{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.995910949568378,
  "eval_steps": 400,
  "global_step": 137,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03634711494775102,
      "grad_norm": 32.58869576990809,
      "learning_rate": 2.857142857142857e-07,
      "logits/chosen": -11.543076515197754,
      "logits/rejected": -11.31420612335205,
      "logps/chosen": -0.5522164106369019,
      "logps/rejected": -0.5735588669776917,
      "loss": 4.805,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -5.522164344787598,
      "rewards/margins": 0.21342459321022034,
      "rewards/rejected": -5.735588550567627,
      "step": 5
    },
    {
      "epoch": 0.07269422989550205,
      "grad_norm": 34.3946259201971,
      "learning_rate": 5.714285714285714e-07,
      "logits/chosen": -10.83501148223877,
      "logits/rejected": -10.637781143188477,
      "logps/chosen": -0.6140329241752625,
      "logps/rejected": -0.6750808954238892,
      "loss": 4.7363,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -6.140329837799072,
      "rewards/margins": 0.6104797124862671,
      "rewards/rejected": -6.7508087158203125,
      "step": 10
    },
    {
      "epoch": 0.10904134484325306,
      "grad_norm": 41.73056692404051,
      "learning_rate": 7.998695344323425e-07,
      "logits/chosen": -10.648634910583496,
      "logits/rejected": -10.217233657836914,
      "logps/chosen": -0.5939905047416687,
      "logps/rejected": -0.6135314702987671,
      "loss": 4.8553,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -5.939904689788818,
      "rewards/margins": 0.1954106092453003,
      "rewards/rejected": -6.135315418243408,
      "step": 15
    },
    {
      "epoch": 0.1453884597910041,
      "grad_norm": 36.02519008680044,
      "learning_rate": 7.953121695121394e-07,
      "logits/chosen": -10.66395378112793,
      "logits/rejected": -10.348527908325195,
      "logps/chosen": -0.5527507066726685,
      "logps/rejected": -0.5806955099105835,
      "loss": 4.6386,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -5.527507781982422,
      "rewards/margins": 0.27944836020469666,
      "rewards/rejected": -5.806955337524414,
      "step": 20
    },
    {
      "epoch": 0.18173557473875512,
      "grad_norm": 40.07097906112277,
      "learning_rate": 7.843163833184991e-07,
      "logits/chosen": -10.465521812438965,
      "logits/rejected": -10.2030668258667,
      "logps/chosen": -0.5359245538711548,
      "logps/rejected": -0.5927993655204773,
      "loss": 4.5869,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -5.359244346618652,
      "rewards/margins": 0.5687500834465027,
      "rewards/rejected": -5.927994728088379,
      "step": 25
    },
    {
      "epoch": 0.21808268968650613,
      "grad_norm": 34.259554554601195,
      "learning_rate": 7.670612634414511e-07,
      "logits/chosen": -10.547082901000977,
      "logits/rejected": -10.052377700805664,
      "logps/chosen": -0.5360420346260071,
      "logps/rejected": -0.5889440774917603,
      "loss": 4.5463,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -5.360420227050781,
      "rewards/margins": 0.5290209054946899,
      "rewards/rejected": -5.889441013336182,
      "step": 30
    },
    {
      "epoch": 0.25442980463425713,
      "grad_norm": 30.666733170155783,
      "learning_rate": 7.438278427948805e-07,
      "logits/chosen": -10.336666107177734,
      "logits/rejected": -10.034899711608887,
      "logps/chosen": -0.6049242615699768,
      "logps/rejected": -0.6828798055648804,
      "loss": 4.3807,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": -6.0492424964904785,
      "rewards/margins": 0.7795557379722595,
      "rewards/rejected": -6.828797817230225,
      "step": 35
    },
    {
      "epoch": 0.2907769195820082,
      "grad_norm": 37.89264733124544,
      "learning_rate": 7.149945224533862e-07,
      "logits/chosen": -10.575347900390625,
      "logits/rejected": -10.193809509277344,
      "logps/chosen": -0.6202062964439392,
      "logps/rejected": -0.6880999803543091,
      "loss": 4.4737,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -6.20206356048584,
      "rewards/margins": 0.678936779499054,
      "rewards/rejected": -6.880999565124512,
      "step": 40
    },
    {
      "epoch": 0.3271240345297592,
      "grad_norm": 36.08127980591517,
      "learning_rate": 6.810309086608129e-07,
      "logits/chosen": -11.216996192932129,
      "logits/rejected": -10.75574016571045,
      "logps/chosen": -0.6237494349479675,
      "logps/rejected": -0.687574028968811,
      "loss": 4.4138,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -6.237493991851807,
      "rewards/margins": 0.6382461190223694,
      "rewards/rejected": -6.875740051269531,
      "step": 45
    },
    {
