{
  "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": 555.6228482194545,
      "learning_rate": 8.333333333333333e-08,
      "logits/chosen": 135.66180419921875,
      "logits/rejected": 110.02157592773438,
      "logps/chosen": -827.0068359375,
      "logps/rejected": -818.9014892578125,
      "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": 676.2488072908993,
      "learning_rate": 4.907293218369498e-07,
      "logits/chosen": 121.71678161621094,
      "logits/rejected": 142.29522705078125,
      "logps/chosen": -812.8253784179688,
      "logps/rejected": -902.03662109375,
      "loss": 0.8916,
      "rewards/accuracies": 0.5208333134651184,
      "rewards/chosen": 1.315455436706543,
      "rewards/margins": 0.254628986120224,
      "rewards/rejected": 1.0608265399932861,
      "step": 10
    },
    {
      "epoch": 0.38,
      "grad_norm": 380.7231206035148,
      "learning_rate": 3.941700805287168e-07,
      "logits/chosen": 132.0984344482422,
      "logits/rejected": 141.592041015625,
      "logps/chosen": -816.7606201171875,
      "logps/rejected": -899.7657470703125,
      "loss": 0.7706,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.9781860113143921,
      "rewards/margins": 1.6455551385879517,
      "rewards/rejected": -2.6237411499023438,
      "step": 20
    },
    {
      "epoch": 0.57,
      "grad_norm": 238.91419418333624,
      "learning_rate": 2.3293939665883228e-07,
      "logits/chosen": 128.60267639160156,
      "logits/rejected": 132.55734252929688,
      "logps/chosen": -847.0579833984375,
      "logps/rejected": -911.5173950195312,
      "loss": 0.6789,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.37434953451156616,
      "rewards/margins": 2.5063283443450928,
      "rewards/rejected": -2.8806777000427246,
      "step": 30
    },
    {
      "epoch": 0.76,
      "grad_norm": 377.40289333938733,
      "learning_rate": 7.936171419533652e-08,
      "logits/chosen": 130.8309783935547,
      "logits/rejected": 126.89227294921875,
      "logps/chosen": -860.4659423828125,
      "logps/rejected": -873.3118896484375,
      "loss": 0.6884,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.2398359328508377,
      "rewards/margins": 1.8883726596832275,
      "rewards/rejected": -2.128208637237549,
      "step": 40
    },
    {
      "epoch": 0.95,
      "grad_norm": 313.23696106113914,
      "learning_rate": 2.328513490917311e-09,
      "logits/chosen": 137.02291870117188,
      "logits/rejected": 137.97555541992188,
      "logps/chosen": -872.61572265625,
      "logps/rejected": -902.748046875,
      "loss": 0.6086,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.32394886016845703,
      "rewards/margins": 1.9923664331436157,
      "rewards/rejected": -2.3163154125213623,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 52,
      "total_flos": 0.0,
      "train_loss": 0.7234916411913358,
      "train_runtime": 575.1359,
      "train_samples_per_second": 11.736,
      "train_steps_per_second": 0.09
    }
  ],
  "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
}