{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5106725734997987,
  "eval_steps": 500,
  "global_step": 317,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008054772452678213,
      "grad_norm": 16.637405395507812,
      "learning_rate": 2.162496240601504e-06,
      "loss": 0.3748,
      "mean_token_accuracy": 0.880865803360939,
      "num_tokens": 744549.0,
      "step": 5
    },
    {
      "epoch": 0.016109544905356425,
      "grad_norm": 3.243762493133545,
      "learning_rate": 4.865616541353383e-06,
      "loss": 0.1971,
      "mean_token_accuracy": 0.9505918800830842,
      "num_tokens": 1486759.0,
      "step": 10
    },
    {
      "epoch": 0.024164317358034634,
      "grad_norm": 1.319171667098999,
      "learning_rate": 7.568736842105265e-06,
      "loss": 0.1227,
      "mean_token_accuracy": 0.9649254441261291,
      "num_tokens": 2246370.0,
      "step": 15
    },
    {
      "epoch": 0.03221908981071285,
      "grad_norm": 0.8977211713790894,
      "learning_rate": 1.0271857142857143e-05,
      "loss": 0.107,
      "mean_token_accuracy": 0.9681479424238205,
      "num_tokens": 3004971.0,
      "step": 20
    },
    {
      "epoch": 0.04027386226339106,
      "grad_norm": 0.8303173780441284,
      "learning_rate": 1.2974977443609024e-05,
      "loss": 0.0905,
      "mean_token_accuracy": 0.9723703324794769,
      "num_tokens": 3767909.0,
      "step": 25
    },
    {
      "epoch": 0.04832863471606927,
      "grad_norm": 0.6402727365493774,
      "learning_rate": 1.5678097744360906e-05,
      "loss": 0.0966,
      "mean_token_accuracy": 0.9714806169271469,
      "num_tokens": 4528962.0,
      "step": 30
    },
    {
      "epoch": 0.056383407168747485,
      "grad_norm": 0.8155502676963806,
      "learning_rate": 1.8381218045112784e-05,
      "loss": 0.1043,
      "mean_token_accuracy": 0.9704061895608902,
      "num_tokens": 5288458.0,
      "step": 35
    },
    {
      "epoch": 0.0644381796214257,
      "grad_norm": 0.5555161833763123,
      "learning_rate": 1.8920205077736123e-05,
      "loss": 0.0986,
      "mean_token_accuracy": 0.9709503561258316,
      "num_tokens": 6038144.0,
      "step": 40
    },
    {
      "epoch": 0.0724929520741039,
      "grad_norm": 0.6285075545310974,
      "learning_rate": 1.8913555947953106e-05,
      "loss": 0.0851,
      "mean_token_accuracy": 0.9755407810211182,
      "num_tokens": 6795467.0,
      "step": 45
    },
    {
      "epoch": 0.08054772452678212,
      "grad_norm": 0.7997414469718933,
      "learning_rate": 1.8901797191659823e-05,
      "loss": 0.0988,
      "mean_token_accuracy": 0.9719499230384827,
      "num_tokens": 7548219.0,
      "step": 50
    },
    {
      "epoch": 0.08860249697946034,
      "grad_norm": 0.8819418549537659,
      "learning_rate": 1.888493728626881e-05,
      "loss": 0.0978,
      "mean_token_accuracy": 0.9702369779348373,
      "num_tokens": 8291120.0,
      "step": 55
    },
    {
      "epoch": 0.09665726943213854,
      "grad_norm": 0.3800303041934967,
      "learning_rate": 1.8862988386838794e-05,
      "loss": 0.0792,
      "mean_token_accuracy": 0.9756836414337158,
      "num_tokens": 9055930.0,
      "step": 60
    },
    {
      "epoch": 0.10471204188481675,
      "grad_norm": 0.6624178886413574,
      "learning_rate": 1.883596631731158e-05,
      "loss": 0.0699,
      "mean_token_accuracy": 0.9781949192285537,
      "num_tokens": 9807045.0,
      "step": 65
    },
    {
      "epoch": 0.11276681433749497,
      "grad_norm": 0.7004092335700989,
      "learning_rate": 1.8803890559103844e-05,
      "loss": 0.0878,
      "mean_token_accuracy": 0.9751201689243316,
      "num_tokens": 10549678.0,
      "step": 70
    },
    {
      "epoch": 0.12082158679017317,
      "grad_norm": 0.5261930227279663,
      "learning_rate": 1.8766784237062124e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.9761551439762115,
      "num_tokens": 11305953.0,
      "step": 75
    },
    {
      "epoch": 0.1288763592428514,
      "grad_norm": 0.5900943875312805,
      "learning_rate": 1.8724674102791083e-05,
      "loss": 0.0795,
