{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 1000,
  "global_step": 125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 0.26106100145598204,
      "learning_rate": 3.846153846153847e-07,
      "logits/chosen": -1.7501684427261353,
      "logits/rejected": -1.5946102142333984,
      "logps/chosen": -84.6928939819336,
      "logps/rejected": -66.21642303466797,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.22496743005560174,
      "learning_rate": 3.846153846153847e-06,
      "logits/chosen": -1.7408065795898438,
      "logits/rejected": -1.716416835784912,
      "logps/chosen": -75.14291381835938,
      "logps/rejected": -78.59541320800781,
      "loss": 0.6926,
      "rewards/accuracies": 0.7222222089767456,
      "rewards/chosen": 0.0007475529564544559,
      "rewards/margins": 0.001586897298693657,
      "rewards/rejected": -0.0008393441094085574,
      "step": 10
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.23134629285918165,
      "learning_rate": 4.9519632010080765e-06,
      "logits/chosen": -1.7608575820922852,
      "logits/rejected": -1.6986463069915771,
      "logps/chosen": -76.15966796875,
      "logps/rejected": -74.84669494628906,
      "loss": 0.6845,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.0132329436019063,
      "rewards/margins": 0.01751554012298584,
      "rewards/rejected": -0.004282594658434391,
      "step": 20
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.2813775096241452,
      "learning_rate": 4.721114089947181e-06,
      "logits/chosen": -1.7035589218139648,
      "logits/rejected": -1.7651374340057373,
      "logps/chosen": -69.0379867553711,
      "logps/rejected": -82.99153900146484,
      "loss": 0.6604,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.040193259716033936,
      "rewards/margins": 0.0666363462805748,
      "rewards/rejected": -0.02644309028983116,
      "step": 30
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.500108908140759,
      "learning_rate": 4.316650805085068e-06,
      "logits/chosen": -1.7027181386947632,
      "logits/rejected": -1.7072885036468506,
      "logps/chosen": -66.51860046386719,
      "logps/rejected": -90.01258850097656,
      "loss": 0.6029,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.08252227306365967,
      "rewards/margins": 0.19697198271751404,
      "rewards/rejected": -0.11444971710443497,
      "step": 40
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5730232284019243,
      "learning_rate": 3.770188363116324e-06,
      "logits/chosen": -1.6690927743911743,
      "logits/rejected": -1.6148641109466553,
      "logps/chosen": -59.45286178588867,
      "logps/rejected": -114.36143493652344,
      "loss": 0.4753,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.16278788447380066,
      "rewards/margins": 0.5389599800109863,
      "rewards/rejected": -0.37617212533950806,
      "step": 50
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.34665392075726903,
      "learning_rate": 3.1244411954180677e-06,
      "logits/chosen": -1.6543304920196533,
      "logits/rejected": -1.4921056032180786,
      "logps/chosen": -53.32585525512695,
      "logps/rejected": -131.92892456054688,
      "loss": 0.3532,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2880913317203522,
      "rewards/margins": 0.8966875076293945,
      "rewards/rejected": -0.60859614610672,
      "step": 60
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.24483374726828422,
      "learning_rate": 2.429884359310328e-06,
      "logits/chosen": -1.5022072792053223,
      "logits/rejected": -1.4189573526382446,
      "logps/chosen": -45.96040725708008,
      "logps/rejected": -179.28762817382812,
      "loss": 0.2707,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3237885534763336,
      "rewards/margins": 1.3718864917755127,
      "rewards/rejected": -1.0480977296829224,
      "step": 70
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.21441653540531172,
      "learning_rate": 1.7408081372259633e-06,
      "logits/chosen": -1.446506381034851,
      "logits/rejected": -1.392620325088501,
      "logps/chosen": -41.029903411865234,
      "logps/rejected": -219.34548950195312,
      "loss": 0.2547,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.32778918743133545,
      "rewards/margins": 1.7608076333999634,
      "rewards/rejected": -1.4330183267593384,
      "step": 80
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.14834153445410317,
      "learning_rate": 1.1110744174509952e-06,
      "logits/chosen": -1.554793119430542,
      "logits/rejected": -1.5076603889465332,
      "logps/chosen": -40.31553268432617,
      "logps/rejected": -228.2432861328125,
      "loss": 0.2351,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.37351807951927185,
      "rewards/margins": 1.922289252281189,
      "rewards/rejected": -1.5487712621688843,
      "step": 90
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1420948746375737,
      "learning_rate": 5.899065604459814e-07,
      "logits/chosen": -1.4002104997634888,
      "logits/rejected": -1.4059101343154907,
      "logps/chosen": -38.817649841308594,
      "logps/rejected": -268.85711669921875,
      "loss": 0.2152,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.371096670627594,
      "rewards/margins": 2.2651829719543457,
      "rewards/rejected": -1.894086241722107,
      "step": 100
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.16015107651366245,
      "learning_rate": 2.1804183734670277e-07,
      "logits/chosen": -1.423669457435608,
      "logits/rejected": -1.459585428237915,
      "logps/chosen": -37.872806549072266,
      "logps/rejected": -281.24542236328125,
      "loss": 0.2042,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.34524857997894287,
      "rewards/margins": 2.3680052757263184,
      "rewards/rejected": -2.022756576538086,
      "step": 110
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.12328447499886092,
      "learning_rate": 2.454718665888589e-08,
      "logits/chosen": -1.530219316482544,
      "logits/rejected": -1.437981367111206,
      "logps/chosen": -39.403289794921875,
      "logps/rejected": -230.40676879882812,
      "loss": 0.2316,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.41614851355552673,
      "rewards/margins": 2.0078394412994385,
      "rewards/rejected": -1.5916908979415894,
      "step": 120
    },
    {
      "epoch": 5.0,
      "step": 125,
      "total_flos": 0.0,
      "train_loss": 0.4001942682266235,
      "train_runtime": 1378.9451,
      "train_samples_per_second": 2.901,
      "train_steps_per_second": 0.091
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}