{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 984,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 70.84053355134651,
      "learning_rate": 2.0202020202020206e-06,
      "loss": 16.5976,
      "step": 1
    },
    {
      "epoch": 0.02,
      "grad_norm": 70.40484875837613,
      "learning_rate": 1.0101010101010101e-05,
      "loss": 16.2878,
      "step": 5
    },
    {
      "epoch": 0.03,
      "grad_norm": 64.13592776682735,
      "learning_rate": 2.0202020202020203e-05,
      "loss": 15.6234,
      "step": 10
    },
    {
      "epoch": 0.05,
      "grad_norm": 48.17657464142415,
      "learning_rate": 3.0303030303030306e-05,
      "loss": 13.1692,
      "step": 15
    },
    {
      "epoch": 0.06,
      "grad_norm": 42.09774383233372,
      "learning_rate": 4.0404040404040405e-05,
      "loss": 9.9677,
      "step": 20
    },
    {
      "epoch": 0.08,
      "grad_norm": 21.40017494750998,
      "learning_rate": 5.050505050505051e-05,
      "loss": 6.438,
      "step": 25
    },
    {
      "epoch": 0.09,
      "grad_norm": 13.47184864955966,
      "learning_rate": 6.060606060606061e-05,
      "loss": 4.5534,
      "step": 30
    },
    {
      "epoch": 0.11,
      "grad_norm": 9.0175134340465,
      "learning_rate": 7.07070707070707e-05,
      "loss": 3.3841,
      "step": 35
    },
    {
      "epoch": 0.12,
      "grad_norm": 7.9069963595872474,
      "learning_rate": 8.080808080808081e-05,
      "loss": 2.4486,
      "step": 40
    },
    {
      "epoch": 0.14,
      "grad_norm": 4.497563859166001,
      "learning_rate": 9.090909090909092e-05,
      "loss": 1.9157,
      "step": 45
    },
    {
      "epoch": 0.15,
      "grad_norm": 3.8365103057277237,
      "learning_rate": 0.00010101010101010102,
      "loss": 1.6253,
      "step": 50
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.6711072114914822,
      "learning_rate": 0.00011111111111111112,
      "loss": 1.4531,
      "step": 55
    },
    {
      "epoch": 0.18,
      "grad_norm": 2.8242292867987038,
      "learning_rate": 0.00012121212121212122,
      "loss": 1.355,
      "step": 60
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.5627627321505486,
      "learning_rate": 0.00013131313131313133,
      "loss": 1.3144,
      "step": 65
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.9610550212854403,
      "learning_rate": 0.0001414141414141414,
      "loss": 1.2403,
      "step": 70
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.4826190020479229,
      "learning_rate": 0.00015151515151515152,
      "loss": 1.2116,
      "step": 75
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.4179410680444433,
      "learning_rate": 0.00016161616161616162,
      "loss": 1.1361,
      "step": 80
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.6440373948698321,
      "learning_rate": 0.00017171717171717173,
      "loss": 1.0898,
      "step": 85
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.4741470072787244,
      "learning_rate": 0.00018181818181818183,
      "loss": 1.1321,
      "step": 90
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.4918119127052252,
      "learning_rate": 0.00019191919191919191,
      "loss": 1.1502,
      "step": 95
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.780737333023736,
      "learning_rate": 0.0001999993699387764,
      "loss": 1.0674,
      "step": 100
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.5713634753942138,
      "learning_rate": 0.0001999773186295914,
      "loss": 1.0976,
      "step": 105
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.4792461465668252,
      "learning_rate": 0.0001999237721983199,
      "loss": 1.0793,
      "step": 110
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.414080214214796,
      "learning_rate": 0.0001998387475133018,
      "loss": 1.1008,
      "step": 115
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.4612237811023239,
      "learning_rate": 0.00019972227135924052,
      "loss": 1.0824,
      "step": 120
