{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9990416866315286,
  "eval_steps": 200,
  "global_step": 3336,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011978917105893628,
      "grad_norm": 476.6510925292969,
      "learning_rate": 8e-07,
      "loss": 11.6475,
      "step": 10
    },
    {
      "epoch": 0.023957834211787255,
      "grad_norm": 74.57940673828125,
      "learning_rate": 1.9999928625229307e-06,
      "loss": 2.3869,
      "step": 20
    },
    {
      "epoch": 0.035936751317680884,
      "grad_norm": 125.54178619384766,
      "learning_rate": 1.999912567076008e-06,
      "loss": 7.1899,
      "step": 30
    },
    {
      "epoch": 0.04791566842357451,
      "grad_norm": 14.804201126098633,
      "learning_rate": 1.999743061523497e-06,
      "loss": 5.0722,
      "step": 40
    },
    {
      "epoch": 0.059894585529468136,
      "grad_norm": 9.310958862304688,
      "learning_rate": 1.999484360988329e-06,
      "loss": 2.9189,
      "step": 50
    },
    {
      "epoch": 0.07187350263536177,
      "grad_norm": 306.2783508300781,
      "learning_rate": 1.999136488551224e-06,
      "loss": 2.8403,
      "step": 60
    },
    {
      "epoch": 0.08385241974125539,
      "grad_norm": 134.76495361328125,
      "learning_rate": 1.9986994752486316e-06,
      "loss": 4.2047,
      "step": 70
    },
    {
      "epoch": 0.09583133684714902,
      "grad_norm": 78.88103485107422,
      "learning_rate": 1.998173360069964e-06,
      "loss": 5.1269,
      "step": 80
    },
    {
      "epoch": 0.10781025395304264,
      "grad_norm": 60.88028335571289,
      "learning_rate": 1.997558189954117e-06,
      "loss": 4.8787,
      "step": 90
    },
    {
      "epoch": 0.11978917105893627,
      "grad_norm": 48.85028076171875,
      "learning_rate": 1.9968540197852784e-06,
      "loss": 2.6971,
      "step": 100
    },
    {
      "epoch": 0.1317680881648299,
      "grad_norm": 82.52726745605469,
      "learning_rate": 1.9960609123880376e-06,
      "loss": 6.6349,
      "step": 110
    },
    {
      "epoch": 0.14374700527072354,
      "grad_norm": 18.934968948364258,
      "learning_rate": 1.9951789385217753e-06,
      "loss": 3.6926,
      "step": 120
    },
    {
      "epoch": 0.15572592237661714,
      "grad_norm": 95.94075012207031,
      "learning_rate": 1.9942081768743535e-06,
      "loss": 5.221,
      "step": 130
    },
    {
      "epoch": 0.16770483948251078,
      "grad_norm": 69.58992767333984,
      "learning_rate": 1.9931487140550935e-06,
      "loss": 5.8621,
      "step": 140
    },
    {
      "epoch": 0.1796837565884044,
      "grad_norm": 18.388065338134766,
      "learning_rate": 1.9920006445870497e-06,
      "loss": 5.2103,
      "step": 150
    },
    {
      "epoch": 0.19166267369429804,
      "grad_norm": 117.37078857421875,
      "learning_rate": 1.9907640708985766e-06,
      "loss": 5.8106,
      "step": 160
    },
    {
      "epoch": 0.20364159080019167,
      "grad_norm": 118.88228607177734,
      "learning_rate": 1.9894391033141887e-06,
      "loss": 4.0891,
      "step": 170
    },
    {
      "epoch": 0.21562050790608528,
      "grad_norm": 17.98489761352539,
      "learning_rate": 1.9880258600447204e-06,
      "loss": 5.7061,
      "step": 180
    },
    {
      "epoch": 0.2275994250119789,
      "grad_norm": 170.75030517578125,
      "learning_rate": 1.986524467176777e-06,
      "loss": 4.2787,
      "step": 190
    },
    {
      "epoch": 0.23957834211787254,
      "grad_norm": 74.85968017578125,
      "learning_rate": 1.9849350586614863e-06,
      "loss": 7.8201,
      "step": 200
    },
    {
      "epoch": 0.23957834211787254,
      "eval_loss": 1.2861307859420776,
      "eval_runtime": 238.5238,
      "eval_samples_per_second": 6.247,
      "eval_steps_per_second": 3.123,
      "step": 200
    },
    {
      "epoch": 0.25155725922376615,
      "grad_norm": 254.58087158203125,
      "learning_rate": 1.983257776302548e-06,
      "loss": 5.8449,
      "step": 210
    },
    {
      "epoch": 0.2635361763296598,
      "grad_norm": 12.087563514709473,
      "learning_rate": 1.9814927697435826e-06,
      "loss": 5.7451,
      "step": 220
    },
    {
      "epoch": 0.2755150934355534,
      "grad_norm": 11.869245529174805,
      "learning_rate": 1.9796401964547794e-06,
      "loss": 6.4206,
      "step": 230
    },
    {
      "epoch": 0.2874940105414471,
      "grad_norm": 18.04955291748047,
      "learning_rate": 1.977700221718848e-06,
      "loss": 3.3466,
      "step": 240
    },
    {
      "epoch": 0.2994729276473407,
      "grad_norm": 40.88437271118164,
      "learning_rate": 1.975673018616273e-06,
      "loss": 4.5986,
      "step": 250
    },
    {
      "epoch": 0.3114518447532343,
      "grad_norm": 17.213830947875977,
      "learning_rate": 1.97355876800987e-06,
      "loss": 3.6809,
      "step": 260
    },
    {
      "epoch": 0.32343076185912795,
      "grad_norm": 31.727243423461914,
      "learning_rate": 1.9713576585286513e-06,
      "loss": 4.692,
      "step": 270
    },
    {
      "epoch": 0.33540967896502155,
      "grad_norm": 95.36638641357422,
      "learning_rate": 1.9690698865509964e-06,
      "loss": 6.1814,
      "step": 280
    },
    {
      "epoch": 0.3473885960709152,
      "grad_norm": 25.670534133911133,
      "learning_rate": 1.966695656187131e-06,
      "loss": 2.8556,
      "step": 290
    },
    {
      "epoch": 0.3593675131768088,
      "grad_norm": 137.57431030273438,
      "learning_rate": 1.9642351792609162e-06,
      "loss": 3.3607,
      "step": 300
    },
    {
      "epoch": 0.3713464302827024,
      "grad_norm": 46.084354400634766,
      "learning_rate": 1.9616886752909523e-06,
      "loss": 6.1352,
      "step": 310
    },
    {
      "epoch": 0.3833253473885961,
      "grad_norm": 110.49552154541016,
      "learning_rate": 1.9590563714709916e-06,
      "loss": 5.8323,
      "step": 320
    },
    {
      "epoch": 0.3953042644944897,
      "grad_norm": 77.74449157714844,
      "learning_rate": 1.9563385026496687e-06,
      "loss": 5.7407,
      "step": 330
    },
    {
      "epoch": 0.40728318160038335,
      "grad_norm": 26.705305099487305,
      "learning_rate": 1.9535353113095493e-06,
      "loss": 5.3508,
      "step": 340
    },
    {
      "epoch": 0.41926209870627695,
      "grad_norm": 146.0390167236328,
      "learning_rate": 1.9506470475454957e-06,
      "loss": 2.9407,
      "step": 350
    },
    {
      "epoch": 0.43124101581217056,
      "grad_norm": 25.178991317749023,
      "learning_rate": 1.947673969042353e-06,
      "loss": 3.0089,
      "step": 360
    },
    {
      "epoch": 0.4432199329180642,
      "grad_norm": 8.458102226257324,
      "learning_rate": 1.9446163410519603e-06,
      "loss": 2.885,
      "step": 370
    },
    {
      "epoch": 0.4551988500239578,
      "grad_norm": 94.5208740234375,
      "learning_rate": 1.9414744363694842e-06,
      "loss": 3.8878,
      "step": 380
    },
    {
      "epoch": 0.4671777671298515,
      "grad_norm": 138.86561584472656,
      "learning_rate": 1.938248535309083e-06,
      "loss": 5.5948,
      "step": 390
    },
    {
      "epoch": 0.4791566842357451,
      "grad_norm": 10.55215072631836,
      "learning_rate": 1.9349389256788943e-06,
      "loss": 2.9242,
      "step": 400
    },
    {
      "epoch": 0.4791566842357451,
      "eval_loss": 1.1180063486099243,
      "eval_runtime": 237.218,
      "eval_samples_per_second": 6.281,
      "eval_steps_per_second": 3.141,
      "step": 400
    },
    {
      "epoch": 0.4911356013416387,
      "grad_norm": 11.329914093017578,
      "learning_rate": 1.931545902755359e-06,
      "loss": 5.7209,
      "step": 410
    },
    {
