{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.362281822051937,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.017028522775649212,
      "grad_norm": 2.3286359310150146,
      "learning_rate": 0.0001988623435722412,
      "loss": 4.1641,
      "mean_token_accuracy": 0.3866006717085838,
      "num_tokens": 4326.0,
      "step": 5
    },
    {
      "epoch": 0.034057045551298425,
      "grad_norm": 1.4926602840423584,
      "learning_rate": 0.00019772468714448238,
      "loss": 2.0591,
      "mean_token_accuracy": 0.6790341928601265,
      "num_tokens": 8676.0,
      "step": 10
    },
    {
      "epoch": 0.05108556832694764,
      "grad_norm": 1.0835968255996704,
      "learning_rate": 0.00019658703071672356,
      "loss": 1.3145,
      "mean_token_accuracy": 0.8256841972470284,
      "num_tokens": 12796.0,
      "step": 15
    },
    {
      "epoch": 0.06811409110259685,
      "grad_norm": 0.7724414467811584,
      "learning_rate": 0.00019544937428896475,
      "loss": 1.3324,
      "mean_token_accuracy": 0.8117582932114601,
      "num_tokens": 17119.0,
      "step": 20
    },
    {
      "epoch": 0.08514261387824607,
      "grad_norm": 0.785762369632721,
      "learning_rate": 0.00019431171786120593,
      "loss": 1.3375,
      "mean_token_accuracy": 0.8029867172241211,
      "num_tokens": 21499.0,
      "step": 25
    },
    {
      "epoch": 0.10217113665389528,
      "grad_norm": 0.6702725291252136,
      "learning_rate": 0.00019317406143344712,
      "loss": 1.2159,
      "mean_token_accuracy": 0.8230627462267875,
      "num_tokens": 25785.0,
      "step": 30
    },
    {
      "epoch": 0.11919965942954448,
      "grad_norm": 0.6287369728088379,
      "learning_rate": 0.0001920364050056883,
      "loss": 1.1634,
      "mean_token_accuracy": 0.8315445825457572,
      "num_tokens": 29980.0,
      "step": 35
    },
    {
      "epoch": 0.1362281822051937,
      "grad_norm": 0.6573143005371094,
      "learning_rate": 0.00019089874857792946,
      "loss": 1.3762,
      "mean_token_accuracy": 0.7971536681056023,
      "num_tokens": 34497.0,
      "step": 40
    },
    {
      "epoch": 0.1532567049808429,
      "grad_norm": 0.6560688614845276,
      "learning_rate": 0.00018976109215017067,
      "loss": 1.1589,
      "mean_token_accuracy": 0.8251897826790809,
      "num_tokens": 38800.0,
      "step": 45
    },
    {
      "epoch": 0.17028522775649213,
      "grad_norm": 0.6654791831970215,
      "learning_rate": 0.00018862343572241183,
      "loss": 1.1498,
      "mean_token_accuracy": 0.8319168403744698,
      "num_tokens": 43055.0,
      "step": 50
    },
    {
      "epoch": 0.18731375053214133,
      "grad_norm": 0.7387417554855347,
      "learning_rate": 0.00018748577929465302,
      "loss": 1.1265,
      "mean_token_accuracy": 0.8290244281291962,
      "num_tokens": 47346.0,
      "step": 55
    },
    {
      "epoch": 0.20434227330779056,
      "grad_norm": 0.5795997977256775,
      "learning_rate": 0.0001863481228668942,
      "loss": 1.0046,
      "mean_token_accuracy": 0.8534704640507698,
      "num_tokens": 51485.0,
      "step": 60
    },
    {
      "epoch": 0.22137079608343976,
      "grad_norm": 0.6834130883216858,
      "learning_rate": 0.0001852104664391354,
      "loss": 1.113,
      "mean_token_accuracy": 0.8345914751291275,
      "num_tokens": 55716.0,
      "step": 65
    },
    {
      "epoch": 0.23839931885908897,