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.5484440969715458,
      "learning_rate": 0.00019957438042876542,
      "loss": 1.0178,
      "step": 125
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.0098511861836643,
      "learning_rate": 0.0001993951213108726,
      "loss": 1.0397,
      "step": 130
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.461268889218997,
      "learning_rate": 0.00019918455047624847,
      "loss": 1.0558,
      "step": 135
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.816402545663251,
      "learning_rate": 0.00019894273425948003,
      "loss": 0.9983,
      "step": 140
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.5043083827737078,
      "learning_rate": 0.0001986697488381581,
      "loss": 0.9565,
      "step": 145
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.5595660341738484,
      "learning_rate": 0.00019836568020887964,
      "loss": 0.9811,
      "step": 150
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.3010086010553632,
      "learning_rate": 0.00019803062416015675,
      "loss": 1.05,
      "step": 155
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.936074537977103,
      "learning_rate": 0.0001976646862422413,
      "loss": 1.0033,
      "step": 160
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.43894566672306,
      "learning_rate": 0.00019726798173387415,
      "loss": 0.9904,
      "step": 165
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.0226681751936604,
      "learning_rate": 0.0001968406356059696,
      "loss": 1.0087,
      "step": 170
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.5765984883843074,
      "learning_rate": 0.00019638278248224688,
      "loss": 1.0074,
      "step": 175
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.4953203184933894,
      "learning_rate": 0.00019589456659682057,
      "loss": 0.9663,
      "step": 180
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.3905440186304812,
      "learning_rate": 0.00019537614174876356,
      "loss": 0.9621,
      "step": 185
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.7108204646804686,
      "learning_rate": 0.0001948276712536569,
      "loss": 0.958,
      "step": 190
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.146705235438316,
      "learning_rate": 0.00019424932789214157,
      "loss": 1.0311,
      "step": 195
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.1584586792685783,
      "learning_rate": 0.00019364129385548892,
      "loss": 1.0171,
      "step": 200
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.629476089763749,
      "learning_rate": 0.00019300376068820602,
      "loss": 0.9775,
      "step": 205
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.2581783118467773,
      "learning_rate": 0.00019233692922769496,
      "loss": 1.0471,
      "step": 210
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.4906993491973393,
      "learning_rate": 0.0001916410095409843,
      "loss": 0.9967,
      "step": 215
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.1893776969242746,
      "learning_rate": 0.00019091622085855363,
      "loss": 0.9619,
      "step": 220
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.4151710347672513,
      "learning_rate": 0.00019016279150527044,
      "loss": 1.0227,
      "step": 225
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.2558733422369428,
      "learning_rate": 0.00018938095882846306,
      "loss": 0.9832,
      "step": 230
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.3320637162932236,
      "learning_rate": 0.00018857096912315062,
      "loss": 0.9802,
      "step": 235
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.2878623504565412,
      "learning_rate": 0.00018773307755445465,
      "loss": 0.9251,
      "step": 240
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.213232935367416,
      "learning_rate": 0.0001868675480772163,
      "loss": 0.98,
      "step": 245
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.3405116881469274,
      "learning_rate": 0.00018597465335284437,