      "epoch": 0.5031145184475323,
      "grad_norm": 121.2057113647461,
      "learning_rate": 1.928069769256879e-06,
      "loss": 4.2294,
      "step": 420
    },
    {
      "epoch": 0.515093435553426,
      "grad_norm": 70.16046905517578,
      "learning_rate": 1.9245108353168055e-06,
      "loss": 5.1172,
      "step": 430
    },
    {
      "epoch": 0.5270723526593196,
      "grad_norm": 55.029964447021484,
      "learning_rate": 1.9208694184557735e-06,
      "loss": 3.8455,
      "step": 440
    },
    {
      "epoch": 0.5390512697652132,
      "grad_norm": 16.75533103942871,
      "learning_rate": 1.9171458435533706e-06,
      "loss": 2.1762,
      "step": 450
    },
    {
      "epoch": 0.5510301868711068,
      "grad_norm": 37.192169189453125,
      "learning_rate": 1.913340442819153e-06,
      "loss": 4.6994,
      "step": 460
    },
    {
      "epoch": 0.5630091039770004,
      "grad_norm": 190.2852020263672,
      "learning_rate": 1.9094535557630067e-06,
      "loss": 8.188,
      "step": 470
    },
    {
      "epoch": 0.5749880210828942,
      "grad_norm": 14.840240478515625,
      "learning_rate": 1.905485529164856e-06,
      "loss": 2.4346,
      "step": 480
    },
    {
      "epoch": 0.5869669381887878,
      "grad_norm": 17.85882568359375,
      "learning_rate": 1.9014367170437255e-06,
      "loss": 5.1088,
      "step": 490
    },
    {
      "epoch": 0.5989458552946814,
      "grad_norm": 98.2167739868164,
      "learning_rate": 1.8973074806261558e-06,
      "loss": 4.4192,
      "step": 500
    },
    {
      "epoch": 0.610924772400575,
      "grad_norm": 72.538330078125,
      "learning_rate": 1.8930981883139734e-06,
      "loss": 4.2753,
      "step": 510
    },
    {
      "epoch": 0.6229036895064686,
      "grad_norm": 121.5967788696289,
      "learning_rate": 1.8888092156514252e-06,
      "loss": 5.0462,
      "step": 520
    },
    {
      "epoch": 0.6348826066123623,
      "grad_norm": 161.8177947998047,
      "learning_rate": 1.8844409452916719e-06,
      "loss": 3.2489,
      "step": 530
    },
    {
      "epoch": 0.6468615237182559,
      "grad_norm": 149.68197631835938,
      "learning_rate": 1.8799937669626481e-06,
      "loss": 4.8399,
      "step": 540
    },
    {
      "epoch": 0.6588404408241495,
      "grad_norm": 87.29440307617188,
      "learning_rate": 1.8754680774322934e-06,
      "loss": 5.3579,
      "step": 550
    },
    {
      "epoch": 0.6708193579300431,
      "grad_norm": 70.05744171142578,
      "learning_rate": 1.8708642804731513e-06,
      "loss": 1.967,
      "step": 560
    },
    {
      "epoch": 0.6827982750359367,
      "grad_norm": 49.84896469116211,
      "learning_rate": 1.866182786826347e-06,
      "loss": 4.1978,
      "step": 570
    },
    {
      "epoch": 0.6947771921418304,
      "grad_norm": 29.61354637145996,
      "learning_rate": 1.861424014164941e-06,
      "loss": 4.025,
      "step": 580
    },
    {
      "epoch": 0.706756109247724,
      "grad_norm": 99.13072204589844,
      "learning_rate": 1.8565883870566666e-06,
      "loss": 4.1162,
      "step": 590
    },
    {
      "epoch": 0.7187350263536176,
      "grad_norm": 130.23606872558594,
      "learning_rate": 1.8516763369260492e-06,
      "loss": 3.0065,
      "step": 600
    },
    {
      "epoch": 0.7187350263536176,
      "eval_loss": 1.104053258895874,
      "eval_runtime": 238.4518,
      "eval_samples_per_second": 6.249,
      "eval_steps_per_second": 3.124,
      "step": 600
    },
    {
      "epoch": 0.7307139434595112,
      "grad_norm": 50.47702407836914,
      "learning_rate": 1.8466883020159161e-06,
      "loss": 4.3503,
      "step": 610
    },
    {
      "epoch": 0.7426928605654048,
      "grad_norm": 17.21928596496582,
      "learning_rate": 1.8416247273482988e-06,
      "loss": 4.4346,
      "step": 620
    },
    {
      "epoch": 0.7546717776712986,
      "grad_norm": 49.47705841064453,
      "learning_rate": 1.8364860646847262e-06,
      "loss": 3.9906,
      "step": 630
    },
    {
      "epoch": 0.7666506947771922,
      "grad_norm": 14.143331527709961,
      "learning_rate": 1.831272772485922e-06,
      "loss": 3.3026,
      "step": 640
    },
    {
      "epoch": 0.7786296118830858,
      "grad_norm": 17.33100128173828,
      "learning_rate": 1.8259853158708997e-06,
      "loss": 6.0244,
      "step": 650
    },
    {
      "epoch": 0.7906085289889794,
      "grad_norm": 10.411093711853027,
      "learning_rate": 1.8206241665754687e-06,
      "loss": 2.8721,
      "step": 660
    },
    {
      "epoch": 0.802587446094873,
      "grad_norm": 181.8240966796875,
      "learning_rate": 1.815189802910143e-06,
      "loss": 5.1721,
      "step": 670
    },
    {
      "epoch": 0.8145663632007667,
      "grad_norm": 39.83287048339844,
      "learning_rate": 1.80968270971747e-06,
      "loss": 4.9115,
      "step": 680
    },
    {
      "epoch": 0.8265452803066603,
      "grad_norm": 39.86928176879883,
      "learning_rate": 1.8041033783287737e-06,
      "loss": 3.8957,
      "step": 690
    },
    {
      "epoch": 0.8385241974125539,
      "grad_norm": 57.7371711730957,
      "learning_rate": 1.7984523065203188e-06,
      "loss": 3.1863,
      "step": 700
    },
    {
      "epoch": 0.8505031145184475,
      "grad_norm": 14.148628234863281,
      "learning_rate": 1.792729998468899e-06,
      "loss": 4.26,
      "step": 710
    },
    {
      "epoch": 0.8624820316243411,
      "grad_norm": 71.03279113769531,
      "learning_rate": 1.7869369647068577e-06,
      "loss": 4.9559,
      "step": 720
    },
    {
      "epoch": 0.8744609487302348,
      "grad_norm": 17.670730590820312,
      "learning_rate": 1.7810737220765372e-06,
      "loss": 3.9867,
      "step": 730
    },
    {
      "epoch": 0.8864398658361284,
      "grad_norm": 14.698404312133789,
      "learning_rate": 1.7751407936841684e-06,
      "loss": 2.7134,
      "step": 740
    },
    {
      "epoch": 0.898418782942022,
      "grad_norm": 66.42393493652344,
      "learning_rate": 1.7691387088532001e-06,
      "loss": 3.2121,
      "step": 750
    },
    {
      "epoch": 0.9103977000479156,
      "grad_norm": 76.34748077392578,
      "learning_rate": 1.7630680030770732e-06,
      "loss": 4.7613,
      "step": 760
    },
    {
      "epoch": 0.9223766171538093,
      "grad_norm": 46.27962112426758,
      "learning_rate": 1.7569292179714465e-06,
      "loss": 3.2976,
      "step": 770
    },
    {
      "epoch": 0.934355534259703,
      "grad_norm": 14.20971965789795,
      "learning_rate": 1.750722901225873e-06,
      "loss": 1.9176,
      "step": 780
    },
    {
      "epoch": 0.9463344513655966,
      "grad_norm": 56.962379455566406,
      "learning_rate": 1.7444496065549384e-06,
      "loss": 1.9859,
      "step": 790
    },
    {
      "epoch": 0.9583133684714902,
      "grad_norm": 32.20167541503906,
      "learning_rate": 1.7381098936488574e-06,
      "loss": 6.9549,
      "step": 800
    },
    {
      "epoch": 0.9583133684714902,
      "eval_loss": 1.0978227853775024,
      "eval_runtime": 238.5511,
      "eval_samples_per_second": 6.246,
      "eval_steps_per_second": 3.123,
      "step": 800
    },
    {
      "epoch": 0.9702922855773838,
      "grad_norm": 82.94165802001953,
      "learning_rate": 1.7317043281235418e-06,
      "loss": 4.1317,
      "step": 810
    },
    {
      "epoch": 0.9822712026832774,
      "grad_norm": 110.44422912597656,
      "learning_rate": 1.725233481470135e-06,
      "loss": 3.2924,
      "step": 820
    },
    {
      "epoch": 0.9942501197891711,
      "grad_norm": 88.77394104003906,
      "learning_rate": 1.7186979310040268e-06,
      "loss": 5.5422,
      "step": 830
    },