      "grad_norm": 0.7211284041404724,
      "learning_rate": 0.00018407281001137657,
      "loss": 1.2132,
      "mean_token_accuracy": 0.8207321017980576,
      "num_tokens": 60030.0,
      "step": 70
    },
    {
      "epoch": 0.2554278416347382,
      "grad_norm": 0.6312271356582642,
      "learning_rate": 0.00018293515358361776,
      "loss": 1.0987,
      "mean_token_accuracy": 0.8333725541830063,
      "num_tokens": 64288.0,
      "step": 75
    },
    {
      "epoch": 0.2724563644103874,
      "grad_norm": 0.7119404077529907,
      "learning_rate": 0.00018179749715585894,
      "loss": 1.0981,
      "mean_token_accuracy": 0.8323698997497558,
      "num_tokens": 68548.0,
      "step": 80
    },
    {
      "epoch": 0.2894848871860366,
      "grad_norm": 0.9000201225280762,
      "learning_rate": 0.00018065984072810013,
      "loss": 1.146,
      "mean_token_accuracy": 0.8253335595130921,
      "num_tokens": 72914.0,
      "step": 85
    },
    {
      "epoch": 0.3065134099616858,
      "grad_norm": 0.6050537824630737,
      "learning_rate": 0.0001795221843003413,
      "loss": 1.0111,
      "mean_token_accuracy": 0.842674246430397,
      "num_tokens": 77168.0,
      "step": 90
    },
    {
      "epoch": 0.32354193273733506,
      "grad_norm": 0.684512197971344,
      "learning_rate": 0.0001783845278725825,
      "loss": 1.1186,
      "mean_token_accuracy": 0.8351956441998482,
      "num_tokens": 81439.0,
      "step": 95
    },
    {
      "epoch": 0.34057045551298426,
      "grad_norm": 0.7323129773139954,
      "learning_rate": 0.00017724687144482368,
      "loss": 1.1287,
      "mean_token_accuracy": 0.8245970487594605,
      "num_tokens": 85773.0,
      "step": 100
    },
    {
      "epoch": 0.35759897828863346,
      "grad_norm": 0.7547741532325745,
      "learning_rate": 0.00017610921501706487,
      "loss": 1.0868,
      "mean_token_accuracy": 0.8298466548323631,
      "num_tokens": 90117.0,
      "step": 105
    },
    {
      "epoch": 0.37462750106428266,
      "grad_norm": 0.7113769054412842,
      "learning_rate": 0.00017497155858930602,
      "loss": 1.1214,
      "mean_token_accuracy": 0.8299670115113258,
      "num_tokens": 94412.0,
      "step": 110
    },
    {
      "epoch": 0.39165602383993187,
      "grad_norm": 0.7486220002174377,
      "learning_rate": 0.00017383390216154724,
      "loss": 1.1452,
      "mean_token_accuracy": 0.8283886179327965,
      "num_tokens": 98782.0,
      "step": 115
    },
    {
      "epoch": 0.4086845466155811,
      "grad_norm": 0.658633828163147,
      "learning_rate": 0.0001726962457337884,
      "loss": 1.0415,
      "mean_token_accuracy": 0.8431179687380791,
      "num_tokens": 103011.0,
      "step": 120
    },
    {
      "epoch": 0.4257130693912303,
      "grad_norm": 0.783794105052948,
      "learning_rate": 0.0001715585893060296,
      "loss": 1.1056,
      "mean_token_accuracy": 0.8254878520965576,
      "num_tokens": 107402.0,
      "step": 125
    },
    {
      "epoch": 0.4427415921668795,
      "grad_norm": 0.7954565286636353,
      "learning_rate": 0.00017042093287827076,
      "loss": 1.0518,
      "mean_token_accuracy": 0.8375606715679169,
      "num_tokens": 111707.0,
      "step": 130
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 0.6913949847221375,
      "learning_rate": 0.00016928327645051198,
      "loss": 1.1104,
