{
  "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": 30.93577209985028,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -2.6151156425476074,
      "logits/real": -2.56895112991333,
      "logps/generated": -100.78079223632812,
      "logps/real": -84.83761596679688,
      "loss": 0.7771,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 53.91196703984471,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -2.4759128093719482,
      "logits/real": -2.5104005336761475,
      "logps/generated": -110.86439514160156,
      "logps/real": -96.45692443847656,
      "loss": 0.7744,
      "rewards/accuracies": 0.1944444477558136,
      "rewards/generated": 0.4696773886680603,
      "rewards/margins": 0.184379443526268,
      "rewards/real": 0.6540568470954895,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 41.533419926208836,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -2.7200608253479004,
      "logits/real": -2.7094995975494385,
      "logps/generated": -111.71916198730469,
      "logps/real": -97.46357727050781,
      "loss": 0.7549,
      "rewards/accuracies": 0.16249999403953552,
      "rewards/generated": 1.2888102531433105,
      "rewards/margins": 0.058855485171079636,
      "rewards/real": 1.347665548324585,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 28.929547003454825,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -2.762415885925293,
      "logits/real": -2.7689743041992188,
      "logps/generated": -84.8138656616211,
      "logps/real": -86.74251556396484,
      "loss": 0.7391,
      "rewards/accuracies": 0.23749999701976776,
      "rewards/generated": 1.5304011106491089,
      "rewards/margins": 0.12500552833080292,
      "rewards/real": 1.6554067134857178,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 27.557827905589537,
      "learning_rate": 2e-07,
      "logits/generated": -2.7791144847869873,
      "logits/real": -2.817704916000366,
      "logps/generated": -101.5577163696289,
      "logps/real": -89.77957153320312,
      "loss": 0.7543,
      "rewards/accuracies": 0.20000000298023224,
      "rewards/generated": 1.3332226276397705,
      "rewards/margins": 0.20582261681556702,
      "rewards/real": 1.5390453338623047,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 38.68815181578532,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -2.7200920581817627,
      "logits/real": -2.7181735038757324,
      "logps/generated": -101.0867691040039,
      "logps/real": -86.67424011230469,
      "loss": 0.7238,
      "rewards/accuracies": 0.23749999701976776,
      "rewards/generated": 1.2783713340759277,
      "rewards/margins": 0.4123215675354004,
      "rewards/real": 1.6906929016113281,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 27.90637817594034,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -2.6357321739196777,
      "logits/real": -2.649296522140503,
      "logps/generated": -121.07977294921875,
      "logps/real": -98.04679107666016,
      "loss": 0.732,
      "rewards/accuracies": 0.2750000059604645,
      "rewards/generated": 0.9751936197280884,
      "rewards/margins": 0.5532658100128174,
      "rewards/real": 1.5284594297409058,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.7473330401605175,
      "train_runtime": 796.1787,
      "train_samples_per_second": 2.511,
      "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
}