    {
      "epoch": 1.0071873502635362,
      "grad_norm": 167.41412353515625,
      "learning_rate": 1.7120982598133456e-06,
      "loss": 3.5133,
      "step": 840
    },
    {
      "epoch": 1.0191662673694297,
      "grad_norm": 16.926607131958008,
      "learning_rate": 1.7054350567069364e-06,
      "loss": 4.2376,
      "step": 850
    },
    {
      "epoch": 1.0311451844753234,
      "grad_norm": 96.12760925292969,
      "learning_rate": 1.698708916161829e-06,
      "loss": 3.6823,
      "step": 860
    },
    {
      "epoch": 1.0431241015812172,
      "grad_norm": 103.28370666503906,
      "learning_rate": 1.6919204382701987e-06,
      "loss": 2.5705,
      "step": 870
    },
    {
      "epoch": 1.0551030186871106,
      "grad_norm": 166.0828857421875,
      "learning_rate": 1.6850702286858298e-06,
      "loss": 2.9061,
      "step": 880
    },
    {
      "epoch": 1.0670819357930044,
      "grad_norm": 136.93392944335938,
      "learning_rate": 1.678158898570078e-06,
      "loss": 2.8635,
      "step": 890
    },
    {
      "epoch": 1.0790608528988979,
      "grad_norm": 14.542271614074707,
      "learning_rate": 1.6711870645373449e-06,
      "loss": 4.2555,
      "step": 900
    },
    {
      "epoch": 1.0910397700047916,
      "grad_norm": 53.567359924316406,
      "learning_rate": 1.6641553486000651e-06,
      "loss": 3.1885,
      "step": 910
    },
    {
      "epoch": 1.1030186871106853,
      "grad_norm": 100.29656982421875,
      "learning_rate": 1.6570643781132118e-06,
      "loss": 4.953,
      "step": 920
    },
    {
      "epoch": 1.1149976042165788,
      "grad_norm": 89.06425476074219,
      "learning_rate": 1.649914785718324e-06,
      "loss": 4.9896,
      "step": 930
    },
    {
      "epoch": 1.1269765213224725,
      "grad_norm": 17.858898162841797,
      "learning_rate": 1.6427072092870651e-06,
      "loss": 1.5295,
      "step": 940
    },
    {
      "epoch": 1.138955438428366,
      "grad_norm": 15.720714569091797,
      "learning_rate": 1.6354422918643133e-06,
      "loss": 3.0117,
      "step": 950
    },
    {
      "epoch": 1.1509343555342597,
      "grad_norm": 13.404827117919922,
      "learning_rate": 1.628120681610789e-06,
      "loss": 2.1361,
      "step": 960
    },
    {
      "epoch": 1.1629132726401532,
      "grad_norm": 130.946044921875,
      "learning_rate": 1.6207430317452297e-06,
      "loss": 3.941,
      "step": 970
    },
    {
      "epoch": 1.174892189746047,
      "grad_norm": 14.730375289916992,
      "learning_rate": 1.613310000486108e-06,
      "loss": 3.2318,
      "step": 980
    },
    {
      "epoch": 1.1868711068519406,
      "grad_norm": 82.65552520751953,
      "learning_rate": 1.6058222509929096e-06,
      "loss": 3.9045,
      "step": 990
    },
    {
      "epoch": 1.1988500239578341,
      "grad_norm": Infinity,
      "learning_rate": 1.5982804513069664e-06,
      "loss": 5.5404,
      "step": 1000
    },
    {
      "epoch": 1.1988500239578341,
      "eval_loss": 1.1037527322769165,
      "eval_runtime": 238.5531,
      "eval_samples_per_second": 6.246,
      "eval_steps_per_second": 3.123,
      "step": 1000
    },
    {
      "epoch": 1.2108289410637278,
      "grad_norm": 67.75409698486328,
      "learning_rate": 1.5914471746978935e-06,
      "loss": 2.6392,
      "step": 1010
    },
    {
      "epoch": 1.2228078581696216,
      "grad_norm": 49.21822738647461,
      "learning_rate": 1.5838045373221053e-06,
      "loss": 4.0259,
      "step": 1020
    },
    {
      "epoch": 1.234786775275515,
      "grad_norm": 228.187744140625,
      "learning_rate": 1.5761098141278849e-06,
      "loss": 5.8343,
      "step": 1030
    },
    {
      "epoch": 1.2467656923814088,
      "grad_norm": 33.68708419799805,
      "learning_rate": 1.5683636916223236e-06,
      "loss": 3.9807,
      "step": 1040
    },
    {
      "epoch": 1.2587446094873023,
      "grad_norm": 140.39187622070312,
      "learning_rate": 1.5605668608982526e-06,
      "loss": 3.9716,
      "step": 1050
    },
    {
      "epoch": 1.270723526593196,
      "grad_norm": 11.902155876159668,
      "learning_rate": 1.5527200175725842e-06,
      "loss": 3.2315,
      "step": 1060
    },
    {
      "epoch": 1.2827024436990895,
      "grad_norm": 144.3499298095703,
      "learning_rate": 1.5448238617242488e-06,
      "loss": 2.6336,
      "step": 1070
    },
    {
      "epoch": 1.2946813608049832,
      "grad_norm": 59.747928619384766,
      "learning_rate": 1.5368790978317395e-06,
      "loss": 3.206,
      "step": 1080
    },
    {
      "epoch": 1.306660277910877,
      "grad_norm": 53.534950256347656,
      "learning_rate": 1.5288864347102545e-06,
      "loss": 4.3036,
      "step": 1090
    },
    {
      "epoch": 1.3186391950167704,
      "grad_norm": 48.62434387207031,
      "learning_rate": 1.520846585448463e-06,
      "loss": 2.4486,
      "step": 1100
    },
    {
      "epoch": 1.3306181121226641,
      "grad_norm": 18.836971282958984,
      "learning_rate": 1.512760267344882e-06,
      "loss": 4.0121,
      "step": 1110
    },
    {
      "epoch": 1.3425970292285578,
      "grad_norm": 224.24917602539062,
      "learning_rate": 1.5046282018438814e-06,
      "loss": 2.8545,
      "step": 1120
    },
    {
      "epoch": 1.3545759463344513,
      "grad_norm": 18.120187759399414,
      "learning_rate": 1.4964511144713174e-06,
      "loss": 3.1619,
      "step": 1130
    },
    {
      "epoch": 1.366554863440345,
      "grad_norm": 39.76359939575195,
      "learning_rate": 1.4882297347698048e-06,
      "loss": 3.0413,
      "step": 1140
    },
    {
      "epoch": 1.3785337805462385,
      "grad_norm": 64.61255645751953,
      "learning_rate": 1.4799647962336255e-06,
      "loss": 3.8001,
      "step": 1150
    },
    {
      "epoch": 1.3905126976521323,
      "grad_norm": 12.653026580810547,
      "learning_rate": 1.471657036243291e-06,
      "loss": 5.532,
      "step": 1160
    },
    {
      "epoch": 1.4024916147580258,
      "grad_norm": 53.37699508666992,
      "learning_rate": 1.4633071959997525e-06,
      "loss": 3.4156,
      "step": 1170
    },
    {
      "epoch": 1.4144705318639195,
      "grad_norm": 37.08938217163086,
      "learning_rate": 1.4549160204582731e-06,
      "loss": 2.5073,
      "step": 1180
    },
    {
      "epoch": 1.4264494489698132,
      "grad_norm": 139.479736328125,
      "learning_rate": 1.4464842582619652e-06,
      "loss": 3.36,
      "step": 1190
    },
    {
      "epoch": 1.4384283660757067,
      "grad_norm": 88.59599304199219,
      "learning_rate": 1.4380126616749975e-06,
      "loss": 5.2213,
      "step": 1200
    },
    {
      "epoch": 1.4384283660757067,
      "eval_loss": 1.1036500930786133,
      "eval_runtime": 238.313,
      "eval_samples_per_second": 6.252,
      "eval_steps_per_second": 3.126,
      "step": 1200
    },
    {
      "epoch": 1.4504072831816004,
      "grad_norm": 129.61988830566406,
      "learning_rate": 1.4295019865154785e-06,
      "loss": 7.1682,
      "step": 1210
    },
    {
      "epoch": 1.462386200287494,
      "grad_norm": 19.823545455932617,
      "learning_rate": 1.4209529920880272e-06,
      "loss": 4.6843,
      "step": 1220
    },
    {
      "epoch": 1.4743651173933876,
      "grad_norm": 107.53630065917969,
      "learning_rate": 1.4123664411160252e-06,
      "loss": 2.4525,
      "step": 1230
    },
    {
      "epoch": 1.4863440344992813,
      "grad_norm": 54.906280517578125,
      "learning_rate": 1.4037430996735722e-06,
      "loss": 5.9388,
      "step": 1240
    },
    {
      "epoch": 1.4983229516051748,
      "grad_norm": 77.29488372802734,
      "learning_rate": 1.3950837371171355e-06,
      "loss": 5.3705,