      "loss": 0.9473,
      "step": 250
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.543540810073074,
      "learning_rate": 0.0001850546746634211,
      "loss": 0.9902,
      "step": 255
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.3710808148280944,
      "learning_rate": 0.00018410790182309169,
      "loss": 0.975,
      "step": 260
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.21692171592665,
      "learning_rate": 0.00018313463308676636,
      "loss": 0.9613,
      "step": 265
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.1934007014753554,
      "learning_rate": 0.00018213517505616338,
      "loss": 0.9618,
      "step": 270
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.60819637926966,
      "learning_rate": 0.00018110984258322237,
      "loss": 0.9939,
      "step": 275
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.2454326978067687,
      "learning_rate": 0.00018005895867091896,
      "loss": 0.9191,
      "step": 280
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.9690417217411257,
      "learning_rate": 0.0001789828543715116,
      "loss": 0.9421,
      "step": 285
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.3978495893847322,
      "learning_rate": 0.0001778818686822523,
      "loss": 0.9757,
      "step": 290
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.2683425234668648,
      "learning_rate": 0.00017675634843859514,
      "loss": 0.9902,
      "step": 295
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.2072894977789046,
      "learning_rate": 0.00017560664820493498,
      "loss": 0.9517,
      "step": 300
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.179285109564365,
      "learning_rate": 0.00017443313016291185,
      "loss": 0.9657,
      "step": 305
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.1314300153647592,
      "learning_rate": 0.00017323616399731533,
      "loss": 0.9438,
      "step": 310
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.0638903666221389,
      "learning_rate": 0.0001720161267796256,
      "loss": 1.0097,
      "step": 315
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.0687942439764777,
      "learning_rate": 0.00017077340284922732,
      "loss": 0.9453,
      "step": 320
    },
    {
      "epoch": 0.99,
      "grad_norm": 1.2396289777933702,
      "learning_rate": 0.000169508383692334,
      "loss": 0.9388,
      "step": 325
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1537501811981201,
      "eval_runtime": 156.3402,
      "eval_samples_per_second": 14.775,
      "eval_steps_per_second": 0.467,
      "step": 328
    },
    {
      "epoch": 1.01,
      "grad_norm": 1.178139096940401,
      "learning_rate": 0.00016822146781866098,
      "loss": 0.9355,
      "step": 330
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.9864464240664622,
      "learning_rate": 0.00016691306063588583,
      "loss": 0.9135,
      "step": 335
    },
    {
      "epoch": 1.04,
      "grad_norm": 1.0449361148940848,
      "learning_rate": 0.00016558357432193578,
      "loss": 0.9088,
      "step": 340
    },
    {
      "epoch": 1.05,
      "grad_norm": 1.02719313804267,
      "learning_rate": 0.00016423342769514228,
      "loss": 0.9361,
      "step": 345
    },
    {
      "epoch": 1.07,
      "grad_norm": 1.1796531987937382,
      "learning_rate": 0.00016286304608230368,
      "loss": 0.9167,
      "step": 350
    },
    {
      "epoch": 1.08,
      "grad_norm": 1.2165440327749395,
      "learning_rate": 0.0001614728611846978,
      "loss": 0.9071,
      "step": 355
    },
    {
      "epoch": 1.1,
      "grad_norm": 1.0680390308725847,
      "learning_rate": 0.0001600633109420861,
      "loss": 0.9423,
      "step": 360
    },
    {
      "epoch": 1.11,
      "grad_norm": 1.1460931817295472,
      "learning_rate": 0.00015863483939475281,
      "loss": 0.9414,
      "step": 365
    },
    {
      "epoch": 1.13,
      "grad_norm": 1.2232830262553624,
      "learning_rate": 0.00015718789654362204,
      "loss": 0.9501,
