{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9363867684478372,
  "eval_steps": 500,
  "global_step": 368,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01272264631043257,
      "grad_norm": 7.66570520401001,
      "learning_rate": 7.142e-06,
      "loss": 1.8805,
      "mean_token_accuracy": 0.6622064590454102,
      "num_tokens": 204352.0,
      "step": 5
    },
    {
      "epoch": 0.02544529262086514,
      "grad_norm": 1.9406589269638062,
      "learning_rate": 1.6069500000000003e-05,
      "loss": 1.3809,
      "mean_token_accuracy": 0.6968963027000428,
      "num_tokens": 408371.0,
      "step": 10
    },
    {
      "epoch": 0.03816793893129771,
      "grad_norm": 1.4358961582183838,
      "learning_rate": 2.4997e-05,
      "loss": 0.8872,
      "mean_token_accuracy": 0.7816788077354431,
      "num_tokens": 612497.0,
      "step": 15
    },
    {
      "epoch": 0.05089058524173028,
      "grad_norm": 0.9055664539337158,
      "learning_rate": 3.3924500000000004e-05,
      "loss": 0.449,
      "mean_token_accuracy": 0.8871559858322143,
      "num_tokens": 816690.0,
      "step": 20
    },
    {
      "epoch": 0.06361323155216285,
      "grad_norm": 1.860472321510315,
      "learning_rate": 3.570240107586499e-05,
      "loss": 0.182,
      "mean_token_accuracy": 0.9483001828193665,
      "num_tokens": 1019984.0,
      "step": 25
    },
    {
      "epoch": 0.07633587786259542,
      "grad_norm": 0.7346654534339905,
      "learning_rate": 3.567154522651094e-05,
      "loss": 0.1147,
      "mean_token_accuracy": 0.9639001488685608,
      "num_tokens": 1223779.0,
      "step": 30
    },
    {
      "epoch": 0.089058524173028,
      "grad_norm": 1.6021146774291992,
      "learning_rate": 3.561701219067847e-05,
      "loss": 0.1031,
      "mean_token_accuracy": 0.9666669845581055,
      "num_tokens": 1427505.0,
      "step": 35
    },
    {
      "epoch": 0.10178117048346055,
      "grad_norm": 0.35029786825180054,
      "learning_rate": 3.553889866638751e-05,
      "loss": 0.0807,
      "mean_token_accuracy": 0.9725039839744568,
      "num_tokens": 1631635.0,
      "step": 40
    },
    {
      "epoch": 0.11450381679389313,
      "grad_norm": 0.255508154630661,
      "learning_rate": 3.543734316459772e-05,
      "loss": 0.0784,
      "mean_token_accuracy": 0.9729812264442443,
      "num_tokens": 1836177.0,
      "step": 45
    },
    {
      "epoch": 0.1272264631043257,
      "grad_norm": 0.27013325691223145,
      "learning_rate": 3.53125257636008e-05,
      "loss": 0.0732,
      "mean_token_accuracy": 0.9745353579521179,
      "num_tokens": 2040117.0,
      "step": 50
    },
    {
      "epoch": 0.13994910941475827,
      "grad_norm": 0.23351548612117767,
      "learning_rate": 3.516466778970551e-05,
      "loss": 0.0684,
      "mean_token_accuracy": 0.9762374997138977,
      "num_tokens": 2244475.0,
      "step": 55
    },
    {
      "epoch": 0.15267175572519084,
      "grad_norm": 0.258129745721817,
      "learning_rate": 3.49940314247817e-05,
      "loss": 0.0689,
      "mean_token_accuracy": 0.9754044890403748,
      "num_tokens": 2448636.0,
      "step": 60
    },
    {
      "epoch": 0.16539440203562342,
      "grad_norm": 0.20423570275306702,
      "learning_rate": 3.480091924135923e-05,
      "loss": 0.0673,
      "mean_token_accuracy": 0.976031219959259,
      "num_tokens": 2652601.0,
      "step": 65
    },
    {
      "epoch": 0.178117048346056,
      "grad_norm": 0.13115821778774261,
      "learning_rate": 3.458567366610612e-05,
      "loss": 0.0596,
      "mean_token_accuracy": 0.9785127282142639,
      "num_tokens": 2857182.0,
      "step": 70
    },
    {
      "epoch": 0.19083969465648856,
      "grad_norm": 0.15491832792758942,
      "learning_rate": 3.434867637263733e-05,
