{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.587443946188341,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35874439461883406,
      "grad_norm": 0.5037005543708801,
      "learning_rate": 9.259259259259259e-07,
      "logits/chosen": 1.851180076599121,
      "logits/rejected": 1.8187497854232788,
      "logps/chosen": -71.38922119140625,
      "logps/rejected": -77.1716537475586,
      "loss": 0.6947,
      "rewards/accuracies": 0.36250001192092896,
      "rewards/chosen": -0.0033944465685635805,
      "rewards/margins": -0.004582419525831938,
      "rewards/rejected": 0.0011879729572683573,
      "step": 10
    },
    {
      "epoch": 0.7174887892376681,
      "grad_norm": 0.4899679124355316,
      "learning_rate": 1.8518518518518519e-06,
      "logits/chosen": 1.804616928100586,
      "logits/rejected": 1.8522075414657593,
      "logps/chosen": -68.71460723876953,
      "logps/rejected": -92.10017395019531,
      "loss": 0.6935,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.0006334683275781572,
      "rewards/margins": 0.002181120216846466,
      "rewards/rejected": -0.0015476513653993607,
      "step": 20
    },
    {
      "epoch": 1.0762331838565022,
      "grad_norm": 0.4610757529735565,
      "learning_rate": 2.7777777777777783e-06,
      "logits/chosen": 1.7368440628051758,
      "logits/rejected": 1.747293472290039,
      "logps/chosen": -86.10456848144531,
      "logps/rejected": -94.18408203125,
      "loss": 0.6923,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.004750942811369896,
      "rewards/margins": -3.7298166716936976e-05,
      "rewards/rejected": -0.0047136456705629826,
      "step": 30
    },
    {
      "epoch": 1.4349775784753362,
      "grad_norm": 0.5546987056732178,
      "learning_rate": 3.7037037037037037e-06,
      "logits/chosen": 1.760947585105896,
      "logits/rejected": 1.7774131298065186,
      "logps/chosen": -72.0299301147461,
      "logps/rejected": -75.53373718261719,
      "loss": 0.6938,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": 0.004539580084383488,
      "rewards/margins": 0.004153046756982803,
      "rewards/rejected": 0.0003865335020236671,
      "step": 40
    },
    {
      "epoch": 1.7937219730941703,
      "grad_norm": 0.5019991397857666,
      "learning_rate": 4.62962962962963e-06,
      "logits/chosen": 1.6594927310943604,
      "logits/rejected": 1.7026996612548828,
      "logps/chosen": -74.72074890136719,
      "logps/rejected": -80.34657287597656,
      "loss": 0.694,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": 0.0014741375343874097,
      "rewards/margins": 0.00692129647359252,
      "rewards/rejected": -0.00544715765863657,
      "step": 50
    },
    {
      "epoch": 2.1524663677130045,
      "grad_norm": 0.5497386455535889,
      "learning_rate": 4.998119881260576e-06,
      "logits/chosen": 1.845393180847168,
      "logits/rejected": 1.8288724422454834,
      "logps/chosen": -85.8925552368164,
      "logps/rejected": -79.84625244140625,
      "loss": 0.6939,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 0.0007343249162659049,
      "rewards/margins": 0.0027797420043498278,
      "rewards/rejected": -0.0020454167388379574,
      "step": 60
    },
    {
      "epoch": 2.5112107623318387,
      "grad_norm": 0.53052818775177,
      "learning_rate": 4.9866405060165044e-06,
      "logits/chosen": 1.7219512462615967,
      "logits/rejected": 1.739393949508667,
      "logps/chosen": -82.69403076171875,
      "logps/rejected": -85.24137878417969,
      "loss": 0.6912,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": 0.004084458574652672,
      "rewards/margins": 8.91583040356636e-05,
      "rewards/rejected": 0.003995300270617008,
      "step": 70
    },
    {
      "epoch": 2.8699551569506725,
      "grad_norm": 0.5697340965270996,
      "learning_rate": 4.964774158361991e-06,
      "logits/chosen": 1.688643217086792,
      "logits/rejected": 1.7375280857086182,
      "logps/chosen": -77.02912902832031,
      "logps/rejected": -84.49137115478516,
      "loss": 0.6923,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.0026642463635653257,
      "rewards/margins": -0.0012511021923273802,
      "rewards/rejected": -0.0014131448697298765,
      "step": 80
    },
    {
      "epoch": 3.2286995515695067,
      "grad_norm": 0.6337418556213379,
      "learning_rate": 4.93261217644956e-06,
      "logits/chosen": 1.7215973138809204,
      "logits/rejected": 1.6856582164764404,
      "logps/chosen": -81.94705963134766,
      "logps/rejected": -79.82647705078125,
      "loss": 0.6916,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": 0.004837460815906525,
      "rewards/margins": 0.0014790964778512716,
      "rewards/rejected": 0.003358363639563322,
      "step": 90
    },
    {
      "epoch": 3.587443946188341,
      "grad_norm": 0.6216289401054382,
      "learning_rate": 4.8902889044347e-06,
      "logits/chosen": 1.8092563152313232,
      "logits/rejected": 1.7624406814575195,
      "logps/chosen": -91.27610778808594,
      "logps/rejected": -76.58167266845703,
      "loss": 0.6879,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.004790940321981907,
      "rewards/margins": 0.018742190673947334,
      "rewards/rejected": -0.013951249420642853,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.383770313203384e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}