{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9992587101556709,
  "eval_steps": 100,
  "global_step": 337,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014825796886582653,
      "grad_norm": 3.8051848965817707,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 0.956,
      "mean_token_accuracy": 0.7389732335227731,
      "step": 5
    },
    {
      "epoch": 0.029651593773165306,
      "grad_norm": 2.3193458934516125,
      "learning_rate": 5.882352941176471e-06,
      "loss": 0.9131,
      "mean_token_accuracy": 0.745571190628745,
      "step": 10
    },
    {
      "epoch": 0.04447739065974796,
      "grad_norm": 1.465130510110654,
      "learning_rate": 8.823529411764707e-06,
      "loss": 0.802,
      "mean_token_accuracy": 0.7669543650793653,
      "step": 15
    },
    {
      "epoch": 0.05930318754633061,
      "grad_norm": 1.1649028225710198,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 0.7568,
      "mean_token_accuracy": 0.7719902319902323,
      "step": 20
    },
    {
      "epoch": 0.07412898443291327,
      "grad_norm": 0.8613760026054987,
      "learning_rate": 1.4705882352941179e-05,
      "loss": 0.7119,
      "mean_token_accuracy": 0.7831028693528694,
      "step": 25
    },
    {
      "epoch": 0.08895478131949593,
      "grad_norm": 0.783478820047515,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 0.6953,
      "mean_token_accuracy": 0.7864308608058606,
      "step": 30
    },
    {
      "epoch": 0.10378057820607858,
      "grad_norm": 0.845141787895257,
      "learning_rate": 1.9999462497359468e-05,
      "loss": 0.6841,
      "mean_token_accuracy": 0.7876358363858365,
      "step": 35
    },
    {
      "epoch": 0.11860637509266123,
      "grad_norm": 0.7223548858937856,
      "learning_rate": 1.9980655971335944e-05,
      "loss": 0.6674,
      "mean_token_accuracy": 0.7906836800335666,
      "step": 40
    },
    {
      "epoch": 0.1334321719792439,
      "grad_norm": 0.8020026503983825,
      "learning_rate": 1.993503206718859e-05,
      "loss": 0.6618,
      "mean_token_accuracy": 0.7926633089133089,
      "step": 45
    },
    {
      "epoch": 0.14825796886582654,
      "grad_norm": 0.7215498275444733,
      "learning_rate": 1.986271337340182e-05,
      "loss": 0.6421,
      "mean_token_accuracy": 0.797378663003663,
      "step": 50
    },
    {
      "epoch": 0.16308376575240918,
      "grad_norm": 0.7852660604263051,
      "learning_rate": 1.976389420563607e-05,
      "loss": 0.646,
      "mean_token_accuracy": 0.796246372287669,
      "step": 55
    },
    {
      "epoch": 0.17790956263899185,
      "grad_norm": 0.7068328201097145,
      "learning_rate": 1.9638840084614182e-05,
      "loss": 0.6349,
      "mean_token_accuracy": 0.7993047924297925,
      "step": 60
    },
    {
      "epoch": 0.1927353595255745,
      "grad_norm": 0.6758929995809626,
      "learning_rate": 1.9487887022684336e-05,
      "loss": 0.6163,
      "mean_token_accuracy": 0.8042696886446887,
      "step": 65
    },
    {
      "epoch": 0.20756115641215717,
      "grad_norm": 0.7716624972063456,
      "learning_rate": 1.9311440620976597e-05,
      "loss": 0.6319,
      "mean_token_accuracy": 0.7993212087850436,
      "step": 70
    },
    {
      "epoch": 0.2223869532987398,
      "grad_norm": 0.6609819990400844,
      "learning_rate": 1.9109974979578852e-05,
      "loss": 0.6356,
      "mean_token_accuracy": 0.798153998778999,
      "step": 75
    },
    {
      "epoch": 0.23721275018532245,
      "grad_norm": 0.7028168784009874,
      "learning_rate": 1.8884031423660492e-05,
      "loss": 0.6337,
      "mean_token_accuracy": 0.7985944050496857,
      "step": 80
    },
    {
      "epoch": 0.2520385470719051,
      "grad_norm": 0.6683430701808383,
      "learning_rate": 1.8634217048966638e-05,
      "loss": 0.6353,
      "mean_token_accuracy": 0.7978753602810393,
      "step": 85
    },
    {
      "epoch": 0.2668643439584878,
      "grad_norm": 0.7314922030102382,
      "learning_rate": 1.836120309059107e-05,
      "loss": 0.6115,
      "mean_token_accuracy": 0.8045245726495727,