      "loss": 0.056,
      "mean_token_accuracy": 0.9797403216362,
      "num_tokens": 3061762.0,
      "step": 75
    },
    {
      "epoch": 0.2035623409669211,
      "grad_norm": 0.18029575049877167,
      "learning_rate": 3.4090347604730814e-05,
      "loss": 0.0567,
      "mean_token_accuracy": 0.9793354153633118,
      "num_tokens": 3265924.0,
      "step": 80
    },
    {
      "epoch": 0.21628498727735368,
      "grad_norm": 0.12074996531009674,
      "learning_rate": 3.381114543115096e-05,
      "loss": 0.0538,
      "mean_token_accuracy": 0.980361008644104,
      "num_tokens": 3470056.0,
      "step": 85
    },
    {
      "epoch": 0.22900763358778625,
      "grad_norm": 0.14074520766735077,
      "learning_rate": 3.351156493340068e-05,
      "loss": 0.0585,
      "mean_token_accuracy": 0.9789868593215942,
      "num_tokens": 3674439.0,
      "step": 90
    },
    {
      "epoch": 0.24173027989821882,
      "grad_norm": 0.11715356260538101,
      "learning_rate": 3.319213732784259e-05,
      "loss": 0.0608,
      "mean_token_accuracy": 0.977981960773468,
      "num_tokens": 3878368.0,
      "step": 95
    },
    {
      "epoch": 0.2544529262086514,
      "grad_norm": 0.160597562789917,
      "learning_rate": 3.285342902374564e-05,
      "loss": 0.0541,
      "mean_token_accuracy": 0.9802710533142089,
      "num_tokens": 4082520.0,
      "step": 100
    },
    {
      "epoch": 0.26717557251908397,
      "grad_norm": 0.13476158678531647,
      "learning_rate": 3.249604061892791e-05,
      "loss": 0.058,
      "mean_token_accuracy": 0.9789228081703186,
      "num_tokens": 4286416.0,
      "step": 105
    },
    {
      "epoch": 0.27989821882951654,
      "grad_norm": 0.13968558609485626,
      "learning_rate": 3.212060583477598e-05,
      "loss": 0.0545,
      "mean_token_accuracy": 0.9799705028533936,
      "num_tokens": 4490360.0,
      "step": 110
    },
    {
      "epoch": 0.2926208651399491,
      "grad_norm": 0.18748225271701813,
      "learning_rate": 3.172779039252981e-05,
      "loss": 0.0521,
      "mean_token_accuracy": 0.9808894634246826,
      "num_tokens": 4694687.0,
      "step": 115
    },
    {
      "epoch": 0.3053435114503817,
      "grad_norm": 0.10537264496088028,
      "learning_rate": 3.131829083282526e-05,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9768295168876648,
      "num_tokens": 4898444.0,
      "step": 120
    },
    {
      "epoch": 0.31806615776081426,
      "grad_norm": 0.17765143513679504,
      "learning_rate": 3.089283328058774e-05,
      "loss": 0.0556,
      "mean_token_accuracy": 0.9799776554107666,
      "num_tokens": 5102850.0,
      "step": 125
    },
    {
      "epoch": 0.33078880407124683,
      "grad_norm": 0.1285717487335205,
      "learning_rate": 3.045217215746697e-05,
      "loss": 0.0549,
      "mean_token_accuracy": 0.9798083662986755,
      "num_tokens": 5307181.0,
      "step": 130
    },
    {
      "epoch": 0.3435114503816794,
      "grad_norm": 0.14681582152843475,
      "learning_rate": 2.9997088844095903e-05,
      "loss": 0.0472,
      "mean_token_accuracy": 0.9827617883682251,
      "num_tokens": 5511359.0,
      "step": 135
    },
    {
      "epoch": 0.356234096692112,
      "grad_norm": 0.10448456555604935,
      "learning_rate": 2.9528390294546044e-05,
      "loss": 0.0597,
      "mean_token_accuracy": 0.9781724691390992,
      "num_tokens": 5715595.0,
      "step": 140
    },
    {
      "epoch": 0.36895674300254455,
      "grad_norm": 0.1462472528219223,
      "learning_rate": 2.9046907605435848e-05,
      "loss": 0.054,
      "mean_token_accuracy": 0.9804641366004944,
      "num_tokens": 5919772.0,
      "step": 145
    },
    {
      "epoch": 0.3816793893129771,
      "grad_norm": 0.1421613097190857,