      "mean_token_accuracy": 0.8326424166560173,
      "num_tokens": 115998.0,
      "step": 135
    },
    {
      "epoch": 0.47679863771817793,
      "grad_norm": 0.7918037176132202,
      "learning_rate": 0.00016814562002275313,
      "loss": 1.0495,
      "mean_token_accuracy": 0.8390962019562721,
      "num_tokens": 120271.0,
      "step": 140
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 0.7816826701164246,
      "learning_rate": 0.00016700796359499432,
      "loss": 1.0463,
      "mean_token_accuracy": 0.8433194428682327,
      "num_tokens": 124461.0,
      "step": 145
    },
    {
      "epoch": 0.5108556832694764,
      "grad_norm": 0.8069042563438416,
      "learning_rate": 0.0001658703071672355,
      "loss": 1.0664,
      "mean_token_accuracy": 0.8278923153877258,
      "num_tokens": 128781.0,
      "step": 150
    },
    {
      "epoch": 0.5278842060451255,
      "grad_norm": 1.0114731788635254,
      "learning_rate": 0.0001647326507394767,
      "loss": 1.1217,
      "mean_token_accuracy": 0.8267972275614739,
      "num_tokens": 133167.0,
      "step": 155
    },
    {
      "epoch": 0.5449127288207748,
      "grad_norm": 0.7900378704071045,
      "learning_rate": 0.00016359499431171787,
      "loss": 0.971,
      "mean_token_accuracy": 0.8421565622091294,
      "num_tokens": 137464.0,
      "step": 160
    },
    {
      "epoch": 0.561941251596424,
      "grad_norm": 0.8059927821159363,
      "learning_rate": 0.00016245733788395906,
      "loss": 1.0457,
      "mean_token_accuracy": 0.8381335601210594,
      "num_tokens": 141760.0,
      "step": 165
    },
    {
      "epoch": 0.5789697743720732,
      "grad_norm": 0.7589371800422668,
      "learning_rate": 0.00016131968145620024,
      "loss": 1.0208,
      "mean_token_accuracy": 0.8342296928167343,
      "num_tokens": 146051.0,
      "step": 170
    },
    {
      "epoch": 0.5959982971477225,
      "grad_norm": 0.692416250705719,
      "learning_rate": 0.00016018202502844143,
      "loss": 0.9837,
      "mean_token_accuracy": 0.8510025009512902,
      "num_tokens": 150270.0,
      "step": 175
    },
    {
      "epoch": 0.6130268199233716,
      "grad_norm": 0.6662923097610474,
      "learning_rate": 0.0001590443686006826,
      "loss": 0.9835,
      "mean_token_accuracy": 0.8463438332080842,
      "num_tokens": 154537.0,
      "step": 180
    },
    {
      "epoch": 0.6300553426990209,
      "grad_norm": 0.7452691793441772,
      "learning_rate": 0.0001579067121729238,
      "loss": 1.0496,
      "mean_token_accuracy": 0.8375737622380257,
      "num_tokens": 158810.0,
      "step": 185
    },
    {
      "epoch": 0.6470838654746701,
      "grad_norm": 0.809700608253479,
      "learning_rate": 0.00015676905574516496,
      "loss": 1.0595,
      "mean_token_accuracy": 0.8294038504362107,
      "num_tokens": 163213.0,
      "step": 190
    },
    {
      "epoch": 0.6641123882503193,
      "grad_norm": 0.7708745002746582,
      "learning_rate": 0.00015563139931740617,
      "loss": 1.0976,
      "mean_token_accuracy": 0.8289313957095146,
      "num_tokens": 167607.0,
      "step": 195
    },
    {
      "epoch": 0.6811409110259685,
      "grad_norm": 0.6482141613960266,
      "learning_rate": 0.00015449374288964733,
      "loss": 1.0701,
      "mean_token_accuracy": 0.8261386752128601,
      "num_tokens": 172024.0,
      "step": 200
    },
    {
      "epoch": 0.6981694338016177,
