{
  "best_metric": 1.1752163171768188,
  "best_model_checkpoint": "./bert_mlm_finetuned/checkpoint-1600",
  "epoch": 0.2048,
  "eval_steps": 100,
  "global_step": 1600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00064,
      "grad_norm": 7.650606632232666,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 6.29,
      "step": 5
    },
    {
      "epoch": 0.00128,
      "grad_norm": 4.541823387145996,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 6.3815,
      "step": 10
    },
    {
      "epoch": 0.00192,
      "grad_norm": 4.245054721832275,
      "learning_rate": 3e-06,
      "loss": 6.2854,
      "step": 15
    },
    {
      "epoch": 0.00256,
      "grad_norm": 4.5587897300720215,
      "learning_rate": 4.000000000000001e-06,
      "loss": 6.0674,
      "step": 20
    },
    {
      "epoch": 0.0032,
      "grad_norm": 3.7703804969787598,
      "learning_rate": 4.800000000000001e-06,
      "loss": 6.2961,
      "step": 25
    },
    {
      "epoch": 0.00384,
      "grad_norm": 3.8425862789154053,
      "learning_rate": 5.8e-06,
      "loss": 6.3326,
      "step": 30
    },
    {
      "epoch": 0.00448,
      "grad_norm": 4.413463115692139,
      "learning_rate": 6.800000000000001e-06,
      "loss": 6.183,
      "step": 35
    },
    {
      "epoch": 0.00512,
      "grad_norm": 4.1980509757995605,
      "learning_rate": 7.800000000000002e-06,
      "loss": 6.2654,
      "step": 40
    },
    {
      "epoch": 0.00576,
      "grad_norm": 3.9166719913482666,
      "learning_rate": 8.8e-06,
      "loss": 6.0916,
      "step": 45
    },
    {
      "epoch": 0.0064,
      "grad_norm": 3.4706904888153076,
      "learning_rate": 9.800000000000001e-06,
      "loss": 6.103,
      "step": 50
    },
    {
      "epoch": 0.00704,
      "grad_norm": 5.138203144073486,
      "learning_rate": 9.999998372356185e-06,
      "loss": 6.2379,
      "step": 55
    },
    {
      "epoch": 0.00768,
      "grad_norm": 3.7806520462036133,
      "learning_rate": 9.999991760055e-06,
      "loss": 6.1776,
      "step": 60
    },
    {
      "epoch": 0.00832,
      "grad_norm": 3.5731871128082275,
      "learning_rate": 9.999980061375427e-06,
      "loss": 6.2082,
      "step": 65
    },
    {
      "epoch": 0.00896,
      "grad_norm": 3.661797285079956,
      "learning_rate": 9.999963276329369e-06,
      "loss": 6.0704,
      "step": 70
    },
    {
      "epoch": 0.0096,
      "grad_norm": 3.6181113719940186,
      "learning_rate": 9.999941404933902e-06,
      "loss": 6.2081,
      "step": 75
    },
    {
      "epoch": 0.01024,
      "grad_norm": 3.3162803649902344,
      "learning_rate": 9.99991444721127e-06,
      "loss": 5.8807,
      "step": 80
    },
    {
      "epoch": 0.01088,
      "grad_norm": 3.6022472381591797,
      "learning_rate": 9.999882403188902e-06,
      "loss": 6.1092,
      "step": 85
    },
    {
      "epoch": 0.01152,
      "grad_norm": 7.291418552398682,
      "learning_rate": 9.999845272899393e-06,
      "loss": 5.7668,
      "step": 90
    },
    {
      "epoch": 0.01216,
      "grad_norm": 3.522437810897827,
      "learning_rate": 9.999803056380517e-06,
      "loss": 6.1621,
      "step": 95
    },
    {
      "epoch": 0.0128,
      "grad_norm": 3.9014439582824707,
      "learning_rate": 9.999755753675216e-06,
      "loss": 6.0573,
      "step": 100
    },
    {
      "epoch": 0.0128,
      "eval_loss": 1.5072969198226929,
      "eval_runtime": 11.1161,
      "eval_samples_per_second": 89.96,
      "eval_steps_per_second": 11.245,
      "step": 100
    },
    {
      "epoch": 0.01344,
      "grad_norm": 3.7579081058502197,
      "learning_rate": 9.999703364831614e-06,
      "loss": 6.1671,
      "step": 105
    },
    {
      "epoch": 0.01408,
      "grad_norm": 3.7058262825012207,
      "learning_rate": 9.999645889903002e-06,
      "loss": 6.1348,
      "step": 110
    },
    {
      "epoch": 0.01472,
      "grad_norm": 5.018667697906494,
      "learning_rate": 9.99958332894785e-06,
      "loss": 5.9376,
      "step": 115
    },
    {
      "epoch": 0.01536,
      "grad_norm": 3.5420188903808594,
      "learning_rate": 9.999515682029798e-06,
      "loss": 5.9961,
      "step": 120
    },
    {
      "epoch": 0.016,
      "grad_norm": 3.5725393295288086,
      "learning_rate": 9.999442949217663e-06,
      "loss": 5.8439,
      "step": 125
    },
    {
      "epoch": 0.01664,
      "grad_norm": 3.8440959453582764,
      "learning_rate": 9.999365130585435e-06,
      "loss": 5.7857,
      "step": 130
    },
    {
      "epoch": 0.01728,
      "grad_norm": 3.4371285438537598,
      "learning_rate": 9.999282226212276e-06,
      "loss": 5.799,
      "step": 135
    },
    {
      "epoch": 0.01792,
      "grad_norm": 3.996847152709961,
      "learning_rate": 9.999194236182523e-06,
      "loss": 6.0022,
      "step": 140
    },
    {
      "epoch": 0.01856,
      "grad_norm": 3.720330238342285,
      "learning_rate": 9.999101160585687e-06,
      "loss": 5.925,
      "step": 145
    },
    {
      "epoch": 0.0192,
      "grad_norm": 3.8822953701019287,
      "learning_rate": 9.99900299951645e-06,
      "loss": 5.8085,
      "step": 150
    },
    {
      "epoch": 0.01984,
      "grad_norm": 3.599283456802368,
      "learning_rate": 9.99889975307467e-06,
      "loss": 5.6533,
      "step": 155
    },
    {
      "epoch": 0.02048,
      "grad_norm": 3.4847381114959717,
      "learning_rate": 9.998791421365376e-06,
      "loss": 5.9021,
      "step": 160
    },
    {
      "epoch": 0.02112,
      "grad_norm": 3.4302055835723877,
      "learning_rate": 9.998678004498774e-06,
      "loss": 5.962,
      "step": 165
    },
    {
      "epoch": 0.02176,
      "grad_norm": 4.561929702758789,
      "learning_rate": 9.99855950259024e-06,
      "loss": 5.9011,
      "step": 170
    },
    {
      "epoch": 0.0224,
      "grad_norm": 4.069271087646484,
