{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.985781990521327,
  "eval_steps": 100,
  "global_step": 52,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 1957.9884167124578,
      "learning_rate": 8.333333333333333e-08,
      "logits/chosen": 124.78954315185547,
      "logits/rejected": 100.39772033691406,
      "logps/chosen": -796.0274658203125,
      "logps/rejected": -794.6148071289062,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.19,
      "grad_norm": 2425.4331055868297,
      "learning_rate": 4.907293218369498e-07,
      "logits/chosen": 113.938720703125,
      "logits/rejected": 135.90719604492188,
      "logps/chosen": -780.183837890625,
      "logps/rejected": -875.3695678710938,
      "loss": 2.1195,
      "rewards/accuracies": 0.4930555522441864,
      "rewards/chosen": 4.653940200805664,
      "rewards/margins": 1.20315420627594,
      "rewards/rejected": 3.4507861137390137,
      "step": 10
    },
    {
      "epoch": 0.38,
      "grad_norm": 1819.3721154286172,
      "learning_rate": 3.941700805287168e-07,
      "logits/chosen": 126.57746887207031,
      "logits/rejected": 137.83840942382812,
      "logps/chosen": -774.68359375,
      "logps/rejected": -861.1171875,
      "loss": 2.0878,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -1.7321250438690186,
      "rewards/margins": 4.814647197723389,
      "rewards/rejected": -6.5467729568481445,
      "step": 20
    },
    {
      "epoch": 0.57,
      "grad_norm": 1588.4595576906577,
      "learning_rate": 2.3293939665883228e-07,
      "logits/chosen": 124.16752624511719,
      "logits/rejected": 127.4310531616211,
      "logps/chosen": -818.1854248046875,
      "logps/rejected": -877.2722778320312,
      "loss": 2.3526,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -3.666670560836792,
      "rewards/margins": 6.680838584899902,
      "rewards/rejected": -10.347509384155273,
      "step": 30
    },
    {
      "epoch": 0.76,
      "grad_norm": 1396.6549591742623,
      "learning_rate": 7.936171419533652e-08,
      "logits/chosen": 123.1187973022461,
      "logits/rejected": 118.75350189208984,
      "logps/chosen": -812.8169555664062,
      "logps/rejected": -826.1077270507812,
      "loss": 1.9731,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": 3.7158901691436768,
      "rewards/margins": 5.847080707550049,
      "rewards/rejected": -2.131190538406372,
      "step": 40
    },
    {
      "epoch": 0.95,
      "grad_norm": 1524.5875251697705,
      "learning_rate": 2.328513490917311e-09,
      "logits/chosen": 133.9492645263672,
      "logits/rejected": 133.85220336914062,
      "logps/chosen": -822.7086791992188,
      "logps/rejected": -851.4788208007812,
      "loss": 1.681,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": 4.888961315155029,
      "rewards/margins": 5.86912727355957,
      "rewards/rejected": -0.9801660776138306,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 52,
      "total_flos": 0.0,
      "train_loss": 1.981816053390503,
      "train_runtime": 574.1146,
      "train_samples_per_second": 11.757,
      "train_steps_per_second": 0.091
    }
  ],
  "logging_steps": 10,
  "max_steps": 52,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}