      "grad_norm": 0.7164883613586426,
      "learning_rate": 0.00015335608646188854,
      "loss": 0.9878,
      "mean_token_accuracy": 0.8418964132666588,
      "num_tokens": 176317.0,
      "step": 205
    },
    {
      "epoch": 0.7151979565772669,
      "grad_norm": 0.710343599319458,
      "learning_rate": 0.0001522184300341297,
      "loss": 0.9306,
      "mean_token_accuracy": 0.8468034327030182,
      "num_tokens": 180558.0,
      "step": 210
    },
    {
      "epoch": 0.7322264793529162,
      "grad_norm": 0.863784909248352,
      "learning_rate": 0.00015108077360637088,
      "loss": 1.0544,
      "mean_token_accuracy": 0.8359945684671402,
      "num_tokens": 184885.0,
      "step": 215
    },
    {
      "epoch": 0.7492550021285653,
      "grad_norm": 0.8381637930870056,
      "learning_rate": 0.00014994311717861207,
      "loss": 1.0522,
      "mean_token_accuracy": 0.836811026930809,
      "num_tokens": 189219.0,
      "step": 220
    },
    {
      "epoch": 0.7662835249042146,
      "grad_norm": 0.683501124382019,
      "learning_rate": 0.00014880546075085325,
      "loss": 0.9427,
      "mean_token_accuracy": 0.8544979348778725,
      "num_tokens": 193437.0,
      "step": 225
    },
    {
      "epoch": 0.7833120476798637,
      "grad_norm": 0.7609812617301941,
      "learning_rate": 0.00014766780432309444,
      "loss": 1.0048,
      "mean_token_accuracy": 0.8403829500079155,
      "num_tokens": 197724.0,
      "step": 230
    },
    {
      "epoch": 0.800340570455513,
      "grad_norm": 0.8223949074745178,
      "learning_rate": 0.00014653014789533562,
      "loss": 1.078,
      "mean_token_accuracy": 0.8296335831284523,
      "num_tokens": 202141.0,
      "step": 235
    },
    {
      "epoch": 0.8173690932311622,
      "grad_norm": 0.9807024002075195,
      "learning_rate": 0.0001453924914675768,
      "loss": 0.9295,
      "mean_token_accuracy": 0.8495418474078178,
      "num_tokens": 206333.0,
      "step": 240
    },
    {
      "epoch": 0.8343976160068114,
      "grad_norm": 0.7864217162132263,
      "learning_rate": 0.000144254835039818,
      "loss": 1.0326,
      "mean_token_accuracy": 0.8422158002853394,
      "num_tokens": 210593.0,
      "step": 245
    },
    {
      "epoch": 0.8514261387824607,
      "grad_norm": 0.8183476328849792,
      "learning_rate": 0.00014311717861205915,
      "loss": 1.0419,
      "mean_token_accuracy": 0.8332764700055122,
      "num_tokens": 214925.0,
      "step": 250
    },
    {
      "epoch": 0.8684546615581098,
      "grad_norm": 0.7906543612480164,
      "learning_rate": 0.00014197952218430036,
      "loss": 0.9278,
      "mean_token_accuracy": 0.8531344652175903,
      "num_tokens": 219106.0,
      "step": 255
    },
    {
      "epoch": 0.885483184333759,
      "grad_norm": 0.7864625453948975,
      "learning_rate": 0.00014084186575654152,
      "loss": 0.9764,
      "mean_token_accuracy": 0.8394758701324463,
      "num_tokens": 223398.0,
      "step": 260
    },
    {
      "epoch": 0.9025117071094083,
      "grad_norm": 0.7231821417808533,
      "learning_rate": 0.00013970420932878273,
      "loss": 1.0357,
      "mean_token_accuracy": 0.8368565306067467,
      "num_tokens": 227720.0,
      "step": 265
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 0.7288718223571777,
      "learning_rate": 0.0001385665529010239,
      "loss": 1.0178,