      "learning_rate": 9.998435915760323e-06,
      "loss": 5.6782,
      "step": 175
    },
    {
      "epoch": 0.02304,
      "grad_norm": 3.5959055423736572,
      "learning_rate": 9.998307244134741e-06,
      "loss": 5.8107,
      "step": 180
    },
    {
      "epoch": 0.02368,
      "grad_norm": 3.5477242469787598,
      "learning_rate": 9.998173487844396e-06,
      "loss": 5.8335,
      "step": 185
    },
    {
      "epoch": 0.02432,
      "grad_norm": 4.488218307495117,
      "learning_rate": 9.998034647025349e-06,
      "loss": 5.8285,
      "step": 190
    },
    {
      "epoch": 0.02496,
      "grad_norm": 3.555074691772461,
      "learning_rate": 9.997890721818844e-06,
      "loss": 5.817,
      "step": 195
    },
    {
      "epoch": 0.0256,
      "grad_norm": 3.6248419284820557,
      "learning_rate": 9.99774171237129e-06,
      "loss": 5.8368,
      "step": 200
    },
    {
      "epoch": 0.0256,
      "eval_loss": 1.440572738647461,
      "eval_runtime": 6.6468,
      "eval_samples_per_second": 150.448,
      "eval_steps_per_second": 18.806,
      "step": 200
    },
    {
      "epoch": 0.02624,
      "grad_norm": 3.432421922683716,
      "learning_rate": 9.997587618834272e-06,
      "loss": 5.7842,
      "step": 205
    },
    {
      "epoch": 0.02688,
      "grad_norm": 3.333038806915283,
      "learning_rate": 9.997428441364546e-06,
      "loss": 5.7173,
      "step": 210
    },
    {
      "epoch": 0.02752,
      "grad_norm": 3.7716541290283203,
      "learning_rate": 9.997264180124038e-06,
      "loss": 5.719,
      "step": 215
    },
    {
      "epoch": 0.02816,
      "grad_norm": 3.345600128173828,
      "learning_rate": 9.99709483527985e-06,
      "loss": 5.8428,
      "step": 220
    },
    {
      "epoch": 0.0288,
      "grad_norm": 3.7677502632141113,
      "learning_rate": 9.99692040700425e-06,
      "loss": 5.7393,
      "step": 225
    },
    {
      "epoch": 0.02944,
      "grad_norm": 11.996383666992188,
      "learning_rate": 9.996740895474682e-06,
      "loss": 5.5566,
      "step": 230
    },
    {
      "epoch": 0.03008,
      "grad_norm": 3.6089084148406982,
      "learning_rate": 9.996556300873758e-06,
      "loss": 5.6939,
      "step": 235
    },
    {
      "epoch": 0.03072,
      "grad_norm": 3.834825038909912,
      "learning_rate": 9.996366623389263e-06,
      "loss": 5.8123,
      "step": 240
    },
    {
      "epoch": 0.03136,
      "grad_norm": 3.570263147354126,
      "learning_rate": 9.99617186321415e-06,
      "loss": 5.6839,
      "step": 245
    },
    {
      "epoch": 0.032,
      "grad_norm": 3.5728812217712402,
      "learning_rate": 9.995972020546545e-06,
      "loss": 5.7764,
      "step": 250
    },
    {
      "epoch": 0.03264,
      "grad_norm": 3.4725637435913086,
      "learning_rate": 9.995767095589743e-06,
      "loss": 5.6879,
      "step": 255
    },
    {
      "epoch": 0.03328,
      "grad_norm": 3.811537742614746,
      "learning_rate": 9.99555708855221e-06,
      "loss": 5.6418,
      "step": 260
    },
    {
      "epoch": 0.03392,
      "grad_norm": 3.494992971420288,
      "learning_rate": 9.99534199964758e-06,
      "loss": 5.6927,
      "step": 265
    },
    {
      "epoch": 0.03456,
      "grad_norm": 3.8107383251190186,
      "learning_rate": 9.995121829094662e-06,
      "loss": 5.5658,
      "step": 270
    },
    {
      "epoch": 0.0352,
      "grad_norm": 3.570551633834839,
      "learning_rate": 9.994896577117425e-06,
      "loss": 5.8131,
      "step": 275
    },
    {
      "epoch": 0.03584,
      "grad_norm": 3.540811538696289,
      "learning_rate": 9.994666243945018e-06,
      "loss": 5.6009,
      "step": 280
    },
    {
      "epoch": 0.03648,
      "grad_norm": 3.7275819778442383,
      "learning_rate": 9.99443082981175e-06,
      "loss": 5.6407,
      "step": 285
    },
    {
      "epoch": 0.03712,
      "grad_norm": 4.194495677947998,
      "learning_rate": 9.994190334957103e-06,
      "loss": 5.8319,
      "step": 290
    },
    {
      "epoch": 0.03776,
      "grad_norm": 3.5107626914978027,
      "learning_rate": 9.993944759625728e-06,
      "loss": 5.5765,
      "step": 295
    },
    {
      "epoch": 0.0384,
      "grad_norm": 3.4100208282470703,
      "learning_rate": 9.993694104067444e-06,
      "loss": 5.7473,
      "step": 300
    },
    {
      "epoch": 0.0384,
      "eval_loss": 1.407908320426941,
      "eval_runtime": 6.6542,
      "eval_samples_per_second": 150.281,
      "eval_steps_per_second": 18.785,
      "step": 300
    },
    {
      "epoch": 0.03904,
      "grad_norm": 3.7727818489074707,
      "learning_rate": 9.993438368537236e-06,
      "loss": 5.6802,
      "step": 305
    },
    {
      "epoch": 0.03968,
      "grad_norm": 3.445909023284912,
      "learning_rate": 9.993177553295258e-06,
      "loss": 5.7484,
      "step": 310
    },
    {
      "epoch": 0.04032,
      "grad_norm": 3.4199888706207275,
      "learning_rate": 9.992911658606832e-06,
      "loss": 5.7648,
      "step": 315
    },
    {
      "epoch": 0.04096,
      "grad_norm": 4.9640655517578125,
      "learning_rate": 9.992640684742445e-06,
      "loss": 5.7922,
      "step": 320
    },
    {
      "epoch": 0.0416,
      "grad_norm": 3.3730976581573486,
      "learning_rate": 9.992364631977754e-06,
      "loss": 5.677,
      "step": 325
    },
    {
      "epoch": 0.04224,
      "grad_norm": 3.540597915649414,
      "learning_rate": 9.99208350059358e-06,
      "loss": 5.5495,
      "step": 330
    },
    {
      "epoch": 0.04288,
      "grad_norm": 3.6853768825531006,
      "learning_rate": 9.991797290875915e-06,
      "loss": 5.4089,
      "step": 335
    },
    {
      "epoch": 0.04352,
      "grad_norm": 3.6380045413970947,
      "learning_rate": 9.991506003115911e-06,
      "loss": 5.4849,
      "step": 340
    },
    {
      "epoch": 0.04416,
      "grad_norm": 3.265488862991333,