      "step": 90
    },
    {
      "epoch": 0.28169014084507044,
      "grad_norm": 0.6365309024889941,
      "learning_rate": 1.8065723119410885e-05,
      "loss": 0.6232,
      "mean_token_accuracy": 0.8015613553113555,
      "step": 95
    },
    {
      "epoch": 0.2965159377316531,
      "grad_norm": 0.6672244460635505,
      "learning_rate": 1.77485710710289e-05,
      "loss": 0.6255,
      "mean_token_accuracy": 0.8006891025641025,
      "step": 100
    },
    {
      "epoch": 0.2965159377316531,
      "eval_loss": 0.6413724422454834,
      "eval_mean_token_accuracy": 0.795018613493309,
      "eval_runtime": 5.4348,
      "eval_samples_per_second": 23.552,
      "eval_steps_per_second": 1.472,
      "step": 100
    },
    {
      "epoch": 0.3113417346182357,
      "grad_norm": 0.75128092844917,
      "learning_rate": 1.741059911251997e-05,
      "loss": 0.6223,
      "mean_token_accuracy": 0.8012797619047619,
      "step": 105
    },
    {
      "epoch": 0.32616753150481836,
      "grad_norm": 0.6526755539743339,
      "learning_rate": 1.7052715352713076e-05,
      "loss": 0.6125,
      "mean_token_accuracy": 0.8042971611721612,
      "step": 110
    },
    {
      "epoch": 0.34099332839140106,
      "grad_norm": 0.5958813027219968,
      "learning_rate": 1.667588140216154e-05,
      "loss": 0.6241,
      "mean_token_accuracy": 0.8009714590964588,
      "step": 115
    },
    {
      "epoch": 0.3558191252779837,
      "grad_norm": 0.6459938604110328,
      "learning_rate": 1.628110978935756e-05,
      "loss": 0.6214,
      "mean_token_accuracy": 0.8017719780219782,
      "step": 120
    },
    {
      "epoch": 0.37064492216456635,
      "grad_norm": 0.7336206637425139,
      "learning_rate": 1.586946124013354e-05,
      "loss": 0.6228,
      "mean_token_accuracy": 0.8008722527472527,
      "step": 125
    },
    {
      "epoch": 0.385470719051149,
      "grad_norm": 0.6924708455089922,
      "learning_rate": 1.5442041827560274e-05,
      "loss": 0.6093,
      "mean_token_accuracy": 0.8049916056166057,
      "step": 130
    },
    {
      "epoch": 0.40029651593773163,
      "grad_norm": 0.6252828305378912,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.6119,
      "mean_token_accuracy": 0.8043566849816848,
      "step": 135
    },
    {
      "epoch": 0.41512231282431433,
      "grad_norm": 0.5543639102481284,
      "learning_rate": 1.4544523495299843e-05,
      "loss": 0.6184,
      "mean_token_accuracy": 0.8021657509157512,
      "step": 140
    },
    {
      "epoch": 0.429948109710897,
      "grad_norm": 0.6721392797389169,
      "learning_rate": 1.4076836149416889e-05,
      "loss": 0.6025,
      "mean_token_accuracy": 0.8070421245421245,
      "step": 145
    },
    {
      "epoch": 0.4447739065974796,
      "grad_norm": 0.7061921312824664,
      "learning_rate": 1.3598194608050011e-05,
      "loss": 0.5859,
      "mean_token_accuracy": 0.8116018009768011,
      "step": 150
    },
    {
      "epoch": 0.45959970348406226,
      "grad_norm": 0.5295415190053663,
      "learning_rate": 1.3109884950114007e-05,
      "loss": 0.6101,
      "mean_token_accuracy": 0.8041674297924299,
      "step": 155
    },
    {
      "epoch": 0.4744255003706449,
      "grad_norm": 0.5162022426859912,
      "learning_rate": 1.2613219232128608e-05,
      "loss": 0.5968,
      "mean_token_accuracy": 0.8079342185592185,
      "step": 160
    },
    {
      "epoch": 0.4892512972572276,
      "grad_norm": 0.5740077528394586,
      "learning_rate": 1.2109531962807333e-05,
      "loss": 0.6052,
      "mean_token_accuracy": 0.8052155664205085,
      "step": 165
    },
    {
      "epoch": 0.5040770941438102,
      "grad_norm": 0.5784618789845658,
      "learning_rate": 1.1600176517318742e-05,
      "loss": 0.598,
      "mean_token_accuracy": 0.8077052808302808,
      "step": 170
    },
    {
      "epoch": 0.5189028910303929,
      "grad_norm": 0.5622095615626541,
      "learning_rate": 1.1086521500854746e-05,
      "loss": 0.5928,
      "mean_token_accuracy": 0.808961385836386,
      "step": 175
