{
  "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": 90.36061761169043,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -0.7607710957527161,
      "logits/real": -0.3239991366863251,
      "logps/generated": -289.2199401855469,
      "logps/real": -291.08062744140625,
      "loss": 0.9696,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 80.57778361752598,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -0.9439048767089844,
      "logits/real": -0.6272310018539429,
      "logps/generated": -260.8728942871094,
      "logps/real": -270.6895446777344,
      "loss": 0.9236,
      "rewards/accuracies": 0.5972222089767456,
      "rewards/generated": 0.07361925393342972,
      "rewards/margins": 0.1317952275276184,
      "rewards/real": 0.20541447401046753,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 184.2526206474729,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -0.9683843851089478,
      "logits/real": -0.5614625215530396,
      "logps/generated": -255.9654998779297,
      "logps/real": -262.6753845214844,
      "loss": 0.7753,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/generated": 0.9276968240737915,
      "rewards/margins": 0.9088083505630493,
      "rewards/real": 1.8365051746368408,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 58.389369897312676,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -1.0661559104919434,
      "logits/real": -0.7151978015899658,
      "logps/generated": -250.697509765625,
      "logps/real": -258.60211181640625,
      "loss": 0.7304,
      "rewards/accuracies": 0.6875,
      "rewards/generated": 1.0640974044799805,
      "rewards/margins": 0.9086271524429321,
      "rewards/real": 1.9727245569229126,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 131.96437613532245,
      "learning_rate": 2e-07,
      "logits/generated": -0.9631511569023132,
      "logits/real": -0.6713980436325073,
      "logps/generated": -252.99539184570312,
      "logps/real": -243.7881317138672,
      "loss": 0.7048,
      "rewards/accuracies": 0.762499988079071,
      "rewards/generated": 1.2877063751220703,
      "rewards/margins": 1.0248647928237915,
      "rewards/real": 2.3125712871551514,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 351.71434098117106,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -1.0679857730865479,
      "logits/real": -0.7241894602775574,
      "logps/generated": -252.9150848388672,
      "logps/real": -244.3249053955078,
      "loss": 0.7103,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/generated": 1.1873204708099365,
      "rewards/margins": 1.0928575992584229,
      "rewards/real": 2.2801780700683594,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 102.7524068481551,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -0.8544826507568359,
      "logits/real": -0.6093840003013611,
      "logps/generated": -257.0255432128906,
      "logps/real": -257.3138122558594,
      "loss": 0.7126,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/generated": 1.4036052227020264,
      "rewards/margins": 1.051663875579834,
      "rewards/real": 2.4552693367004395,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.7647512930054818,
      "train_runtime": 925.9529,
      "train_samples_per_second": 2.151,
      "train_steps_per_second": 0.067
    }
  ],
  "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
}