{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9902995720399428,
  "eval_steps": 100,
  "global_step": 218,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.2727272727272729e-07,
      "logits/chosen": -2.779836893081665,
      "logits/rejected": -2.772892951965332,
      "logps/chosen": -67.39044952392578,
      "logps/rejected": -65.7892074584961,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.09,
      "learning_rate": 2.2727272727272728e-06,
      "logits/chosen": -2.764805793762207,
      "logits/rejected": -2.7586262226104736,
      "logps/chosen": -63.05746841430664,
      "logps/rejected": -64.96013641357422,
      "loss": 0.6864,
      "rewards/accuracies": 0.6006944179534912,
      "rewards/chosen": 0.009925955906510353,
      "rewards/margins": 0.013827367685735226,
      "rewards/rejected": -0.0039014113135635853,
      "step": 10
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.5454545454545455e-06,
      "logits/chosen": -2.7655444145202637,
      "logits/rejected": -2.7531120777130127,
      "logps/chosen": -60.976318359375,
      "logps/rejected": -64.35781860351562,
      "loss": 0.5979,
      "rewards/accuracies": 0.8500000238418579,
      "rewards/chosen": 0.06535812467336655,
      "rewards/margins": 0.22067773342132568,
      "rewards/rejected": -0.15531960129737854,
      "step": 20
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.795918367346939e-06,
      "logits/chosen": -2.7331104278564453,
      "logits/rejected": -2.722367763519287,
      "logps/chosen": -63.54418182373047,
      "logps/rejected": -77.56448364257812,
      "loss": 0.3997,
      "rewards/accuracies": 0.878125011920929,
      "rewards/chosen": -0.11322204768657684,
      "rewards/margins": 0.9385285377502441,
      "rewards/rejected": -1.0517505407333374,
      "step": 30
    },
    {
      "epoch": 0.37,
      "learning_rate": 4.540816326530613e-06,
      "logits/chosen": -2.685359477996826,
      "logits/rejected": -2.6721653938293457,
      "logps/chosen": -67.8324966430664,
      "logps/rejected": -89.94172668457031,
      "loss": 0.2768,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -0.5856838226318359,
      "rewards/margins": 1.7414783239364624,
      "rewards/rejected": -2.327162265777588,
      "step": 40
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.2857142857142855e-06,
      "logits/chosen": -2.660297393798828,
      "logits/rejected": -2.6442055702209473,
      "logps/chosen": -72.59104919433594,
      "logps/rejected": -93.20745849609375,
      "loss": 0.2475,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -0.7966881394386292,
      "rewards/margins": 2.058647632598877,
      "rewards/rejected": -2.8553357124328613,
      "step": 50
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.03061224489796e-06,
      "logits/chosen": -2.643165111541748,
      "logits/rejected": -2.63153076171875,
      "logps/chosen": -72.09125518798828,
      "logps/rejected": -99.36156463623047,
      "loss": 0.2392,
      "rewards/accuracies": 0.903124988079071,
      "rewards/chosen": -1.083268404006958,
      "rewards/margins": 2.3197226524353027,
      "rewards/rejected": -3.4029908180236816,
      "step": 60
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.7755102040816327e-06,
      "logits/chosen": -2.638352632522583,
      "logits/rejected": -2.6236445903778076,
      "logps/chosen": -77.40001678466797,
      "logps/rejected": -107.3912124633789,
      "loss": 0.2044,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -1.5230330228805542,
      "rewards/margins": 2.718259572982788,
      "rewards/rejected": -4.241292476654053,
      "step": 70
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.5204081632653062e-06,
      "logits/chosen": -2.6177425384521484,
      "logits/rejected": -2.597738265991211,
      "logps/chosen": -75.93782043457031,
      "logps/rejected": -108.6824951171875,
