{
  "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": -2.8084447383880615,
      "logits/rejected": -2.7193427085876465,
      "logps/chosen": -297.1988220214844,
      "logps/pi_response": -75.58995056152344,
      "logps/ref_response": -75.58995056152344,
      "logps/rejected": -175.9150848388672,
      "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": -2.746209144592285,
      "logits/rejected": -2.70629620552063,
      "logps/chosen": -238.48106384277344,
      "logps/pi_response": -73.80045318603516,
      "logps/ref_response": -73.4278335571289,
      "logps/rejected": -189.3667755126953,
      "loss": 0.6894,
      "rewards/accuracies": 0.5555555820465088,
      "rewards/chosen": 0.00897198636084795,
      "rewards/margins": 0.007072604261338711,
      "rewards/rejected": 0.0018993823323398829,
      "step": 10
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.187457503795526e-07,
      "logits/chosen": -2.723341703414917,
      "logits/rejected": -2.6848156452178955,
      "logps/chosen": -232.9024658203125,
      "logps/pi_response": -85.4902572631836,
      "logps/ref_response": -74.08830261230469,
      "logps/rejected": -171.67996215820312,
      "loss": 0.6557,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": 0.01758940890431404,
      "rewards/margins": 0.08052871376276016,
      "rewards/rejected": -0.06293930113315582,
      "step": 20
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.8691164100062034e-07,
      "logits/chosen": -2.6599957942962646,
      "logits/rejected": -2.6350595951080322,
      "logps/chosen": -276.35589599609375,
      "logps/pi_response": -125.26264953613281,
      "logps/ref_response": -80.98223876953125,
      "logps/rejected": -198.37667846679688,
      "loss": 0.6167,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.09377123415470123,
      "rewards/margins": 0.23797473311424255,
      "rewards/rejected": -0.331745982170105,
      "step": 30
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4248369943086995e-07,
      "logits/chosen": -2.6382107734680176,
      "logits/rejected": -2.6134047508239746,
      "logps/chosen": -240.15499877929688,
      "logps/pi_response": -123.49603271484375,
      "logps/ref_response": -68.78013610839844,
      "logps/rejected": -209.35964965820312,
      "loss": 0.595,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.21807746589183807,
      "rewards/margins": 0.22044435143470764,
      "rewards/rejected": -0.43852177262306213,
      "step": 40
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.473909705816111e-08,
      "logits/chosen": -2.6552937030792236,
      "logits/rejected": -2.6305758953094482,
      "logps/chosen": -277.57086181640625,
      "logps/pi_response": -141.43019104003906,
      "logps/ref_response": -70.03689575195312,
      "logps/rejected": -226.5507354736328,
      "loss": 0.5707,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.29111790657043457,
      "rewards/margins": 0.29376304149627686,
      "rewards/rejected": -0.5848809480667114,
      "step": 50
    },
    {
      "epoch": 0.99,
      "step": 59,
      "total_flos": 0.0,
      "train_loss": 0.6155199842937922,
      "train_runtime": 3481.3266,
      "train_samples_per_second": 4.39,
      "train_steps_per_second": 0.017
    }
  ],
  "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
}