,
      "rewards/margins": 2.2641992568969727,
      "rewards/rejected": -3.497448444366455,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.3287574052810669,
      "learning_rate": 4.415111107797445e-05,
      "logits/chosen": -1.3382649421691895,
      "logits/rejected": -1.30353581905365,
      "logps/chosen": -36.43912887573242,
      "logps/rejected": -70.29068756103516,
      "loss": 0.26097993850708007,
      "rewards/accuracies": 0.8687500357627869,
      "rewards/chosen": -1.6973193883895874,
      "rewards/margins": 3.017174005508423,
      "rewards/rejected": -4.714493751525879,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.3445669710636139,
      "learning_rate": 3.920161866827889e-05,
      "logits/chosen": -1.4549230337142944,
      "logits/rejected": -1.4277979135513306,
      "logps/chosen": -41.87653732299805,
      "logps/rejected": -82.90950012207031,
      "loss": 0.2709719896316528,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -2.2981889247894287,
      "rewards/margins": 3.5901012420654297,
      "rewards/rejected": -5.8882904052734375,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.4734262526035309,
      "learning_rate": 3.323434381080199e-05,
      "logits/chosen": -1.4583113193511963,
      "logits/rejected": -1.4419019222259521,
      "logps/chosen": -46.35929489135742,
      "logps/rejected": -82.18633270263672,
      "loss": 0.4293703079223633,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -2.7741143703460693,
      "rewards/margins": 3.1638267040252686,
      "rewards/rejected": -5.937941074371338,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.14828024804592133,
      "learning_rate": 2.667694110977506e-05,
      "logits/chosen": -1.3716756105422974,
      "logits/rejected": -1.3496290445327759,
      "logps/chosen": -37.190982818603516,
      "logps/rejected": -80.84061431884766,
      "loss": 0.22459855079650878,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -1.9332157373428345,
      "rewards/margins": 3.8292477130889893,
      "rewards/rejected": -5.762463569641113,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.1763170063495636,
      "learning_rate": 1.9999357655598893e-05,
      "logits/chosen": -1.355380892753601,
      "logits/rejected": -1.3304802179336548,
      "logps/chosen": -39.43621826171875,
      "logps/rejected": -82.64469909667969,
      "loss": 0.10246082544326782,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -2.0429646968841553,
      "rewards/margins": 3.7551822662353516,
      "rewards/rejected": -5.798146724700928,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.08582232892513275,
      "learning_rate": 1.3680153490759073e-05,
      "logits/chosen": -1.3984593152999878,
      "logits/rejected": -1.3604938983917236,
      "logps/chosen": -35.92609405517578,
      "logps/rejected": -94.2962875366211,
      "loss": 0.07779874801635742,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -1.6332558393478394,
      "rewards/margins": 5.143055438995361,
      "rewards/rejected": -6.776310920715332,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.11894813179969788,
      "learning_rate": 8.172204814310742e-06,
      "logits/chosen": -1.4377853870391846,
      "logits/rejected": -1.4119752645492554,
      "logps/chosen": -33.87895965576172,
      "logps/rejected": -80.11370086669922,
      "loss": 0.09446020126342773,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -1.5904297828674316,
      "rewards/margins": 4.260800838470459,
      "rewards/rejected": -5.851230144500732,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.0858096107840538,
      "learning_rate": 3.8702478614051355e-06,
      "logits/chosen": -1.4167169332504272,
      "logits/rejected": -1.3845655918121338,
      "logps/chosen": -38.7296257019043,
      "logps/rejected": -86.01542663574219,
      "loss": 0.08398746848106384,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -2.0574593544006348,
      "rewards/margins": 4.346583843231201,
      "rewards/rejected": -6.404043674468994,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.08499918133020401,
      "learning_rate": 1.0825894793143721e-06,
      "logits/chosen": -1.4523096084594727,
      "logits/rejected": -1.4183976650238037,
      "logps/chosen": -36.14750289916992,
      "logps/rejected": -87.8320083618164,
      "loss": 0.06980652809143066,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.8463581800460815,
      "rewards/margins": 4.731562614440918,
      "rewards/rejected": -6.577920436859131,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.13648074865341187,
      "learning_rate": 9.011808227865625e-09,
      "logits/chosen": -1.44231116771698,
      "logits/rejected": -1.4165829420089722,
      "logps/chosen": -36.7476921081543,
      "logps/rejected": -86.03533172607422,
      "loss": 0.081082022190094,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -1.7559616565704346,
      "rewards/margins": 4.578075885772705,
      "rewards/rejected": -6.334037780761719,
      "step": 130
    },
    {
      "epoch": 2.0,
      "step": 130,
      "total_flos": 1.008142593959854e+17,
      "train_loss": 0.22079544021533085,
      "train_runtime": 3380.4206,
      "train_samples_per_second": 0.615,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 10,
  "max_steps": 130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1.008142593959854e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}