      "learning_rate": 9.991209637609887e-06,
      "loss": 5.523,
      "step": 345
    },
    {
      "epoch": 0.0448,
      "grad_norm": 3.2634189128875732,
      "learning_rate": 9.990908194659332e-06,
      "loss": 5.5664,
      "step": 350
    },
    {
      "epoch": 0.04544,
      "grad_norm": 3.569810152053833,
      "learning_rate": 9.990601674570895e-06,
      "loss": 5.5059,
      "step": 355
    },
    {
      "epoch": 0.04608,
      "grad_norm": 3.580211877822876,
      "learning_rate": 9.990290077656393e-06,
      "loss": 5.4079,
      "step": 360
    },
    {
      "epoch": 0.04672,
      "grad_norm": 3.4860317707061768,
      "learning_rate": 9.989973404232805e-06,
      "loss": 5.6858,
      "step": 365
    },
    {
      "epoch": 0.04736,
      "grad_norm": 4.026730060577393,
      "learning_rate": 9.989651654622277e-06,
      "loss": 5.5662,
      "step": 370
    },
    {
      "epoch": 0.048,
      "grad_norm": 3.364692449569702,
      "learning_rate": 9.989324829152119e-06,
      "loss": 5.5304,
      "step": 375
    },
    {
      "epoch": 0.04864,
      "grad_norm": 3.611964464187622,
      "learning_rate": 9.9889929281548e-06,
      "loss": 5.3911,
      "step": 380
    },
    {
      "epoch": 0.04928,
      "grad_norm": 3.2946035861968994,
      "learning_rate": 9.988655951967958e-06,
      "loss": 5.4102,
      "step": 385
    },
    {
      "epoch": 0.04992,
      "grad_norm": 3.963909864425659,
      "learning_rate": 9.98831390093439e-06,
      "loss": 5.549,
      "step": 390
    },
    {
      "epoch": 0.05056,
      "grad_norm": 3.2876341342926025,
      "learning_rate": 9.987966775402056e-06,
      "loss": 5.5388,
      "step": 395
    },
    {
      "epoch": 0.0512,
      "grad_norm": 3.8467471599578857,
      "learning_rate": 9.98761457572408e-06,
      "loss": 5.454,
      "step": 400
    },
    {
      "epoch": 0.0512,
      "eval_loss": 1.3826359510421753,
      "eval_runtime": 7.0199,
      "eval_samples_per_second": 142.452,
      "eval_steps_per_second": 17.807,
      "step": 400
    },
    {
      "epoch": 0.05184,
      "grad_norm": 3.675231695175171,
      "learning_rate": 9.987257302258748e-06,
      "loss": 5.674,
      "step": 405
    },
    {
      "epoch": 0.05248,
      "grad_norm": 3.787940263748169,
      "learning_rate": 9.986894955369504e-06,
      "loss": 5.5466,
      "step": 410
    },
    {
      "epoch": 0.05312,
      "grad_norm": 3.677966833114624,
      "learning_rate": 9.986527535424956e-06,
      "loss": 5.4762,
      "step": 415
    },
    {
      "epoch": 0.05376,
      "grad_norm": 3.5083606243133545,
      "learning_rate": 9.986155042798874e-06,
      "loss": 5.3145,
      "step": 420
    },
    {
      "epoch": 0.0544,
      "grad_norm": 3.536379098892212,
      "learning_rate": 9.98577747787018e-06,
      "loss": 5.3769,
      "step": 425
    },
    {
      "epoch": 0.05504,
      "grad_norm": 3.5448412895202637,
      "learning_rate": 9.98539484102297e-06,
      "loss": 5.3996,
      "step": 430
    },
    {
      "epoch": 0.05568,
      "grad_norm": 3.359647274017334,
      "learning_rate": 9.985007132646489e-06,
      "loss": 5.3114,
      "step": 435
    },
    {
      "epoch": 0.05632,
      "grad_norm": 3.3419110774993896,
      "learning_rate": 9.984614353135143e-06,
      "loss": 5.4383,
      "step": 440
    },
    {
      "epoch": 0.05696,
      "grad_norm": 3.558025360107422,
      "learning_rate": 9.984216502888496e-06,
      "loss": 5.5239,
      "step": 445
    },
    {
      "epoch": 0.0576,
      "grad_norm": 3.6349422931671143,
      "learning_rate": 9.983813582311277e-06,
      "loss": 5.5639,
      "step": 450
    },
    {
      "epoch": 0.05824,
      "grad_norm": 3.2916922569274902,
      "learning_rate": 9.983405591813362e-06,
      "loss": 5.3886,
      "step": 455
    },
    {
      "epoch": 0.05888,
      "grad_norm": 3.32891845703125,
      "learning_rate": 9.982992531809796e-06,
      "loss": 5.526,
      "step": 460
    },
    {
      "epoch": 0.05952,
      "grad_norm": 3.8752880096435547,
      "learning_rate": 9.982574402720773e-06,
      "loss": 5.6599,
      "step": 465
    },
    {
      "epoch": 0.06016,
      "grad_norm": 3.604433536529541,
      "learning_rate": 9.982151204971646e-06,
      "loss": 5.4567,
      "step": 470
    },
    {
      "epoch": 0.0608,
      "grad_norm": 3.3058159351348877,
      "learning_rate": 9.981722938992926e-06,
      "loss": 5.4981,
      "step": 475
    },
    {
      "epoch": 0.06144,
      "grad_norm": 3.7341926097869873,
      "learning_rate": 9.981289605220276e-06,
      "loss": 5.3278,
      "step": 480
    },
    {
      "epoch": 0.06208,
      "grad_norm": 3.51798415184021,
      "learning_rate": 9.980851204094519e-06,
      "loss": 5.5029,
      "step": 485
    },
    {
      "epoch": 0.06272,
      "grad_norm": 3.6541428565979004,
      "learning_rate": 9.980407736061629e-06,
      "loss": 5.3987,
      "step": 490
    },
    {
      "epoch": 0.06336,
      "grad_norm": 3.420767307281494,
      "learning_rate": 9.979959201572736e-06,
      "loss": 5.405,
      "step": 495
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.7169559001922607,
      "learning_rate": 9.979505601084124e-06,
      "loss": 5.498,
      "step": 500
    },
    {
      "epoch": 0.064,
      "eval_loss": 1.3493109941482544,
      "eval_runtime": 7.1309,
      "eval_samples_per_second": 140.234,
      "eval_steps_per_second": 17.529,
      "step": 500
    },
    {
      "epoch": 0.06464,
      "grad_norm": 4.536627769470215,
      "learning_rate": 9.97904693505723e-06,
      "loss": 5.5237,
      "step": 505
    },
    {
      "epoch": 0.06528,
      "grad_norm": 3.204948902130127,
      "learning_rate": 9.978583203958649e-06,
      "loss": 5.3746,
      "step": 510
    },
    {
      "epoch": 0.06592,
      "grad_norm": 3.4658005237579346,