      "step": 370
    },
    {
      "epoch": 1.14,
      "grad_norm": 1.1171762766168023,
      "learning_rate": 0.00015572293820849753,
      "loss": 0.9278,
      "step": 375
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.134580557741046,
      "learning_rate": 0.00015424042588446882,
      "loss": 0.8877,
      "step": 380
    },
    {
      "epoch": 1.17,
      "grad_norm": 1.1467717113467715,
      "learning_rate": 0.00015274082659653,
      "loss": 0.8968,
      "step": 385
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.0245879848355788,
      "learning_rate": 0.0001512246127524561,
      "loss": 0.933,
      "step": 390
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.9149640789208816,
      "learning_rate": 0.0001496922619939842,
      "loss": 0.9562,
      "step": 395
    },
    {
      "epoch": 1.22,
      "grad_norm": 1.5241591126822251,
      "learning_rate": 0.00014814425704634508,
      "loss": 0.9906,
      "step": 400
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.0478360565843916,
      "learning_rate": 0.00014658108556619417,
      "loss": 0.9131,
      "step": 405
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.1580130995552973,
      "learning_rate": 0.00014500323998798843,
      "loss": 0.9385,
      "step": 410
    },
    {
      "epoch": 1.27,
      "grad_norm": 1.247960363943888,
      "learning_rate": 0.00014341121736885843,
      "loss": 0.8527,
      "step": 415
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.1576786962267551,
      "learning_rate": 0.00014180551923202405,
      "loss": 0.9051,
      "step": 420
    },
    {
      "epoch": 1.3,
      "grad_norm": 1.0608499549231114,
      "learning_rate": 0.00014018665140880332,
      "loss": 0.908,
      "step": 425
    },
    {
      "epoch": 1.31,
      "grad_norm": 1.1012203287352098,
      "learning_rate": 0.000138555123879264,
      "loss": 0.9342,
      "step": 430
    },
    {
      "epoch": 1.33,
      "grad_norm": 1.0806519524832878,
      "learning_rate": 0.00013691145061156844,
      "loss": 0.9136,
      "step": 435
    },
    {
      "epoch": 1.34,
      "grad_norm": 1.0359931963939946,
      "learning_rate": 0.00013525614940006184,
      "loss": 0.8949,
      "step": 440
    },
    {
      "epoch": 1.36,
      "grad_norm": 1.3063581428657738,
      "learning_rate": 0.00013358974170215538,
      "loss": 0.9174,
      "step": 445
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.0718302654560807,
      "learning_rate": 0.00013191275247405527,
      "loss": 0.8404,
      "step": 450
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.9302638253667144,
      "learning_rate": 0.00013022571000538953,
      "loss": 0.9538,
      "step": 455
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.002831930422535,
      "learning_rate": 0.00012852914575278498,
      "loss": 0.9497,
      "step": 460
    },
    {
      "epoch": 1.42,
      "grad_norm": 1.0473392521685814,
      "learning_rate": 0.0001268235941724463,
      "loss": 0.9065,
      "step": 465
    },
    {
      "epoch": 1.43,
      "grad_norm": 1.077577179849463,
      "learning_rate": 0.00012510959255179006,
      "loss": 0.9666,
      "step": 470
    },
    {
      "epoch": 1.45,
      "grad_norm": 1.0906580828243861,
      "learning_rate": 0.00012338768084018718,
      "loss": 0.9149,
      "step": 475
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.1044210032346624,
      "learning_rate": 0.00012165840147886656,
      "loss": 0.9367,
      "step": 480
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.20379457933143,
      "learning_rate": 0.00011992229923003377,
      "loss": 0.9065,
      "step": 485
    },
    {
      "epoch": 1.49,
      "grad_norm": 1.2198024955592683,
      "learning_rate": 0.00011817992100525872,
      "loss": 0.9161,
      "step": 490
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.9478289622764081,
      "learning_rate": 0.00011643181569318595,
      "loss": 0.9285,
      "step": 495
    },
    {
      "epoch": 1.52,
