{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9978994991113266,
  "eval_steps": 500,
  "global_step": 386,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025852318629827112,
      "grad_norm": 0.19946137070655823,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.4855,
      "mean_token_accuracy": 0.8692543655633926,
      "step": 1
    },
    {
      "epoch": 0.012926159314913557,
      "grad_norm": 0.18048785626888275,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.5257,
      "mean_token_accuracy": 0.8587683830410242,
      "step": 5
    },
    {
      "epoch": 0.025852318629827113,
      "grad_norm": 0.18388661742210388,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.5986,
      "mean_token_accuracy": 0.838620126247406,
      "step": 10
    },
    {
      "epoch": 0.038778477944740666,
      "grad_norm": 0.13041158020496368,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.5464,
      "mean_token_accuracy": 0.8496200017631054,
      "step": 15
    },
    {
      "epoch": 0.051704637259654226,
      "grad_norm": 0.10899022966623306,
      "learning_rate": 0.00010256410256410256,
      "loss": 0.4321,
      "mean_token_accuracy": 0.8813652157783508,
      "step": 20
    },
    {
      "epoch": 0.06463079657456779,
      "grad_norm": 0.12952959537506104,
      "learning_rate": 0.00012820512820512823,
      "loss": 0.4429,
      "mean_token_accuracy": 0.8760499283671379,
      "step": 25
    },
    {
      "epoch": 0.07755695588948133,
      "grad_norm": 0.11217343807220459,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.4426,
      "mean_token_accuracy": 0.8728497698903084,
      "step": 30
    },
    {
      "epoch": 0.09048311520439489,
      "grad_norm": 0.11570142209529877,
      "learning_rate": 0.0001794871794871795,
      "loss": 0.4326,
      "mean_token_accuracy": 0.8739698991179466,
      "step": 35
    },
    {
      "epoch": 0.10340927451930845,
      "grad_norm": 0.11375421285629272,
      "learning_rate": 0.00019999590166142655,
      "loss": 0.4398,
      "mean_token_accuracy": 0.8683982357382775,
      "step": 40
    },
    {
      "epoch": 0.11633543383422201,
      "grad_norm": 0.09867997467517853,
      "learning_rate": 0.00019985249508066755,
      "loss": 0.3217,
      "mean_token_accuracy": 0.9007599778473377,
      "step": 45
    },
    {
      "epoch": 0.12926159314913557,
      "grad_norm": 0.10740496963262558,
      "learning_rate": 0.00019950450737506824,
      "loss": 0.3468,
      "mean_token_accuracy": 0.8914886720478534,
      "step": 50
    },
    {
      "epoch": 0.14218775246404913,
      "grad_norm": 0.10870860517024994,
      "learning_rate": 0.00019895265151345518,
      "loss": 0.297,
      "mean_token_accuracy": 0.9054083719849586,
      "step": 55
    },
    {
      "epoch": 0.15511391177896267,
      "grad_norm": 0.1147221028804779,
      "learning_rate": 0.00019819805815653768,
      "loss": 0.2912,
      "mean_token_accuracy": 0.9063926823437214,
      "step": 60
    },
    {
      "epoch": 0.16804007109387623,
      "grad_norm": 0.1086113452911377,
      "learning_rate": 0.00019724227334037256,
      "loss": 0.2709,
      "mean_token_accuracy": 0.9122237786650658,
      "step": 65
    },
    {
      "epoch": 0.18096623040878979,
      "grad_norm": 0.11263403296470642,
      "learning_rate": 0.00019608725530879375,
      "loss": 0.2775,
      "mean_token_accuracy": 0.9088832184672355,
      "step": 70
    },
    {
      "epoch": 0.19389238972370335,
      "grad_norm": 0.11787361651659012,
      "learning_rate": 0.00019473537050129704,
      "loss": 0.2758,