      "learning_rate": 9.978114408260118e-06,
      "loss": 5.4567,
      "step": 515
    },
    {
      "epoch": 0.06656,
      "grad_norm": 4.932333469390869,
      "learning_rate": 9.977640548438534e-06,
      "loss": 5.1959,
      "step": 520
    },
    {
      "epoch": 0.0672,
      "grad_norm": 3.4697563648223877,
      "learning_rate": 9.977161624975948e-06,
      "loss": 5.4013,
      "step": 525
    },
    {
      "epoch": 0.06784,
      "grad_norm": 3.441819667816162,
      "learning_rate": 9.976677638359553e-06,
      "loss": 5.4899,
      "step": 530
    },
    {
      "epoch": 0.06848,
      "grad_norm": 3.4293930530548096,
      "learning_rate": 9.9761885890817e-06,
      "loss": 5.3569,
      "step": 535
    },
    {
      "epoch": 0.06912,
      "grad_norm": 3.5388574600219727,
      "learning_rate": 9.975694477639885e-06,
      "loss": 5.2739,
      "step": 540
    },
    {
      "epoch": 0.06976,
      "grad_norm": 3.735548973083496,
      "learning_rate": 9.97519530453676e-06,
      "loss": 5.4253,
      "step": 545
    },
    {
      "epoch": 0.0704,
      "grad_norm": 3.33503794670105,
      "learning_rate": 9.974691070280121e-06,
      "loss": 5.1569,
      "step": 550
    },
    {
      "epoch": 0.07104,
      "grad_norm": 3.5171401500701904,
      "learning_rate": 9.974181775382915e-06,
      "loss": 5.3242,
      "step": 555
    },
    {
      "epoch": 0.07168,
      "grad_norm": 3.565356969833374,
      "learning_rate": 9.973667420363233e-06,
      "loss": 5.3893,
      "step": 560
    },
    {
      "epoch": 0.07232,
      "grad_norm": 3.172163248062134,
      "learning_rate": 9.973148005744319e-06,
      "loss": 5.3824,
      "step": 565
    },
    {
      "epoch": 0.07296,
      "grad_norm": 3.517838716506958,
      "learning_rate": 9.972623532054564e-06,
      "loss": 5.2673,
      "step": 570
    },
    {
      "epoch": 0.0736,
      "grad_norm": 3.328416585922241,
      "learning_rate": 9.9720939998275e-06,
      "loss": 5.2649,
      "step": 575
    },
    {
      "epoch": 0.07424,
      "grad_norm": 3.475539445877075,
      "learning_rate": 9.971559409601807e-06,
      "loss": 5.3318,
      "step": 580
    },
    {
      "epoch": 0.07488,
      "grad_norm": 3.492013692855835,
      "learning_rate": 9.971019761921317e-06,
      "loss": 5.2735,
      "step": 585
    },
    {
      "epoch": 0.07552,
      "grad_norm": 3.474803924560547,
      "learning_rate": 9.970475057334997e-06,
      "loss": 5.3722,
      "step": 590
    },
    {
      "epoch": 0.07616,
      "grad_norm": 3.4162726402282715,
      "learning_rate": 9.96992529639696e-06,
      "loss": 5.3901,
      "step": 595
    },
    {
      "epoch": 0.0768,
      "grad_norm": 3.3643155097961426,
      "learning_rate": 9.969370479666473e-06,
      "loss": 5.2384,
      "step": 600
    },
    {
      "epoch": 0.0768,
      "eval_loss": 1.3373793363571167,
      "eval_runtime": 6.5847,
      "eval_samples_per_second": 151.867,
      "eval_steps_per_second": 18.983,
      "step": 600
    },
    {
      "epoch": 0.07744,
      "grad_norm": 3.44301176071167,
      "learning_rate": 9.968810607707933e-06,
      "loss": 5.2322,
      "step": 605
    },
    {
      "epoch": 0.07808,
      "grad_norm": 3.422262668609619,
      "learning_rate": 9.968245681090887e-06,
      "loss": 5.1708,
      "step": 610
    },
    {
      "epoch": 0.07872,
      "grad_norm": 3.2879252433776855,
      "learning_rate": 9.96767570039002e-06,
      "loss": 5.2291,
      "step": 615
    },
    {
      "epoch": 0.07936,
      "grad_norm": 3.6026480197906494,
      "learning_rate": 9.967100666185163e-06,
      "loss": 5.4241,
      "step": 620
    },
    {
      "epoch": 0.08,
      "grad_norm": 3.3642101287841797,
      "learning_rate": 9.966520579061286e-06,
      "loss": 5.4473,
      "step": 625
    },
    {
      "epoch": 0.08064,
      "grad_norm": 3.5968470573425293,
      "learning_rate": 9.965935439608493e-06,
      "loss": 5.3982,
      "step": 630
    },
    {
      "epoch": 0.08128,
      "grad_norm": 3.352083206176758,
      "learning_rate": 9.96534524842204e-06,
      "loss": 5.3953,
      "step": 635
    },
    {
      "epoch": 0.08192,
      "grad_norm": 3.3571720123291016,
      "learning_rate": 9.964750006102311e-06,
      "loss": 5.3159,
      "step": 640
    },
    {
      "epoch": 0.08256,
      "grad_norm": 3.486246109008789,
      "learning_rate": 9.964149713254833e-06,
      "loss": 5.211,
      "step": 645
    },
    {
      "epoch": 0.0832,
      "grad_norm": 3.674906015396118,
      "learning_rate": 9.96354437049027e-06,
      "loss": 5.3374,
      "step": 650
    },
    {
      "epoch": 0.08384,
      "grad_norm": 3.590810537338257,
      "learning_rate": 9.962933978424426e-06,
      "loss": 5.2194,
      "step": 655
    },
    {
      "epoch": 0.08448,
      "grad_norm": 3.551786184310913,
      "learning_rate": 9.962318537678238e-06,
      "loss": 5.1187,
      "step": 660
    },
    {
      "epoch": 0.08512,
      "grad_norm": 3.5391581058502197,
      "learning_rate": 9.961698048877776e-06,
      "loss": 5.2001,
      "step": 665
    },
    {
      "epoch": 0.08576,
      "grad_norm": 3.6105592250823975,
      "learning_rate": 9.961072512654255e-06,
      "loss": 5.2758,
      "step": 670
    },
    {
      "epoch": 0.0864,
      "grad_norm": 3.7463858127593994,
      "learning_rate": 9.960441929644017e-06,
      "loss": 5.2137,
      "step": 675
    },
    {
      "epoch": 0.08704,
      "grad_norm": 3.9237470626831055,
      "learning_rate": 9.959806300488538e-06,
      "loss": 5.2047,
      "step": 680
    },
    {
      "epoch": 0.08768,
      "grad_norm": 3.392827272415161,
      "learning_rate": 9.95916562583443e-06,
      "loss": 5.3071,
      "step": 685
    },
    {
      "epoch": 0.08832,
      "grad_norm": 3.221484661102295,
