{
  "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": 73.50294131392823,
      "learning_rate": 2.857142857142857e-07,
      "logits/chosen": -10.441385269165039,
      "logits/rejected": -9.520084381103516,
      "logps/chosen": -0.9301475286483765,
      "logps/rejected": -1.348732590675354,
      "loss": 3.7321,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -9.301475524902344,
      "rewards/margins": 4.185849189758301,
      "rewards/rejected": -13.487324714660645,
      "step": 5
    },
    {
      "epoch": 0.07269422989550205,
      "grad_norm": 49.41339223972053,
      "learning_rate": 5.714285714285714e-07,
      "logits/chosen": -10.112415313720703,
      "logits/rejected": -9.321776390075684,
      "logps/chosen": -1.0208427906036377,
      "logps/rejected": -1.3525886535644531,
      "loss": 3.5537,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -10.208429336547852,
      "rewards/margins": 3.317457914352417,
      "rewards/rejected": -13.525888442993164,
      "step": 10
    },
    {
      "epoch": 0.10904134484325306,
      "grad_norm": 56.21490730930501,
      "learning_rate": 7.998695344323425e-07,
      "logits/chosen": -10.125561714172363,
      "logits/rejected": -9.403921127319336,
      "logps/chosen": -1.0114651918411255,
      "logps/rejected": -1.4378917217254639,
      "loss": 3.688,
      "rewards/accuracies": 0.8187500238418579,
      "rewards/chosen": -10.114651679992676,
      "rewards/margins": 4.264266014099121,
      "rewards/rejected": -14.37891674041748,
      "step": 15
    },
    {
      "epoch": 0.1453884597910041,
      "grad_norm": 49.2098026266453,
      "learning_rate": 7.953121695121394e-07,
      "logits/chosen": -9.852151870727539,
      "logits/rejected": -8.917947769165039,
      "logps/chosen": -0.917424201965332,
      "logps/rejected": -1.2778595685958862,
      "loss": 3.3675,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -9.17424201965332,
      "rewards/margins": 3.6043541431427,
      "rewards/rejected": -12.778596878051758,
      "step": 20
    },
    {
      "epoch": 0.18173557473875512,
      "grad_norm": 65.87909071485771,
      "learning_rate": 7.843163833184991e-07,
      "logits/chosen": -10.04487419128418,
      "logits/rejected": -9.234810829162598,
      "logps/chosen": -1.029579520225525,
      "logps/rejected": -1.2623463869094849,
      "loss": 3.3763,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -10.295795440673828,
      "rewards/margins": 2.3276684284210205,
      "rewards/rejected": -12.623464584350586,
      "step": 25
    },
    {
      "epoch": 0.21808268968650613,
      "grad_norm": 53.32133564579517,
      "learning_rate": 7.670612634414511e-07,
      "logits/chosen": -9.879186630249023,
      "logits/rejected": -8.854208946228027,
      "logps/chosen": -0.9565294981002808,
      "logps/rejected": -1.3326746225357056,
      "loss": 3.2331,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -9.565296173095703,
      "rewards/margins": 3.7614502906799316,
      "rewards/rejected": -13.326745986938477,
      "step": 30
    },
    {
      "epoch": 0.25442980463425713,
      "grad_norm": 47.904474253960814,
      "learning_rate": 7.438278427948805e-07,
      "logits/chosen": -10.390070915222168,
      "logits/rejected": -9.588186264038086,
      "logps/chosen": -0.997881293296814,
      "logps/rejected": -1.4048808813095093,
      "loss": 3.1431,
      "rewards/accuracies": 0.84375,
      "rewards/chosen": -9.978813171386719,
      "rewards/margins": 4.069996356964111,
      "rewards/rejected": -14.048810005187988,
      "step": 35
    },
    {
      "epoch": 0.2907769195820082,
      "grad_norm": 52.83351276305546,
      "learning_rate": 7.149945224533862e-07,
      "logits/chosen": -10.330121040344238,
      "logits/rejected": -9.436741828918457,
      "logps/chosen": -1.2074191570281982,
      "logps/rejected": -1.6147959232330322,
      "loss": 3.079,
      "rewards/accuracies": 0.8062499761581421,
      "rewards/chosen": -12.074190139770508,
      "rewards/margins": 4.07376766204834,
      "rewards/rejected": -16.147958755493164,
      "step": 40
    },
    {
      "epoch": 0.3271240345297592,
      "grad_norm": 50.87463012185036,
      "learning_rate": 6.810309086608129e-07,
      "logits/chosen": -10.859004974365234,
      "logits/rejected": -9.854225158691406,
      "logps/chosen": -1.2316290140151978,
      "logps/rejected": -1.8107597827911377,
      "loss": 2.9332,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": -12.316289901733398,
      "rewards/margins": 5.79130744934082,
      "rewards/rejected": -18.10759735107422,
      "step": 45
    },
    {
      "epoch": 0.36347114947751025,
      "grad_norm": 96.65968234360786,
