{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9921671018276762,
  "eval_steps": 100,
  "global_step": 95,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "logits/chosen": 0.8826487064361572,
      "logits/rejected": 0.921362042427063,
      "logps/chosen": -36.58121871948242,
      "logps/rejected": -54.902320861816406,
      "loss": 2500.0,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-06,
      "logits/chosen": 0.8914464116096497,
      "logits/rejected": 0.8742997050285339,
      "logps/chosen": -87.8135986328125,
      "logps/rejected": -96.38023376464844,
      "loss": 2502.7986,
      "rewards/accuracies": 0.2152777761220932,
      "rewards/chosen": -0.00011578062549233437,
      "rewards/margins": -2.257831147289835e-05,
      "rewards/rejected": -9.320233948528767e-05,
      "step": 10
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.83118057351089e-06,
      "logits/chosen": 0.7811827659606934,
      "logits/rejected": 0.8477146029472351,
      "logps/chosen": -91.7614517211914,
      "logps/rejected": -85.11201477050781,
      "loss": 2501.6135,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -3.986436786362901e-05,
      "rewards/margins": -0.0002826174022629857,
      "rewards/rejected": 0.00024275311443489045,
      "step": 20
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.3475222930516484e-06,
      "logits/chosen": 0.8648529052734375,
      "logits/rejected": 0.8481088876724243,
      "logps/chosen": -85.29056549072266,
      "logps/rejected": -78.01969146728516,
      "loss": 2497.0559,
      "rewards/accuracies": 0.2750000059604645,
      "rewards/chosen": 0.0002869653981178999,
      "rewards/margins": 0.00010402966290712357,
      "rewards/rejected": 0.00018293573521077633,
      "step": 30
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.6143458894413463e-06,
      "logits/chosen": 0.7654550075531006,
      "logits/rejected": 0.8423868417739868,
      "logps/chosen": -122.76900482177734,
      "logps/rejected": -108.79447174072266,
      "loss": 2500.084,
      "rewards/accuracies": 0.28125,
      "rewards/chosen": -0.00036479695700109005,
      "rewards/margins": -0.0001517109922133386,
      "rewards/rejected": -0.0002130859502358362,
      "step": 40
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.730670898658255e-06,
      "logits/chosen": 0.8405305743217468,
      "logits/rejected": 0.869471549987793,
      "logps/chosen": -65.78740692138672,
      "logps/rejected": -71.66648864746094,
      "loss": 2500.4,
      "rewards/accuracies": 0.21875,
      "rewards/chosen": -0.00032314873533323407,
      "rewards/margins": -0.00028933738940395415,
      "rewards/rejected": -3.381132773938589e-05,
      "step": 50
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8158425248197931e-06,
      "logits/chosen": 0.8164768218994141,
      "logits/rejected": 0.8862431645393372,
      "logps/chosen": -117.6532974243164,
      "logps/rejected": -110.94734191894531,
      "loss": 2499.3979,
      "rewards/accuracies": 0.26875001192092896,
      "rewards/chosen": -0.0002028214803431183,
      "rewards/margins": -4.244589217705652e-05,
      "rewards/rejected": -0.0001603755954420194,
      "step": 60
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.934134090518593e-07,
      "logits/chosen": 0.8050645589828491,
      "logits/rejected": 0.8623224496841431,
      "logps/chosen": -82.68321990966797,
      "logps/rejected": -83.93122863769531,
      "loss": 2500.1014,
      "rewards/accuracies": 0.30000001192092896,
      "rewards/chosen": -0.00020013593893963844,
      "rewards/margins": 3.215682227164507e-05,
      "rewards/rejected": -0.00023229271755553782,
      "step": 70
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.7445716067596506e-07,
      "logits/chosen": 0.76527339220047,
      "logits/rejected": 0.8293226957321167,
      "logps/chosen": -105.8263931274414,
      "logps/rejected": -104.97891998291016,
      "loss": 2496.7385,
      "rewards/accuracies": 0.2750000059604645,
      "rewards/chosen": -0.00012336719373706728,
      "rewards/margins": -3.703986237724166e-07,
      "rewards/rejected": -0.00012299678928684443,
      "step": 80
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.256725079024554e-08,
      "logits/chosen": 0.8350859880447388,
      "logits/rejected": 0.855826199054718,
      "logps/chosen": -77.66065979003906,
      "logps/rejected": -90.00727081298828,
      "loss": 2495.8914,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 0.00030836346559226513,
      "rewards/margins": 0.0007544533582404256,
      "rewards/rejected": -0.00044608983444049954,
      "step": 90
    },
    {
      "epoch": 0.99,
      "step": 95,
      "total_flos": 0.0,
      "train_loss": 2499.3611225328946,
      "train_runtime": 1147.2033,
      "train_samples_per_second": 5.329,
      "train_steps_per_second": 0.083
    }
  ],
  "logging_steps": 10,
  "max_steps": 95,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}