      "mean_token_accuracy": 0.909870583564043,
      "step": 75
    },
    {
      "epoch": 0.2068185490386169,
      "grad_norm": 0.0966123640537262,
      "learning_rate": 0.00019318938870459984,
      "loss": 0.2113,
      "mean_token_accuracy": 0.9297552116215229,
      "step": 80
    },
    {
      "epoch": 0.21974470835353047,
      "grad_norm": 0.09169968962669373,
      "learning_rate": 0.00019145247737780961,
      "loss": 0.2097,
      "mean_token_accuracy": 0.9294927291572094,
      "step": 85
    },
    {
      "epoch": 0.23267086766844403,
      "grad_norm": 0.10205108672380447,
      "learning_rate": 0.0001895281951628281,
      "loss": 0.2354,
      "mean_token_accuracy": 0.9218288294970989,
      "step": 90
    },
    {
      "epoch": 0.24559702698335756,
      "grad_norm": 0.11714070290327072,
      "learning_rate": 0.00018742048459328682,
      "loss": 0.2286,
      "mean_token_accuracy": 0.9224011048674583,
      "step": 95
    },
    {
      "epoch": 0.25852318629827115,
      "grad_norm": 0.08743036538362503,
      "learning_rate": 0.00018513366401695276,
      "loss": 0.2524,
      "mean_token_accuracy": 0.9145891763269901,
      "step": 100
    },
    {
      "epoch": 0.2714493456131847,
      "grad_norm": 0.12086658179759979,
      "learning_rate": 0.00018267241874815314,
      "loss": 0.2355,
      "mean_token_accuracy": 0.9212763957679272,
      "step": 105
    },
    {
      "epoch": 0.28437550492809827,
      "grad_norm": 0.10271570086479187,
      "learning_rate": 0.0001800417914683471,
      "loss": 0.2033,
      "mean_token_accuracy": 0.9312141306698323,
      "step": 110
    },
    {
      "epoch": 0.2973016642430118,
      "grad_norm": 0.11228019744157791,
      "learning_rate": 0.0001772471718945119,
      "loss": 0.1578,
      "mean_token_accuracy": 0.946045958250761,
      "step": 115
    },
    {
      "epoch": 0.31022782355792533,
      "grad_norm": 0.1075127124786377,
      "learning_rate": 0.00017429428573651024,
      "loss": 0.2109,
      "mean_token_accuracy": 0.928734278678894,
      "step": 120
    },
    {
      "epoch": 0.3231539828728389,
      "grad_norm": 0.16245290637016296,
      "learning_rate": 0.00017118918296606537,
      "loss": 0.2544,
      "mean_token_accuracy": 0.9145586542785168,
      "step": 125
    },
    {
      "epoch": 0.33608014218775245,
      "grad_norm": 0.0984274297952652,
      "learning_rate": 0.0001679382254213768,
      "loss": 0.2398,
      "mean_token_accuracy": 0.9193581290543079,
      "step": 130
    },
    {
      "epoch": 0.34900630150266604,
      "grad_norm": 0.14143706858158112,
      "learning_rate": 0.00016454807377277398,
      "loss": 0.2187,
      "mean_token_accuracy": 0.9259690448641777,
      "step": 135
    },
    {
      "epoch": 0.36193246081757957,
      "grad_norm": 0.10752697288990021,
      "learning_rate": 0.0001610256738761125,
      "loss": 0.2205,
      "mean_token_accuracy": 0.9248595975339413,
      "step": 140
    },
    {
      "epoch": 0.37485862013249316,
      "grad_norm": 0.15562467277050018,
      "learning_rate": 0.00015737824254187275,
      "loss": 0.2399,
      "mean_token_accuracy": 0.9187141321599483,
      "step": 145
    },
    {
      "epoch": 0.3877847794474067,
      "grad_norm": 0.10149288177490234,
      "learning_rate": 0.00015361325274911779,
      "loss": 0.1906,
      "mean_token_accuracy": 0.9351025439798832,
      "step": 150
    },
    {
      "epoch": 0.4007109387623202,
      "grad_norm": 0.13106182217597961,
      "learning_rate": 0.00014973841833460457,
      "loss": 0.2366,