      "mean_token_accuracy": 0.9775821447372437,
      "num_tokens": 12057281.0,
      "step": 80
    },
    {
      "epoch": 0.1369311316955296,
      "grad_norm": 0.5025807023048401,
      "learning_rate": 1.8677590515367078e-05,
      "loss": 0.0981,
      "mean_token_accuracy": 0.9728454887866974,
      "num_tokens": 12811732.0,
      "step": 85
    },
    {
      "epoch": 0.1449859041482078,
      "grad_norm": 0.4680890738964081,
      "learning_rate": 1.8625567419450963e-05,
      "loss": 0.0644,
      "mean_token_accuracy": 0.9785055041313171,
      "num_tokens": 13557606.0,
      "step": 90
    },
    {
      "epoch": 0.15304067660088602,
      "grad_norm": 0.4929722249507904,
      "learning_rate": 1.8568642320815828e-05,
      "loss": 0.0744,
      "mean_token_accuracy": 0.9765197962522507,
      "num_tokens": 14312535.0,
      "step": 95
    },
    {
      "epoch": 0.16109544905356424,
      "grad_norm": 0.6782490611076355,
      "learning_rate": 1.850685625930747e-05,
      "loss": 0.0841,
      "mean_token_accuracy": 0.9740508913993835,
      "num_tokens": 15068862.0,
      "step": 100
    },
    {
      "epoch": 0.16915022150624245,
      "grad_norm": 0.6594260334968567,
      "learning_rate": 1.844025377925687e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9786866426467895,
      "num_tokens": 15823311.0,
      "step": 105
    },
    {
      "epoch": 0.17720499395892067,
      "grad_norm": 0.4855263829231262,
      "learning_rate": 1.8368882897366254e-05,
      "loss": 0.1069,
      "mean_token_accuracy": 0.9722075581550598,
      "num_tokens": 16578166.0,
      "step": 110
    },
    {
      "epoch": 0.18525976641159886,
      "grad_norm": 0.32182013988494873,
      "learning_rate": 1.8292795068091676e-05,
      "loss": 0.0899,
      "mean_token_accuracy": 0.9738094180822372,
      "num_tokens": 17336439.0,
      "step": 115
    },
    {
      "epoch": 0.19331453886427707,
      "grad_norm": 0.5268917083740234,
      "learning_rate": 1.82120451465472e-05,
      "loss": 0.0883,
      "mean_token_accuracy": 0.9742991536855697,
      "num_tokens": 18083985.0,
      "step": 120
    },
    {
      "epoch": 0.2013693113169553,
      "grad_norm": 0.642301082611084,
      "learning_rate": 1.8126691348957424e-05,
      "loss": 0.0739,
      "mean_token_accuracy": 0.9780176162719727,
      "num_tokens": 18833940.0,
      "step": 125
    },
    {
      "epoch": 0.2094240837696335,
      "grad_norm": 0.713036835193634,
      "learning_rate": 1.8036795210686777e-05,
      "loss": 0.0763,
      "mean_token_accuracy": 0.976913234591484,
      "num_tokens": 19581823.0,
      "step": 130
    },
    {
      "epoch": 0.21747885622231172,
      "grad_norm": 0.5190884470939636,
      "learning_rate": 1.7942421541875942e-05,
      "loss": 0.0966,
      "mean_token_accuracy": 0.9732217401266098,
      "num_tokens": 20338712.0,
      "step": 135
    },
    {
      "epoch": 0.22553362867498994,
      "grad_norm": 0.3933056890964508,
      "learning_rate": 1.7843638380717322e-05,
      "loss": 0.0736,
      "mean_token_accuracy": 0.977441081404686,
      "num_tokens": 21089716.0,
      "step": 140
    },
    {
      "epoch": 0.23358840112766815,
      "grad_norm": 0.5459633469581604,
      "learning_rate": 1.774051694440328e-05,
      "loss": 0.0778,
      "mean_token_accuracy": 0.9770717680454254,
      "num_tokens": 21837845.0,
      "step": 145
    },
    {
      "epoch": 0.24164317358034634,
      "grad_norm": 0.48511502146720886,
      "learning_rate": 1.763313157778247e-05,
      "loss": 0.0742,
      "mean_token_accuracy": 0.9780830204486847,
      "num_tokens": 22585463.0,
      "step": 150
    },
    {
      "epoch": 0.24969794603302456,
      "grad_norm": 0.5405490398406982,
      "learning_rate": 1.752155969976134e-05,
      "loss": 0.0636,
      "mean_token_accuracy": 0.9794131308794022,
      "num_tokens": 23353228.0,
      "step": 155
    },
    {
      "epoch": 0.2577527184857028,
      "grad_norm": 0.5847693681716919,
      "learning_rate": 1.740588174748935e-05,
      "loss": 0.1,
      "mean_token_accuracy": 0.9715630292892456,
