{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9842931937172775,
  "eval_steps": 100,
  "global_step": 47,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 1.708532048983374,
      "learning_rate": 1e-07,
      "logits/chosen": 1.0625,
      "logits/rejected": 0.77734375,
      "logps/chosen": -95.5,
      "logps/rejected": -175.0,
      "loss": 0.7942,
      "rewards/accuracies": 0.5546875,
      "rewards/chosen": 0.271484375,
      "rewards/margins": 0.216796875,
      "rewards/rejected": 0.055419921875,
      "step": 1
    },
    {
      "epoch": 0.21,
      "grad_norm": 2.287090620471851,
      "learning_rate": 4.82718437161051e-07,
      "logits/chosen": 0.9296875,
      "logits/rejected": 0.77734375,
      "logps/chosen": -106.0,
      "logps/rejected": -175.0,
      "loss": 0.9342,
      "rewards/accuracies": 0.4921875,
      "rewards/chosen": -0.01361083984375,
      "rewards/margins": -0.0277099609375,
      "rewards/rejected": 0.01385498046875,
      "step": 10
    },
    {
      "epoch": 0.42,
      "grad_norm": 2.0110436551886433,
      "learning_rate": 3.584709347793895e-07,
      "logits/chosen": 0.9296875,
      "logits/rejected": 0.78125,
      "logps/chosen": -107.5,
      "logps/rejected": -177.0,
      "loss": 0.9423,
      "rewards/accuracies": 0.49687498807907104,
      "rewards/chosen": 0.0016021728515625,
      "rewards/margins": -0.02197265625,
      "rewards/rejected": 0.023681640625,
      "step": 20
    },
    {
      "epoch": 0.63,
      "grad_norm": 2.2429733590663514,
      "learning_rate": 1.763112063972739e-07,
      "logits/chosen": 0.94921875,
      "logits/rejected": 0.796875,
      "logps/chosen": -106.5,
      "logps/rejected": -175.0,
      "loss": 0.9318,
      "rewards/accuracies": 0.4703125059604645,
      "rewards/chosen": -0.02490234375,
      "rewards/margins": -0.03515625,
      "rewards/rejected": 0.01031494140625,
      "step": 30
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.8865669259605762,
      "learning_rate": 3.349364905389032e-08,
      "logits/chosen": 1.0,
      "logits/rejected": 0.85546875,
      "logps/chosen": -107.5,
      "logps/rejected": -176.0,
      "loss": 0.9092,
      "rewards/accuracies": 0.5023437738418579,
      "rewards/chosen": -0.0751953125,
      "rewards/margins": 0.006683349609375,
      "rewards/rejected": -0.08154296875,
      "step": 40
    },
    {
      "epoch": 0.98,
      "step": 47,
      "total_flos": 0.0,
      "train_loss": 0.9213893159906915,
      "train_runtime": 2106.8268,
      "train_samples_per_second": 2.9,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 47,
  "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
}