{
  "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": -2.7495079040527344,
      "logits/rejected": -2.7064833641052246,
      "logps/chosen": -164.81121826171875,
      "logps/rejected": -156.92835998535156,
      "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": -2.7017762660980225,
      "logits/rejected": -2.6830601692199707,
      "logps/chosen": -139.487548828125,
      "logps/rejected": -143.1040802001953,
      "loss": 0.6932,
      "rewards/accuracies": 0.3958333432674408,
      "rewards/chosen": 0.004722592420876026,
      "rewards/margins": -0.0002141917502740398,
      "rewards/rejected": 0.004936783574521542,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.994932636402032e-06,
      "logits/chosen": -2.735475540161133,
      "logits/rejected": -2.686282157897949,
      "logps/chosen": -130.50253295898438,
      "logps/rejected": -130.809326171875,
      "loss": 0.692,
      "rewards/accuracies": 0.49687498807907104,
      "rewards/chosen": 0.08099536597728729,
      "rewards/margins": 0.001809996203519404,
      "rewards/rejected": 0.0791853740811348,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.905416503522124e-06,
      "logits/chosen": -2.7524123191833496,
      "logits/rejected": -2.7140748500823975,
      "logps/chosen": -123.51778411865234,
      "logps/rejected": -126.41548919677734,
      "loss": 0.6893,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": 0.14591369032859802,
      "rewards/margins": 0.008296089246869087,
      "rewards/rejected": 0.13761760294437408,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.707922373336524e-06,
      "logits/chosen": -2.6890153884887695,
      "logits/rejected": -2.6654117107391357,
      "logps/chosen": -132.10757446289062,
      "logps/rejected": -136.89508056640625,
      "loss": 0.6877,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": 0.09638135135173798,
      "rewards/margins": 0.012408262118697166,
      "rewards/rejected": 0.08397307246923447,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.411315662967732e-06,
      "logits/chosen": -2.6580519676208496,
      "logits/rejected": -2.691502809524536,
      "logps/chosen": -122.78013610839844,
      "logps/rejected": -132.18321228027344,
      "loss": 0.6865,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": 0.1647396981716156,
      "rewards/margins": 0.004126954823732376,
      "rewards/rejected": 0.16061276197433472,
      "step": 50
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.028910905897229e-06,
      "logits/chosen": -2.655439615249634,
      "logits/rejected": -2.6334452629089355,
      "logps/chosen": -124.12255859375,
      "logps/rejected": -132.15664672851562,
      "loss": 0.6837,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.14119209349155426,
      "rewards/margins": 0.026216480880975723,
      "rewards/rejected": 0.11497560888528824,
      "step": 60
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.577874068920446e-06,
      "logits/chosen": -2.6378352642059326,
      "logits/rejected": -2.601242780685425,
      "logps/chosen": -144.02236938476562,
      "logps/rejected": -147.04486083984375,
      "loss": 0.6856,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": -0.013076819479465485,
      "rewards/margins": 0.030333761125802994,
      "rewards/rejected": -0.04341058060526848,
      "step": 70
    },
    {
      "epoch": 0.48,
      "learning_rate": 3.0784519801008546e-06,
      "logits/chosen": -2.613574504852295,
      "logits/rejected": -2.547489643096924,
      "logps/chosen": -128.6609649658203,
      "logps/rejected": -128.91639709472656,
      "loss": 0.6795,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": 0.1631721556186676,
      "rewards/margins": 0.03735139220952988,
      "rewards/rejected": 0.12582075595855713,
      "step": 80
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.553063458334059e-06,
      "logits/chosen": -2.6456406116485596,
      "logits/rejected": -2.6266322135925293,
      "logps/chosen": -124.13682556152344,
      "logps/rejected": -135.78231811523438,
      "loss": 0.6817,
      "rewards/accuracies": 0.5718749761581421,
      "rewards/chosen": 0.15800026059150696,