      "learning_rate": 9.958519906333438e-06,
      "loss": 5.183,
      "step": 690
    },
    {
      "epoch": 0.08896,
      "grad_norm": 3.5143983364105225,
      "learning_rate": 9.957869142642437e-06,
      "loss": 5.3171,
      "step": 695
    },
    {
      "epoch": 0.0896,
      "grad_norm": 3.497072696685791,
      "learning_rate": 9.957213335423433e-06,
      "loss": 5.1784,
      "step": 700
    },
    {
      "epoch": 0.0896,
      "eval_loss": 1.2988511323928833,
      "eval_runtime": 6.9763,
      "eval_samples_per_second": 143.342,
      "eval_steps_per_second": 17.918,
      "step": 700
    },
    {
      "epoch": 0.09024,
      "grad_norm": 3.3822438716888428,
      "learning_rate": 9.956552485343566e-06,
      "loss": 5.1732,
      "step": 705
    },
    {
      "epoch": 0.09088,
      "grad_norm": 3.3949694633483887,
      "learning_rate": 9.955886593075101e-06,
      "loss": 5.2725,
      "step": 710
    },
    {
      "epoch": 0.09152,
      "grad_norm": 3.2577288150787354,
      "learning_rate": 9.955215659295438e-06,
      "loss": 5.2207,
      "step": 715
    },
    {
      "epoch": 0.09216,
      "grad_norm": 3.769519567489624,
      "learning_rate": 9.954539684687103e-06,
      "loss": 5.2152,
      "step": 720
    },
    {
      "epoch": 0.0928,
      "grad_norm": 3.3824892044067383,
      "learning_rate": 9.953858669937746e-06,
      "loss": 5.2085,
      "step": 725
    },
    {
      "epoch": 0.09344,
      "grad_norm": 3.771742105484009,
      "learning_rate": 9.953172615740152e-06,
      "loss": 5.1575,
      "step": 730
    },
    {
      "epoch": 0.09408,
      "grad_norm": 3.7706689834594727,
      "learning_rate": 9.952481522792226e-06,
      "loss": 4.9608,
      "step": 735
    },
    {
      "epoch": 0.09472,
      "grad_norm": 3.8110334873199463,
      "learning_rate": 9.951785391797001e-06,
      "loss": 5.21,
      "step": 740
    },
    {
      "epoch": 0.09536,
      "grad_norm": 3.3012993335723877,
      "learning_rate": 9.951084223462636e-06,
      "loss": 5.2475,
      "step": 745
    },
    {
      "epoch": 0.096,
      "grad_norm": 3.6353518962860107,
      "learning_rate": 9.950378018502415e-06,
      "loss": 5.0985,
      "step": 750
    },
    {
      "epoch": 0.09664,
      "grad_norm": 3.369378089904785,
      "learning_rate": 9.949666777634743e-06,
      "loss": 5.1986,
      "step": 755
    },
    {
      "epoch": 0.09728,
      "grad_norm": 3.2247676849365234,
      "learning_rate": 9.948950501583147e-06,
      "loss": 5.3192,
      "step": 760
    },
    {
      "epoch": 0.09792,
      "grad_norm": 3.6966888904571533,
      "learning_rate": 9.948229191076284e-06,
      "loss": 5.1654,
      "step": 765
    },
    {
      "epoch": 0.09856,
      "grad_norm": 3.5823962688446045,
      "learning_rate": 9.947502846847921e-06,
      "loss": 5.1351,
      "step": 770
    },
    {
      "epoch": 0.0992,
      "grad_norm": 3.5258729457855225,
      "learning_rate": 9.946771469636955e-06,
      "loss": 5.1745,
      "step": 775
    },
    {
      "epoch": 0.09984,
      "grad_norm": 3.42067813873291,
      "learning_rate": 9.946035060187398e-06,
      "loss": 5.1569,
      "step": 780
    },
    {
      "epoch": 0.10048,
      "grad_norm": 3.9832825660705566,
      "learning_rate": 9.945293619248383e-06,
      "loss": 4.9796,
      "step": 785
    },
    {
      "epoch": 0.10112,
      "grad_norm": 3.742013692855835,
      "learning_rate": 9.944547147574162e-06,
      "loss": 5.1625,
      "step": 790
    },
    {
      "epoch": 0.10176,
      "grad_norm": 3.3150367736816406,
      "learning_rate": 9.943795645924104e-06,
      "loss": 5.099,
      "step": 795
    },
    {
      "epoch": 0.1024,
      "grad_norm": 3.359069585800171,
      "learning_rate": 9.943039115062691e-06,
      "loss": 5.1877,
      "step": 800
    },
    {
      "epoch": 0.1024,
      "eval_loss": 1.2946017980575562,
      "eval_runtime": 7.4306,
      "eval_samples_per_second": 134.579,
      "eval_steps_per_second": 16.822,
      "step": 800
    },
    {
      "epoch": 0.10304,
      "grad_norm": 3.703000545501709,
      "learning_rate": 9.94227755575953e-06,
      "loss": 5.1581,
      "step": 805
    },
    {
      "epoch": 0.10368,
      "grad_norm": 3.5370070934295654,
      "learning_rate": 9.941510968789334e-06,
      "loss": 5.2402,
      "step": 810
    },
    {
      "epoch": 0.10432,
      "grad_norm": 3.5010828971862793,
      "learning_rate": 9.940739354931936e-06,
      "loss": 5.1828,
      "step": 815
    },
    {
      "epoch": 0.10496,
      "grad_norm": 3.4637820720672607,
      "learning_rate": 9.93996271497228e-06,
      "loss": 5.1792,
      "step": 820
    },
    {
      "epoch": 0.1056,
      "grad_norm": 3.409712076187134,
      "learning_rate": 9.939181049700427e-06,
      "loss": 5.0721,
      "step": 825
    },
    {
      "epoch": 0.10624,
      "grad_norm": 3.589414596557617,
      "learning_rate": 9.938394359911545e-06,
      "loss": 5.234,
      "step": 830
    },
    {
      "epoch": 0.10688,
      "grad_norm": 3.444977045059204,
      "learning_rate": 9.937602646405918e-06,
      "loss": 4.9763,
      "step": 835
    },
    {
      "epoch": 0.10752,
      "grad_norm": 3.3560900688171387,
      "learning_rate": 9.936805909988935e-06,
      "loss": 5.2006,
      "step": 840
    },
    {
      "epoch": 0.10816,
      "grad_norm": 3.345703601837158,
      "learning_rate": 9.9360041514711e-06,
      "loss": 5.0287,
      "step": 845
    },
    {
      "epoch": 0.1088,
      "grad_norm": 3.492363691329956,
      "learning_rate": 9.935197371668024e-06,
      "loss": 5.0908,
      "step": 850
    },
    {
      "epoch": 0.10944,
      "grad_norm": 7.459951400756836,
      "learning_rate": 9.934385571400425e-06,
      "loss": 5.1735,
      "step": 855
    },
    {
      "epoch": 0.11008,
      "grad_norm": 3.5033841133117676,