      "learning_rate": 2.8553494542229622e-05,
      "loss": 0.0493,
      "mean_token_accuracy": 0.9818941712379455,
      "num_tokens": 6123654.0,
      "step": 150
    },
    {
      "epoch": 0.3944020356234097,
      "grad_norm": 0.1263999044895172,
      "learning_rate": 2.804902602533992e-05,
      "loss": 0.0525,
      "mean_token_accuracy": 0.980873703956604,
      "num_tokens": 6327633.0,
      "step": 155
    },
    {
      "epoch": 0.4071246819338422,
      "grad_norm": 0.13942250609397888,
      "learning_rate": 2.753439657871804e-05,
      "loss": 0.0558,
      "mean_token_accuracy": 0.97970951795578,
      "num_tokens": 6531407.0,
      "step": 160
    },
    {
      "epoch": 0.4198473282442748,
      "grad_norm": 0.13747908174991608,
      "learning_rate": 2.7010518743683477e-05,
      "loss": 0.0503,
      "mean_token_accuracy": 0.9813982605934143,
      "num_tokens": 6735388.0,
      "step": 165
    },
    {
      "epoch": 0.43256997455470736,
      "grad_norm": 0.08825346827507019,
      "learning_rate": 2.6478321460804976e-05,
      "loss": 0.0501,
      "mean_token_accuracy": 0.9816948294639587,
      "num_tokens": 6938927.0,
      "step": 170
    },
    {
      "epoch": 0.44529262086513993,
      "grad_norm": 0.19031524658203125,
      "learning_rate": 2.5938748422702418e-05,
      "loss": 0.0538,
      "mean_token_accuracy": 0.9802047729492187,
      "num_tokens": 7143289.0,
      "step": 175
    },
    {
      "epoch": 0.4580152671755725,
      "grad_norm": 0.10854300111532211,
      "learning_rate": 2.539275640069035e-05,
      "loss": 0.0472,
      "mean_token_accuracy": 0.9826422333717346,
      "num_tokens": 7347460.0,
      "step": 180
    },
    {
      "epoch": 0.4707379134860051,
      "grad_norm": 0.11002419143915176,
      "learning_rate": 2.484131354823041e-05,
      "loss": 0.0535,
      "mean_token_accuracy": 0.9802260994911194,
      "num_tokens": 7551434.0,
      "step": 185
    },
    {
      "epoch": 0.48346055979643765,
      "grad_norm": 0.1230173334479332,
      "learning_rate": 2.4285397684200836e-05,
      "loss": 0.058,
      "mean_token_accuracy": 0.9790604472160339,
      "num_tokens": 7755531.0,
      "step": 190
    },
    {
      "epoch": 0.4961832061068702,
      "grad_norm": 0.12328363955020905,
      "learning_rate": 2.3725994559027345e-05,
      "loss": 0.0494,
      "mean_token_accuracy": 0.982027554512024,
      "num_tokens": 7959120.0,
      "step": 195
    },
    {
      "epoch": 0.5089058524173028,
      "grad_norm": 0.14120548963546753,
      "learning_rate": 2.3164096106749705e-05,
      "loss": 0.0559,
      "mean_token_accuracy": 0.9795668363571167,
      "num_tokens": 8163304.0,
      "step": 200
    },
    {
      "epoch": 0.5216284987277354,
      "grad_norm": 0.13413408398628235,
      "learning_rate": 2.260069868612358e-05,
      "loss": 0.0522,
      "mean_token_accuracy": 0.9810293197631836,
      "num_tokens": 8367304.0,
      "step": 205
    },
    {
      "epoch": 0.5343511450381679,
      "grad_norm": 0.1026059240102768,
      "learning_rate": 2.203680131387643e-05,
      "loss": 0.0482,
      "mean_token_accuracy": 0.9819956064224243,
      "num_tokens": 8571809.0,
      "step": 210
    },
    {
      "epoch": 0.5470737913486005,
      "grad_norm": 0.11920855939388275,
      "learning_rate": 2.1473403893250302e-05,
      "loss": 0.0543,
      "mean_token_accuracy": 0.9803555488586426,
      "num_tokens": 8775987.0,
      "step": 215
    },
    {
      "epoch": 0.5597964376590331,
      "grad_norm": 0.15539394319057465,
      "learning_rate": 2.0911505440972662e-05,
      "loss": 0.0585,
      "mean_token_accuracy": 0.9789112687110901,
      "num_tokens": 8980105.0,
      "step": 220
    },
    {
      "epoch": 0.5725190839694656,