      "epoch": 0.36347114947751025,
      "grad_norm": 34.2641539929766,
      "learning_rate": 6.424901643866552e-07,
      "logits/chosen": -11.69914722442627,
      "logits/rejected": -11.246435165405273,
      "logps/chosen": -0.6562625169754028,
      "logps/rejected": -0.7224141359329224,
      "loss": 4.3367,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -6.562624931335449,
      "rewards/margins": 0.6615163087844849,
      "rewards/rejected": -7.2241411209106445,
      "step": 50
    },
    {
      "epoch": 0.39981826442526125,
      "grad_norm": 36.53529303407606,
      "learning_rate": 6e-07,
      "logits/chosen": -11.810195922851562,
      "logits/rejected": -11.363134384155273,
      "logps/chosen": -0.7002917528152466,
      "logps/rejected": -0.778505802154541,
      "loss": 4.3706,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -7.002917289733887,
      "rewards/margins": 0.7821400761604309,
      "rewards/rejected": -7.785058498382568,
      "step": 55
    },
    {
      "epoch": 0.43616537937301225,
      "grad_norm": 35.761908817299414,
      "learning_rate": 5.542524497952543e-07,
      "logits/chosen": -11.559731483459473,
      "logits/rejected": -11.183916091918945,
      "logps/chosen": -0.7286056280136108,
      "logps/rejected": -0.8559492230415344,
      "loss": 4.2217,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -7.2860565185546875,
      "rewards/margins": 1.273435354232788,
      "rewards/rejected": -8.559491157531738,
      "step": 60
    },
    {
      "epoch": 0.4725124943207633,
      "grad_norm": 231.37525001147708,
      "learning_rate": 5.059926008786647e-07,
      "logits/chosen": -11.819048881530762,
      "logits/rejected": -11.407302856445312,
      "logps/chosen": -0.734784722328186,
      "logps/rejected": -0.8366082906723022,
      "loss": 4.1901,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -7.347847938537598,
      "rewards/margins": 1.0182350873947144,
      "rewards/rejected": -8.366083145141602,
      "step": 65
    },
    {
      "epoch": 0.5088596092685143,
      "grad_norm": 51.233160095021326,
      "learning_rate": 4.5600645798745166e-07,
      "logits/chosen": -11.959329605102539,
      "logits/rejected": -11.445878028869629,
      "logps/chosen": -0.7511547803878784,
      "logps/rejected": -0.8529723882675171,
      "loss": 4.2912,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -7.511547088623047,
      "rewards/margins": 1.0181769132614136,
      "rewards/rejected": -8.52972412109375,
      "step": 70
    },
    {
      "epoch": 0.5452067242162654,
      "grad_norm": 59.733803732439824,
      "learning_rate": 4.051081418863895e-07,
      "logits/chosen": -12.409022331237793,
      "logits/rejected": -11.8023042678833,
      "logps/chosen": -0.8404130935668945,
      "logps/rejected": -0.9404484629631042,
      "loss": 4.0174,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -8.404130935668945,
      "rewards/margins": 1.0003544092178345,
      "rewards/rejected": -9.404484748840332,
      "step": 75
    },
    {
      "epoch": 0.5815538391640164,
      "grad_norm": 60.192585986207376,
      "learning_rate": 3.541266298406398e-07,
      "logits/chosen": -12.791674613952637,
      "logits/rejected": -12.519620895385742,
      "logps/chosen": -0.8719121217727661,
      "logps/rejected": -1.016026258468628,
      "loss": 4.1563,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -8.719121932983398,
      "rewards/margins": 1.4411423206329346,
      "rewards/rejected": -10.160263061523438,
      "step": 80
    },
    {
      "epoch": 0.6179009541117674,
      "grad_norm": 63.01387669368681,
      "learning_rate": 3.0389225412181565e-07,
      "logits/chosen": -13.761955261230469,
      "logits/rejected": -13.322006225585938,
      "logps/chosen": -0.8667750358581543,
      "logps/rejected": -1.0224027633666992,
      "loss": 4.0181,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -8.667750358581543,
      "rewards/margins": 1.5562773942947388,
      "rewards/rejected": -10.224027633666992,
      "step": 85
    },
    {
      "epoch": 0.6542480690595184,
      "grad_norm": 58.79093252453101,
      "learning_rate": 2.5522317844515273e-07,
      "logits/chosen": -13.937658309936523,
      "logits/rejected": -13.547172546386719,
      "logps/chosen": -0.9299012422561646,
      "logps/rejected": -1.087494134902954,
      "loss": 3.8751,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -9.29901123046875,
      "rewards/margins": 1.5759302377700806,
      "rewards/rejected": -10.874940872192383,
      "step": 90
    },
    {
      "epoch": 0.6905951840072694,
      "grad_norm": 90.56773515906556,
      "learning_rate": 2.0891207259509476e-07,
      "logits/chosen": -14.276998519897461,
      "logits/rejected": -13.726056098937988,