      "learning_rate": 9.933568751494131e-06,
      "loss": 5.053,
      "step": 860
    },
    {
      "epoch": 0.11072,
      "grad_norm": 3.5542259216308594,
      "learning_rate": 9.93274691278007e-06,
      "loss": 5.1463,
      "step": 865
    },
    {
      "epoch": 0.11136,
      "grad_norm": 3.3819243907928467,
      "learning_rate": 9.931920056094285e-06,
      "loss": 5.0397,
      "step": 870
    },
    {
      "epoch": 0.112,
      "grad_norm": 3.406768798828125,
      "learning_rate": 9.931088182277915e-06,
      "loss": 5.179,
      "step": 875
    },
    {
      "epoch": 0.11264,
      "grad_norm": 5.960773944854736,
      "learning_rate": 9.930251292177206e-06,
      "loss": 5.217,
      "step": 880
    },
    {
      "epoch": 0.11328,
      "grad_norm": 3.5821049213409424,
      "learning_rate": 9.929409386643511e-06,
      "loss": 5.0374,
      "step": 885
    },
    {
      "epoch": 0.11392,
      "grad_norm": 3.3204903602600098,
      "learning_rate": 9.928562466533279e-06,
      "loss": 5.1856,
      "step": 890
    },
    {
      "epoch": 0.11456,
      "grad_norm": 4.022350788116455,
      "learning_rate": 9.927710532708064e-06,
      "loss": 5.1051,
      "step": 895
    },
    {
      "epoch": 0.1152,
      "grad_norm": 3.3810718059539795,
      "learning_rate": 9.926853586034515e-06,
      "loss": 5.1691,
      "step": 900
    },
    {
      "epoch": 0.1152,
      "eval_loss": 1.2660380601882935,
      "eval_runtime": 6.8853,
      "eval_samples_per_second": 145.238,
      "eval_steps_per_second": 18.155,
      "step": 900
    },
    {
      "epoch": 0.11584,
      "grad_norm": 3.5757713317871094,
      "learning_rate": 9.92599162738439e-06,
      "loss": 5.1505,
      "step": 905
    },
    {
      "epoch": 0.11648,
      "grad_norm": 3.38582706451416,
      "learning_rate": 9.925124657634537e-06,
      "loss": 5.0915,
      "step": 910
    },
    {
      "epoch": 0.11712,
      "grad_norm": 3.4189300537109375,
      "learning_rate": 9.924252677666905e-06,
      "loss": 5.1992,
      "step": 915
    },
    {
      "epoch": 0.11776,
      "grad_norm": 3.4118812084198,
      "learning_rate": 9.92337568836854e-06,
      "loss": 5.1334,
      "step": 920
    },
    {
      "epoch": 0.1184,
      "grad_norm": 3.5167789459228516,
      "learning_rate": 9.922493690631583e-06,
      "loss": 5.1003,
      "step": 925
    },
    {
      "epoch": 0.11904,
      "grad_norm": 3.546893358230591,
      "learning_rate": 9.921606685353268e-06,
      "loss": 5.1346,
      "step": 930
    },
    {
      "epoch": 0.11968,
      "grad_norm": 3.1576385498046875,
      "learning_rate": 9.920714673435931e-06,
      "loss": 4.9601,
      "step": 935
    },
    {
      "epoch": 0.12032,
      "grad_norm": 3.4227495193481445,
      "learning_rate": 9.91981765578699e-06,
      "loss": 5.0087,
      "step": 940
    },
    {
      "epoch": 0.12096,
      "grad_norm": 3.4890694618225098,
      "learning_rate": 9.918915633318964e-06,
      "loss": 5.1319,
      "step": 945
    },
    {
      "epoch": 0.1216,
      "grad_norm": 3.7377865314483643,
      "learning_rate": 9.918008606949459e-06,
      "loss": 5.0618,
      "step": 950
    },
    {
      "epoch": 0.12224,
      "grad_norm": 3.793402671813965,
      "learning_rate": 9.917096577601172e-06,
      "loss": 4.9998,
      "step": 955
    },
    {
      "epoch": 0.12288,
      "grad_norm": 3.404918909072876,
      "learning_rate": 9.916179546201889e-06,
      "loss": 5.0865,
      "step": 960
    },
    {
      "epoch": 0.12352,
      "grad_norm": 3.6076908111572266,
      "learning_rate": 9.915257513684488e-06,
      "loss": 5.0004,
      "step": 965
    },
    {
      "epoch": 0.12416,
      "grad_norm": 3.631777286529541,
      "learning_rate": 9.914330480986932e-06,
      "loss": 5.2806,
      "step": 970
    },
    {
      "epoch": 0.1248,
      "grad_norm": 3.323333501815796,
      "learning_rate": 9.913398449052266e-06,
      "loss": 5.07,
      "step": 975
    },
    {
      "epoch": 0.12544,
      "grad_norm": 3.6380035877227783,
      "learning_rate": 9.912461418828628e-06,
      "loss": 5.0559,
      "step": 980
    },
    {
      "epoch": 0.12608,
      "grad_norm": 3.7685458660125732,
      "learning_rate": 9.911519391269238e-06,
      "loss": 5.0497,
      "step": 985
    },
    {
      "epoch": 0.12672,
      "grad_norm": 3.4882941246032715,
      "learning_rate": 9.910572367332397e-06,
      "loss": 5.0388,
      "step": 990
    },
    {
      "epoch": 0.12736,
      "grad_norm": 3.27787184715271,
      "learning_rate": 9.909620347981493e-06,
      "loss": 5.0285,
      "step": 995
    },
    {
      "epoch": 0.128,
      "grad_norm": 3.388284921646118,
      "learning_rate": 9.908663334184994e-06,
      "loss": 5.1426,
      "step": 1000
    },
    {
      "epoch": 0.128,
      "eval_loss": 1.2478246688842773,
      "eval_runtime": 9.3123,
      "eval_samples_per_second": 107.384,
      "eval_steps_per_second": 13.423,
      "step": 1000
    },
    {
      "epoch": 0.12864,
      "grad_norm": 3.4602177143096924,
      "learning_rate": 9.907701326916448e-06,
      "loss": 4.8852,
      "step": 1005
    },
    {
      "epoch": 0.12928,
      "grad_norm": 3.7464816570281982,
      "learning_rate": 9.906734327154481e-06,
      "loss": 4.9129,
      "step": 1010
    },
    {
      "epoch": 0.12992,
      "grad_norm": 6.138649940490723,
      "learning_rate": 9.905762335882804e-06,
      "loss": 5.1037,
      "step": 1015
    },
    {
      "epoch": 0.13056,
      "grad_norm": 3.5933375358581543,
      "learning_rate": 9.904785354090198e-06,
      "loss": 4.9644,
      "step": 1020
    },
    {
      "epoch": 0.1312,
      "grad_norm": 3.6777257919311523,
      "learning_rate": 9.903803382770528e-06,
      "loss": 5.0575,
      "step": 1025
    },
    {
      "epoch": 0.13184,
      "grad_norm": 3.4429285526275635,
