{
  "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": 43.21725757428246,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -0.9537944793701172,
      "logits/real": -1.0051143169403076,
      "logps/generated": -316.6651611328125,
      "logps/real": -291.36328125,
      "loss": 0.9684,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 40.965171560031784,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -0.8021091818809509,
      "logits/real": -0.7525854706764221,
      "logps/generated": -295.2068176269531,
      "logps/real": -279.92877197265625,
      "loss": 0.9499,
      "rewards/accuracies": 0.5555555820465088,
      "rewards/generated": -0.0021832569036632776,
      "rewards/margins": 0.04766163229942322,
      "rewards/real": 0.045478373765945435,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 44.279849646311796,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -0.7875266671180725,
      "logits/real": -0.7331272959709167,
      "logps/generated": -282.63128662109375,
      "logps/real": -270.755126953125,
      "loss": 0.9017,
      "rewards/accuracies": 0.6875,
      "rewards/generated": 0.061399005353450775,
      "rewards/margins": 0.26890262961387634,
      "rewards/real": 0.3303016722202301,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 54.547186171140005,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -0.8387455940246582,
      "logits/real": -0.7834113836288452,
      "logps/generated": -286.93084716796875,
      "logps/real": -276.5713806152344,
      "loss": 0.8875,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/generated": 0.2545087933540344,
      "rewards/margins": 0.18184302747249603,
      "rewards/real": 0.43635183572769165,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 43.815629015597885,
      "learning_rate": 2e-07,
      "logits/generated": -0.882081151008606,
      "logits/real": -0.8120096325874329,
      "logps/generated": -298.81744384765625,
      "logps/real": -270.52203369140625,
      "loss": 0.8268,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/generated": 0.1146630272269249,
      "rewards/margins": 0.46714526414871216,
      "rewards/real": 0.5818082690238953,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 34.532344880201805,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -0.8125056028366089,
      "logits/real": -0.8328151702880859,
      "logps/generated": -292.03656005859375,
      "logps/real": -276.155029296875,
      "loss": 0.8497,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/generated": 0.2003205567598343,
      "rewards/margins": 0.3035683035850525,
      "rewards/real": 0.5038889050483704,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 32.27958496678707,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -0.8254791498184204,
      "logits/real": -0.8004539608955383,
      "logps/generated": -292.7684020996094,
      "logps/real": -276.4676513671875,
      "loss": 0.8181,
      "rewards/accuracies": 0.75,
      "rewards/generated": 0.117733433842659,
      "rewards/margins": 0.36951756477355957,
      "rewards/real": 0.487250953912735,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.8712061278281673,
      "train_runtime": 704.489,
      "train_samples_per_second": 2.835,
      "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
}