{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 81,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 4.271874539371896,
      "learning_rate": 5.555555555555555e-08,
      "logits/chosen": 0.23574040830135345,
      "logits/rejected": -0.9465489983558655,
      "logps/chosen": -461.99664306640625,
      "logps/rejected": -1076.3154296875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.12,
      "grad_norm": 4.053815107503334,
      "learning_rate": 4.997620553954645e-07,
      "logits/chosen": 0.3158790171146393,
      "logits/rejected": -0.6447792649269104,
      "logps/chosen": -491.0052185058594,
      "logps/rejected": -1057.68212890625,
      "loss": 0.693,
      "rewards/accuracies": 0.4444444477558136,
      "rewards/chosen": 0.0007979909423738718,
      "rewards/margins": 0.00010702761210268363,
      "rewards/rejected": 0.0006909631192684174,
      "step": 10
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.8468513788685024,
      "learning_rate": 4.717527082945554e-07,
      "logits/chosen": 0.3795907199382782,
      "logits/rejected": -0.6666702032089233,
      "logps/chosen": -426.83050537109375,
      "logps/rejected": -1043.7567138671875,
      "loss": 0.6899,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 0.00618498120456934,
      "rewards/margins": 0.005285660736262798,
      "rewards/rejected": 0.0008993210503831506,
      "step": 20
    },
    {
      "epoch": 0.37,
      "grad_norm": 3.809120262255183,
      "learning_rate": 4.0219035725218013e-07,
      "logits/chosen": 0.4218289256095886,
      "logits/rejected": -0.7673497796058655,
      "logps/chosen": -476.8456115722656,
      "logps/rejected": -1008.1292114257812,
      "loss": 0.6821,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.019507069140672684,
      "rewards/margins": 0.0191163569688797,
      "rewards/rejected": 0.00039071092032827437,
      "step": 30
    },
    {
      "epoch": 0.49,
      "grad_norm": 3.8969795803158,
      "learning_rate": 3.041099034845257e-07,
      "logits/chosen": 0.24538719654083252,
      "logits/rejected": -0.5925542712211609,
      "logps/chosen": -496.660400390625,
      "logps/rejected": -976.67041015625,
      "loss": 0.6779,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 0.027163496240973473,
      "rewards/margins": 0.030033668503165245,
      "rewards/rejected": -0.0028701708652079105,
      "step": 40
    },
    {
      "epoch": 0.62,
      "grad_norm": 3.9669240189360293,
      "learning_rate": 1.9589009651547428e-07,
      "logits/chosen": 0.4538944661617279,
      "logits/rejected": -0.61766117811203,
      "logps/chosen": -444.9410095214844,
      "logps/rejected": -980.8577270507812,
      "loss": 0.6697,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 0.04584876075387001,
      "rewards/margins": 0.04866841062903404,
      "rewards/rejected": -0.002819649875164032,
      "step": 50
    },
    {
      "epoch": 0.74,
      "grad_norm": 3.761586308179101,
      "learning_rate": 9.780964274781983e-08,
      "logits/chosen": 0.3308876156806946,
      "logits/rejected": -0.6796020269393921,
      "logps/chosen": -471.8916931152344,
      "logps/rejected": -997.9669799804688,
      "loss": 0.6645,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": 0.05425606295466423,
      "rewards/margins": 0.06382595747709274,
      "rewards/rejected": -0.009569892659783363,
      "step": 60
    },
    {
      "epoch": 0.86,
      "grad_norm": 3.8973053098411667,
      "learning_rate": 2.824729170544457e-08,
      "logits/chosen": 0.3237437307834625,
      "logits/rejected": -0.6007438898086548,
      "logps/chosen": -452.1813049316406,
      "logps/rejected": -974.1201171875,
      "loss": 0.6626,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": 0.041471779346466064,
      "rewards/margins": 0.04470660537481308,
      "rewards/rejected": -0.003234823001548648,
      "step": 70
    },
    {
      "epoch": 0.99,
      "grad_norm": 3.6208468629029484,
      "learning_rate": 2.3794460453555044e-10,
      "logits/chosen": 0.3552563190460205,
      "logits/rejected": -0.7167419195175171,
      "logps/chosen": -452.73162841796875,
      "logps/rejected": -1027.007568359375,
      "loss": 0.665,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": 0.05553414672613144,
      "rewards/margins": 0.059998393058776855,
      "rewards/rejected": -0.00446424912661314,
      "step": 80
    },
    {
      "epoch": 1.0,
      "step": 81,
      "total_flos": 0.0,
      "train_loss": 0.6753969060050117,
      "train_runtime": 1121.4796,
      "train_samples_per_second": 4.608,
      "train_steps_per_second": 0.072
    }
  ],
  "logging_steps": 10,
  "max_steps": 81,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}