      "step": 1250
    },
    {
      "epoch": 1.5103018687110685,
      "grad_norm": 115.9428482055664,
      "learning_rate": 1.3863891260169114e-06,
      "loss": 4.0317,
      "step": 1260
    },
    {
      "epoch": 1.522280785816962,
      "grad_norm": 19.207189559936523,
      "learning_rate": 1.3776600420878973e-06,
      "loss": 3.8767,
      "step": 1270
    },
    {
      "epoch": 1.5342597029228557,
      "grad_norm": 83.13814544677734,
      "learning_rate": 1.3688972641206837e-06,
      "loss": 4.5835,
      "step": 1280
    },
    {
      "epoch": 1.5462386200287495,
      "grad_norm": 222.7005157470703,
      "learning_rate": 1.3601015739119733e-06,
      "loss": 3.3379,
      "step": 1290
    },
    {
      "epoch": 1.558217537134643,
      "grad_norm": 51.51054382324219,
      "learning_rate": 1.35127375619483e-06,
      "loss": 5.4397,
      "step": 1300
    },
    {
      "epoch": 1.5701964542405367,
      "grad_norm": 109.09092712402344,
      "learning_rate": 1.3424145985686662e-06,
      "loss": 3.1896,
      "step": 1310
    },
    {
      "epoch": 1.5821753713464304,
      "grad_norm": 11.890337944030762,
      "learning_rate": 1.333524891428976e-06,
      "loss": 4.4828,
      "step": 1320
    },
    {
      "epoch": 1.5941542884523239,
      "grad_norm": 18.87173843383789,
      "learning_rate": 1.324605427896817e-06,
      "loss": 2.4719,
      "step": 1330
    },
    {
      "epoch": 1.6061332055582176,
      "grad_norm": 110.26778411865234,
      "learning_rate": 1.3156570037480497e-06,
      "loss": 3.4721,
      "step": 1340
    },
    {
      "epoch": 1.6181121226641113,
      "grad_norm": 19.194913864135742,
      "learning_rate": 1.3066804173423397e-06,
      "loss": 4.3532,
      "step": 1350
    },
    {
      "epoch": 1.6300910397700048,
      "grad_norm": 89.65567016601562,
      "learning_rate": 1.297676469551931e-06,
      "loss": 4.1742,
      "step": 1360
    },
    {
      "epoch": 1.6420699568758983,
      "grad_norm": 15.179718017578125,
      "learning_rate": 1.2886459636901927e-06,
      "loss": 4.2612,
      "step": 1370
    },
    {
      "epoch": 1.654048873981792,
      "grad_norm": 14.557687759399414,
      "learning_rate": 1.2795897054399498e-06,
      "loss": 5.2594,
      "step": 1380
    },
    {
      "epoch": 1.6660277910876857,
      "grad_norm": 20.148624420166016,
      "learning_rate": 1.2705085027816008e-06,
      "loss": 3.3919,
      "step": 1390
    },
    {
      "epoch": 1.6780067081935792,
      "grad_norm": 9.0452241897583,
      "learning_rate": 1.261403165921032e-06,
      "loss": 4.3208,
      "step": 1400
    },
    {
      "epoch": 1.6780067081935792,
      "eval_loss": 1.0974289178848267,
      "eval_runtime": 238.4129,
      "eval_samples_per_second": 6.25,
      "eval_steps_per_second": 3.125,
      "step": 1400
    },
    {
      "epoch": 1.689985625299473,
      "grad_norm": 18.7681941986084,
      "learning_rate": 1.2522745072173336e-06,
      "loss": 2.5784,
      "step": 1410
    },
    {
      "epoch": 1.7019645424053667,
      "grad_norm": 112.4189224243164,
      "learning_rate": 1.243123341110321e-06,
      "loss": 4.0173,
      "step": 1420
    },
    {
      "epoch": 1.7139434595112601,
      "grad_norm": 13.051095008850098,
      "learning_rate": 1.2339504840478738e-06,
      "loss": 3.1098,
      "step": 1430
    },
    {
      "epoch": 1.7259223766171539,
      "grad_norm": 16.918392181396484,
      "learning_rate": 1.224756754413092e-06,
      "loss": 3.1983,
      "step": 1440
    },
    {
      "epoch": 1.7379012937230476,
      "grad_norm": 167.8545379638672,
      "learning_rate": 1.2155429724512838e-06,
      "loss": 4.8368,
      "step": 1450
    },
    {
      "epoch": 1.749880210828941,
      "grad_norm": 19.237834930419922,
      "learning_rate": 1.206309960196784e-06,
      "loss": 3.1809,
      "step": 1460
    },
    {
      "epoch": 1.7618591279348346,
      "grad_norm": 95.0063247680664,
      "learning_rate": 1.1970585413996132e-06,
      "loss": 3.9006,
      "step": 1470
    },
    {
      "epoch": 1.7738380450407283,
      "grad_norm": 55.70530319213867,
      "learning_rate": 1.1877895414519858e-06,
      "loss": 3.3394,
      "step": 1480
    },
    {
      "epoch": 1.785816962146622,
      "grad_norm": 23.173871994018555,
      "learning_rate": 1.1785037873146697e-06,
      "loss": 2.4079,
      "step": 1490
    },
    {
      "epoch": 1.7977958792525155,
      "grad_norm": 120.7456283569336,
      "learning_rate": 1.1692021074432054e-06,
      "loss": 4.2111,
      "step": 1500
    },
    {
      "epoch": 1.8097747963584092,
      "grad_norm": 89.36892700195312,
      "learning_rate": 1.1598853317139958e-06,
      "loss": 1.8205,
      "step": 1510
    },
    {
      "epoch": 1.821753713464303,
      "grad_norm": 66.96819305419922,
      "learning_rate": 1.150554291350263e-06,
      "loss": 4.6707,
      "step": 1520
    },
    {
      "epoch": 1.8337326305701964,
      "grad_norm": 52.6048583984375,
      "learning_rate": 1.1412098188478914e-06,
      "loss": 2.2611,
      "step": 1530
    },
    {
      "epoch": 1.8457115476760901,
      "grad_norm": 100.39757537841797,
      "learning_rate": 1.1318527479011513e-06,
      "loss": 3.3554,
      "step": 1540
    },
    {
      "epoch": 1.8576904647819839,
      "grad_norm": 12.364606857299805,
      "learning_rate": 1.1224839133283208e-06,
      "loss": 2.7868,
      "step": 1550
    },
    {
      "epoch": 1.8696693818878773,
      "grad_norm": 93.31403350830078,
      "learning_rate": 1.1131041509972032e-06,
      "loss": 3.7607,
      "step": 1560
    },
    {
      "epoch": 1.8816482989937708,
      "grad_norm": 77.30158233642578,
      "learning_rate": 1.1037142977505548e-06,
      "loss": 3.28,
      "step": 1570
    },
    {
      "epoch": 1.8936272160996646,
      "grad_norm": 13.82420825958252,
      "learning_rate": 1.0943151913314211e-06,
      "loss": 3.3544,
      "step": 1580
    },
    {
      "epoch": 1.9056061332055583,
      "grad_norm": 16.398128509521484,
      "learning_rate": 1.084907670308397e-06,
      "loss": 2.7871,
      "step": 1590
    },
    {
      "epoch": 1.9175850503114518,
      "grad_norm": 19.750925064086914,
      "learning_rate": 1.0754925740008098e-06,
      "loss": 4.1985,
      "step": 1600
    },
    {
      "epoch": 1.9175850503114518,
      "eval_loss": 1.1042989492416382,
      "eval_runtime": 238.4482,
      "eval_samples_per_second": 6.249,
      "eval_steps_per_second": 3.124,
      "step": 1600
    },
    {
      "epoch": 1.9295639674173455,
      "grad_norm": 16.436159133911133,
      "learning_rate": 1.066070742403839e-06,
      "loss": 3.9566,
      "step": 1610
    },
    {
      "epoch": 1.9415428845232392,
      "grad_norm": 67.2239761352539,
      "learning_rate": 1.056643016113572e-06,
      "loss": 4.0604,
      "step": 1620
    },
    {
      "epoch": 1.9535218016291327,
      "grad_norm": 52.419456481933594,
      "learning_rate": 1.047210236252008e-06,
      "loss": 4.4566,
      "step": 1630
    },
    {
      "epoch": 1.9655007187350264,
      "grad_norm": 102.50648498535156,
      "learning_rate": 1.0377732443920155e-06,
      "loss": 2.5929,
      "step": 1640
    },
    {
      "epoch": 1.9774796358409201,
      "grad_norm": 85.2761459350586,
      "learning_rate": 1.0283328824822498e-06,
      "loss": 3.278,
      "step": 1650
    },
    {
      "epoch": 1.9894585529468136,
      "grad_norm": 111.09942626953125,
      "learning_rate": 1.0188899927720324e-06,
      "loss": 2.1727,
      "step": 1660
    },
    {
      "epoch": 2.0023957834211785,
      "grad_norm": 88.2173080444336,