      "mean_token_accuracy": 0.9191612683236599,
      "step": 155
    },
    {
      "epoch": 0.4136370980772338,
      "grad_norm": 0.12485964596271515,
      "learning_rate": 0.0001457616781884173,
      "loss": 0.2532,
      "mean_token_accuracy": 0.9145220316946506,
      "step": 160
    },
    {
      "epoch": 0.42656325739214734,
      "grad_norm": 0.12415427714586258,
      "learning_rate": 0.0001416911799885049,
      "loss": 0.1772,
      "mean_token_accuracy": 0.9388307243585586,
      "step": 165
    },
    {
      "epoch": 0.43948941670706093,
      "grad_norm": 0.12156961858272552,
      "learning_rate": 0.0001375352635074461,
      "loss": 0.1974,
      "mean_token_accuracy": 0.9322396464645862,
      "step": 170
    },
    {
      "epoch": 0.45241557602197446,
      "grad_norm": 0.08770665526390076,
      "learning_rate": 0.00013330244352564527,
      "loss": 0.2159,
      "mean_token_accuracy": 0.9263734519481659,
      "step": 175
    },
    {
      "epoch": 0.46534173533688805,
      "grad_norm": 0.11970090866088867,
      "learning_rate": 0.00012900139238596598,
      "loss": 0.1788,
      "mean_token_accuracy": 0.9383759558200836,
      "step": 180
    },
    {
      "epoch": 0.4782678946518016,
      "grad_norm": 0.0777999758720398,
      "learning_rate": 0.00012464092222554552,
      "loss": 0.1921,
      "mean_token_accuracy": 0.9340388782322406,
      "step": 185
    },
    {
      "epoch": 0.4911940539667151,
      "grad_norm": 0.1088053435087204,
      "learning_rate": 0.00012022996692119424,
      "loss": 0.2428,
      "mean_token_accuracy": 0.9170919217169284,
      "step": 190
    },
    {
      "epoch": 0.5041202132816287,
      "grad_norm": 0.12958504259586334,
      "learning_rate": 0.00011577756378537033,
      "loss": 0.1805,
      "mean_token_accuracy": 0.9373964861035347,
      "step": 195
    },
    {
      "epoch": 0.5170463725965423,
      "grad_norm": 0.08942475914955139,
      "learning_rate": 0.00011129283505023274,
      "loss": 0.1757,
      "mean_token_accuracy": 0.9391221977770329,
      "step": 200
    },
    {
      "epoch": 0.5299725319114558,
      "grad_norm": 0.12055882811546326,
      "learning_rate": 0.00010678496917770719,
      "loss": 0.2274,
      "mean_token_accuracy": 0.9221370957791806,
      "step": 205
    },
    {
      "epoch": 0.5428986912263694,
      "grad_norm": 0.09946688264608383,
      "learning_rate": 0.00010226320203385878,
      "loss": 0.2235,
      "mean_token_accuracy": 0.9239851593971252,
      "step": 210
    },
    {
      "epoch": 0.5558248505412829,
      "grad_norm": 0.08823594450950623,
      "learning_rate": 9.773679796614124e-05,
      "loss": 0.1757,
      "mean_token_accuracy": 0.9392839625477791,
      "step": 215
    },
    {
      "epoch": 0.5687510098561965,
      "grad_norm": 0.09366550296545029,
      "learning_rate": 9.321503082229282e-05,
      "loss": 0.215,
      "mean_token_accuracy": 0.9264558620750905,
      "step": 220
    },
    {
      "epoch": 0.58167716917111,
      "grad_norm": 0.1250247210264206,
      "learning_rate": 8.87071649497673e-05,
      "loss": 0.203,
      "mean_token_accuracy": 0.9299200311303139,
      "step": 225
    },
    {
      "epoch": 0.5946033284860236,
      "grad_norm": 0.07920438051223755,
      "learning_rate": 8.422243621462969e-05,
      "loss": 0.1777,
      "mean_token_accuracy": 0.93870100826025,
      "step": 230
    },
    {
      "epoch": 0.6075294878009372,
      "grad_norm": 0.09220347553491592,
      "learning_rate": 7.97700330788058e-05,
      "loss": 0.2382,
      "mean_token_accuracy": 0.9188789471983909,