      "grad_norm": 0.1551484316587448,
      "learning_rate": 2.0352102315799165e-05,
      "loss": 0.0536,
      "mean_token_accuracy": 0.9803162574768066,
      "num_tokens": 9184510.0,
      "step": 225
    },
    {
      "epoch": 0.5852417302798982,
      "grad_norm": 0.12535728514194489,
      "learning_rate": 1.9796186451769593e-05,
      "loss": 0.0497,
      "mean_token_accuracy": 0.9818916082382202,
      "num_tokens": 9388438.0,
      "step": 230
    },
    {
      "epoch": 0.5979643765903307,
      "grad_norm": 0.13271310925483704,
      "learning_rate": 1.924474359930965e-05,
      "loss": 0.0524,
      "mean_token_accuracy": 0.9809573411941528,
      "num_tokens": 9592608.0,
      "step": 235
    },
    {
      "epoch": 0.6106870229007634,
      "grad_norm": 0.11600778251886368,
      "learning_rate": 1.869875157729759e-05,
      "loss": 0.0552,
      "mean_token_accuracy": 0.9797375440597534,
      "num_tokens": 9796999.0,
      "step": 240
    },
    {
      "epoch": 0.6234096692111959,
      "grad_norm": 0.12414652854204178,
      "learning_rate": 1.815917853919503e-05,
      "loss": 0.048,
      "mean_token_accuracy": 0.9824962496757508,
      "num_tokens": 10000279.0,
      "step": 245
    },
    {
      "epoch": 0.6361323155216285,
      "grad_norm": 0.1270761489868164,
      "learning_rate": 1.7626981256316527e-05,
      "loss": 0.0487,
      "mean_token_accuracy": 0.9822483897209168,
      "num_tokens": 10204018.0,
      "step": 250
    },
    {
      "epoch": 0.648854961832061,
      "grad_norm": 0.09779980033636093,
      "learning_rate": 1.7103103421281967e-05,
      "loss": 0.0516,
      "mean_token_accuracy": 0.9810034275054932,
      "num_tokens": 10408010.0,
      "step": 255
    },
    {
      "epoch": 0.6615776081424937,
      "grad_norm": 0.11059050261974335,
      "learning_rate": 1.658847397466009e-05,
      "loss": 0.0525,
      "mean_token_accuracy": 0.9808456897735596,
      "num_tokens": 10611936.0,
      "step": 260
    },
    {
      "epoch": 0.6743002544529262,
      "grad_norm": 0.09004539996385574,
      "learning_rate": 1.6084005457770385e-05,
      "loss": 0.0496,
      "mean_token_accuracy": 0.9818039417266846,
      "num_tokens": 10816288.0,
      "step": 265
    },
    {
      "epoch": 0.6870229007633588,
      "grad_norm": 0.10434029996395111,
      "learning_rate": 1.5590592394564157e-05,
      "loss": 0.0482,
      "mean_token_accuracy": 0.9822108745574951,
      "num_tokens": 11020673.0,
      "step": 270
    },
    {
      "epoch": 0.6997455470737913,
      "grad_norm": 0.09719809144735336,
      "learning_rate": 1.5109109705453962e-05,
      "loss": 0.0527,
      "mean_token_accuracy": 0.9806708216667175,
      "num_tokens": 11224646.0,
      "step": 275
    },
    {
      "epoch": 0.712468193384224,
      "grad_norm": 0.10297023504972458,
      "learning_rate": 1.46404111559041e-05,
      "loss": 0.0488,
      "mean_token_accuracy": 0.9819470047950745,
      "num_tokens": 11429157.0,
      "step": 280
    },
    {
      "epoch": 0.7251908396946565,
      "grad_norm": 0.1588681936264038,
      "learning_rate": 1.4185327842533032e-05,
      "loss": 0.0568,
      "mean_token_accuracy": 0.9791428089141846,
      "num_tokens": 11633534.0,
      "step": 285
    },
    {
      "epoch": 0.7379134860050891,
      "grad_norm": 0.08974728733301163,
      "learning_rate": 1.3744666719412261e-05,
      "loss": 0.0486,
      "mean_token_accuracy": 0.982176947593689,
      "num_tokens": 11837304.0,
      "step": 290
    },
    {
      "epoch": 0.7506361323155216,
      "grad_norm": 0.12151776254177094,
      "learning_rate": 1.3319209167174745e-05,
      "loss": 0.0456,
      "mean_token_accuracy": 0.9831613540649414,
      "num_tokens": 12041445.0,
      "step": 295
    },