      "learning_rate": 1.009445417736213e-06,
      "loss": 4.7098,
      "step": 1670
    },
    {
      "epoch": 2.0143747005270725,
      "grad_norm": 15.37248420715332,
      "learning_rate": 1e-06,
      "loss": 3.3523,
      "step": 1680
    },
    {
      "epoch": 2.026353617632966,
      "grad_norm": 428.4352111816406,
      "learning_rate": 9.905545822637871e-07,
      "loss": 4.4776,
      "step": 1690
    },
    {
      "epoch": 2.0383325347388594,
      "grad_norm": 16.143062591552734,
      "learning_rate": 9.811100072279673e-07,
      "loss": 3.2249,
      "step": 1700
    },
    {
      "epoch": 2.0503114518447534,
      "grad_norm": 142.73899841308594,
      "learning_rate": 9.716671175177506e-07,
      "loss": 3.6488,
      "step": 1710
    },
    {
      "epoch": 2.062290368950647,
      "grad_norm": 86.692626953125,
      "learning_rate": 9.622267556079844e-07,
      "loss": 2.3491,
      "step": 1720
    },
    {
      "epoch": 2.0742692860565404,
      "grad_norm": 127.98802947998047,
      "learning_rate": 9.527897637479921e-07,
      "loss": 4.828,
      "step": 1730
    },
    {
      "epoch": 2.0862482031624343,
      "grad_norm": 14.027091026306152,
      "learning_rate": 9.433569838864282e-07,
      "loss": 4.176,
      "step": 1740
    },
    {
      "epoch": 2.098227120268328,
      "grad_norm": 16.68279266357422,
      "learning_rate": 9.33929257596161e-07,
      "loss": 2.8037,
      "step": 1750
    },
    {
      "epoch": 2.1102060373742213,
      "grad_norm": 68.17903900146484,
      "learning_rate": 9.245074259991904e-07,
      "loss": 3.448,
      "step": 1760
    },
    {
      "epoch": 2.122184954480115,
      "grad_norm": 90.32327270507812,
      "learning_rate": 9.150923296916032e-07,
      "loss": 3.3154,
      "step": 1770
    },
    {
      "epoch": 2.1341638715860087,
      "grad_norm": 218.0877227783203,
      "learning_rate": 9.056848086685789e-07,
      "loss": 3.8818,
      "step": 1780
    },
    {
      "epoch": 2.1461427886919022,
      "grad_norm": 117.9521484375,
      "learning_rate": 8.96285702249445e-07,
      "loss": 3.7345,
      "step": 1790
    },
    {
      "epoch": 2.1581217057977957,
      "grad_norm": 14.993733406066895,
      "learning_rate": 8.868958490027966e-07,
      "loss": 3.3709,
      "step": 1800
    },
    {
      "epoch": 2.1581217057977957,
      "eval_loss": 1.1172066926956177,
      "eval_runtime": 238.6675,
      "eval_samples_per_second": 6.243,
      "eval_steps_per_second": 3.121,
      "step": 1800
    },
    {
      "epoch": 2.1701006229036897,
      "grad_norm": 273.74365234375,
      "learning_rate": 8.775160866716791e-07,
      "loss": 4.06,
      "step": 1810
    },
    {
      "epoch": 2.182079540009583,
      "grad_norm": 11.733575820922852,
      "learning_rate": 8.681472520988488e-07,
      "loss": 2.4028,
      "step": 1820
    },
    {
      "epoch": 2.1940584571154766,
      "grad_norm": 270.543212890625,
      "learning_rate": 8.587901811521087e-07,
      "loss": 5.1853,
      "step": 1830
    },
    {
      "epoch": 2.2060373742213706,
      "grad_norm": 18.38782501220703,
      "learning_rate": 8.494457086497368e-07,
      "loss": 1.9458,
      "step": 1840
    },
    {
      "epoch": 2.218016291327264,
      "grad_norm": 127.68826293945312,
      "learning_rate": 8.401146682860041e-07,
      "loss": 3.2338,
      "step": 1850
    },
    {
      "epoch": 2.2299952084331576,
      "grad_norm": 186.4503173828125,
      "learning_rate": 8.307978925567945e-07,
      "loss": 2.5094,
      "step": 1860
    },
    {
      "epoch": 2.241974125539051,
      "grad_norm": 15.038443565368652,
      "learning_rate": 8.214962126853307e-07,
      "loss": 2.6388,
      "step": 1870
    },
    {
      "epoch": 2.253953042644945,
      "grad_norm": 74.26438903808594,
      "learning_rate": 8.122104585480143e-07,
      "loss": 2.24,
      "step": 1880
    },
    {
      "epoch": 2.2659319597508385,
      "grad_norm": 57.38971710205078,
      "learning_rate": 8.029414586003866e-07,
      "loss": 4.3915,
      "step": 1890
    },
    {
      "epoch": 2.277910876856732,
      "grad_norm": 221.86866760253906,
      "learning_rate": 7.93690039803216e-07,
      "loss": 3.0979,
      "step": 1900
    },
    {
      "epoch": 2.289889793962626,
      "grad_norm": 66.43515014648438,
      "learning_rate": 7.844570275487159e-07,
      "loss": 2.0459,
      "step": 1910
    },
    {
      "epoch": 2.3018687110685194,
      "grad_norm": 174.6913604736328,
      "learning_rate": 7.752432455869081e-07,
      "loss": 4.3113,
      "step": 1920
    },
    {
      "epoch": 2.313847628174413,
      "grad_norm": 12.269379615783691,
      "learning_rate": 7.660495159521264e-07,
      "loss": 2.6802,
      "step": 1930
    },
    {
      "epoch": 2.3258265452803064,
      "grad_norm": 14.982316970825195,
      "learning_rate": 7.56876658889679e-07,
      "loss": 3.2325,
      "step": 1940
    },
    {
      "epoch": 2.3378054623862004,
      "grad_norm": 75.86939239501953,
      "learning_rate": 7.477254927826664e-07,
      "loss": 1.0064,
      "step": 1950
    },
    {
      "epoch": 2.349784379492094,
      "grad_norm": 12.59216594696045,
      "learning_rate": 7.38596834078968e-07,
      "loss": 2.3464,
      "step": 1960
    },
    {
      "epoch": 2.361763296597988,
      "grad_norm": 44.80192184448242,
      "learning_rate": 7.294914972183992e-07,
      "loss": 3.9336,
      "step": 1970
    },
    {
      "epoch": 2.3737422137038813,
      "grad_norm": 227.7502899169922,
      "learning_rate": 7.204102945600502e-07,
      "loss": 3.3652,
      "step": 1980
    },
    {
      "epoch": 2.3857211308097748,
      "grad_norm": 202.1083526611328,
      "learning_rate": 7.113540363098072e-07,
      "loss": 3.0293,
      "step": 1990
    },
    {
      "epoch": 2.3977000479156683,
      "grad_norm": 148.12112426757812,
      "learning_rate": 7.02323530448069e-07,
      "loss": 3.7548,
      "step": 2000
    },
    {
      "epoch": 2.3977000479156683,
      "eval_loss": 1.1212018728256226,
      "eval_runtime": 237.7587,
      "eval_samples_per_second": 6.267,
      "eval_steps_per_second": 3.133,
      "step": 2000
    },
    {
      "epoch": 2.409678965021562,
      "grad_norm": 107.33174133300781,
      "learning_rate": 6.933195826576603e-07,
      "loss": 3.9499,
      "step": 2010
    },
    {
      "epoch": 2.4216578821274557,
      "grad_norm": 15.232194900512695,
      "learning_rate": 6.843429962519504e-07,
      "loss": 4.3203,
      "step": 2020
    },
    {
      "epoch": 2.433636799233349,
      "grad_norm": 14.64511489868164,
      "learning_rate": 6.75394572103183e-07,
      "loss": 4.5243,
      "step": 2030
    },
    {
      "epoch": 2.445615716339243,
      "grad_norm": 16.1613712310791,
      "learning_rate": 6.664751085710239e-07,
      "loss": 3.5644,
      "step": 2040
    },
    {
      "epoch": 2.4575946334451366,
      "grad_norm": 61.8858642578125,
      "learning_rate": 6.575854014313338e-07,
      "loss": 3.7972,
      "step": 2050
    },
    {
      "epoch": 2.46957355055103,
      "grad_norm": 52.300514221191406,
      "learning_rate": 6.487262438051701e-07,
      "loss": 3.6956,
      "step": 2060
    },
    {
      "epoch": 2.4815524676569236,
      "grad_norm": 208.3380126953125,
      "learning_rate": 6.398984260880266e-07,
      "loss": 3.6895,
      "step": 2070
    },
    {
      "epoch": 2.4935313847628175,
      "grad_norm": 18.773216247558594,
      "learning_rate": 6.311027358793166e-07,
      "loss": 3.0383,
      "step": 2080
    },
    {
      "epoch": 2.505510301868711,
      "grad_norm": 19.703941345214844,
      "learning_rate": 6.223399579121029e-07,
