{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9992429977289932,
  "eval_steps": 100,
  "global_step": 165,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.9411764705882356e-07,
      "logits/chosen": -2.737081289291382,
      "logits/rejected": -2.680964708328247,
      "logps/chosen": -126.38134765625,
      "logps/rejected": -136.25076293945312,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9411764705882355e-06,
      "logits/chosen": -2.731968879699707,
      "logits/rejected": -2.708989143371582,
      "logps/chosen": -118.21139526367188,
      "logps/rejected": -111.94728088378906,
      "loss": 0.6932,
      "rewards/accuracies": 0.4583333432674408,
      "rewards/chosen": -1.655664891586639e-05,
      "rewards/margins": -7.096579793142155e-05,
      "rewards/rejected": 5.4409170843428e-05,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.994932636402032e-06,
      "logits/chosen": -2.744499683380127,
      "logits/rejected": -2.7339892387390137,
      "logps/chosen": -113.14430236816406,
      "logps/rejected": -123.68851470947266,
      "loss": 0.6929,
      "rewards/accuracies": 0.5218750238418579,
      "rewards/chosen": 0.007225497625768185,
      "rewards/margins": 0.0004294503596611321,
      "rewards/rejected": 0.0067960480228066444,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.905416503522124e-06,
      "logits/chosen": -2.772244930267334,
      "logits/rejected": -2.7886576652526855,
      "logps/chosen": -115.32562255859375,
      "logps/rejected": -122.47587585449219,
      "loss": 0.6925,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": 0.024915488436818123,
      "rewards/margins": 0.00023288575175683945,
      "rewards/rejected": 0.024682600051164627,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.707922373336524e-06,
      "logits/chosen": -2.740614175796509,
      "logits/rejected": -2.727870464324951,
      "logps/chosen": -111.22251892089844,
      "logps/rejected": -113.79164123535156,
      "loss": 0.691,
      "rewards/accuracies": 0.596875011920929,
      "rewards/chosen": 0.052265096455812454,
      "rewards/margins": 0.003552838694304228,
      "rewards/rejected": 0.04871225729584694,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.411315662967732e-06,
      "logits/chosen": -2.775801181793213,
      "logits/rejected": -2.7031962871551514,
      "logps/chosen": -109.81254577636719,
      "logps/rejected": -113.3335952758789,
      "loss": 0.6897,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": 0.0849301889538765,
      "rewards/margins": 0.006556935608386993,
      "rewards/rejected": 0.0783732533454895,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.028910905897229e-06,
      "logits/chosen": -2.756106376647949,
      "logits/rejected": -2.7039153575897217,
      "logps/chosen": -110.66825103759766,
      "logps/rejected": -110.25054931640625,
      "loss": 0.689,
      "rewards/accuracies": 0.596875011920929,
      "rewards/chosen": 0.09365645796060562,
      "rewards/margins": 0.008471069857478142,
      "rewards/rejected": 0.08518538624048233,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.577874068920446e-06,
      "logits/chosen": -2.7151947021484375,
      "logits/rejected": -2.706727981567383,
      "logps/chosen": -109.93450927734375,
      "logps/rejected": -113.0154800415039,
      "loss": 0.6885,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": 0.07744868099689484,
      "rewards/margins": 0.004513105843216181,
      "rewards/rejected": 0.0729355737566948,
      "step": 70
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.0784519801008546e-06,
      "logits/chosen": -2.692235231399536,
      "logits/rejected": -2.6538617610931396,
      "logps/chosen": -111.87430572509766,
      "logps/rejected": -115.57984924316406,
      "loss": 0.6867,
      "rewards/accuracies": 0.596875011920929,
      "rewards/chosen": 0.07563529908657074,
      "rewards/margins": 0.015832457691431046,
      "rewards/rejected": 0.059802841395139694,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.553063458334059e-06,
      "logits/chosen": -2.738804340362549,
      "logits/rejected": -2.690701723098755,
      "logps/chosen": -120.409423828125,
      "logps/rejected": -118.25407409667969,
