{
  "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": 107.88710762985947,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -2.752808094024658,
      "logits/real": -2.2284693717956543,
      "logps/generated": -121.13522338867188,
      "logps/real": -153.264892578125,
      "loss": 0.8501,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 91.94956537605988,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -2.6014535427093506,
      "logits/real": -2.6951539516448975,
      "logps/generated": -133.1143035888672,
      "logps/real": -140.10543823242188,
      "loss": 0.7861,
      "rewards/accuracies": 0.5416666865348816,
      "rewards/generated": 0.36728113889694214,
      "rewards/margins": 0.08309322595596313,
      "rewards/real": 0.45037439465522766,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 81.49584102213795,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -2.740352153778076,
      "logits/real": -2.7934975624084473,
      "logps/generated": -124.69525146484375,
      "logps/real": -122.46867370605469,
      "loss": 0.8205,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/generated": 0.6012061834335327,
      "rewards/margins": 0.2280128300189972,
      "rewards/real": 0.8292189836502075,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 86.29390865351938,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -2.7653279304504395,
      "logits/real": -2.861011028289795,
      "logps/generated": -104.97627258300781,
      "logps/real": -110.78263854980469,
      "loss": 0.7926,
      "rewards/accuracies": 0.6875,
      "rewards/generated": 1.0582153797149658,
      "rewards/margins": 0.4253184199333191,
      "rewards/real": 1.4835339784622192,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 73.40988849274181,
      "learning_rate": 2e-07,
      "logits/generated": -2.899013042449951,
      "logits/real": -2.9045419692993164,
      "logps/generated": -116.1448745727539,
      "logps/real": -126.13529968261719,
      "loss": 0.8216,
      "rewards/accuracies": 0.574999988079071,
      "rewards/generated": 1.988576889038086,
      "rewards/margins": 0.31346917152404785,
      "rewards/real": 2.302046060562134,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 87.84851470741809,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -2.8664755821228027,
      "logits/real": -2.9287781715393066,
      "logps/generated": -113.2905502319336,
      "logps/real": -129.95413208007812,
      "loss": 0.803,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/generated": 2.1775314807891846,
      "rewards/margins": 0.2134767472743988,
      "rewards/real": 2.391007900238037,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 73.6437417177821,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -2.788062572479248,
      "logits/real": -2.8471646308898926,
      "logps/generated": -119.0680923461914,
      "logps/real": -124.4272232055664,
      "loss": 0.7742,
      "rewards/accuracies": 0.625,
      "rewards/generated": 2.202199935913086,
      "rewards/margins": 0.361619770526886,
      "rewards/real": 2.5638198852539062,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.8044227688543258,
      "train_runtime": 744.0688,
      "train_samples_per_second": 2.688,
      "train_steps_per_second": 0.083
    }
  ],
  "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
}