      "loss": 2.5712,
      "step": 2090
    },
    {
      "epoch": 2.5174892189746045,
      "grad_norm": 63.63347625732422,
      "learning_rate": 6.136108739830886e-07,
      "loss": 2.2939,
      "step": 2100
    },
    {
      "epoch": 2.5294681360804985,
      "grad_norm": 23.118276596069336,
      "learning_rate": 6.049162628828644e-07,
      "loss": 3.329,
      "step": 2110
    },
    {
      "epoch": 2.541447053186392,
      "grad_norm": 16.502140045166016,
      "learning_rate": 5.962569003264276e-07,
      "loss": 4.0458,
      "step": 2120
    },
    {
      "epoch": 2.5534259702922855,
      "grad_norm": 11.94331169128418,
      "learning_rate": 5.876335588839746e-07,
      "loss": 3.7107,
      "step": 2130
    },
    {
      "epoch": 2.565404887398179,
      "grad_norm": 20.692808151245117,
      "learning_rate": 5.79047007911973e-07,
      "loss": 2.3799,
      "step": 2140
    },
    {
      "epoch": 2.577383804504073,
      "grad_norm": 170.5503692626953,
      "learning_rate": 5.704980134845213e-07,
      "loss": 2.808,
      "step": 2150
    },
    {
      "epoch": 2.5893627216099664,
      "grad_norm": 17.668119430541992,
      "learning_rate": 5.619873383250029e-07,
      "loss": 2.4657,
      "step": 2160
    },
    {
      "epoch": 2.6013416387158603,
      "grad_norm": 12.804652214050293,
      "learning_rate": 5.535157417380346e-07,
      "loss": 4.2857,
      "step": 2170
    },
    {
      "epoch": 2.613320555821754,
      "grad_norm": 231.0024871826172,
      "learning_rate": 5.450839795417266e-07,
      "loss": 5.443,
      "step": 2180
    },
    {
      "epoch": 2.6252994729276473,
      "grad_norm": 156.8734893798828,
      "learning_rate": 5.366928040002476e-07,
      "loss": 4.17,
      "step": 2190
    },
    {
      "epoch": 2.637278390033541,
      "grad_norm": 49.93342590332031,
      "learning_rate": 5.283429637567091e-07,
      "loss": 3.2694,
      "step": 2200
    },
    {
      "epoch": 2.637278390033541,
      "eval_loss": 1.1175537109375,
      "eval_runtime": 238.0084,
      "eval_samples_per_second": 6.26,
      "eval_steps_per_second": 3.13,
      "step": 2200
    },
    {
      "epoch": 2.6492573071394347,
      "grad_norm": 11.659183502197266,
      "learning_rate": 5.200352037663745e-07,
      "loss": 1.2186,
      "step": 2210
    },
    {
      "epoch": 2.6612362242453282,
      "grad_norm": 16.97669219970703,
      "learning_rate": 5.117702652301952e-07,
      "loss": 3.9984,
      "step": 2220
    },
    {
      "epoch": 2.6732151413512217,
      "grad_norm": 49.165199279785156,
      "learning_rate": 5.035488855286823e-07,
      "loss": 2.916,
      "step": 2230
    },
    {
      "epoch": 2.6851940584571157,
      "grad_norm": 14.636496543884277,
      "learning_rate": 4.953717981561186e-07,
      "loss": 3.1995,
      "step": 2240
    },
    {
      "epoch": 2.697172975563009,
      "grad_norm": 13.993462562561035,
      "learning_rate": 4.872397326551179e-07,
      "loss": 3.1531,
      "step": 2250
    },
    {
      "epoch": 2.7091518926689027,
      "grad_norm": 14.866923332214355,
      "learning_rate": 4.791534145515368e-07,
      "loss": 2.7951,
      "step": 2260
    },
    {
      "epoch": 2.721130809774796,
      "grad_norm": 12.874122619628906,
      "learning_rate": 4.711135652897452e-07,
      "loss": 4.0197,
      "step": 2270
    },
    {
      "epoch": 2.73310972688069,
      "grad_norm": 166.74929809570312,
      "learning_rate": 4.6312090216826074e-07,
      "loss": 4.324,
      "step": 2280
    },
    {
      "epoch": 2.7450886439865836,
      "grad_norm": 169.29539489746094,
      "learning_rate": 4.551761382757513e-07,
      "loss": 4.1737,
      "step": 2290
    },
    {
      "epoch": 2.757067561092477,
      "grad_norm": 12.698741912841797,
      "learning_rate": 4.4727998242741627e-07,
      "loss": 3.2982,
      "step": 2300
    },
    {
      "epoch": 2.769046478198371,
      "grad_norm": 97.41565704345703,
      "learning_rate": 4.394331391017474e-07,
      "loss": 3.0522,
      "step": 2310
    },
    {
      "epoch": 2.7810253953042645,
      "grad_norm": 13.397387504577637,
      "learning_rate": 4.316363083776766e-07,
      "loss": 5.0599,
      "step": 2320
    },
    {
      "epoch": 2.793004312410158,
      "grad_norm": 14.772340774536133,
      "learning_rate": 4.2389018587211524e-07,
      "loss": 2.43,
      "step": 2330
    },
    {
      "epoch": 2.8049832295160515,
      "grad_norm": 218.20303344726562,
      "learning_rate": 4.1619546267789453e-07,
      "loss": 5.5137,
      "step": 2340
    },
    {
      "epoch": 2.8169621466219454,
      "grad_norm": 312.639892578125,
      "learning_rate": 4.0855282530210676e-07,
      "loss": 4.4751,
      "step": 2350
    },
    {
      "epoch": 2.828941063727839,
      "grad_norm": 293.28387451171875,
      "learning_rate": 4.0096295560485547e-07,
      "loss": 4.1398,
      "step": 2360
    },
    {
      "epoch": 2.840919980833733,
      "grad_norm": 84.95050811767578,
      "learning_rate": 3.934265307384239e-07,
      "loss": 3.7418,
      "step": 2370
    },
    {
      "epoch": 2.8528988979396264,
      "grad_norm": 108.16950988769531,
      "learning_rate": 3.8594422308685793e-07,
      "loss": 3.392,
      "step": 2380
    },
    {
      "epoch": 2.86487781504552,
      "grad_norm": 258.593505859375,
      "learning_rate": 3.785167002059799e-07,
      "loss": 4.247,
      "step": 2390
    },
    {
      "epoch": 2.8768567321514134,
      "grad_norm": 247.61813354492188,
      "learning_rate": 3.7114462476382966e-07,
      "loss": 3.9058,
      "step": 2400
    },
    {
      "epoch": 2.8768567321514134,
      "eval_loss": 1.125927209854126,
      "eval_runtime": 237.8939,
      "eval_samples_per_second": 6.263,
      "eval_steps_per_second": 3.132,
      "step": 2400
    },
    {
      "epoch": 2.8888356492573073,
      "grad_norm": 82.83909606933594,
      "learning_rate": 3.6382865448154187e-07,
      "loss": 3.9744,
      "step": 2410
    },
    {
      "epoch": 2.900814566363201,
      "grad_norm": 289.5834655761719,
      "learning_rate": 3.5656944207466633e-07,
      "loss": 4.8423,
      "step": 2420
    },
    {
      "epoch": 2.9127934834690943,
      "grad_norm": 11.12870979309082,
      "learning_rate": 3.4936763519493495e-07,
      "loss": 1.8868,
      "step": 2430
    },
    {
      "epoch": 2.924772400574988,
      "grad_norm": 259.54205322265625,
      "learning_rate": 3.4222387637247806e-07,
      "loss": 5.073,
      "step": 2440
    },
    {
      "epoch": 2.9367513176808817,
      "grad_norm": 260.9941711425781,
      "learning_rate": 3.351388029585007e-07,
      "loss": 3.9769,
      "step": 2450
    },
    {
      "epoch": 2.948730234786775,
      "grad_norm": 138.75767517089844,
      "learning_rate": 3.281130470684166e-07,
      "loss": 1.8905,
      "step": 2460
    },
    {
      "epoch": 2.9607091518926687,
      "grad_norm": 212.35589599609375,
      "learning_rate": 3.2114723552545606e-07,
      "loss": 5.6336,
      "step": 2470
    },
    {
      "epoch": 2.9726880689985626,
      "grad_norm": 77.375,
      "learning_rate": 3.142419898047399e-07,
      "loss": 2.5325,
      "step": 2480
    },
    {
      "epoch": 2.984666986104456,
      "grad_norm": 301.5721740722656,
      "learning_rate": 3.073979259778332e-07,
      "loss": 2.8451,
      "step": 2490
    },
    {
      "epoch": 2.9966459032103496,
      "grad_norm": 23.281728744506836,
      "learning_rate": 3.006156546577796e-07,
      "loss": 4.271,
      "step": 2500
    },
    {
      "epoch": 3.009583133684715,
      "grad_norm": 60.59037399291992,
      "learning_rate": 2.9389578094462607e-07,