      "num_tokens": 24104796.0,
      "step": 160
    },
    {
      "epoch": 0.26580749093838096,
      "grad_norm": 0.6486185193061829,
      "learning_rate": 1.728618111836827e-05,
      "loss": 0.0826,
      "mean_token_accuracy": 0.9757199078798294,
      "num_tokens": 24856010.0,
      "step": 165
    },
    {
      "epoch": 0.2738622633910592,
      "grad_norm": 0.30762332677841187,
      "learning_rate": 1.7162544109927246e-05,
      "loss": 0.09,
      "mean_token_accuracy": 0.9746719419956207,
      "num_tokens": 25605837.0,
      "step": 170
    },
    {
      "epoch": 0.2819170358437374,
      "grad_norm": 0.5456633567810059,
      "learning_rate": 1.7035059857607073e-05,
      "loss": 0.0794,
      "mean_token_accuracy": 0.9768573999404907,
      "num_tokens": 26358650.0,
      "step": 175
    },
    {
      "epoch": 0.2899718082964156,
      "grad_norm": 0.5899956226348877,
      "learning_rate": 1.6903820270498478e-05,
      "loss": 0.0748,
      "mean_token_accuracy": 0.9787687987089158,
      "num_tokens": 27104563.0,
      "step": 180
    },
    {
      "epoch": 0.29802658074909383,
      "grad_norm": 0.5380017161369324,
      "learning_rate": 1.6768919965080727e-05,
      "loss": 0.0803,
      "mean_token_accuracy": 0.9778407573699951,
      "num_tokens": 27851642.0,
      "step": 185
    },
    {
      "epoch": 0.30608135320177204,
      "grad_norm": 0.579448938369751,
      "learning_rate": 1.6630456197008404e-05,
      "loss": 0.0726,
      "mean_token_accuracy": 0.9781927406787873,
      "num_tokens": 28597261.0,
      "step": 190
    },
    {
      "epoch": 0.31413612565445026,
      "grad_norm": 0.5752624273300171,
      "learning_rate": 1.6488528790995466e-05,
      "loss": 0.0854,
      "mean_token_accuracy": 0.9745152711868286,
      "num_tokens": 29362166.0,
      "step": 195
    },
    {
      "epoch": 0.3221908981071285,
      "grad_norm": 0.5112693905830383,
      "learning_rate": 1.634324006884713e-05,
      "loss": 0.0783,
      "mean_token_accuracy": 0.9761517316102981,
      "num_tokens": 30126668.0,
      "step": 200
    },
    {
      "epoch": 0.3302456705598067,
      "grad_norm": 0.346615731716156,
      "learning_rate": 1.619469477569153e-05,
      "loss": 0.0687,
      "mean_token_accuracy": 0.9790632575750351,
      "num_tokens": 30878060.0,
      "step": 205
    },
    {
      "epoch": 0.3383004430124849,
      "grad_norm": 0.5523544549942017,
      "learning_rate": 1.604300000446427e-05,
      "loss": 0.0639,
      "mean_token_accuracy": 0.9795442938804626,
      "num_tokens": 31630658.0,
      "step": 210
    },
    {
      "epoch": 0.3463552154651631,
      "grad_norm": 0.5302470326423645,
      "learning_rate": 1.588826511870035e-05,
      "loss": 0.0734,
      "mean_token_accuracy": 0.9772966742515564,
      "num_tokens": 32396923.0,
      "step": 215
    },
    {
      "epoch": 0.35440998791784134,
      "grad_norm": 0.38110172748565674,
      "learning_rate": 1.5730601673689106e-05,
      "loss": 0.0704,
      "mean_token_accuracy": 0.9794994920492173,
      "num_tokens": 33141640.0,
      "step": 220
    },
    {
      "epoch": 0.36246476037051956,
      "grad_norm": 0.369781494140625,
      "learning_rate": 1.5570123336049007e-05,
      "loss": 0.0765,
      "mean_token_accuracy": 0.9773078709840775,
      "num_tokens": 33901915.0,
      "step": 225
    },
    {
      "epoch": 0.3705195328231977,
      "grad_norm": 0.5634907484054565,
      "learning_rate": 1.5406945801780347e-05,
      "loss": 0.0684,
      "mean_token_accuracy": 0.978277200460434,
      "num_tokens": 34664064.0,
      "step": 230
    },
    {
      "epoch": 0.37857430527587593,
      "grad_norm": 0.5715321898460388,
      "learning_rate": 1.5241186712854794e-05,
      "loss": 0.0723,
      "mean_token_accuracy": 0.9781462490558624,
      "num_tokens": 35410955.0,
      "step": 235
    },
    {
      "epoch": 0.38662907772855415,
      "grad_norm": 0.4872036576271057,
      "learning_rate": 1.5072965572402066e-05,
      "loss": 0.0902,
      "mean_token_accuracy": 0.9755937904119492,
