{
  "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": 228.67711231286552,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -3.182648181915283,
      "logits/real": -3.236078977584839,
      "logps/generated": -231.1735382080078,
      "logps/real": -216.56446838378906,
      "loss": 0.914,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 104.82372723323877,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -3.0453360080718994,
      "logits/real": -3.0438425540924072,
      "logps/generated": -241.46511840820312,
      "logps/real": -229.99560546875,
      "loss": 0.842,
      "rewards/accuracies": 0.5694444179534912,
      "rewards/generated": -0.5092797875404358,
      "rewards/margins": 0.3939414620399475,
      "rewards/real": -0.11533831059932709,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 160.6471422793089,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -3.0548062324523926,
      "logits/real": -2.970714569091797,
      "logps/generated": -253.4170379638672,
      "logps/real": -231.45254516601562,
      "loss": 0.7195,
      "rewards/accuracies": 0.800000011920929,
      "rewards/generated": -2.3348288536071777,
      "rewards/margins": 1.3843599557876587,
      "rewards/real": -0.9504690170288086,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 146.55739161612283,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -3.0757832527160645,
      "logits/real": -3.142075300216675,
      "logps/generated": -258.6484680175781,
      "logps/real": -223.58065795898438,
      "loss": 0.5971,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/generated": -1.7125927209854126,
      "rewards/margins": 1.9861854314804077,
      "rewards/real": 0.2735927999019623,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 102.71447615342244,
      "learning_rate": 2e-07,
      "logits/generated": -2.938872814178467,
      "logits/real": -2.982527017593384,
      "logps/generated": -247.6828155517578,
      "logps/real": -228.6986541748047,
      "loss": 0.5605,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/generated": -2.5453975200653076,
      "rewards/margins": 1.9871200323104858,
      "rewards/real": -0.558277428150177,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 128.82542081084267,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -2.9596309661865234,
      "logits/real": -3.0371811389923096,
      "logps/generated": -258.96600341796875,
      "logps/real": -224.0094757080078,
      "loss": 0.6154,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/generated": -2.521024465560913,
      "rewards/margins": 2.7490367889404297,
      "rewards/real": 0.22801223397254944,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 140.9790674306219,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -2.8980393409729004,
      "logits/real": -3.0217809677124023,
      "logps/generated": -256.95941162109375,
      "logps/real": -225.6009063720703,
      "loss": 0.491,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/generated": -2.6129565238952637,
      "rewards/margins": 2.7330925464630127,
      "rewards/real": 0.12013578414916992,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.648834865900778,
      "train_runtime": 793.7409,
      "train_samples_per_second": 2.518,
      "train_steps_per_second": 0.078
    }
  ],
  "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
}