{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.998324958123953,
  "eval_steps": 100,
  "global_step": 149,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.3333333333333335e-07,
      "logits/chosen": -2.491262435913086,
      "logits/rejected": -2.5593011379241943,
      "logps/chosen": -151.13595581054688,
      "logps/rejected": -176.25180053710938,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.3333333333333333e-06,
      "logits/chosen": -2.657362699508667,
      "logits/rejected": -2.6224162578582764,
      "logps/chosen": -177.82272338867188,
      "logps/rejected": -179.98106384277344,
      "loss": 0.6925,
      "rewards/accuracies": 0.4930555522441864,
      "rewards/chosen": -0.013637593947350979,
      "rewards/margins": 0.004961313679814339,
      "rewards/rejected": -0.018598908558487892,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.982842942906386e-06,
      "logits/chosen": -2.702573299407959,
      "logits/rejected": -2.6558778285980225,
      "logps/chosen": -193.16390991210938,
      "logps/rejected": -188.91493225097656,
      "loss": 0.694,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -0.10607340186834335,
      "rewards/margins": 0.0018767903093248606,
      "rewards/rejected": -0.10795019567012787,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.846996204000967e-06,
      "logits/chosen": -2.6284031867980957,
      "logits/rejected": -2.585899829864502,
      "logps/chosen": -186.4259490966797,
      "logps/rejected": -187.31175231933594,
      "loss": 0.6914,
      "rewards/accuracies": 0.5406249761581421,
      "rewards/chosen": -0.0701591819524765,
      "rewards/margins": 0.008776131086051464,
      "rewards/rejected": -0.07893531024456024,
      "step": 30
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.582735470385229e-06,
      "logits/chosen": -2.6113476753234863,
      "logits/rejected": -2.6379783153533936,
      "logps/chosen": -169.52566528320312,
      "logps/rejected": -175.36927795410156,
      "loss": 0.6898,
      "rewards/accuracies": 0.528124988079071,
      "rewards/chosen": 0.0019457591697573662,
      "rewards/margins": 0.011439744383096695,
      "rewards/rejected": -0.009493985213339329,
      "step": 40
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.204519553876095e-06,
      "logits/chosen": -2.6095616817474365,
      "logits/rejected": -2.5940325260162354,
      "logps/chosen": -160.98004150390625,
      "logps/rejected": -164.4163360595703,
      "loss": 0.6926,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": 0.015476897358894348,
      "rewards/margins": 0.006927810609340668,
      "rewards/rejected": 0.008549087680876255,
      "step": 50
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7330422317447686e-06,
      "logits/chosen": -2.662424087524414,
      "logits/rejected": -2.6424801349639893,
      "logps/chosen": -180.03500366210938,
      "logps/rejected": -174.4924774169922,
      "loss": 0.6925,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": 0.0384332574903965,
      "rewards/margins": 0.005103477276861668,
      "rewards/rejected": 0.03332977741956711,
      "step": 60
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1941000034687516e-06,
      "logits/chosen": -2.6400465965270996,
      "logits/rejected": -2.6263554096221924,
      "logps/chosen": -167.74423217773438,
      "logps/rejected": -171.72691345214844,
      "loss": 0.6919,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.05802411586046219,
      "rewards/margins": 0.008442241698503494,
      "rewards/rejected": -0.06646636128425598,
      "step": 70
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6171806561748503e-06,
      "logits/chosen": -2.6121268272399902,
      "logits/rejected": -2.5906381607055664,
      "logps/chosen": -180.24801635742188,
      "logps/rejected": -179.08216857910156,
      "loss": 0.6883,
      "rewards/accuracies": 0.534375011920929,
      "rewards/chosen": -0.025833597406744957,
      "rewards/margins": 0.0072801136411726475,
      "rewards/rejected": -0.03311371058225632,
      "step": 80
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0338498642707977e-06,
      "logits/chosen": -2.663304090499878,
      "logits/rejected": -2.61620831489563,
      "logps/chosen": -175.11361694335938,
      "logps/rejected": -171.87161254882812,
      "loss": 0.6862,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.03003253974020481,
      "rewards/margins": 0.019972536712884903,
      "rewards/rejected": -0.05000507831573486,
      "step": 90
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4760240991587338e-06,
      "logits/chosen": -2.612849712371826,
      "logits/rejected": -2.590980052947998,
      "logps/chosen": -158.46450805664062,
      "logps/rejected": -165.57606506347656,
      "loss": 0.686,
      "rewards/accuracies": 0.596875011920929,
      "rewards/chosen": -0.00229537021368742,
      "rewards/margins": 0.028590286150574684,
      "rewards/rejected": -0.03088565543293953,
      "step": 100
    },
    {
      "epoch": 0.67,
      "eval_logits/chosen": -2.5771172046661377,
      "eval_logits/rejected": -2.4848170280456543,
      "eval_logps/chosen": -307.7010192871094,
      "eval_logps/rejected": -301.19329833984375,
      "eval_loss": 0.683651864528656,
      "eval_rewards/accuracies": 0.5720000267028809,
      "eval_rewards/chosen": -0.09298302233219147,
      "eval_rewards/margins": 0.020600860938429832,
      "eval_rewards/rejected": -0.11358388513326645,
      "eval_runtime": 382.0082,
      "eval_samples_per_second": 5.235,
      "eval_steps_per_second": 0.654,
      "step": 100
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.742243453755202e-07,
      "logits/chosen": -2.6324477195739746,
      "logits/rejected": -2.6193971633911133,
      "logps/chosen": -169.78958129882812,
      "logps/rejected": -178.68710327148438,
      "loss": 0.69,
      "rewards/accuracies": 0.528124988079071,
      "rewards/chosen": -0.028186390176415443,
      "rewards/margins": 0.006812813691794872,
      "rewards/rejected": -0.03499920293688774,
      "step": 110
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.559061696656199e-07,
      "logits/chosen": -2.6105129718780518,
      "logits/rejected": -2.5832369327545166,
      "logps/chosen": -177.0843505859375,
      "logps/rejected": -179.75782775878906,
      "loss": 0.6852,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": -0.021660596132278442,
      "rewards/margins": 0.016608651727437973,
      "rewards/rejected": -0.038269251585006714,
      "step": 120
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4395751190352924e-07,
      "logits/chosen": -2.590940475463867,
      "logits/rejected": -2.5727646350860596,
      "logps/chosen": -174.79104614257812,
      "logps/rejected": -179.25521850585938,
      "loss": 0.687,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.03851151093840599,
      "rewards/margins": 0.011401178315281868,
      "rewards/rejected": -0.04991268739104271,
      "step": 130
    },
    {
      "epoch": 0.94,
      "learning_rate": 5.544639001763719e-08,
      "logits/chosen": -2.6073455810546875,
      "logits/rejected": -2.6085193157196045,
      "logps/chosen": -189.36952209472656,
      "logps/rejected": -190.8089141845703,
      "loss": 0.688,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.04472886398434639,
      "rewards/margins": 0.012851757928729057,
      "rewards/rejected": -0.0575806125998497,
      "step": 140
    },
    {
      "epoch": 1.0,
      "step": 149,
      "total_flos": 0.0,
      "train_loss": 0.6894321513656002,
      "train_runtime": 6902.8391,
      "train_samples_per_second": 2.767,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 149,
  "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
}