{
  "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": 11.613288464777222,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -1.1354284286499023,
      "logits/real": -0.6185171008110046,
      "logps/generated": -118.00320434570312,
      "logps/real": -200.14151000976562,
      "loss": 0.8305,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 9.733030230104115,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -1.043811321258545,
      "logits/real": -1.071096420288086,
      "logps/generated": -150.50152587890625,
      "logps/real": -150.53155517578125,
      "loss": 0.8213,
      "rewards/accuracies": 0.0833333358168602,
      "rewards/generated": 0.08699116855859756,
      "rewards/margins": -0.0006038660649210215,
      "rewards/real": 0.08638730645179749,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 8.782709887193477,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -1.3196420669555664,
      "logits/real": -1.299203872680664,
      "logps/generated": -141.1020050048828,
      "logps/real": -135.54541015625,
      "loss": 0.8026,
      "rewards/accuracies": 0.1875,
      "rewards/generated": 0.5164750814437866,
      "rewards/margins": 0.023883480578660965,
      "rewards/real": 0.5403585433959961,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 6.004735799495323,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -1.1754536628723145,
      "logits/real": -1.196388602256775,
      "logps/generated": -128.12631225585938,
      "logps/real": -131.55551147460938,
      "loss": 0.8034,
      "rewards/accuracies": 0.17499999701976776,
      "rewards/generated": 0.9843085408210754,
      "rewards/margins": 0.01862453855574131,
      "rewards/real": 1.002933144569397,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 7.699712198334635,
      "learning_rate": 2e-07,
      "logits/generated": -1.1779353618621826,
      "logits/real": -1.3112047910690308,
      "logps/generated": -155.4976806640625,
      "logps/real": -132.09780883789062,
      "loss": 0.7895,
      "rewards/accuracies": 0.20000000298023224,
      "rewards/generated": 1.2571728229522705,
      "rewards/margins": 0.08223126828670502,
      "rewards/real": 1.3394043445587158,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 6.175849275981129,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -1.3791444301605225,
      "logits/real": -1.37796151638031,
      "logps/generated": -131.0447998046875,
      "logps/real": -130.96212768554688,
      "loss": 0.7898,
      "rewards/accuracies": 0.13750000298023224,
      "rewards/generated": 1.49892258644104,
      "rewards/margins": 0.029646193608641624,
      "rewards/real": 1.5285687446594238,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 7.696425445812134,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -1.4152286052703857,
      "logits/real": -1.4222019910812378,
      "logps/generated": -129.74317932128906,
      "logps/real": -120.62994384765625,
      "loss": 0.7694,
      "rewards/accuracies": 0.3125,
      "rewards/generated": 1.5588010549545288,
      "rewards/margins": 0.09574465453624725,
      "rewards/real": 1.654545783996582,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.7963649149863951,
      "train_runtime": 722.2719,
      "train_samples_per_second": 2.768,
      "train_steps_per_second": 0.086
    }
  ],
  "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
}