      "loss": 4.768,
      "step": 2510
    },
    {
      "epoch": 3.0215620507906085,
      "grad_norm": 12.599560737609863,
      "learning_rate": 2.872389043714343e-07,
      "loss": 3.4648,
      "step": 2520
    },
    {
      "epoch": 3.033540967896502,
      "grad_norm": 60.7338981628418,
      "learning_rate": 2.806456188507943e-07,
      "loss": 4.1664,
      "step": 2530
    },
    {
      "epoch": 3.045519885002396,
      "grad_norm": 124.4323959350586,
      "learning_rate": 2.7411651262183465e-07,
      "loss": 3.2584,
      "step": 2540
    },
    {
      "epoch": 3.0574988021082894,
      "grad_norm": 10.696157455444336,
      "learning_rate": 2.676521681977425e-07,
      "loss": 1.6209,
      "step": 2550
    },
    {
      "epoch": 3.069477719214183,
      "grad_norm": 78.30728149414062,
      "learning_rate": 2.612531623137922e-07,
      "loss": 2.9967,
      "step": 2560
    },
    {
      "epoch": 3.081456636320077,
      "grad_norm": 12.110499382019043,
      "learning_rate": 2.5492006587589033e-07,
      "loss": 1.8501,
      "step": 2570
    },
    {
      "epoch": 3.0934355534259703,
      "grad_norm": 142.67459106445312,
      "learning_rate": 2.4865344390964016e-07,
      "loss": 6.7426,
      "step": 2580
    },
    {
      "epoch": 3.105414470531864,
      "grad_norm": 84.46195220947266,
      "learning_rate": 2.424538555099326e-07,
      "loss": 3.011,
      "step": 2590
    },
    {
      "epoch": 3.1173933876377578,
      "grad_norm": 287.4181213378906,
      "learning_rate": 2.3632185379106383e-07,
      "loss": 4.2906,
      "step": 2600
    },
    {
      "epoch": 3.1173933876377578,
      "eval_loss": 1.1396493911743164,
      "eval_runtime": 240.643,
      "eval_samples_per_second": 6.192,
      "eval_steps_per_second": 3.096,
      "step": 2600
    },
    {
      "epoch": 3.1293723047436512,
      "grad_norm": 17.018083572387695,
      "learning_rate": 2.302579858373881e-07,
      "loss": 2.1286,
      "step": 2610
    },
    {
      "epoch": 3.1413512218495447,
      "grad_norm": 13.265679359436035,
      "learning_rate": 2.2426279265450708e-07,
      "loss": 2.9965,
      "step": 2620
    },
    {
      "epoch": 3.1533301389554382,
      "grad_norm": 156.8545684814453,
      "learning_rate": 2.183368091210037e-07,
      "loss": 3.7899,
      "step": 2630
    },
    {
      "epoch": 3.165309056061332,
      "grad_norm": 17.818618774414062,
      "learning_rate": 2.1248056394072078e-07,
      "loss": 4.1165,
      "step": 2640
    },
    {
      "epoch": 3.1772879731672257,
      "grad_norm": 15.270909309387207,
      "learning_rate": 2.0669457959559177e-07,
      "loss": 3.1192,
      "step": 2650
    },
    {
      "epoch": 3.189266890273119,
      "grad_norm": 78.96019744873047,
      "learning_rate": 2.0097937229902485e-07,
      "loss": 5.3403,
      "step": 2660
    },
    {
      "epoch": 3.201245807379013,
      "grad_norm": 376.2461242675781,
      "learning_rate": 1.9533545194984791e-07,
      "loss": 3.9551,
      "step": 2670
    },
    {
      "epoch": 3.2132247244849066,
      "grad_norm": 16.849010467529297,
      "learning_rate": 1.8976332208681744e-07,
      "loss": 5.6715,
      "step": 2680
    },
    {
      "epoch": 3.2252036415908,
      "grad_norm": 171.6584930419922,
      "learning_rate": 1.8426347984369273e-07,
      "loss": 5.2323,
      "step": 2690
    },
    {
      "epoch": 3.237182558696694,
      "grad_norm": 120.8387680053711,
      "learning_rate": 1.788364159048833e-07,
      "loss": 4.2853,
      "step": 2700
    },
    {
      "epoch": 3.2491614758025875,
      "grad_norm": 13.419800758361816,
      "learning_rate": 1.734826144616698e-07,
      "loss": 2.9811,
      "step": 2710
    },
    {
      "epoch": 3.261140392908481,
      "grad_norm": 196.2430877685547,
      "learning_rate": 1.6820255316900756e-07,
      "loss": 4.3565,
      "step": 2720
    },
    {
      "epoch": 3.2731193100143745,
      "grad_norm": 15.63242244720459,
      "learning_rate": 1.6299670310290915e-07,
      "loss": 2.4933,
      "step": 2730
    },
    {
      "epoch": 3.2850982271202684,
      "grad_norm": 60.21677017211914,
      "learning_rate": 1.5786552871841774e-07,
      "loss": 2.6513,
      "step": 2740
    },
    {
      "epoch": 3.297077144226162,
      "grad_norm": 223.22097778320312,
      "learning_rate": 1.528094878081677e-07,
      "loss": 3.3477,
      "step": 2750
    },
    {
      "epoch": 3.3090560613320554,
      "grad_norm": 95.59901428222656,
      "learning_rate": 1.478290314615427e-07,
      "loss": 3.0678,
      "step": 2760
    },
    {
      "epoch": 3.3210349784379494,
      "grad_norm": 36.64326095581055,
      "learning_rate": 1.4292460402442995e-07,
      "loss": 3.4483,
      "step": 2770
    },
    {
      "epoch": 3.333013895543843,
      "grad_norm": 233.20370483398438,
      "learning_rate": 1.3809664305957625e-07,
      "loss": 2.9447,
      "step": 2780
    },
    {
      "epoch": 3.3449928126497364,
      "grad_norm": 14.824277877807617,
      "learning_rate": 1.3334557930754963e-07,
      "loss": 2.547,
      "step": 2790
    },
    {
      "epoch": 3.35697172975563,
      "grad_norm": 208.10154724121094,
      "learning_rate": 1.2867183664831038e-07,
      "loss": 6.0572,
      "step": 2800
    },
    {
      "epoch": 3.35697172975563,
      "eval_loss": 1.147083044052124,
      "eval_runtime": 239.357,
      "eval_samples_per_second": 6.225,
      "eval_steps_per_second": 3.113,
      "step": 2800
    },
    {
      "epoch": 3.368950646861524,
      "grad_norm": 167.9786834716797,
      "learning_rate": 1.2407583206339256e-07,
      "loss": 2.6004,
      "step": 2810
    },
    {
      "epoch": 3.3809295639674173,
      "grad_norm": 236.07484436035156,
      "learning_rate": 1.195579755987024e-07,
      "loss": 2.3534,
      "step": 2820
    },
    {
      "epoch": 3.3929084810733112,
      "grad_norm": 38.18111801147461,
      "learning_rate": 1.1511867032793321e-07,
      "loss": 2.5498,
      "step": 2830
    },
    {
      "epoch": 3.4048873981792047,
      "grad_norm": 136.2451934814453,
      "learning_rate": 1.107583123166066e-07,
      "loss": 5.3208,
      "step": 2840
    },
    {
      "epoch": 3.416866315285098,
      "grad_norm": 316.2095947265625,
      "learning_rate": 1.0647729058673427e-07,
      "loss": 4.3772,
      "step": 2850
    },
    {
      "epoch": 3.4288452323909917,
      "grad_norm": 14.228015899658203,
      "learning_rate": 1.0227598708211172e-07,
      "loss": 2.2948,
      "step": 2860
    },
    {
      "epoch": 3.4408241494968856,
      "grad_norm": 52.19302749633789,
      "learning_rate": 9.81547766342401e-08,
      "loss": 2.223,
      "step": 2870
    },
    {
      "epoch": 3.452803066602779,
      "grad_norm": 18.04366683959961,
      "learning_rate": 9.411402692888715e-08,
      "loss": 3.6671,
      "step": 2880
    },
    {
      "epoch": 3.4647819837086726,
      "grad_norm": 12.00094985961914,
      "learning_rate": 9.015409847328037e-08,
      "loss": 2.3488,
      "step": 2890
    },
    {
      "epoch": 3.4767609008145666,
      "grad_norm": 88.41423797607422,
      "learning_rate": 8.62753445639457e-08,
      "loss": 3.2758,
      "step": 2900
    },
    {
      "epoch": 3.48873981792046,
      "grad_norm": 56.944828033447266,
      "learning_rate": 8.247811125518489e-08,
      "loss": 1.576,
      "step": 2910
    },
    {
      "epoch": 3.5007187350263536,
      "grad_norm": 323.6489562988281,
      "learning_rate": 7.876273732820327e-08,
      "loss": 4.9528,
      "step": 2920
    },
    {
      "epoch": 3.512697652132247,
      "grad_norm": 10.932809829711914,