      "grad_norm": 1.3556113362942024,
      "learning_rate": 0.00011467853398662236,
      "loss": 0.9149,
      "step": 500
    },
    {
      "epoch": 1.54,
      "grad_norm": 1.0405819610764953,
      "learning_rate": 0.00011292062820905651,
      "loss": 0.8987,
      "step": 505
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.0272942715335152,
      "learning_rate": 0.00011115865214066414,
      "loss": 0.9296,
      "step": 510
    },
    {
      "epoch": 1.57,
      "grad_norm": 1.235836010680135,
      "learning_rate": 0.00010939316084385489,
      "loss": 0.907,
      "step": 515
    },
    {
      "epoch": 1.59,
      "grad_norm": 1.141603178243694,
      "learning_rate": 0.00010762471048841501,
      "loss": 0.9424,
      "step": 520
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.9315641865978828,
      "learning_rate": 0.00010585385817630137,
      "loss": 0.8801,
      "step": 525
    },
    {
      "epoch": 1.62,
      "grad_norm": 1.2155191029089636,
      "learning_rate": 0.00010408116176614167,
      "loss": 0.9832,
      "step": 530
    },
    {
      "epoch": 1.63,
      "grad_norm": 1.03073958762087,
      "learning_rate": 0.00010230717969749616,
      "loss": 0.9226,
      "step": 535
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.9081025102798024,
      "learning_rate": 0.00010053247081493685,
      "loss": 0.9383,
      "step": 540
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.9719608179885268,
      "learning_rate": 9.875759419199848e-05,
      "loss": 0.9139,
      "step": 545
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.9765085879374533,
      "learning_rate": 9.698310895505784e-05,
      "loss": 0.9546,
      "step": 550
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.9934690602517904,
      "learning_rate": 9.520957410719632e-05,
      "loss": 0.9085,
      "step": 555
    },
    {
      "epoch": 1.71,
      "grad_norm": 1.0535937418956107,
      "learning_rate": 9.343754835210106e-05,
      "loss": 0.9058,
      "step": 560
    },
    {
      "epoch": 1.72,
      "grad_norm": 1.023947025769762,
      "learning_rate": 9.166758991806098e-05,
      "loss": 0.9229,
      "step": 565
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.9292604570424309,
      "learning_rate": 8.990025638211179e-05,
      "loss": 0.9222,
      "step": 570
    },
    {
      "epoch": 1.75,
      "grad_norm": 1.114153037641386,
      "learning_rate": 8.813610449438692e-05,
      "loss": 0.8626,
      "step": 575
    },
    {
      "epoch": 1.77,
      "grad_norm": 1.040325901090811,
      "learning_rate": 8.637569000272835e-05,
      "loss": 0.9408,
      "step": 580
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.8817859269222795,
      "learning_rate": 8.461956747761374e-05,
      "loss": 0.8877,
      "step": 585
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.1284180630525456,
      "learning_rate": 8.286829013745382e-05,
      "loss": 0.9207,
      "step": 590
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.9343742775094561,
      "learning_rate": 8.11224096743163e-05,
      "loss": 0.8933,
      "step": 595
    },
    {
      "epoch": 1.83,
      "grad_norm": 1.0538577448642068,
      "learning_rate": 7.938247608013021e-05,
      "loss": 0.9298,
      "step": 600
    },
    {
      "epoch": 1.84,
      "grad_norm": 1.0018046757269787,
      "learning_rate": 7.764903747342603e-05,
      "loss": 0.9026,
      "step": 605
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.9082963605695086,
      "learning_rate": 7.592263992666604e-05,
      "loss": 0.8603,
      "step": 610
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.9123839165404105,
      "learning_rate": 7.420382729421883e-05,
      "loss": 0.9466,
      "step": 615
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.8612729466211886,
      "learning_rate": 7.249314104103315e-05,
      "loss": 0.9053,
      "step": 620
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.9068865831245562,
      "learning_rate": 7.079112007206394e-05,