      "mean_token_accuracy": 0.842984040081501,
      "num_tokens": 232021.0,
      "step": 270
    },
    {
      "epoch": 0.9365687526607067,
      "grad_norm": 0.7466130256652832,
      "learning_rate": 0.0001374288964732651,
      "loss": 0.9974,
      "mean_token_accuracy": 0.8431279867887497,
      "num_tokens": 236312.0,
      "step": 275
    },
    {
      "epoch": 0.9535972754363559,
      "grad_norm": 0.7876543402671814,
      "learning_rate": 0.00013629124004550626,
      "loss": 1.0057,
      "mean_token_accuracy": 0.8389794036746026,
      "num_tokens": 240594.0,
      "step": 280
    },
    {
      "epoch": 0.9706257982120051,
      "grad_norm": 0.738293468952179,
      "learning_rate": 0.00013515358361774744,
      "loss": 0.9129,
      "mean_token_accuracy": 0.8528255730867386,
      "num_tokens": 244859.0,
      "step": 285
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 0.8145989775657654,
      "learning_rate": 0.00013401592718998863,
      "loss": 0.9188,
      "mean_token_accuracy": 0.8555046319961548,
      "num_tokens": 249060.0,
      "step": 290
    },
    {
      "epoch": 1.0046828437633035,
      "grad_norm": 0.693233072757721,
      "learning_rate": 0.00013287827076222981,
      "loss": 1.0648,
      "mean_token_accuracy": 0.8477524280548095,
      "num_tokens": 253361.0,
      "step": 295
    },
    {
      "epoch": 1.0217113665389528,
      "grad_norm": 0.8171095848083496,
      "learning_rate": 0.000131740614334471,
      "loss": 0.883,
      "mean_token_accuracy": 0.8561580404639244,
      "num_tokens": 257633.0,
      "step": 300
    },
    {
      "epoch": 1.038739889314602,
      "grad_norm": 0.7343988418579102,
      "learning_rate": 0.00013060295790671218,
      "loss": 0.8591,
      "mean_token_accuracy": 0.8583228841423989,
      "num_tokens": 261975.0,
      "step": 305
    },
    {
      "epoch": 1.055768412090251,
      "grad_norm": 0.6689726114273071,
      "learning_rate": 0.00012946530147895337,
      "loss": 0.8397,
      "mean_token_accuracy": 0.8668102487921715,
      "num_tokens": 266174.0,
      "step": 310
    },
    {
      "epoch": 1.0727969348659003,
      "grad_norm": 0.8284767270088196,
      "learning_rate": 0.00012832764505119455,
      "loss": 0.9108,
      "mean_token_accuracy": 0.8551412716507911,
      "num_tokens": 270416.0,
      "step": 315
    },
    {
      "epoch": 1.0898254576415496,
      "grad_norm": 0.7800877690315247,
      "learning_rate": 0.0001271899886234357,
      "loss": 0.923,
      "mean_token_accuracy": 0.8521544948220253,
      "num_tokens": 274802.0,
      "step": 320
    },
    {
      "epoch": 1.1068539804171988,
      "grad_norm": 0.803035318851471,
      "learning_rate": 0.00012605233219567692,
      "loss": 0.9009,
      "mean_token_accuracy": 0.8583568409085274,
      "num_tokens": 279061.0,
      "step": 325
    },
    {
      "epoch": 1.123882503192848,
      "grad_norm": 0.7928335666656494,
      "learning_rate": 0.00012491467576791808,
      "loss": 0.8677,
      "mean_token_accuracy": 0.860838033258915,
      "num_tokens": 283290.0,
      "step": 330
    },
    {
      "epoch": 1.1409110259684971,
      "grad_norm": 0.82757967710495,
      "learning_rate": 0.0001237770193401593,
      "loss": 0.9326,
      "mean_token_accuracy": 0.8526211172342301,
      "num_tokens": 287627.0,
      "step": 335
    },
    {
      "epoch": 1.1579395487441464,