      "learning_rate": 9.902816422922727e-06,
      "loss": 4.8722,
      "step": 1030
    },
    {
      "epoch": 0.13248,
      "grad_norm": 3.7400121688842773,
      "learning_rate": 9.90182447555081e-06,
      "loss": 4.9521,
      "step": 1035
    },
    {
      "epoch": 0.13312,
      "grad_norm": 3.2183690071105957,
      "learning_rate": 9.900827541663862e-06,
      "loss": 5.0314,
      "step": 1040
    },
    {
      "epoch": 0.13376,
      "grad_norm": 3.563539505004883,
      "learning_rate": 9.899825622276041e-06,
      "loss": 4.9471,
      "step": 1045
    },
    {
      "epoch": 0.1344,
      "grad_norm": 3.3289413452148438,
      "learning_rate": 9.898818718406578e-06,
      "loss": 5.0223,
      "step": 1050
    },
    {
      "epoch": 0.13504,
      "grad_norm": 3.3363258838653564,
      "learning_rate": 9.89780683107977e-06,
      "loss": 4.8883,
      "step": 1055
    },
    {
      "epoch": 0.13568,
      "grad_norm": 3.5950427055358887,
      "learning_rate": 9.896789961324991e-06,
      "loss": 4.9488,
      "step": 1060
    },
    {
      "epoch": 0.13632,
      "grad_norm": 3.2444112300872803,
      "learning_rate": 9.895768110176677e-06,
      "loss": 4.9408,
      "step": 1065
    },
    {
      "epoch": 0.13696,
      "grad_norm": 3.2985880374908447,
      "learning_rate": 9.894741278674337e-06,
      "loss": 4.9875,
      "step": 1070
    },
    {
      "epoch": 0.1376,
      "grad_norm": 3.474818229675293,
      "learning_rate": 9.89370946786254e-06,
      "loss": 5.0526,
      "step": 1075
    },
    {
      "epoch": 0.13824,
      "grad_norm": 4.721025466918945,
      "learning_rate": 9.892672678790926e-06,
      "loss": 5.1362,
      "step": 1080
    },
    {
      "epoch": 0.13888,
      "grad_norm": 3.84086012840271,
      "learning_rate": 9.891630912514197e-06,
      "loss": 4.9631,
      "step": 1085
    },
    {
      "epoch": 0.13952,
      "grad_norm": 3.487732172012329,
      "learning_rate": 9.890584170092115e-06,
      "loss": 4.9211,
      "step": 1090
    },
    {
      "epoch": 0.14016,
      "grad_norm": 3.398810625076294,
      "learning_rate": 9.889532452589512e-06,
      "loss": 4.9814,
      "step": 1095
    },
    {
      "epoch": 0.1408,
      "grad_norm": 3.3263680934906006,
      "learning_rate": 9.888475761076273e-06,
      "loss": 4.9985,
      "step": 1100
    },
    {
      "epoch": 0.1408,
      "eval_loss": 1.2442607879638672,
      "eval_runtime": 6.5582,
      "eval_samples_per_second": 152.481,
      "eval_steps_per_second": 19.06,
      "step": 1100
    },
    {
      "epoch": 0.14144,
      "grad_norm": 3.4481613636016846,
      "learning_rate": 9.887414096627348e-06,
      "loss": 5.0169,
      "step": 1105
    },
    {
      "epoch": 0.14208,
      "grad_norm": 3.2736401557922363,
      "learning_rate": 9.886347460322744e-06,
      "loss": 5.0703,
      "step": 1110
    },
    {
      "epoch": 0.14272,
      "grad_norm": 3.2973997592926025,
      "learning_rate": 9.885275853247526e-06,
      "loss": 4.9957,
      "step": 1115
    },
    {
      "epoch": 0.14336,
      "grad_norm": 3.6516940593719482,
      "learning_rate": 9.884199276491817e-06,
      "loss": 5.0162,
      "step": 1120
    },
    {
      "epoch": 0.144,
      "grad_norm": 3.1835155487060547,
      "learning_rate": 9.883117731150792e-06,
      "loss": 4.9765,
      "step": 1125
    },
    {
      "epoch": 0.14464,
      "grad_norm": 3.21928334236145,
      "learning_rate": 9.882031218324681e-06,
      "loss": 5.0611,
      "step": 1130
    },
    {
      "epoch": 0.14528,
      "grad_norm": 4.601723670959473,
      "learning_rate": 9.880939739118772e-06,
      "loss": 5.0637,
      "step": 1135
    },
    {
      "epoch": 0.14592,
      "grad_norm": 3.2973368167877197,
      "learning_rate": 9.879843294643402e-06,
      "loss": 4.9621,
      "step": 1140
    },
    {
      "epoch": 0.14656,
      "grad_norm": 3.4781899452209473,
      "learning_rate": 9.878741886013959e-06,
      "loss": 4.9482,
      "step": 1145
    },
    {
      "epoch": 0.1472,
      "grad_norm": 3.5175704956054688,
      "learning_rate": 9.877635514350878e-06,
      "loss": 4.8594,
      "step": 1150
    },
    {
      "epoch": 0.14784,
      "grad_norm": 3.4302468299865723,
      "learning_rate": 9.87652418077965e-06,
      "loss": 4.8865,
      "step": 1155
    },
    {
      "epoch": 0.14848,
      "grad_norm": 3.464651346206665,
      "learning_rate": 9.875407886430806e-06,
      "loss": 4.9922,
      "step": 1160
    },
    {
      "epoch": 0.14912,
      "grad_norm": 4.064827919006348,
      "learning_rate": 9.87428663243993e-06,
      "loss": 4.9592,
      "step": 1165
    },
    {
      "epoch": 0.14976,
      "grad_norm": 3.654902458190918,
      "learning_rate": 9.873160419947645e-06,
      "loss": 4.9286,
      "step": 1170
    },
    {
      "epoch": 0.1504,
      "grad_norm": 3.395596981048584,
      "learning_rate": 9.872029250099626e-06,
      "loss": 5.0057,
      "step": 1175
    },
    {
      "epoch": 0.15104,
      "grad_norm": 3.745281457901001,
      "learning_rate": 9.870893124046582e-06,
      "loss": 4.8671,
      "step": 1180
    },
    {
      "epoch": 0.15168,
      "grad_norm": 3.449518918991089,
      "learning_rate": 9.869752042944271e-06,
      "loss": 4.8306,
      "step": 1185
    },
    {
      "epoch": 0.15232,
      "grad_norm": 3.1926662921905518,
      "learning_rate": 9.868606007953487e-06,
      "loss": 5.0347,
      "step": 1190
    },
    {
      "epoch": 0.15296,
      "grad_norm": 3.4620425701141357,
      "learning_rate": 9.86745502024007e-06,
      "loss": 4.857,
      "step": 1195
    },
    {
      "epoch": 0.1536,
      "grad_norm": 3.5597681999206543,
      "learning_rate": 9.866299080974886e-06,
      "loss": 4.9225,
      "step": 1200
    },
    {
      "epoch": 0.1536,