      "step": 235
    },
    {
      "epoch": 0.6204556471158507,
      "grad_norm": 0.09347136318683624,
      "learning_rate": 7.535907777445449e-05,
      "loss": 0.1768,
      "mean_token_accuracy": 0.9390825219452381,
      "step": 240
    },
    {
      "epoch": 0.6333818064307642,
      "grad_norm": 0.06848734617233276,
      "learning_rate": 7.099860761403403e-05,
      "loss": 0.1909,
      "mean_token_accuracy": 0.9346477761864662,
      "step": 245
    },
    {
      "epoch": 0.6463079657456778,
      "grad_norm": 0.08231879770755768,
      "learning_rate": 6.669755647435474e-05,
      "loss": 0.2094,
      "mean_token_accuracy": 0.9281139463186264,
      "step": 250
    },
    {
      "epoch": 0.6592341250605914,
      "grad_norm": 0.14479972422122955,
      "learning_rate": 6.24647364925539e-05,
      "loss": 0.1897,
      "mean_token_accuracy": 0.9346370972692967,
      "step": 255
    },
    {
      "epoch": 0.6721602843755049,
      "grad_norm": 0.10493209958076477,
      "learning_rate": 5.830882001149517e-05,
      "loss": 0.1977,
      "mean_token_accuracy": 0.9318774163722991,
      "step": 260
    },
    {
      "epoch": 0.6850864436904185,
      "grad_norm": 0.08989942073822021,
      "learning_rate": 5.423832181158274e-05,
      "loss": 0.1786,
      "mean_token_accuracy": 0.9378982990980148,
      "step": 265
    },
    {
      "epoch": 0.6980126030053321,
      "grad_norm": 0.0723404809832573,
      "learning_rate": 5.0261581665395475e-05,
      "loss": 0.1854,
      "mean_token_accuracy": 0.9353665545582771,
      "step": 270
    },
    {
      "epoch": 0.7109387623202456,
      "grad_norm": 0.08017778396606445,
      "learning_rate": 4.6386747250882224e-05,
      "loss": 0.2185,
      "mean_token_accuracy": 0.9242644280195236,
      "step": 275
    },
    {
      "epoch": 0.7238649216351591,
      "grad_norm": 0.09016852080821991,
      "learning_rate": 4.2621757458127285e-05,
      "loss": 0.1912,
      "mean_token_accuracy": 0.9338557474315167,
      "step": 280
    },
    {
      "epoch": 0.7367910809500727,
      "grad_norm": 0.0839182510972023,
      "learning_rate": 3.8974326123887515e-05,
      "loss": 0.1764,
      "mean_token_accuracy": 0.9384186826646328,
      "step": 285
    },
    {
      "epoch": 0.7497172402649863,
      "grad_norm": 0.08571181446313858,
      "learning_rate": 3.5451926227225997e-05,
      "loss": 0.1984,
      "mean_token_accuracy": 0.931569704413414,
      "step": 290
    },
    {
      "epoch": 0.7626433995798998,
      "grad_norm": 0.09723369032144547,
      "learning_rate": 3.20617745786232e-05,
      "loss": 0.2095,
      "mean_token_accuracy": 0.9279029227793216,
      "step": 295
    },
    {
      "epoch": 0.7755695588948134,
      "grad_norm": 0.08322826772928238,
      "learning_rate": 2.8810817033934656e-05,
      "loss": 0.214,
      "mean_token_accuracy": 0.926049928367138,
      "step": 300
    },
    {
      "epoch": 0.788495718209727,
      "grad_norm": 0.0943559929728508,
      "learning_rate": 2.5705714263489776e-05,
      "loss": 0.2025,
      "mean_token_accuracy": 0.9304434671998024,
      "step": 305
    },
    {
      "epoch": 0.8014218775246404,
      "grad_norm": 0.12426702678203583,
      "learning_rate": 2.275282810548811e-05,
      "loss": 0.2255,
      "mean_token_accuracy": 0.92267579510808,
      "step": 310
    },
    {
      "epoch": 0.814348036839554,
      "grad_norm": 0.07339806854724884,
      "learning_rate": 1.9958208531652877e-05,
      "loss": 0.2119,
      "mean_token_accuracy": 0.9270739153027534,
      "step": 315
    },
    {