      "loss": 0.6841,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.05362895876169205,
      "rewards/margins": 0.01961613819003105,
      "rewards/rejected": 0.034012824296951294,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.025292943281429e-06,
      "logits/chosen": -2.718660593032837,
      "logits/rejected": -2.7189323902130127,
      "logps/chosen": -108.1891098022461,
      "logps/rejected": -112.934814453125,
      "loss": 0.6837,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": 0.06428461521863937,
      "rewards/margins": 0.0179769154638052,
      "rewards/rejected": 0.04630769044160843,
      "step": 100
    },
    {
      "epoch": 0.61,
      "eval_logits/chosen": -2.7115368843078613,
      "eval_logits/rejected": -2.6278061866760254,
      "eval_logps/chosen": -286.1498718261719,
      "eval_logps/rejected": -263.6365966796875,
      "eval_loss": 0.6740216016769409,
      "eval_rewards/accuracies": 0.6359999775886536,
      "eval_rewards/chosen": 0.008544988930225372,
      "eval_rewards/margins": 0.03750109300017357,
      "eval_rewards/rejected": -0.028956104069948196,
      "eval_runtime": 383.9881,
      "eval_samples_per_second": 5.208,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5188318011445907e-06,
      "logits/chosen": -2.7399725914001465,
      "logits/rejected": -2.710850954055786,
      "logps/chosen": -123.49687194824219,
      "logps/rejected": -121.06239318847656,
      "loss": 0.6835,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": 0.031930722296237946,
      "rewards/margins": 0.025511348620057106,
      "rewards/rejected": 0.00641937181353569,
      "step": 110
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0564148305586296e-06,
      "logits/chosen": -2.733646869659424,
      "logits/rejected": -2.7130093574523926,
      "logps/chosen": -116.14371490478516,
      "logps/rejected": -121.31622314453125,
      "loss": 0.6835,
      "rewards/accuracies": 0.559374988079071,
      "rewards/chosen": 0.04128889739513397,
      "rewards/margins": 0.021482665091753006,
      "rewards/rejected": 0.019806232303380966,
      "step": 120
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.587997083462197e-07,
      "logits/chosen": -2.706106424331665,
      "logits/rejected": -2.683042049407959,
      "logps/chosen": -121.27201080322266,
      "logps/rejected": -124.78038024902344,
      "loss": 0.6784,
      "rewards/accuracies": 0.621874988079071,
      "rewards/chosen": 0.032023753970861435,
      "rewards/margins": 0.03133785352110863,
      "rewards/rejected": 0.0006858977722004056,
      "step": 130
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.438351873250492e-07,
      "logits/chosen": -2.6774215698242188,
      "logits/rejected": -2.66115140914917,
      "logps/chosen": -111.2270736694336,
      "logps/rejected": -116.47705078125,
      "loss": 0.6774,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": 0.032939545810222626,
      "rewards/margins": 0.04166686534881592,
      "rewards/rejected": -0.008727315813302994,
      "step": 140
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.2565987432367032e-07,
      "logits/chosen": -2.7032761573791504,
      "logits/rejected": -2.6723108291625977,
      "logps/chosen": -109.5132064819336,
      "logps/rejected": -114.572509765625,
      "loss": 0.6802,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.037231095135211945,
      "rewards/margins": 0.029751187190413475,
      "rewards/rejected": 0.007479907013475895,
      "step": 150
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4067554877743861e-08,
      "logits/chosen": -2.6688144207000732,
      "logits/rejected": -2.6238021850585938,
      "logps/chosen": -110.0870361328125,
      "logps/rejected": -108.6930160522461,
      "loss": 0.68,
      "rewards/accuracies": 0.5843750238418579,
      "rewards/chosen": 0.050410233438014984,
      "rewards/margins": 0.03611644357442856,
      "rewards/rejected": 0.014293788000941277,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 165,
      "total_flos": 0.0,
      "train_loss": 0.685761218359976,
      "train_runtime": 6660.773,
      "train_samples_per_second": 3.173,
      "train_steps_per_second": 0.025
    }
  ],
  "logging_steps": 10,
  "max_steps": 165,
  "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
}