      "loss": 0.9099,
      "step": 625
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.9654244669285843,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.922,
      "step": 630
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.8439836577250905,
      "learning_rate": 6.741521578888272e-05,
      "loss": 0.9283,
      "step": 635
    },
    {
      "epoch": 1.95,
      "grad_norm": 1.0735235072492841,
      "learning_rate": 6.574239596105951e-05,
      "loss": 0.9292,
      "step": 640
    },
    {
      "epoch": 1.97,
      "grad_norm": 1.1672700848634472,
      "learning_rate": 6.408036805520801e-05,
      "loss": 0.8986,
      "step": 645
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.952568318833893,
      "learning_rate": 6.242965564780071e-05,
      "loss": 0.8812,
      "step": 650
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1182556181299996,
      "learning_rate": 6.079077875067136e-05,
      "loss": 0.9232,
      "step": 655
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.1174074411392212,
      "eval_runtime": 155.7531,
      "eval_samples_per_second": 14.831,
      "eval_steps_per_second": 0.469,
      "step": 656
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.0745550951971172,
      "learning_rate": 5.916425364719975e-05,
      "loss": 0.8609,
      "step": 660
    },
    {
      "epoch": 2.03,
      "grad_norm": 0.9414624320547766,
      "learning_rate": 5.755059272967054e-05,
      "loss": 0.8921,
      "step": 665
    },
    {
      "epoch": 2.04,
      "grad_norm": 0.9382686352429901,
      "learning_rate": 5.5950304337858176e-05,
      "loss": 0.8712,
      "step": 670
    },
    {
      "epoch": 2.06,
      "grad_norm": 1.0370060367806875,
      "learning_rate": 5.436389259888841e-05,
      "loss": 0.9143,
      "step": 675
    },
    {
      "epoch": 2.07,
      "grad_norm": 1.0449428824200535,
      "learning_rate": 5.279185726842658e-05,
      "loss": 0.8609,
      "step": 680
    },
    {
      "epoch": 2.09,
      "grad_norm": 1.2140796388499904,
      "learning_rate": 5.1234693573243554e-05,
      "loss": 0.8814,
      "step": 685
    },
    {
      "epoch": 2.1,
      "grad_norm": 1.0118415907930658,
      "learning_rate": 4.969289205520778e-05,
      "loss": 0.8797,
      "step": 690
    },
    {
      "epoch": 2.12,
      "grad_norm": 1.0784940351527341,
      "learning_rate": 4.816693841675368e-05,
      "loss": 0.8898,
      "step": 695
    },
    {
      "epoch": 2.13,
      "grad_norm": 1.231043264244889,
      "learning_rate": 4.6657313367874256e-05,
      "loss": 0.8772,
      "step": 700
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.93091465787962,
      "learning_rate": 4.516449247468666e-05,
      "loss": 0.9097,
      "step": 705
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.1113936820092343,
      "learning_rate": 4.368894600961792e-05,
      "loss": 0.8757,
      "step": 710
    },
    {
      "epoch": 2.18,
      "grad_norm": 0.9533707648230687,
      "learning_rate": 4.223113880325865e-05,
      "loss": 0.872,
      "step": 715
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.0584356613150987,
      "learning_rate": 4.0791530097930676e-05,
      "loss": 0.848,
      "step": 720
    },
    {
      "epoch": 2.21,
      "grad_norm": 1.0955157548141539,
      "learning_rate": 3.937057340301551e-05,
      "loss": 0.8831,
      "step": 725
    },
    {
      "epoch": 2.23,
      "grad_norm": 1.0974350105007893,
      "learning_rate": 3.7968716352088406e-05,
      "loss": 0.8965,
      "step": 730
    },
    {
      "epoch": 2.24,
      "grad_norm": 1.07417322379521,
      "learning_rate": 3.658640056190378e-05,
      "loss": 0.8024,
      "step": 735
    },
    {
      "epoch": 2.26,
      "grad_norm": 0.932073914233341,
      "learning_rate": 3.52240614932758e-05,
      "loss": 0.859,
      "step": 740
    },
    {
      "epoch": 2.27,
      "grad_norm": 0.960220324993126,
      "learning_rate": 3.388212831389854e-05,
      "loss": 0.8955,
      "step": 745