    },
    {
      "epoch": 0.5337286879169756,
      "grad_norm": 0.555421977804962,
      "learning_rate": 1.0569947071276847e-05,
      "loss": 0.6079,
      "mean_token_accuracy": 0.8038957570207568,
      "step": 180
    },
    {
      "epoch": 0.5485544848035582,
      "grad_norm": 0.5591254267329707,
      "learning_rate": 1.0051841230721065e-05,
      "loss": 0.5996,
      "mean_token_accuracy": 0.8062667887667889,
      "step": 185
    },
    {
      "epoch": 0.5633802816901409,
      "grad_norm": 0.5480628578430322,
      "learning_rate": 9.533596096125826e-06,
      "loss": 0.584,
      "mean_token_accuracy": 0.812239774114774,
      "step": 190
    },
    {
      "epoch": 0.5782060785767235,
      "grad_norm": 0.5514593108031277,
      "learning_rate": 9.016604158703654e-06,
      "loss": 0.6084,
      "mean_token_accuracy": 0.8041910866910866,
      "step": 195
    },
    {
      "epoch": 0.5930318754633062,
      "grad_norm": 0.5363039052592674,
      "learning_rate": 8.502254542407186e-06,
      "loss": 0.6039,
      "mean_token_accuracy": 0.8054059829059831,
      "step": 200
    },
    {
      "epoch": 0.5930318754633062,
      "eval_loss": 0.6165258288383484,
      "eval_mean_token_accuracy": 0.8018075907910054,
      "eval_runtime": 5.2218,
      "eval_samples_per_second": 24.513,
      "eval_steps_per_second": 1.532,
      "step": 200
    },
    {
      "epoch": 0.6078576723498889,
      "grad_norm": 0.5869561564258384,
      "learning_rate": 7.991929271442817e-06,
      "loss": 0.5845,
      "mean_token_accuracy": 0.810370879120879,
      "step": 205
    },
    {
      "epoch": 0.6226834692364714,
      "grad_norm": 0.5126220770273285,
      "learning_rate": 7.48699955686089e-06,
      "loss": 0.6053,
      "mean_token_accuracy": 0.8049973686472104,
      "step": 210
    },
    {
      "epoch": 0.6375092661230541,
      "grad_norm": 0.5283766104356306,
      "learning_rate": 6.988822112200157e-06,
      "loss": 0.5809,
      "mean_token_accuracy": 0.8124908424908428,
      "step": 215
    },
    {
      "epoch": 0.6523350630096367,
      "grad_norm": 0.5538487666398877,
      "learning_rate": 6.498735508086094e-06,
      "loss": 0.6041,
      "mean_token_accuracy": 0.8046634615384616,
      "step": 220
    },
    {
      "epoch": 0.6671608598962194,
      "grad_norm": 0.502142536336529,
      "learning_rate": 6.018056575578075e-06,
      "loss": 0.5831,
      "mean_token_accuracy": 0.8119062881562883,
      "step": 225
    },
    {
      "epoch": 0.6819866567828021,
      "grad_norm": 0.5468779254684961,
      "learning_rate": 5.548076867929331e-06,
      "loss": 0.5938,
      "mean_token_accuracy": 0.8081074481074481,
      "step": 230
    },
    {
      "epoch": 0.6968124536693847,
      "grad_norm": 0.4985379501273117,
      "learning_rate": 5.090059190266779e-06,
      "loss": 0.5963,
      "mean_token_accuracy": 0.8077899877899876,
      "step": 235
    },
    {
      "epoch": 0.7116382505559674,
      "grad_norm": 0.5151452293487124,
      "learning_rate": 4.645234206515171e-06,
      "loss": 0.5899,
      "mean_token_accuracy": 0.8092720264856231,
      "step": 240
    },
    {
      "epoch": 0.72646404744255,
      "grad_norm": 0.4949458523296326,
      "learning_rate": 4.214797132682597e-06,
      "loss": 0.5921,
      "mean_token_accuracy": 0.8082945650477681,
      "step": 245
    },
    {
      "epoch": 0.7412898443291327,
      "grad_norm": 0.47351353980603134,
      "learning_rate": 3.799904525392251e-06,
      "loss": 0.5906,
      "mean_token_accuracy": 0.8085080891330888,
      "step": 250
    },
    {
      "epoch": 0.7561156412157154,
      "grad_norm": 0.48563113738070895,
      "learning_rate": 3.401671174289469e-06,
      "loss": 0.5822,
      "mean_token_accuracy": 0.8119319261486757,
      "step": 255
    },
    {
      "epoch": 0.770941438102298,
      "grad_norm": 0.4624658793689265,
      "learning_rate": 3.021167106673928e-06,
      "loss": 0.5895,
      "mean_token_accuracy": 0.8094238400488403,
      "step": 260
    },
    {
