{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.992,
  "eval_steps": 200,
  "global_step": 62,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016,
      "grad_norm": 22.116561494440536,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -0.9303410053253174,
      "logits/real": -0.6472625732421875,
      "logps/generated": -159.71011352539062,
      "logps/real": -190.3052978515625,
      "loss": 0.8876,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 19.166988301842625,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -0.7766941785812378,
      "logits/real": -0.7624438405036926,
      "logps/generated": -174.18157958984375,
      "logps/real": -182.86370849609375,
      "loss": 0.8573,
      "rewards/accuracies": 0.5555555820465088,
      "rewards/generated": 0.08411566913127899,
      "rewards/margins": 0.02031405083835125,
      "rewards/real": 0.1044297143816948,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 20.3933839095788,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -0.7564908862113953,
      "logits/real": -0.7665221691131592,
      "logps/generated": -163.1689910888672,
      "logps/real": -164.83465576171875,
      "loss": 0.8524,
      "rewards/accuracies": 0.512499988079071,
      "rewards/generated": 0.45903801918029785,
      "rewards/margins": -0.008176992647349834,
      "rewards/real": 0.45086097717285156,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 19.977285440131215,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -0.8268370628356934,
      "logits/real": -0.6981123685836792,
      "logps/generated": -146.0032958984375,
      "logps/real": -154.56153869628906,
      "loss": 0.8307,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/generated": 0.660775363445282,
      "rewards/margins": 0.10715095698833466,
      "rewards/real": 0.7679262757301331,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 19.443118690010486,
      "learning_rate": 2e-07,
      "logits/generated": -0.7490435242652893,
      "logits/real": -0.6841712594032288,
      "logps/generated": -165.5490264892578,
      "logps/real": -177.98483276367188,
      "loss": 0.8378,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/generated": 0.8964563608169556,
      "rewards/margins": 0.039561666548252106,
      "rewards/real": 0.9360179901123047,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 20.705818914648916,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -0.7975619435310364,
      "logits/real": -0.6225983500480652,
      "logps/generated": -163.07777404785156,
      "logps/real": -180.90386962890625,
      "loss": 0.8413,
      "rewards/accuracies": 0.625,
      "rewards/generated": 1.017493486404419,
      "rewards/margins": 0.0801156610250473,
      "rewards/real": 1.097609043121338,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 20.424310682494383,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -0.6620000600814819,
      "logits/real": -0.5432597398757935,
      "logps/generated": -169.11434936523438,
      "logps/real": -175.96197509765625,
      "loss": 0.8309,
      "rewards/accuracies": 0.5625,
      "rewards/generated": 1.11982262134552,
      "rewards/margins": 0.07809869945049286,
      "rewards/real": 1.1979211568832397,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.8412376757591001,
      "train_runtime": 702.1132,
      "train_samples_per_second": 2.849,
      "train_steps_per_second": 0.088
    }
  ],
  "logging_steps": 10,
  "max_steps": 62,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "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
}