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.9981005413964692,
      "learning_rate": 3.256102376314824e-05,
      "loss": 0.94,
      "step": 750
    },
    {
      "epoch": 2.3,
      "grad_norm": 1.0397017457559017,
      "learning_rate": 3.126116401891085e-05,
      "loss": 0.8587,
      "step": 755
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.9448895278738276,
      "learning_rate": 2.9982958566476705e-05,
      "loss": 0.8707,
      "step": 760
    },
    {
      "epoch": 2.33,
      "grad_norm": 1.0610274165270486,
      "learning_rate": 2.872681006954315e-05,
      "loss": 0.8825,
      "step": 765
    },
    {
      "epoch": 2.35,
      "grad_norm": 1.0210002250645476,
      "learning_rate": 2.749311424336659e-05,
      "loss": 0.8915,
      "step": 770
    },
    {
      "epoch": 2.36,
      "grad_norm": 0.9667083953810022,
      "learning_rate": 2.6282259730103e-05,
      "loss": 0.8727,
      "step": 775
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.9251306242277755,
      "learning_rate": 2.5094627976376927e-05,
      "loss": 0.8887,
      "step": 780
    },
    {
      "epoch": 2.39,
      "grad_norm": 0.9383384405043022,
      "learning_rate": 2.393059311311715e-05,
      "loss": 0.8894,
      "step": 785
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.8717419970020719,
      "learning_rate": 2.2790521837697034e-05,
      "loss": 0.8279,
      "step": 790
    },
    {
      "epoch": 2.42,
      "grad_norm": 0.9486113739115168,
      "learning_rate": 2.167477329841633e-05,
      "loss": 0.8545,
      "step": 795
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.9988727624604463,
      "learning_rate": 2.0583698981361577e-05,
      "loss": 0.8624,
      "step": 800
    },
    {
      "epoch": 2.45,
      "grad_norm": 0.9856023900634717,
      "learning_rate": 1.9517642599679807e-05,
      "loss": 0.8953,
      "step": 805
    },
    {
      "epoch": 2.47,
      "grad_norm": 1.146341438442503,
      "learning_rate": 1.8476939985301256e-05,
      "loss": 0.8583,
      "step": 810
    },
    {
      "epoch": 2.48,
      "grad_norm": 0.9934930089425651,
      "learning_rate": 1.7461918983144588e-05,
      "loss": 0.8236,
      "step": 815
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.0570151368189196,
      "learning_rate": 1.6472899347838356e-05,
      "loss": 0.8521,
      "step": 820
    },
    {
      "epoch": 2.52,
      "grad_norm": 0.940846894356948,
      "learning_rate": 1.5510192642991073e-05,
      "loss": 0.8412,
      "step": 825
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.9182797173905372,
      "learning_rate": 1.4574102143041512e-05,
      "loss": 0.8292,
      "step": 830
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.0514937418330759,
      "learning_rate": 1.3664922737720586e-05,
      "loss": 0.8463,
      "step": 835
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.9672420613383418,
      "learning_rate": 1.2782940839154111e-05,
      "loss": 0.8677,
      "step": 840
    },
    {
      "epoch": 2.58,
      "grad_norm": 1.0215480963434165,
      "learning_rate": 1.192843429163677e-05,
      "loss": 0.8555,
      "step": 845
    },
    {
      "epoch": 2.59,
      "grad_norm": 1.133368133421988,
      "learning_rate": 1.1101672284104624e-05,
      "loss": 0.8732,
      "step": 850
    },
    {
      "epoch": 2.61,
      "grad_norm": 0.9880449036656486,
      "learning_rate": 1.0302915265334723e-05,
      "loss": 0.8832,
      "step": 855
    },
    {
      "epoch": 2.62,
      "grad_norm": 0.934408626896597,
      "learning_rate": 9.532414861897632e-06,
      "loss": 0.8527,
      "step": 860
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.90008489968748,
      "learning_rate": 8.790413798889452e-06,
      "loss": 0.8611,
      "step": 865
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.9910034914102378,
      "learning_rate": 8.077145823467924e-06,
      "loss": 0.932,
      "step": 870
    },
    {
      "epoch": 2.67,
      "grad_norm": 1.1138535634921578,