      "num_tokens": 36157704.0,
      "step": 240
    },
    {
      "epoch": 0.39468385018123237,
      "grad_norm": 0.3256291449069977,
      "learning_rate": 1.4902403658554765e-05,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9753373205661774,
      "num_tokens": 36917283.0,
      "step": 245
    },
    {
      "epoch": 0.4027386226339106,
      "grad_norm": 0.441094309091568,
      "learning_rate": 1.4729623937013588e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9773626685142517,
      "num_tokens": 37673237.0,
      "step": 250
    },
    {
      "epoch": 0.4107933950865888,
      "grad_norm": 0.3978787064552307,
      "learning_rate": 1.4554750972395824e-05,
      "loss": 0.0724,
      "mean_token_accuracy": 0.9792384207248688,
      "num_tokens": 38420567.0,
      "step": 255
    },
    {
      "epoch": 0.418848167539267,
      "grad_norm": 0.5710241794586182,
      "learning_rate": 1.437791083843118e-05,
      "loss": 0.0699,
      "mean_token_accuracy": 0.979467561841011,
      "num_tokens": 39183545.0,
      "step": 260
    },
    {
      "epoch": 0.42690293999194523,
      "grad_norm": 0.43882453441619873,
      "learning_rate": 1.4199231027069605e-05,
      "loss": 0.0613,
      "mean_token_accuracy": 0.9806316882371903,
      "num_tokens": 39944325.0,
      "step": 265
    },
    {
      "epoch": 0.43495771244462345,
      "grad_norm": 0.41558167338371277,
      "learning_rate": 1.4018840356566632e-05,
      "loss": 0.0723,
      "mean_token_accuracy": 0.9787821143865585,
      "num_tokens": 40693954.0,
      "step": 270
    },
    {
      "epoch": 0.44301248489730166,
      "grad_norm": 0.49668020009994507,
      "learning_rate": 1.383686887861255e-05,
      "loss": 0.0785,
      "mean_token_accuracy": 0.9763137608766556,
      "num_tokens": 41450905.0,
      "step": 275
    },
    {
      "epoch": 0.4510672573499799,
      "grad_norm": 0.3873698115348816,
      "learning_rate": 1.3653447784572342e-05,
      "loss": 0.0641,
      "mean_token_accuracy": 0.9808479577302933,
      "num_tokens": 42192006.0,
      "step": 280
    },
    {
      "epoch": 0.4591220298026581,
      "grad_norm": 0.49236777424812317,
      "learning_rate": 1.3468709310903972e-05,
      "loss": 0.0764,
      "mean_token_accuracy": 0.977475494146347,
      "num_tokens": 42969954.0,
      "step": 285
    },
    {
      "epoch": 0.4671768022553363,
      "grad_norm": 0.6129188537597656,
      "learning_rate": 1.3282786643823185e-05,
      "loss": 0.0855,
      "mean_token_accuracy": 0.9768295049667358,
      "num_tokens": 43722838.0,
      "step": 290
    },
    {
      "epoch": 0.4752315747080145,
      "grad_norm": 0.4230325520038605,
      "learning_rate": 1.3095813823283666e-05,
      "loss": 0.0765,
      "mean_token_accuracy": 0.9784682154655456,
      "num_tokens": 44468533.0,
      "step": 295
    },
    {
      "epoch": 0.4832863471606927,
      "grad_norm": 0.5834869742393494,
      "learning_rate": 1.2907925646341617e-05,
      "loss": 0.0858,
      "mean_token_accuracy": 0.9760559558868408,
      "num_tokens": 45218879.0,
      "step": 300
    },
    {
      "epoch": 0.4913411196133709,
      "grad_norm": 0.4738158881664276,
      "learning_rate": 1.2719257569974546e-05,
      "loss": 0.0633,
      "mean_token_accuracy": 0.9806985557079315,
      "num_tokens": 45986211.0,
      "step": 305
    },
    {
      "epoch": 0.4993958920660491,
      "grad_norm": 0.4312545955181122,
      "learning_rate": 1.252994561342429e-05,
      "loss": 0.0785,
      "mean_token_accuracy": 0.9778055489063263,
      "num_tokens": 46748814.0,
      "step": 310
    },
    {
      "epoch": 0.5074506645187273,
      "grad_norm": 0.7503715753555298,
      "learning_rate": 1.2340126260134662e-05,
      "loss": 0.0643,
      "mean_token_accuracy": 0.979890787601471,
      "num_tokens": 47506536.0,
      "step": 315
    }
  ],
  "logging_steps": 5,
  "max_steps": 620,
  "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": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.166197393225941e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}