      "learning_rate": 6.424901643866552e-07,
      "logits/chosen": -11.214125633239746,
      "logits/rejected": -10.290956497192383,
      "logps/chosen": -1.4425190687179565,
      "logps/rejected": -1.994137167930603,
      "loss": 2.9568,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": -14.425191879272461,
      "rewards/margins": 5.516180515289307,
      "rewards/rejected": -19.941370010375977,
      "step": 50
    },
    {
      "epoch": 0.39981826442526125,
      "grad_norm": 76.49645663435781,
      "learning_rate": 6e-07,
      "logits/chosen": -11.809680938720703,
      "logits/rejected": -10.86604118347168,
      "logps/chosen": -1.511315107345581,
      "logps/rejected": -2.095630168914795,
      "loss": 2.7726,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": -15.113149642944336,
      "rewards/margins": 5.843151092529297,
      "rewards/rejected": -20.956302642822266,
      "step": 55
    },
    {
      "epoch": 0.43616537937301225,
      "grad_norm": 52.07543926650269,
      "learning_rate": 5.542524497952543e-07,
      "logits/chosen": -12.52156925201416,
      "logits/rejected": -11.367462158203125,
      "logps/chosen": -1.417328119277954,
      "logps/rejected": -1.9500234127044678,
      "loss": 2.7633,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -14.1732816696167,
      "rewards/margins": 5.3269548416137695,
      "rewards/rejected": -19.500234603881836,
      "step": 60
    },
    {
      "epoch": 0.4725124943207633,
      "grad_norm": 86.00453174970828,
      "learning_rate": 5.059926008786647e-07,
      "logits/chosen": -13.00245189666748,
      "logits/rejected": -11.82313346862793,
      "logps/chosen": -1.3697974681854248,
      "logps/rejected": -1.979061484336853,
      "loss": 2.6871,
      "rewards/accuracies": 0.893750011920929,
      "rewards/chosen": -13.697975158691406,
      "rewards/margins": 6.092639923095703,
      "rewards/rejected": -19.79061508178711,
      "step": 65
    },
    {
      "epoch": 0.5088596092685143,
      "grad_norm": 74.14149953928305,
      "learning_rate": 4.5600645798745166e-07,
      "logits/chosen": -13.302050590515137,
      "logits/rejected": -12.503473281860352,
      "logps/chosen": -1.3789888620376587,
      "logps/rejected": -1.9646472930908203,
      "loss": 2.6182,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -13.789888381958008,
      "rewards/margins": 5.856583595275879,
      "rewards/rejected": -19.646472930908203,
      "step": 70
    },
    {
      "epoch": 0.5452067242162654,
      "grad_norm": 82.09661278070892,
      "learning_rate": 4.051081418863895e-07,
      "logits/chosen": -14.177574157714844,
      "logits/rejected": -12.784692764282227,
      "logps/chosen": -1.523525595664978,
      "logps/rejected": -2.243549346923828,
      "loss": 2.4206,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -15.235257148742676,
      "rewards/margins": 7.200235843658447,
      "rewards/rejected": -22.43549156188965,
      "step": 75
    },
    {
      "epoch": 0.5815538391640164,
      "grad_norm": 78.17477380695215,
      "learning_rate": 3.541266298406398e-07,
      "logits/chosen": -13.6174955368042,
      "logits/rejected": -12.198015213012695,
      "logps/chosen": -1.413405418395996,
      "logps/rejected": -2.035590648651123,
      "loss": 2.3449,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -14.134053230285645,
      "rewards/margins": 6.221851825714111,
      "rewards/rejected": -20.355907440185547,
      "step": 80
    },
    {
      "epoch": 0.6179009541117674,
      "grad_norm": 161.18580193098623,
      "learning_rate": 3.0389225412181565e-07,
      "logits/chosen": -14.64953899383545,
      "logits/rejected": -13.436065673828125,
      "logps/chosen": -1.577237844467163,
      "logps/rejected": -2.26001238822937,
      "loss": 2.3647,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -15.772378921508789,
      "rewards/margins": 6.827744483947754,
      "rewards/rejected": -22.60012435913086,
      "step": 85
    },
    {
      "epoch": 0.6542480690595184,
      "grad_norm": 110.64212883100511,
      "learning_rate": 2.5522317844515273e-07,
      "logits/chosen": -15.540824890136719,
      "logits/rejected": -14.182299613952637,
      "logps/chosen": -1.7375078201293945,
      "logps/rejected": -2.5354347229003906,
      "loss": 2.2847,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -17.375080108642578,
      "rewards/margins": 7.979270935058594,
      "rewards/rejected": -25.35434913635254,
      "step": 90
    },
    {
      "epoch": 0.6905951840072694,
      "grad_norm": 94.27522984793,
      "learning_rate": 2.0891207259509476e-07,
      "logits/chosen": -16.05794906616211,
      "logits/rejected": -14.873092651367188,