      "learning_rate": 7.51295542608834e-08,
      "loss": 3.6918,
      "step": 2930
    },
    {
      "epoch": 3.524676569238141,
      "grad_norm": 12.945392608642578,
      "learning_rate": 7.157888619821106e-08,
      "loss": 1.7161,
      "step": 2940
    },
    {
      "epoch": 3.5366554863440345,
      "grad_norm": 358.42498779296875,
      "learning_rate": 6.811104992335648e-08,
      "loss": 4.4565,
      "step": 2950
    },
    {
      "epoch": 3.548634403449928,
      "grad_norm": 231.78411865234375,
      "learning_rate": 6.47263548294108e-08,
      "loss": 3.5586,
      "step": 2960
    },
    {
      "epoch": 3.560613320555822,
      "grad_norm": 231.84341430664062,
      "learning_rate": 6.142510289178337e-08,
      "loss": 3.4724,
      "step": 2970
    },
    {
      "epoch": 3.5725922376617154,
      "grad_norm": 198.78810119628906,
      "learning_rate": 5.8207588641260185e-08,
      "loss": 2.5415,
      "step": 2980
    },
    {
      "epoch": 3.584571154767609,
      "grad_norm": 76.49454498291016,
      "learning_rate": 5.507409913772543e-08,
      "loss": 3.7494,
      "step": 2990
    },
    {
      "epoch": 3.5965500718735024,
      "grad_norm": 262.359130859375,
      "learning_rate": 5.202491394455155e-08,
      "loss": 4.0544,
      "step": 3000
    },
    {
      "epoch": 3.5965500718735024,
      "eval_loss": 1.1526151895523071,
      "eval_runtime": 238.4822,
      "eval_samples_per_second": 6.248,
      "eval_steps_per_second": 3.124,
      "step": 3000
    },
    {
      "epoch": 3.6085289889793963,
      "grad_norm": 14.773038864135742,
      "learning_rate": 4.9060305103657e-08,
      "loss": 3.8126,
      "step": 3010
    },
    {
      "epoch": 3.62050790608529,
      "grad_norm": 288.3175354003906,
      "learning_rate": 4.61805371112356e-08,
      "loss": 2.2371,
      "step": 3020
    },
    {
      "epoch": 3.6324868231911838,
      "grad_norm": 15.71839427947998,
      "learning_rate": 4.3661497350331423e-08,
      "loss": 2.8255,
      "step": 3030
    },
    {
      "epoch": 3.6444657402970773,
      "grad_norm": 47.34138870239258,
      "learning_rate": 4.094362852900846e-08,
      "loss": 2.4564,
      "step": 3040
    },
    {
      "epoch": 3.6564446574029708,
      "grad_norm": 133.03443908691406,
      "learning_rate": 3.8311324709047524e-08,
      "loss": 3.9076,
      "step": 3050
    },
    {
      "epoch": 3.6684235745088642,
      "grad_norm": 85.79124450683594,
      "learning_rate": 3.57648207390836e-08,
      "loss": 1.6264,
      "step": 3060
    },
    {
      "epoch": 3.680402491614758,
      "grad_norm": 11.73284912109375,
      "learning_rate": 3.3304343812869175e-08,
      "loss": 2.2377,
      "step": 3070
    },
    {
      "epoch": 3.6923814087206517,
      "grad_norm": 103.52379608154297,
      "learning_rate": 3.0930113449003536e-08,
      "loss": 2.5226,
      "step": 3080
    },
    {
      "epoch": 3.704360325826545,
      "grad_norm": 121.46673583984375,
      "learning_rate": 2.8642341471348585e-08,
      "loss": 5.8129,
      "step": 3090
    },
    {
      "epoch": 3.716339242932439,
      "grad_norm": 39.363887786865234,
      "learning_rate": 2.644123199013004e-08,
      "loss": 1.8684,
      "step": 3100
    },
    {
      "epoch": 3.7283181600383326,
      "grad_norm": 22.17119598388672,
      "learning_rate": 2.432698138372713e-08,
      "loss": 4.5753,
      "step": 3110
    },
    {
      "epoch": 3.740297077144226,
      "grad_norm": 123.43873596191406,
      "learning_rate": 2.2299778281151927e-08,
      "loss": 3.4706,
      "step": 3120
    },
    {
      "epoch": 3.7522759942501196,
      "grad_norm": 62.01347351074219,
      "learning_rate": 2.03598035452206e-08,
      "loss": 1.0962,
      "step": 3130
    },
    {
      "epoch": 3.7642549113560135,
      "grad_norm": 106.05194091796875,
      "learning_rate": 1.8507230256417316e-08,
      "loss": 4.0847,
      "step": 3140
    },
    {
      "epoch": 3.776233828461907,
      "grad_norm": 14.505044937133789,
      "learning_rate": 1.674222369745182e-08,
      "loss": 2.7005,
      "step": 3150
    },
    {
      "epoch": 3.7882127455678005,
      "grad_norm": 74.14771270751953,
      "learning_rate": 1.5064941338513548e-08,
      "loss": 4.5833,
      "step": 3160
    },
    {
      "epoch": 3.8001916626736945,
      "grad_norm": 109.61576080322266,
      "learning_rate": 1.3475532823222779e-08,
      "loss": 3.3511,
      "step": 3170
    },
    {
      "epoch": 3.812170579779588,
      "grad_norm": 142.658447265625,
      "learning_rate": 1.1974139955279294e-08,
      "loss": 4.0569,
      "step": 3180
    },
    {
      "epoch": 3.8241494968854814,
      "grad_norm": 374.00701904296875,
      "learning_rate": 1.0560896685811061e-08,
      "loss": 3.657,
      "step": 3190
    },
    {
      "epoch": 3.836128413991375,
      "grad_norm": 191.76698303222656,
      "learning_rate": 9.235929101423457e-09,
      "loss": 3.2204,
      "step": 3200
    },
    {
      "epoch": 3.836128413991375,
      "eval_loss": 1.1512279510498047,
      "eval_runtime": 239.5597,
      "eval_samples_per_second": 6.22,
      "eval_steps_per_second": 3.11,
      "step": 3200
    },
    {
      "epoch": 3.848107331097269,
      "grad_norm": 89.6385726928711,
      "learning_rate": 7.99935541295016e-09,
      "loss": 4.2172,
      "step": 3210
    },
    {
      "epoch": 3.8600862482031624,
      "grad_norm": 52.579227447509766,
      "learning_rate": 6.8512859449064705e-09,
      "loss": 3.1792,
      "step": 3220
    },
    {
      "epoch": 3.8720651653090563,
      "grad_norm": 159.96087646484375,
      "learning_rate": 5.791823125646522e-09,
      "loss": 4.6078,
      "step": 3230
    },
    {
      "epoch": 3.88404408241495,
      "grad_norm": 248.06581115722656,
      "learning_rate": 4.8210614782245866e-09,
      "loss": 4.6227,
      "step": 3240
    },
    {
      "epoch": 3.8960229995208433,
      "grad_norm": 10.945176124572754,
      "learning_rate": 3.939087611962377e-09,
      "loss": 1.8884,
      "step": 3250
    },
    {
      "epoch": 3.908001916626737,
      "grad_norm": 279.2950439453125,
      "learning_rate": 3.1459802147214554e-09,
      "loss": 3.5358,
      "step": 3260
    },
    {
      "epoch": 3.9199808337326307,
      "grad_norm": 278.5760192871094,
      "learning_rate": 2.441810045883175e-09,
      "loss": 4.0377,
      "step": 3270
    },
    {
      "epoch": 3.9319597508385242,
      "grad_norm": 20.70584487915039,
      "learning_rate": 1.8266399300355118e-09,
      "loss": 4.9587,
      "step": 3280
    },
    {
      "epoch": 3.9439386679444177,
      "grad_norm": 238.9534912109375,
      "learning_rate": 1.300524751368326e-09,
      "loss": 3.8602,
      "step": 3290
    },
    {
      "epoch": 3.9559175850503117,
      "grad_norm": 16.01015281677246,
      "learning_rate": 8.635114487760553e-10,
      "loss": 1.4575,
      "step": 3300
    },
    {
      "epoch": 3.967896502156205,
      "grad_norm": 14.110264778137207,
      "learning_rate": 5.156390116707321e-10,
      "loss": 1.1673,
      "step": 3310
    },
    {
      "epoch": 3.9798754192620986,
      "grad_norm": 43.93592834472656,
      "learning_rate": 2.56938476502655e-10,
      "loss": 3.8769,
      "step": 3320
    },
    {
      "epoch": 3.991854336367992,
      "grad_norm": 9.19588851928711,
      "learning_rate": 8.743292399204793e-11,
      "loss": 2.4475,
      "step": 3330
    }
  ],
  "logging_steps": 10,
  "max_steps": 3336,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 300,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.982695446856335e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}