      "rewards/margins": 0.03171014413237572,
      "rewards/rejected": 0.12629011273384094,
      "step": 90
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.025292943281429e-06,
      "logits/chosen": -2.659860134124756,
      "logits/rejected": -2.62748122215271,
      "logps/chosen": -129.3069305419922,
      "logps/rejected": -131.54519653320312,
      "loss": 0.6816,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": 0.12287310510873795,
      "rewards/margins": 0.0398445650935173,
      "rewards/rejected": 0.08302854001522064,
      "step": 100
    },
    {
      "epoch": 0.61,
      "eval_logits/chosen": -2.6012067794799805,
      "eval_logits/rejected": -2.5080597400665283,
      "eval_logps/chosen": -295.2945556640625,
      "eval_logps/rejected": -276.2926330566406,
      "eval_loss": 0.6669759750366211,
      "eval_rewards/accuracies": 0.6259999871253967,
      "eval_rewards/chosen": -0.0072901868261396885,
      "eval_rewards/margins": 0.05662847310304642,
      "eval_rewards/rejected": -0.06391866505146027,
      "eval_runtime": 383.8552,
      "eval_samples_per_second": 5.21,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5188318011445907e-06,
      "logits/chosen": -2.6604905128479004,
      "logits/rejected": -2.6317429542541504,
      "logps/chosen": -140.4075927734375,
      "logps/rejected": -148.9583282470703,
      "loss": 0.674,
      "rewards/accuracies": 0.609375,
      "rewards/chosen": 0.03068632446229458,
      "rewards/margins": 0.04105384275317192,
      "rewards/rejected": -0.01036751363426447,
      "step": 110
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.0564148305586296e-06,
      "logits/chosen": -2.6386797428131104,
      "logits/rejected": -2.65537428855896,
      "logps/chosen": -133.14720153808594,
      "logps/rejected": -141.28182983398438,
      "loss": 0.6802,
      "rewards/accuracies": 0.578125,
      "rewards/chosen": 0.07068979740142822,
      "rewards/margins": 0.035736724734306335,
      "rewards/rejected": 0.034953076392412186,
      "step": 120
    },
    {
      "epoch": 0.79,
      "learning_rate": 6.587997083462197e-07,
      "logits/chosen": -2.6442322731018066,
      "logits/rejected": -2.62335467338562,
      "logps/chosen": -145.12142944335938,
      "logps/rejected": -146.34092712402344,
      "loss": 0.6763,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.05981719493865967,
      "rewards/margins": 0.03406576067209244,
      "rewards/rejected": 0.02575143240392208,
      "step": 130
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.438351873250492e-07,
      "logits/chosen": -2.591649293899536,
      "logits/rejected": -2.575448513031006,
      "logps/chosen": -136.5642852783203,
      "logps/rejected": -135.15036010742188,
      "loss": 0.6799,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.02418154664337635,
      "rewards/margins": 0.014547958970069885,
      "rewards/rejected": 0.009633589535951614,
      "step": 140
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.2565987432367032e-07,
      "logits/chosen": -2.628605842590332,
      "logits/rejected": -2.6137516498565674,
      "logps/chosen": -131.29632568359375,
      "logps/rejected": -132.41908264160156,
      "loss": 0.681,
      "rewards/accuracies": 0.590624988079071,
      "rewards/chosen": 0.05179730802774429,
      "rewards/margins": 0.021983899176120758,
      "rewards/rejected": 0.029813403263688087,
      "step": 150
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.4067554877743861e-08,
      "logits/chosen": -2.5783798694610596,
      "logits/rejected": -2.5735974311828613,
      "logps/chosen": -126.33128356933594,
      "logps/rejected": -133.07003784179688,
      "loss": 0.6789,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.06782820075750351,
      "rewards/margins": 0.04517916589975357,
      "rewards/rejected": 0.022649036720395088,
      "step": 160
    },
    {
      "epoch": 1.0,
      "step": 165,
      "total_flos": 0.0,
      "train_loss": 0.6834371660694931,
      "train_runtime": 6903.7645,
      "train_samples_per_second": 3.061,
      "train_steps_per_second": 0.024
    }
  ],
  "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
}