      "epoch": 0.8272741961544676,
      "grad_norm": 0.07293356955051422,
      "learning_rate": 1.73275812518469e-05,
      "loss": 0.1728,
      "mean_token_accuracy": 0.9397997766733169,
      "step": 320
    },
    {
      "epoch": 0.8402003554693812,
      "grad_norm": 0.06353917717933655,
      "learning_rate": 1.4866335983047264e-05,
      "loss": 0.2043,
      "mean_token_accuracy": 0.9285617105662822,
      "step": 325
    },
    {
      "epoch": 0.8531265147842947,
      "grad_norm": 0.0852632075548172,
      "learning_rate": 1.2579515406713193e-05,
      "loss": 0.1894,
      "mean_token_accuracy": 0.9348805241286755,
      "step": 330
    },
    {
      "epoch": 0.8660526740992083,
      "grad_norm": 0.10367190837860107,
      "learning_rate": 1.0471804837171916e-05,
      "loss": 0.2155,
      "mean_token_accuracy": 0.9256119452416897,
      "step": 335
    },
    {
      "epoch": 0.8789788334141219,
      "grad_norm": 0.07647697627544403,
      "learning_rate": 8.547522622190385e-06,
      "loss": 0.191,
      "mean_token_accuracy": 0.9336939886212349,
      "step": 340
    },
    {
      "epoch": 0.8919049927290353,
      "grad_norm": 0.09329604357481003,
      "learning_rate": 6.810611295400171e-06,
      "loss": 0.162,
      "mean_token_accuracy": 0.9432654656469822,
      "step": 345
    },
    {
      "epoch": 0.9048311520439489,
      "grad_norm": 0.10007894039154053,
      "learning_rate": 5.264629498702967e-06,
      "loss": 0.2196,
      "mean_token_accuracy": 0.9224743604660034,
      "step": 350
    },
    {
      "epoch": 0.9177573113588625,
      "grad_norm": 0.09568954259157181,
      "learning_rate": 3.91274469120626e-06,
      "loss": 0.2326,
      "mean_token_accuracy": 0.9203760787844658,
      "step": 355
    },
    {
      "epoch": 0.9306834706737761,
      "grad_norm": 0.09242628514766693,
      "learning_rate": 2.7577266596274576e-06,
      "loss": 0.217,
      "mean_token_accuracy": 0.9255997397005558,
      "step": 360
    },
    {
      "epoch": 0.9436096299886896,
      "grad_norm": 0.08556357026100159,
      "learning_rate": 1.8019418434623404e-06,
      "loss": 0.1704,
      "mean_token_accuracy": 0.9403167776763439,
      "step": 365
    },
    {
      "epoch": 0.9565357893036032,
      "grad_norm": 0.08201264590024948,
      "learning_rate": 1.0473484865448525e-06,
      "loss": 0.1877,
      "mean_token_accuracy": 0.9344814352691173,
      "step": 370
    },
    {
      "epoch": 0.9694619486185168,
      "grad_norm": 0.08764708787202835,
      "learning_rate": 4.954926249317815e-07,
      "loss": 0.1614,
      "mean_token_accuracy": 0.943453174829483,
      "step": 375
    },
    {
      "epoch": 0.9823881079334302,
      "grad_norm": 0.09879707545042038,
      "learning_rate": 1.4750491933247512e-07,
      "loss": 0.1984,
      "mean_token_accuracy": 0.9314216762781143,
      "step": 380
    },
    {
      "epoch": 0.9953142672483438,
      "grad_norm": 0.0833079144358635,
      "learning_rate": 4.0983385734660875e-09,
      "loss": 0.1962,
      "mean_token_accuracy": 0.9319176472723484,
      "step": 385
    },
    {
      "epoch": 0.9978994991113266,
      "mean_token_accuracy": 0.9182563126087189,
      "step": 386,
      "total_flos": 1.112161775475753e+17,
      "train_loss": 0.24216800009339584,
      "train_runtime": 6297.1418,
      "train_samples_per_second": 0.983,
      "train_steps_per_second": 0.061
    }
  ],
  "logging_steps": 5,
  "max_steps": 386,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.112161775475753e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}