      "logps/chosen": -0.9327268600463867,
      "logps/rejected": -1.1215860843658447,
      "loss": 3.8327,
      "rewards/accuracies": 0.793749988079071,
      "rewards/chosen": -9.327268600463867,
      "rewards/margins": 1.8885917663574219,
      "rewards/rejected": -11.215860366821289,
      "step": 95
    },
    {
      "epoch": 0.7269422989550205,
      "grad_norm": 68.83137467247752,
      "learning_rate": 1.6571320226872206e-07,
      "logits/chosen": -15.408729553222656,
      "logits/rejected": -14.846705436706543,
      "logps/chosen": -0.994869589805603,
      "logps/rejected": -1.1596581935882568,
      "loss": 3.8063,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -9.948695182800293,
      "rewards/margins": 1.64788818359375,
      "rewards/rejected": -11.596583366394043,
      "step": 100
    },
    {
      "epoch": 0.7632894139027715,
      "grad_norm": 81.97650507265567,
      "learning_rate": 1.2633014440382787e-07,
      "logits/chosen": -15.801300048828125,
      "logits/rejected": -15.365028381347656,
      "logps/chosen": -1.0137146711349487,
      "logps/rejected": -1.173123836517334,
      "loss": 3.7038,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": -10.137146949768066,
      "rewards/margins": 1.5940909385681152,
      "rewards/rejected": -11.731237411499023,
      "step": 105
    },
    {
      "epoch": 0.7996365288505225,
      "grad_norm": 76.7946168478378,
      "learning_rate": 9.14043280712228e-08,
      "logits/chosen": -16.20096206665039,
      "logits/rejected": -15.696843147277832,
      "logps/chosen": -1.1373329162597656,
      "logps/rejected": -1.354536771774292,
      "loss": 3.6719,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -11.373329162597656,
      "rewards/margins": 2.1720378398895264,
      "rewards/rejected": -13.545367240905762,
      "step": 110
    },
    {
      "epoch": 0.8359836437982735,
      "grad_norm": 85.18089495737438,
      "learning_rate": 6.150458756494239e-08,
      "logits/chosen": -16.4575138092041,
      "logits/rejected": -16.271873474121094,
      "logps/chosen": -1.0763649940490723,
      "logps/rejected": -1.265742540359497,
      "loss": 3.621,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -10.763648986816406,
      "rewards/margins": 1.8937755823135376,
      "rewards/rejected": -12.657424926757812,
      "step": 115
    },
    {
      "epoch": 0.8723307587460245,
      "grad_norm": 80.36432752317246,
      "learning_rate": 3.711789783843522e-08,
      "logits/chosen": -16.593286514282227,
      "logits/rejected": -16.234304428100586,
      "logps/chosen": -1.041198492050171,
      "logps/rejected": -1.273951768875122,
      "loss": 3.481,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -10.411985397338867,
      "rewards/margins": 2.327531337738037,
      "rewards/rejected": -12.739517211914062,
      "step": 120
    },
    {
      "epoch": 0.9086778736937755,
      "grad_norm": 81.72943922815173,
      "learning_rate": 1.8641443178027784e-08,
      "logits/chosen": -16.43305206298828,
      "logits/rejected": -16.25759506225586,
      "logps/chosen": -1.142892599105835,
      "logps/rejected": -1.3480346202850342,
      "loss": 3.6377,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -11.428924560546875,
      "rewards/margins": 2.051419734954834,
      "rewards/rejected": -13.480344772338867,
      "step": 125
    },
    {
      "epoch": 0.9450249886415266,
      "grad_norm": 74.32925754994847,
      "learning_rate": 6.376148290617145e-09,
      "logits/chosen": -16.8632869720459,
      "logits/rejected": -16.60898208618164,
      "logps/chosen": -1.1211764812469482,
      "logps/rejected": -1.3164139986038208,
      "loss": 3.4601,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -11.21176528930664,
      "rewards/margins": 1.9523746967315674,
      "rewards/rejected": -13.164140701293945,
      "step": 130
    },
    {
      "epoch": 0.9813721035892776,
      "grad_norm": 82.1235499731324,
      "learning_rate": 5.217771643080127e-10,
      "logits/chosen": -16.644031524658203,
      "logits/rejected": -16.64177131652832,
      "logps/chosen": -1.1249492168426514,
      "logps/rejected": -1.297629714012146,
      "loss": 3.5925,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": -11.249493598937988,
      "rewards/margins": 1.726805329322815,
      "rewards/rejected": -12.976297378540039,
      "step": 135
    },
    {
      "epoch": 0.995910949568378,
      "step": 137,
      "total_flos": 0.0,
      "train_loss": 4.128190910729178,
      "train_runtime": 2318.2745,
      "train_samples_per_second": 7.594,
      "train_steps_per_second": 0.059
    }
  ],
  "logging_steps": 5,
  "max_steps": 137,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}