{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 50,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 1.25e-07,
      "logits/generated": -2.9780185222625732,
      "logits/real": -2.891969919204712,
      "logps/generated": -75.28430938720703,
      "logps/real": -112.39949035644531,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.928571428571428e-07,
      "logits/generated": -3.0109641551971436,
      "logits/real": -2.735833168029785,
      "logps/generated": -129.67689514160156,
      "logps/real": -122.11038970947266,
      "loss": 0.1661,
      "rewards/accuracies": 0.875,
      "rewards/generated": -2.9385464191436768,
      "rewards/margins": 3.736133098602295,
      "rewards/real": 0.7975864410400391,
      "step": 10
    },
    {
      "epoch": 1.25,
      "learning_rate": 2.1428571428571426e-07,
      "logits/generated": -2.9212605953216553,
      "logits/real": -2.739665985107422,
      "logps/generated": -155.23023986816406,
      "logps/real": -139.42665100097656,
      "loss": 0.0224,
      "rewards/accuracies": 1.0,
      "rewards/generated": -5.471407890319824,
      "rewards/margins": 6.653175354003906,
      "rewards/real": 1.1817679405212402,
      "step": 20
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.571428571428571e-08,
      "logits/generated": -2.919872760772705,
      "logits/real": -2.64644455909729,
      "logps/generated": -149.81588745117188,
      "logps/real": -110.94522857666016,
      "loss": 0.0084,
      "rewards/accuracies": 0.987500011920929,
      "rewards/generated": -5.761648178100586,
      "rewards/margins": 7.198714256286621,
      "rewards/real": 1.4370663166046143,
      "step": 30
    },
    {
      "epoch": 2.0,
      "step": 32,
      "total_flos": 0.0,
      "train_loss": 0.07844111844315194,
      "train_runtime": 153.525,
      "train_samples_per_second": 6.514,
      "train_steps_per_second": 0.208
    }
  ],
  "logging_steps": 10,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}