{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 43.88695656143784,
      "learning_rate": 3.125e-08,
      "logits/chosen": -2.2037925720214844,
      "logits/rejected": -1.493311882019043,
      "logps/chosen": -129.30540466308594,
      "logps/rejected": -155.7332000732422,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "grad_norm": 29.303706464347986,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -2.2856006622314453,
      "logits/rejected": -1.7261626720428467,
      "logps/chosen": -151.80892944335938,
      "logps/rejected": -181.124755859375,
      "loss": 0.682,
      "rewards/accuracies": 0.5833333134651184,
      "rewards/chosen": -0.04080594703555107,
      "rewards/margins": 0.03990985453128815,
      "rewards/rejected": -0.08071580529212952,
      "step": 10
    },
    {
      "epoch": 0.13,
      "grad_norm": 30.19368756466062,
      "learning_rate": 4.989935734988097e-07,
      "logits/chosen": -2.2370126247406006,
      "logits/rejected": -1.4457467794418335,
      "logps/chosen": -218.135009765625,
      "logps/rejected": -258.29510498046875,
      "loss": 0.6503,
      "rewards/accuracies": 0.6156250238418579,
      "rewards/chosen": -0.7151989340782166,
      "rewards/margins": 0.19002024829387665,
      "rewards/rejected": -0.9052190780639648,
      "step": 20
    },
    {
      "epoch": 0.19,
      "grad_norm": 30.744279703525947,
      "learning_rate": 4.877641290737883e-07,
      "logits/chosen": -2.1132819652557373,
      "logits/rejected": -1.3941830396652222,
      "logps/chosen": -199.1055145263672,
      "logps/rejected": -239.34249877929688,
      "loss": 0.6229,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.5117620229721069,
      "rewards/margins": 0.264423668384552,
      "rewards/rejected": -0.7761856913566589,
      "step": 30
    },
    {
      "epoch": 0.26,
      "grad_norm": 26.98732429420661,
      "learning_rate": 4.646121984004665e-07,
      "logits/chosen": -2.2066855430603027,
      "logits/rejected": -1.2292633056640625,
      "logps/chosen": -188.8787841796875,
      "logps/rejected": -265.1071472167969,
      "loss": 0.5738,
      "rewards/accuracies": 0.753125011920929,
      "rewards/chosen": -0.45722055435180664,
      "rewards/margins": 0.5506829619407654,
      "rewards/rejected": -1.0079034566879272,
      "step": 40
    },
    {
      "epoch": 0.32,
      "grad_norm": 24.079733843730505,
      "learning_rate": 4.3069871595684787e-07,
      "logits/chosen": -1.9566110372543335,
      "logits/rejected": -1.0668319463729858,
      "logps/chosen": -214.89682006835938,
      "logps/rejected": -296.9588623046875,
      "loss": 0.5684,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.6292861104011536,
      "rewards/margins": 0.594062089920044,
      "rewards/rejected": -1.2233481407165527,
      "step": 50
    },
    {
      "epoch": 0.38,
      "grad_norm": 33.30951831902919,
      "learning_rate": 3.877242453630256e-07,
      "logits/chosen": -2.0148699283599854,
      "logits/rejected": -1.5277093648910522,
      "logps/chosen": -219.0657958984375,
      "logps/rejected": -274.61724853515625,
      "loss": 0.5746,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.6577062010765076,
      "rewards/margins": 0.4176337718963623,
      "rewards/rejected": -1.075339913368225,
      "step": 60
    },
    {
      "epoch": 0.45,
      "grad_norm": 33.78144466638135,
      "learning_rate": 3.378437060203357e-07,
      "logits/chosen": -1.801613450050354,
      "logits/rejected": -1.3443340063095093,
      "logps/chosen": -202.5223388671875,
      "logps/rejected": -285.18939208984375,
      "loss": 0.5695,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.6388754844665527,
      "rewards/margins": 0.5537781715393066,
      "rewards/rejected": -1.1926535367965698,
      "step": 70
    },
    {
      "epoch": 0.51,
      "grad_norm": 27.821012779285784,
      "learning_rate": 2.8355831645441387e-07,
      "logits/chosen": -1.8170337677001953,