      "loss": 0.1696,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -1.198141098022461,
      "rewards/margins": 3.195936918258667,
      "rewards/rejected": -4.394078254699707,
      "step": 80
    },
    {
      "epoch": 0.82,
      "learning_rate": 3.2653061224489794e-06,
      "logits/chosen": -2.6300315856933594,
      "logits/rejected": -2.6183547973632812,
      "logps/chosen": -80.39871978759766,
      "logps/rejected": -113.6379165649414,
      "loss": 0.1925,
      "rewards/accuracies": 0.8968750238418579,
      "rewards/chosen": -1.5853779315948486,
      "rewards/margins": 3.1853203773498535,
      "rewards/rejected": -4.770698547363281,
      "step": 90
    },
    {
      "epoch": 0.91,
      "learning_rate": 3.0102040816326534e-06,
      "logits/chosen": -2.6064066886901855,
      "logits/rejected": -2.594722270965576,
      "logps/chosen": -81.89433288574219,
      "logps/rejected": -115.00162506103516,
      "loss": 0.1534,
      "rewards/accuracies": 0.9468749761581421,
      "rewards/chosen": -1.6648212671279907,
      "rewards/margins": 3.4960713386535645,
      "rewards/rejected": -5.160892486572266,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -2.6083757877349854,
      "eval_logits/rejected": -2.5937814712524414,
      "eval_logps/chosen": -80.02838134765625,
      "eval_logps/rejected": -119.6004409790039,
      "eval_loss": 0.15436382591724396,
      "eval_rewards/accuracies": 0.939497709274292,
      "eval_rewards/chosen": -1.6908209323883057,
      "eval_rewards/margins": 3.7731716632843018,
      "eval_rewards/rejected": -5.463992595672607,
      "eval_runtime": 295.1725,
      "eval_samples_per_second": 2.968,
      "eval_steps_per_second": 2.968,
      "step": 109
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.7551020408163266e-06,
      "logits/chosen": -2.6033756732940674,
      "logits/rejected": -2.582958221435547,
      "logps/chosen": -80.38322448730469,
      "logps/rejected": -116.6507797241211,
      "loss": 0.138,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -1.656935691833496,
      "rewards/margins": 3.698695659637451,
      "rewards/rejected": -5.3556318283081055,
      "step": 110
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.5e-06,
      "logits/chosen": -2.604905843734741,
      "logits/rejected": -2.591404438018799,
      "logps/chosen": -80.86669921875,
      "logps/rejected": -119.6518783569336,
      "loss": 0.1395,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": -1.803047776222229,
      "rewards/margins": 3.7810962200164795,
      "rewards/rejected": -5.58414363861084,
      "step": 120
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.244897959183674e-06,
      "logits/chosen": -2.590282440185547,
      "logits/rejected": -2.576897144317627,
      "logps/chosen": -83.53189086914062,
      "logps/rejected": -131.67037963867188,
      "loss": 0.1004,
      "rewards/accuracies": 0.981249988079071,
      "rewards/chosen": -1.9269969463348389,
      "rewards/margins": 4.388735294342041,
      "rewards/rejected": -6.315732002258301,
      "step": 130
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.989795918367347e-06,
      "logits/chosen": -2.578322410583496,
      "logits/rejected": -2.5577735900878906,
      "logps/chosen": -85.14395141601562,
      "logps/rejected": -126.8256607055664,
      "loss": 0.1422,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -2.2692408561706543,
      "rewards/margins": 4.1861348152160645,
      "rewards/rejected": -6.455375671386719,
      "step": 140
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.7346938775510206e-06,
      "logits/chosen": -2.5812907218933105,
      "logits/rejected": -2.5663979053497314,
      "logps/chosen": -87.79288482666016,
      "logps/rejected": -131.3497314453125,
      "loss": 0.1284,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -2.4120209217071533,