      "grad_norm": 0.8021914958953857,
      "learning_rate": 0.00012263936291240045,
      "loss": 0.9066,
      "mean_token_accuracy": 0.8505649596452713,
      "num_tokens": 291975.0,
      "step": 340
    },
    {
      "epoch": 1.1749680715197957,
      "grad_norm": 0.7565475106239319,
      "learning_rate": 0.00012150170648464165,
      "loss": 0.8812,
      "mean_token_accuracy": 0.8562570214271545,
      "num_tokens": 296232.0,
      "step": 345
    },
    {
      "epoch": 1.191996594295445,
      "grad_norm": 0.8126978278160095,
      "learning_rate": 0.00012036405005688282,
      "loss": 0.905,
      "mean_token_accuracy": 0.8556828230619431,
      "num_tokens": 300591.0,
      "step": 350
    },
    {
      "epoch": 1.2090251170710942,
      "grad_norm": 0.8174765706062317,
      "learning_rate": 0.000119226393629124,
      "loss": 0.9861,
      "mean_token_accuracy": 0.8424012079834938,
      "num_tokens": 304998.0,
      "step": 355
    },
    {
      "epoch": 1.2260536398467432,
      "grad_norm": 0.8921035528182983,
      "learning_rate": 0.00011808873720136519,
      "loss": 0.7724,
      "mean_token_accuracy": 0.8735315829515458,
      "num_tokens": 309137.0,
      "step": 360
    },
    {
      "epoch": 1.2430821626223925,
      "grad_norm": 0.855992317199707,
      "learning_rate": 0.00011695108077360638,
      "loss": 0.8922,
      "mean_token_accuracy": 0.854039640724659,
      "num_tokens": 313389.0,
      "step": 365
    },
    {
      "epoch": 1.2601106853980417,
      "grad_norm": 0.9408860206604004,
      "learning_rate": 0.00011581342434584756,
      "loss": 0.9254,
      "mean_token_accuracy": 0.8507854476571083,
      "num_tokens": 317743.0,
      "step": 370
    },
    {
      "epoch": 1.277139208173691,
      "grad_norm": 0.7847197651863098,
      "learning_rate": 0.00011467576791808873,
      "loss": 0.8563,
      "mean_token_accuracy": 0.869862625002861,
      "num_tokens": 321873.0,
      "step": 375
    },
    {
      "epoch": 1.29416773094934,
      "grad_norm": 0.8556840419769287,
      "learning_rate": 0.00011353811149032993,
      "loss": 0.8489,
      "mean_token_accuracy": 0.8591933220624923,
      "num_tokens": 326124.0,
      "step": 380
    },
    {
      "epoch": 1.3111962537249893,
      "grad_norm": 0.8849514126777649,
      "learning_rate": 0.0001124004550625711,
      "loss": 0.8526,
      "mean_token_accuracy": 0.8600789621472359,
      "num_tokens": 330405.0,
      "step": 385
    },
    {
      "epoch": 1.3282247765006385,
      "grad_norm": 0.8099865913391113,
      "learning_rate": 0.00011126279863481229,
      "loss": 0.9309,
      "mean_token_accuracy": 0.8511071890592575,
      "num_tokens": 334701.0,
      "step": 390
    },
    {
      "epoch": 1.3452532992762878,
      "grad_norm": 0.8668401837348938,
      "learning_rate": 0.00011012514220705347,
      "loss": 0.8671,
      "mean_token_accuracy": 0.8605618432164193,
      "num_tokens": 338951.0,
      "step": 395
    },
    {
      "epoch": 1.362281822051937,
      "grad_norm": 0.8669872283935547,
      "learning_rate": 0.00010898748577929466,
      "loss": 0.8796,
      "mean_token_accuracy": 0.8592276558279991,
      "num_tokens": 343203.0,
      "step": 400
    }
  ],
  "logging_steps": 5,
  "max_steps": 879,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3216867264285696.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}