      "logps/chosen": -1.8003565073013306,
      "logps/rejected": -2.508544683456421,
      "loss": 2.1888,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -18.003564834594727,
      "rewards/margins": 7.081883907318115,
      "rewards/rejected": -25.085451126098633,
      "step": 95
    },
    {
      "epoch": 0.7269422989550205,
      "grad_norm": 108.23339503673007,
      "learning_rate": 1.6571320226872206e-07,
      "logits/chosen": -15.491415023803711,
      "logits/rejected": -14.448579788208008,
      "logps/chosen": -1.7869226932525635,
      "logps/rejected": -2.6193735599517822,
      "loss": 2.235,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -17.869226455688477,
      "rewards/margins": 8.324507713317871,
      "rewards/rejected": -26.193735122680664,
      "step": 100
    },
    {
      "epoch": 0.7632894139027715,
      "grad_norm": 98.94867860284775,
      "learning_rate": 1.2633014440382787e-07,
      "logits/chosen": -16.772117614746094,
      "logits/rejected": -15.896406173706055,
      "logps/chosen": -1.7917287349700928,
      "logps/rejected": -2.5536069869995117,
      "loss": 1.8876,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": -17.917287826538086,
      "rewards/margins": 7.61877965927124,
      "rewards/rejected": -25.536067962646484,
      "step": 105
    },
    {
      "epoch": 0.7996365288505225,
      "grad_norm": 116.37014336849657,
      "learning_rate": 9.14043280712228e-08,
      "logits/chosen": -16.703330993652344,
      "logits/rejected": -15.419268608093262,
      "logps/chosen": -1.8466812372207642,
      "logps/rejected": -2.60697340965271,
      "loss": 2.0518,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -18.466812133789062,
      "rewards/margins": 7.602923393249512,
      "rewards/rejected": -26.06973648071289,
      "step": 110
    },
    {
      "epoch": 0.8359836437982735,
      "grad_norm": 88.56805919880858,
      "learning_rate": 6.150458756494239e-08,
      "logits/chosen": -17.041057586669922,
      "logits/rejected": -15.86296558380127,
      "logps/chosen": -1.9040367603302002,
      "logps/rejected": -2.855522632598877,
      "loss": 1.8644,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": -19.040367126464844,
      "rewards/margins": 9.514856338500977,
      "rewards/rejected": -28.555225372314453,
      "step": 115
    },
    {
      "epoch": 0.8723307587460245,
      "grad_norm": 119.49778162125348,
      "learning_rate": 3.711789783843522e-08,
      "logits/chosen": -16.915624618530273,
      "logits/rejected": -15.977205276489258,
      "logps/chosen": -1.9863815307617188,
      "logps/rejected": -2.819209575653076,
      "loss": 2.0238,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": -19.863813400268555,
      "rewards/margins": 8.328282356262207,
      "rewards/rejected": -28.192096710205078,
      "step": 120
    },
    {
      "epoch": 0.9086778736937755,
      "grad_norm": 100.49399716119115,
      "learning_rate": 1.8641443178027784e-08,
      "logits/chosen": -17.482728958129883,
      "logits/rejected": -16.11745262145996,
      "logps/chosen": -1.8876680135726929,
      "logps/rejected": -2.652832269668579,
      "loss": 2.0136,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": -18.876678466796875,
      "rewards/margins": 7.6516432762146,
      "rewards/rejected": -26.528324127197266,
      "step": 125
    },
    {
      "epoch": 0.9450249886415266,
      "grad_norm": 78.59767084378562,
      "learning_rate": 6.376148290617145e-09,
      "logits/chosen": -16.445621490478516,
      "logits/rejected": -15.228750228881836,
      "logps/chosen": -1.947279691696167,
      "logps/rejected": -2.799992084503174,
      "loss": 1.9014,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": -19.472797393798828,
      "rewards/margins": 8.527124404907227,
      "rewards/rejected": -27.999919891357422,
      "step": 130
    },
    {
      "epoch": 0.9813721035892776,
      "grad_norm": 106.19866039415072,
      "learning_rate": 5.217771643080127e-10,
      "logits/chosen": -17.454696655273438,
      "logits/rejected": -16.37453842163086,
      "logps/chosen": -1.9225728511810303,
      "logps/rejected": -2.7656478881835938,
      "loss": 1.9868,
      "rewards/accuracies": 0.875,
      "rewards/chosen": -19.225727081298828,
      "rewards/margins": 8.430750846862793,
      "rewards/rejected": -27.656478881835938,
      "step": 135
    },
    {
      "epoch": 0.995910949568378,
      "step": 137,
      "total_flos": 0.0,
      "train_loss": 2.6376401560149922,
      "train_runtime": 2565.593,
      "train_samples_per_second": 6.862,
      "train_steps_per_second": 0.053
    }
  ],
  "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
}