    {
      "epoch": 0.7633587786259542,
      "grad_norm": 0.14409556984901428,
      "learning_rate": 1.2909709607470192e-05,
      "loss": 0.0557,
      "mean_token_accuracy": 0.9793829798698426,
      "num_tokens": 12245544.0,
      "step": 300
    },
    {
      "epoch": 0.7760814249363868,
      "grad_norm": 0.12963253259658813,
      "learning_rate": 1.2516894165224026e-05,
      "loss": 0.0541,
      "mean_token_accuracy": 0.980055284500122,
      "num_tokens": 12450041.0,
      "step": 305
    },
    {
      "epoch": 0.7888040712468194,
      "grad_norm": 0.09643556922674179,
      "learning_rate": 1.2141459381072098e-05,
      "loss": 0.0506,
      "mean_token_accuracy": 0.9816014289855957,
      "num_tokens": 12654456.0,
      "step": 310
    },
    {
      "epoch": 0.8015267175572519,
      "grad_norm": 0.07658141851425171,
      "learning_rate": 1.178407097625436e-05,
      "loss": 0.0527,
      "mean_token_accuracy": 0.9805378794670105,
      "num_tokens": 12858221.0,
      "step": 315
    },
    {
      "epoch": 0.8142493638676844,
      "grad_norm": 0.09057999402284622,
      "learning_rate": 1.1445362672157415e-05,
      "loss": 0.0555,
      "mean_token_accuracy": 0.9796663165092468,
      "num_tokens": 13061611.0,
      "step": 320
    },
    {
      "epoch": 0.8269720101781171,
      "grad_norm": 0.10103996098041534,
      "learning_rate": 1.1125935066599316e-05,
      "loss": 0.0481,
      "mean_token_accuracy": 0.9822862863540649,
      "num_tokens": 13266008.0,
      "step": 325
    },
    {
      "epoch": 0.8396946564885496,
      "grad_norm": 0.10731196403503418,
      "learning_rate": 1.0826354568849048e-05,
      "loss": 0.0519,
      "mean_token_accuracy": 0.9808366537094116,
      "num_tokens": 13469127.0,
      "step": 330
    },
    {
      "epoch": 0.8524173027989822,
      "grad_norm": 0.09134743362665176,
      "learning_rate": 1.0547152395269189e-05,
      "loss": 0.0534,
      "mean_token_accuracy": 0.9803672075271607,
      "num_tokens": 13672890.0,
      "step": 335
    },
    {
      "epoch": 0.8651399491094147,
      "grad_norm": 0.09591839462518692,
      "learning_rate": 1.0288823627362674e-05,
      "loss": 0.0543,
      "mean_token_accuracy": 0.9801756381988526,
      "num_tokens": 13876613.0,
      "step": 340
    },
    {
      "epoch": 0.8778625954198473,
      "grad_norm": 0.09613614529371262,
      "learning_rate": 1.0051826333893878e-05,
      "loss": 0.0535,
      "mean_token_accuracy": 0.9803989171981812,
      "num_tokens": 14080986.0,
      "step": 345
    },
    {
      "epoch": 0.8905852417302799,
      "grad_norm": 0.09382142871618271,
      "learning_rate": 9.836580758640768e-06,
      "loss": 0.0538,
      "mean_token_accuracy": 0.9806716322898865,
      "num_tokens": 14284897.0,
      "step": 350
    },
    {
      "epoch": 0.9033078880407125,
      "grad_norm": 0.13422974944114685,
      "learning_rate": 9.643468575218302e-06,
      "loss": 0.0485,
      "mean_token_accuracy": 0.9820788860321045,
      "num_tokens": 14489073.0,
      "step": 355
    },
    {
      "epoch": 0.916030534351145,
      "grad_norm": 0.12756891548633575,
      "learning_rate": 9.472832210294495e-06,
      "loss": 0.0502,
      "mean_token_accuracy": 0.9815787076950073,
      "num_tokens": 14693873.0,
      "step": 360
    },
    {
      "epoch": 0.9287531806615776,
      "grad_norm": 0.11220107972621918,
      "learning_rate": 9.324974236399208e-06,
      "loss": 0.0477,
      "mean_token_accuracy": 0.9823699235916138,
      "num_tokens": 14897591.0,
      "step": 365
    }
  ],
  "logging_steps": 5,
  "max_steps": 393,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.512720333608059e+17,
  "train_batch_size": 20,
  "trial_name": null,
  "trial_params": null
}