      "learning_rate": 7.392835631216766e-06,
      "loss": 0.878,
      "step": 875
    },
    {
      "epoch": 2.68,
      "grad_norm": 1.1422833234291747,
      "learning_rate": 6.7376987953614246e-06,
      "loss": 0.8607,
      "step": 880
    },
    {
      "epoch": 2.7,
      "grad_norm": 1.1003844781363714,
      "learning_rate": 6.111941698858681e-06,
      "loss": 0.912,
      "step": 885
    },
    {
      "epoch": 2.71,
      "grad_norm": 1.137696707362488,
      "learning_rate": 5.5157614693812865e-06,
      "loss": 0.8416,
      "step": 890
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.9640103484120158,
      "learning_rate": 4.9493459172183285e-06,
      "loss": 0.8452,
      "step": 895
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.9223029933937373,
      "learning_rate": 4.412873476110702e-06,
      "loss": 0.8568,
      "step": 900
    },
    {
      "epoch": 2.76,
      "grad_norm": 0.9007733331158723,
      "learning_rate": 3.906513147040425e-06,
      "loss": 0.8713,
      "step": 905
    },
    {
      "epoch": 2.77,
      "grad_norm": 1.0288738184384674,
      "learning_rate": 3.43042444499152e-06,
      "loss": 0.9235,
      "step": 910
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.0184524502121122,
      "learning_rate": 2.984757348699152e-06,
      "loss": 0.846,
      "step": 915
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.1092808565987327,
      "learning_rate": 2.569652253402999e-06,
      "loss": 0.8701,
      "step": 920
    },
    {
      "epoch": 2.82,
      "grad_norm": 0.9711566457168485,
      "learning_rate": 2.1852399266194314e-06,
      "loss": 0.84,
      "step": 925
    },
    {
      "epoch": 2.84,
      "grad_norm": 1.0636030655856668,
      "learning_rate": 1.8316414669469539e-06,
      "loss": 0.8756,
      "step": 930
    },
    {
      "epoch": 2.85,
      "grad_norm": 1.062584489248108,
      "learning_rate": 1.5089682659172321e-06,
      "loss": 0.848,
      "step": 935
    },
    {
      "epoch": 2.87,
      "grad_norm": 1.109498548513231,
      "learning_rate": 1.2173219729043507e-06,
      "loss": 0.8488,
      "step": 940
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.9140768109712519,
      "learning_rate": 9.567944631029169e-07,
      "loss": 0.8614,
      "step": 945
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.9332248724423268,
      "learning_rate": 7.274678085852693e-07,
      "loss": 0.8839,
      "step": 950
    },
    {
      "epoch": 2.91,
      "grad_norm": 0.9421812832473311,
      "learning_rate": 5.294142524469359e-07,
      "loss": 0.8691,
      "step": 955
    },
    {
      "epoch": 2.93,
      "grad_norm": 1.0718790701802063,
      "learning_rate": 3.6269618604847233e-07,
      "loss": 0.8243,
      "step": 960
    },
    {
      "epoch": 2.94,
      "grad_norm": 0.996460942018122,
      "learning_rate": 2.2736612936065106e-07,
      "loss": 0.842,
      "step": 965
    },
    {
      "epoch": 2.96,
      "grad_norm": 1.1337948192404557,
      "learning_rate": 1.2346671441958447e-07,
      "loss": 0.8646,
      "step": 970
    },
    {
      "epoch": 2.97,
      "grad_norm": 0.9562746029081519,
      "learning_rate": 5.103067189662358e-08,
      "loss": 0.8854,
      "step": 975
    },
    {
      "epoch": 2.99,
      "grad_norm": 0.9192847747786613,
      "learning_rate": 1.0080820787450319e-08,
      "loss": 0.8492,
      "step": 980
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.1201515197753906,
      "eval_runtime": 155.5537,
      "eval_samples_per_second": 14.85,
      "eval_steps_per_second": 0.469,
      "step": 984
    },
    {
      "epoch": 3.0,
      "step": 984,
      "total_flos": 2528700107063296.0,
      "train_loss": 1.2712977070633957,
      "train_runtime": 11725.5058,
      "train_samples_per_second": 5.365,
      "train_steps_per_second": 0.084
    }
  ],
  "logging_steps": 5,
  "max_steps": 984,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2528700107063296.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}