{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9874476987447699,
  "eval_steps": 500,
  "global_step": 59,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 8.333333333333333e-08,
      "logits/chosen": -2.8012306690216064,
      "logits/rejected": -2.7200045585632324,
      "logps/chosen": -308.75421142578125,
      "logps/pi_response": -233.9468231201172,
      "logps/ref_response": -233.9468231201172,
      "logps/rejected": -311.24969482421875,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.930057285201027e-07,
      "logits/chosen": -2.7711262702941895,
      "logits/rejected": -2.709352731704712,
      "logps/chosen": -263.98291015625,
      "logps/pi_response": -193.27850341796875,
      "logps/ref_response": -193.61187744140625,
      "logps/rejected": -278.3876037597656,
      "loss": 0.69,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": 0.011274036020040512,
      "rewards/margins": 0.006776580587029457,
      "rewards/rejected": 0.004497454967349768,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.187457503795526e-07,
      "logits/chosen": -2.811276912689209,
      "logits/rejected": -2.751648426055908,
      "logps/chosen": -266.3084411621094,
      "logps/pi_response": -196.0972137451172,
      "logps/ref_response": -193.69471740722656,
      "logps/rejected": -272.2674255371094,
      "loss": 0.6601,
      "rewards/accuracies": 0.765625,
      "rewards/chosen": 0.045443661510944366,
      "rewards/margins": 0.06504254043102264,
      "rewards/rejected": -0.019598882645368576,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.8691164100062034e-07,
      "logits/chosen": -2.8319926261901855,
      "logits/rejected": -2.7600412368774414,
      "logps/chosen": -287.2323303222656,
      "logps/pi_response": -224.0599365234375,
      "logps/ref_response": -200.99618530273438,
      "logps/rejected": -287.2210998535156,
      "loss": 0.6115,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -0.010728769935667515,
      "rewards/margins": 0.23644748330116272,
      "rewards/rejected": -0.24717621505260468,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4248369943086995e-07,
      "logits/chosen": -2.7870166301727295,
      "logits/rejected": -2.7307822704315186,
      "logps/chosen": -258.9184265136719,
      "logps/pi_response": -226.122314453125,
      "logps/ref_response": -185.12936401367188,
      "logps/rejected": -311.3656311035156,
      "loss": 0.5706,
      "rewards/accuracies": 0.809374988079071,
      "rewards/chosen": -0.15448638796806335,
      "rewards/margins": 0.28851866722106934,
      "rewards/rejected": -0.4430050253868103,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.473909705816111e-08,
      "logits/chosen": -2.794811964035034,
      "logits/rejected": -2.7504866123199463,
      "logps/chosen": -283.52069091796875,
      "logps/pi_response": -245.2189483642578,
      "logps/ref_response": -190.3854522705078,
      "logps/rejected": -315.20977783203125,
      "loss": 0.5375,
      "rewards/accuracies": 0.8031250238418579,
      "rewards/chosen": -0.2589899003505707,
      "rewards/margins": 0.3519485890865326,
      "rewards/rejected": -0.6109384298324585,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.5997810929508532,
      "train_runtime": 3304.1575,
      "train_samples_per_second": 4.625,
      "train_steps_per_second": 0.018
    }
  ],
  "logging_steps": 10,
  "max_steps": 59,
  "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
}