      "logits/rejected": -1.3328382968902588,
      "logps/chosen": -208.6138153076172,
      "logps/rejected": -285.3661804199219,
      "loss": 0.5502,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.6443996429443359,
      "rewards/margins": 0.5184130072593689,
      "rewards/rejected": -1.1628127098083496,
      "step": 80
    },
    {
      "epoch": 0.58,
      "grad_norm": 32.68944554362572,
      "learning_rate": 2.2759017277414164e-07,
      "logits/chosen": -1.6956002712249756,
      "logits/rejected": -0.9904757738113403,
      "logps/chosen": -215.2491912841797,
      "logps/rejected": -314.622314453125,
      "loss": 0.549,
      "rewards/accuracies": 0.7593749761581421,
      "rewards/chosen": -0.688922107219696,
      "rewards/margins": 0.7578561305999756,
      "rewards/rejected": -1.4467782974243164,
      "step": 90
    },
    {
      "epoch": 0.64,
      "grad_norm": 24.987088937796276,
      "learning_rate": 1.7274575140626315e-07,
      "logits/chosen": -1.6817023754119873,
      "logits/rejected": -0.9368717074394226,
      "logps/chosen": -212.26806640625,
      "logps/rejected": -296.8437805175781,
      "loss": 0.5617,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.7037585973739624,
      "rewards/margins": 0.5949375629425049,
      "rewards/rejected": -1.2986961603164673,
      "step": 100
    },
    {
      "epoch": 0.7,
      "grad_norm": 120.0167955365787,
      "learning_rate": 1.2177518064852348e-07,
      "logits/chosen": -1.8496421575546265,
      "logits/rejected": -0.7083995342254639,
      "logps/chosen": -204.84237670898438,
      "logps/rejected": -285.4545593261719,
      "loss": 0.5655,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.6622025370597839,
      "rewards/margins": 0.6420421600341797,
      "rewards/rejected": -1.3042447566986084,
      "step": 110
    },
    {
      "epoch": 0.77,
      "grad_norm": 29.13926345567332,
      "learning_rate": 7.723433775328384e-08,
      "logits/chosen": -1.6169201135635376,
      "logits/rejected": -0.7033487558364868,
      "logps/chosen": -219.37203979492188,
      "logps/rejected": -313.266845703125,
      "loss": 0.5701,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.7050714492797852,
      "rewards/margins": 0.6713167428970337,
      "rewards/rejected": -1.3763883113861084,
      "step": 120
    },
    {
      "epoch": 0.83,
      "grad_norm": 24.14450890897982,
      "learning_rate": 4.1356686569674335e-08,
      "logits/chosen": -1.6877450942993164,
      "logits/rejected": -0.4587131440639496,
      "logps/chosen": -211.0887451171875,
      "logps/rejected": -298.6360168457031,
      "loss": 0.5568,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.6507588028907776,
      "rewards/margins": 0.6841335296630859,
      "rewards/rejected": -1.3348922729492188,
      "step": 130
    },
    {
      "epoch": 0.9,
      "grad_norm": 26.54072268887061,
      "learning_rate": 1.5941282340065697e-08,
      "logits/chosen": -1.6877959966659546,
      "logits/rejected": -1.1048305034637451,
      "logps/chosen": -205.51443481445312,
      "logps/rejected": -284.08270263671875,
      "loss": 0.5438,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.6199182271957397,
      "rewards/margins": 0.6091288328170776,
      "rewards/rejected": -1.2290470600128174,
      "step": 140
    },
    {
      "epoch": 0.96,
      "grad_norm": 28.93586626169144,
      "learning_rate": 2.2625595580163247e-09,
      "logits/chosen": -1.6776434183120728,
      "logits/rejected": -0.7655587792396545,
      "logps/chosen": -212.2307891845703,
      "logps/rejected": -279.22686767578125,
      "loss": 0.5468,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.6327279806137085,
      "rewards/margins": 0.518112063407898,
      "rewards/rejected": -1.150840163230896,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.5795359783447706,
      "train_runtime": 30147.7887,
      "train_samples_per_second": 0.663,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}