      "epoch": 0.7857672349888807,
      "grad_norm": 0.4887471221196979,
      "learning_rate": 2.6594147124053983e-06,
      "loss": 0.5933,
      "mean_token_accuracy": 0.8082280219780221,
      "step": 265
    },
    {
      "epoch": 0.8005930318754633,
      "grad_norm": 0.4755865942641489,
      "learning_rate": 2.317385996808195e-06,
      "loss": 0.5796,
      "mean_token_accuracy": 0.8127861721611721,
      "step": 270
    },
    {
      "epoch": 0.815418828762046,
      "grad_norm": 0.47035132329171114,
      "learning_rate": 1.9959999689556407e-06,
      "loss": 0.5967,
      "mean_token_accuracy": 0.8072924297924298,
      "step": 275
    },
    {
      "epoch": 0.8302446256486287,
      "grad_norm": 0.472860512590234,
      "learning_rate": 1.6961201723520248e-06,
      "loss": 0.578,
      "mean_token_accuracy": 0.8127960927960929,
      "step": 280
    },
    {
      "epoch": 0.8450704225352113,
      "grad_norm": 0.4614371078147505,
      "learning_rate": 1.4185523646469822e-06,
      "loss": 0.5901,
      "mean_token_accuracy": 0.8086108107426412,
      "step": 285
    },
    {
      "epoch": 0.859896219421794,
      "grad_norm": 0.43931896763540745,
      "learning_rate": 1.1640423526166987e-06,
      "loss": 0.5834,
      "mean_token_accuracy": 0.8111981074481076,
      "step": 290
    },
    {
      "epoch": 0.8747220163083765,
      "grad_norm": 0.4456895234302146,
      "learning_rate": 9.332739882292752e-07,
      "loss": 0.5842,
      "mean_token_accuracy": 0.8116254578754578,
      "step": 295
    },
    {
      "epoch": 0.8895478131949592,
      "grad_norm": 0.43426061796264454,
      "learning_rate": 7.268673311786378e-07,
      "loss": 0.572,
      "mean_token_accuracy": 0.8145253357753358,
      "step": 300
    },
    {
      "epoch": 0.8895478131949592,
      "eval_loss": 0.6066814064979553,
      "eval_mean_token_accuracy": 0.8039550018326881,
      "eval_runtime": 5.5504,
      "eval_samples_per_second": 23.061,
      "eval_steps_per_second": 1.441,
      "step": 300
    },
    {
      "epoch": 0.9043736100815419,
      "grad_norm": 0.45927392759405916,
      "learning_rate": 5.453769828241872e-07,
      "loss": 0.5848,
      "mean_token_accuracy": 0.8108882783882784,
      "step": 305
    },
    {
      "epoch": 0.9191994069681245,
      "grad_norm": 0.44015278909822164,
      "learning_rate": 3.8929059601275463e-07,
      "loss": 0.5883,
      "mean_token_accuracy": 0.8098282967032967,
      "step": 310
    },
    {
      "epoch": 0.9340252038547072,
      "grad_norm": 0.46373639838618946,
      "learning_rate": 2.5902756478688674e-07,
      "loss": 0.5893,
      "mean_token_accuracy": 0.8085653235653236,
      "step": 315
    },
    {
      "epoch": 0.9488510007412898,
      "grad_norm": 0.4453169356064979,
      "learning_rate": 1.5493789750014032e-07,
      "loss": 0.5934,
      "mean_token_accuracy": 0.8079464285714286,
      "step": 320
    },
    {
      "epoch": 0.9636767976278725,
      "grad_norm": 0.44080298077757607,
      "learning_rate": 7.730127636723539e-08,
      "loss": 0.5835,
      "mean_token_accuracy": 0.8108723706229716,
      "step": 325
    },
    {
      "epoch": 0.9785025945144552,
      "grad_norm": 0.42873433987601206,
      "learning_rate": 2.6326305976001054e-08,
      "loss": 0.5817,
      "mean_token_accuracy": 0.8121665140415143,
      "step": 330
    },
    {
      "epoch": 0.9933283914010378,
      "grad_norm": 0.44474948977305667,
      "learning_rate": 2.149952780321485e-09,
      "loss": 0.5933,
      "mean_token_accuracy": 0.8080050057957257,
      "step": 335
    },
    {
      "epoch": 0.9992587101556709,
      "mean_token_accuracy": 0.8132173382173382,
      "step": 337,
      "total_flos": 176608216350720.0,
      "train_loss": 0.6238384791580789,
      "train_runtime": 3258.6411,
      "train_samples_per_second": 6.624,
      "train_steps_per_second": 0.103
    }
  ],
  "logging_steps": 5,
  "max_steps": 337,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 176608216350720.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}