{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 200,
  "global_step": 125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008,
      "grad_norm": 147.89808677022958,
      "learning_rate": 3.846153846153846e-08,
      "logits/generated": -2.517829418182373,
      "logits/real": -2.19148588180542,
      "logps/generated": -122.74644470214844,
      "logps/real": -171.45494079589844,
      "loss": 0.843,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "grad_norm": 134.47376195877482,
      "learning_rate": 3.8461538461538463e-07,
      "logits/generated": -2.4292914867401123,
      "logits/real": -2.433361530303955,
      "logps/generated": -128.08148193359375,
      "logps/real": -129.51199340820312,
      "loss": 0.8235,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/generated": -0.045434772968292236,
      "rewards/margins": 0.06507062166929245,
      "rewards/real": 0.019635863602161407,
      "step": 10
    },
    {
      "epoch": 0.16,
      "grad_norm": 116.28359724719947,
      "learning_rate": 4.6874999999999996e-07,
      "logits/generated": -2.5721659660339355,
      "logits/real": -2.5940754413604736,
      "logps/generated": -140.29403686523438,
      "logps/real": -152.1209259033203,
      "loss": 0.7834,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/generated": -0.023977667093276978,
      "rewards/margins": 0.11755923926830292,
      "rewards/real": 0.09358155727386475,
      "step": 20
    },
    {
      "epoch": 0.24,
      "grad_norm": 125.42421242327684,
      "learning_rate": 4.2410714285714283e-07,
      "logits/generated": -2.451955556869507,
      "logits/real": -2.5324044227600098,
      "logps/generated": -132.75013732910156,
      "logps/real": -126.1029052734375,
      "loss": 0.8007,
      "rewards/accuracies": 0.699999988079071,
      "rewards/generated": 0.14888708293437958,
      "rewards/margins": 0.19555671513080597,
      "rewards/real": 0.34444382786750793,
      "step": 30
    },
    {
      "epoch": 0.32,
      "grad_norm": 120.25166417318326,
      "learning_rate": 3.794642857142857e-07,
      "logits/generated": -2.4365317821502686,
      "logits/real": -2.4784128665924072,
      "logps/generated": -121.6016845703125,
      "logps/real": -121.78703308105469,
      "loss": 0.7746,
      "rewards/accuracies": 0.699999988079071,
      "rewards/generated": 0.10297657549381256,
      "rewards/margins": 0.5478044152259827,
      "rewards/real": 0.650780975818634,
      "step": 40
    },
    {
      "epoch": 0.4,
      "grad_norm": 175.1395451582952,
      "learning_rate": 3.348214285714285e-07,
      "logits/generated": -2.4824845790863037,
      "logits/real": -2.628711223602295,
      "logps/generated": -108.2540054321289,
      "logps/real": -110.33675384521484,
      "loss": 0.8005,
      "rewards/accuracies": 0.625,
      "rewards/generated": 0.8452979326248169,
      "rewards/margins": 0.32963427901268005,
      "rewards/real": 1.1749321222305298,
      "step": 50
    },
    {
      "epoch": 0.48,
      "grad_norm": 136.08698116076778,
      "learning_rate": 2.9017857142857143e-07,
      "logits/generated": -2.5444846153259277,
      "logits/real": -2.5586628913879395,
      "logps/generated": -103.48173522949219,
      "logps/real": -113.9021224975586,
      "loss": 0.7874,
      "rewards/accuracies": 0.699999988079071,
      "rewards/generated": 0.6939820051193237,
      "rewards/margins": 0.3397112190723419,
      "rewards/real": 1.0336930751800537,
      "step": 60
    },
    {
      "epoch": 0.56,
      "grad_norm": 146.30984170549294,
      "learning_rate": 2.4553571428571425e-07,
      "logits/generated": -2.645383358001709,
      "logits/real": -2.6484053134918213,
      "logps/generated": -125.13639831542969,
      "logps/real": -122.11183166503906,
      "loss": 0.8997,
      "rewards/accuracies": 0.5,
      "rewards/generated": 1.455406665802002,
      "rewards/margins": -0.3015662729740143,
      "rewards/real": 1.1538405418395996,
      "step": 70
    },
    {
      "epoch": 0.64,
      "grad_norm": 84.4045941576579,
      "learning_rate": 2.0089285714285714e-07,
      "logits/generated": -2.6049187183380127,
      "logits/real": -2.5537009239196777,
      "logps/generated": -118.47220611572266,
      "logps/real": -142.0023193359375,
      "loss": 0.7769,
      "rewards/accuracies": 0.699999988079071,
      "rewards/generated": 0.9211456179618835,
      "rewards/margins": 0.8597875833511353,
      "rewards/real": 1.7809330224990845,
      "step": 80
    },
    {
      "epoch": 0.72,
      "grad_norm": 127.03498544690872,
      "learning_rate": 1.5624999999999999e-07,
      "logits/generated": -2.618711233139038,
      "logits/real": -2.6723551750183105,
      "logps/generated": -114.33888244628906,
      "logps/real": -143.75564575195312,
      "loss": 0.7321,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/generated": 1.1004014015197754,
      "rewards/margins": 0.39606350660324097,
      "rewards/real": 1.4964649677276611,
      "step": 90
    },
    {
      "epoch": 0.8,
      "grad_norm": 122.17610215112185,
      "learning_rate": 1.1160714285714285e-07,
      "logits/generated": -2.5757622718811035,
      "logits/real": -2.5916008949279785,
      "logps/generated": -127.0289306640625,
      "logps/real": -129.51431274414062,
      "loss": 0.8665,
      "rewards/accuracies": 0.625,
      "rewards/generated": 1.3618738651275635,
      "rewards/margins": 0.11677682399749756,
      "rewards/real": 1.4786508083343506,
      "step": 100
    },
    {
      "epoch": 0.88,
      "grad_norm": 133.71019821071505,
      "learning_rate": 6.696428571428571e-08,
      "logits/generated": -2.4835140705108643,
      "logits/real": -2.498384952545166,
      "logps/generated": -125.1689682006836,
      "logps/real": -121.4903564453125,
      "loss": 0.7741,
      "rewards/accuracies": 0.550000011920929,
      "rewards/generated": 1.3967167139053345,
      "rewards/margins": 0.36560097336769104,
      "rewards/real": 1.7623176574707031,
      "step": 110
    },
    {
      "epoch": 0.96,
      "grad_norm": 87.60022795114257,
      "learning_rate": 2.2321428571428572e-08,
      "logits/generated": -2.532543897628784,
      "logits/real": -2.6096348762512207,
      "logps/generated": -124.0304183959961,
      "logps/real": -137.7096405029297,
      "loss": 0.8181,
      "rewards/accuracies": 0.625,
      "rewards/generated": 1.4160971641540527,
      "rewards/margins": 0.37906622886657715,
      "rewards/real": 1.7951635122299194,
      "step": 120
    },
    {
      "epoch": 1.0,
      "step": 125,
      "total_flos": 0.0,
      "train_loss": 0.7958872079849243,
      "train_runtime": 757.4298,
      "train_samples_per_second": 2.641,
      "train_steps_per_second": 0.165
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "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
}