      "rewards/margins": 4.293553352355957,
      "rewards/rejected": -6.705574035644531,
      "step": 150
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.479591836734694e-06,
      "logits/chosen": -2.596059560775757,
      "logits/rejected": -2.58156156539917,
      "logps/chosen": -83.30199432373047,
      "logps/rejected": -132.67092895507812,
      "loss": 0.1066,
      "rewards/accuracies": 0.9593750238418579,
      "rewards/chosen": -2.065701484680176,
      "rewards/margins": 4.5842509269714355,
      "rewards/rejected": -6.6499528884887695,
      "step": 160
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.2244897959183673e-06,
      "logits/chosen": -2.5813376903533936,
      "logits/rejected": -2.569676160812378,
      "logps/chosen": -82.94264221191406,
      "logps/rejected": -132.35789489746094,
      "loss": 0.0994,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -2.1285476684570312,
      "rewards/margins": 4.649580955505371,
      "rewards/rejected": -6.778128147125244,
      "step": 170
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.69387755102041e-07,
      "logits/chosen": -2.575303077697754,
      "logits/rejected": -2.5673909187316895,
      "logps/chosen": -86.48652648925781,
      "logps/rejected": -134.35574340820312,
      "loss": 0.1224,
      "rewards/accuracies": 0.956250011920929,
      "rewards/chosen": -2.278262138366699,
      "rewards/margins": 4.526266098022461,
      "rewards/rejected": -6.80452823638916,
      "step": 180
    },
    {
      "epoch": 1.73,
      "learning_rate": 7.142857142857143e-07,
      "logits/chosen": -2.5673184394836426,
      "logits/rejected": -2.5392918586730957,
      "logps/chosen": -85.58756256103516,
      "logps/rejected": -128.51121520996094,
      "loss": 0.1345,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": -2.157008409500122,
      "rewards/margins": 4.34613037109375,
      "rewards/rejected": -6.503138542175293,
      "step": 190
    },
    {
      "epoch": 1.83,
      "learning_rate": 4.591836734693878e-07,
      "logits/chosen": -2.5774478912353516,
      "logits/rejected": -2.556039333343506,
      "logps/chosen": -84.05760955810547,
      "logps/rejected": -132.11903381347656,
      "loss": 0.1088,
      "rewards/accuracies": 0.9593750238418579,
      "rewards/chosen": -2.128478527069092,
      "rewards/margins": 4.659018039703369,
      "rewards/rejected": -6.787497043609619,
      "step": 200
    },
    {
      "epoch": 1.92,
      "learning_rate": 2.0408163265306121e-07,
      "logits/chosen": -2.563249111175537,
      "logits/rejected": -2.5445797443389893,
      "logps/chosen": -83.83250427246094,
      "logps/rejected": -133.15673828125,
      "loss": 0.0995,
      "rewards/accuracies": 0.965624988079071,
      "rewards/chosen": -2.065305709838867,
      "rewards/margins": 4.883781909942627,
      "rewards/rejected": -6.949087619781494,
      "step": 210
    },
    {
      "epoch": 1.99,
      "eval_logits/chosen": -2.5726864337921143,
      "eval_logits/rejected": -2.556718587875366,
      "eval_logps/chosen": -85.11767578125,
      "eval_logps/rejected": -133.85464477539062,
      "eval_loss": 0.12992651760578156,
      "eval_rewards/accuracies": 0.9520547986030579,
      "eval_rewards/chosen": -2.199751138687134,
      "eval_rewards/margins": 4.689663887023926,
      "eval_rewards/rejected": -6.889414310455322,
      "eval_runtime": 288.5007,
      "eval_samples_per_second": 3.036,
      "eval_steps_per_second": 3.036,
      "step": 218
    },
    {
      "epoch": 1.99,
      "step": 218,
      "total_flos": 0.0,
      "train_loss": 0.20939183043777396,
      "train_runtime": 5162.9911,
      "train_samples_per_second": 1.358,
      "train_steps_per_second": 0.042
    }
  ],
  "logging_steps": 10,
  "max_steps": 218,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}