      "eval_loss": 1.2185124158859253,
      "eval_runtime": 7.9383,
      "eval_samples_per_second": 125.972,
      "eval_steps_per_second": 15.746,
      "step": 1200
    },
    {
      "epoch": 0.15424,
      "grad_norm": 3.5934455394744873,
      "learning_rate": 9.865138191333852e-06,
      "loss": 4.7654,
      "step": 1205
    },
    {
      "epoch": 0.15488,
      "grad_norm": 3.8588831424713135,
      "learning_rate": 9.863972352497912e-06,
      "loss": 4.9993,
      "step": 1210
    },
    {
      "epoch": 0.15552,
      "grad_norm": 3.58868408203125,
      "learning_rate": 9.86280156565305e-06,
      "loss": 4.8217,
      "step": 1215
    },
    {
      "epoch": 0.15616,
      "grad_norm": 3.5407521724700928,
      "learning_rate": 9.861625831990278e-06,
      "loss": 4.875,
      "step": 1220
    },
    {
      "epoch": 0.1568,
      "grad_norm": 3.4974656105041504,
      "learning_rate": 9.860445152705644e-06,
      "loss": 5.0627,
      "step": 1225
    },
    {
      "epoch": 0.15744,
      "grad_norm": 3.655677556991577,
      "learning_rate": 9.859259529000228e-06,
      "loss": 4.8015,
      "step": 1230
    },
    {
      "epoch": 0.15808,
      "grad_norm": 3.55148983001709,
      "learning_rate": 9.858068962080136e-06,
      "loss": 5.1209,
      "step": 1235
    },
    {
      "epoch": 0.15872,
      "grad_norm": 3.4331536293029785,
      "learning_rate": 9.856873453156506e-06,
      "loss": 4.9739,
      "step": 1240
    },
    {
      "epoch": 0.15936,
      "grad_norm": 3.374394655227661,
      "learning_rate": 9.855673003445502e-06,
      "loss": 4.8138,
      "step": 1245
    },
    {
      "epoch": 0.16,
      "grad_norm": 3.5296385288238525,
      "learning_rate": 9.854467614168315e-06,
      "loss": 5.0274,
      "step": 1250
    },
    {
      "epoch": 0.16064,
      "grad_norm": 3.6533989906311035,
      "learning_rate": 9.85325728655116e-06,
      "loss": 4.9979,
      "step": 1255
    },
    {
      "epoch": 0.16128,
      "grad_norm": 3.3504199981689453,
      "learning_rate": 9.852042021825272e-06,
      "loss": 4.8317,
      "step": 1260
    },
    {
      "epoch": 0.16192,
      "grad_norm": 3.614529609680176,
      "learning_rate": 9.850821821226918e-06,
      "loss": 4.9413,
      "step": 1265
    },
    {
      "epoch": 0.16256,
      "grad_norm": 3.4821839332580566,
      "learning_rate": 9.849596685997376e-06,
      "loss": 4.904,
      "step": 1270
    },
    {
      "epoch": 0.1632,
      "grad_norm": 3.3400087356567383,
      "learning_rate": 9.848366617382951e-06,
      "loss": 4.9039,
      "step": 1275
    },
    {
      "epoch": 0.16384,
      "grad_norm": 4.062397003173828,
      "learning_rate": 9.847131616634963e-06,
      "loss": 4.7378,
      "step": 1280
    },
    {
      "epoch": 0.16448,
      "grad_norm": 3.689796209335327,
      "learning_rate": 9.845891685009751e-06,
      "loss": 4.8799,
      "step": 1285
    },
    {
      "epoch": 0.16512,
      "grad_norm": 3.509657621383667,
      "learning_rate": 9.84464682376867e-06,
      "loss": 4.8513,
      "step": 1290
    },
    {
      "epoch": 0.16576,
      "grad_norm": 3.4828646183013916,
      "learning_rate": 9.843397034178088e-06,
      "loss": 5.0151,
      "step": 1295
    },
    {
      "epoch": 0.1664,
      "grad_norm": 3.394510507583618,
      "learning_rate": 9.842142317509387e-06,
      "loss": 4.7585,
      "step": 1300
    },
    {
      "epoch": 0.1664,
      "eval_loss": 1.2179418802261353,
      "eval_runtime": 6.7952,
      "eval_samples_per_second": 147.163,
      "eval_steps_per_second": 18.395,
      "step": 1300
    },
    {
      "epoch": 0.16704,
      "grad_norm": 3.4089293479919434,
      "learning_rate": 9.840882675038962e-06,
      "loss": 4.7646,
      "step": 1305
    },
    {
      "epoch": 0.16768,
      "grad_norm": 3.1607353687286377,
      "learning_rate": 9.83961810804822e-06,
      "loss": 4.9528,
      "step": 1310
    },
    {
      "epoch": 0.16832,
      "grad_norm": 3.30869197845459,
      "learning_rate": 9.838348617823573e-06,
      "loss": 5.0086,
      "step": 1315
    },
    {
      "epoch": 0.16896,
      "grad_norm": 3.6550564765930176,
      "learning_rate": 9.837074205656452e-06,
      "loss": 4.8675,
      "step": 1320
    },
    {
      "epoch": 0.1696,
      "grad_norm": 3.6141419410705566,
      "learning_rate": 9.835794872843281e-06,
      "loss": 4.8885,
      "step": 1325
    },
    {
      "epoch": 0.17024,
      "grad_norm": 3.4006361961364746,
      "learning_rate": 9.834510620685497e-06,
      "loss": 4.7784,
      "step": 1330
    },
    {
      "epoch": 0.17088,
      "grad_norm": 3.4397149085998535,
      "learning_rate": 9.833221450489543e-06,
      "loss": 4.929,
      "step": 1335
    },
    {
      "epoch": 0.17152,
      "grad_norm": 3.613502025604248,
      "learning_rate": 9.83192736356686e-06,
      "loss": 4.8763,
      "step": 1340
    },
    {
      "epoch": 0.17216,
      "grad_norm": 3.613837957382202,
      "learning_rate": 9.830628361233896e-06,
      "loss": 4.8765,
      "step": 1345
    },
    {
      "epoch": 0.1728,
      "grad_norm": 3.775621175765991,
      "learning_rate": 9.829324444812096e-06,
      "loss": 4.8103,
      "step": 1350
    },
    {
      "epoch": 0.17344,
      "grad_norm": 3.6856908798217773,
      "learning_rate": 9.828015615627904e-06,
      "loss": 4.8867,
      "step": 1355
    },
    {
      "epoch": 0.17408,
      "grad_norm": 3.3510427474975586,
      "learning_rate": 9.826701875012763e-06,
      "loss": 4.7708,
      "step": 1360
    },
    {
      "epoch": 0.17472,
      "grad_norm": 3.342366933822632,
      "learning_rate": 9.82538322430311e-06,
      "loss": 4.8404,
      "step": 1365
    },
    {
      "epoch": 0.17536,
      "grad_norm": 3.5898385047912598,
      "learning_rate": 9.824059664840378e-06,
      "loss": 4.8205,
      "step": 1370
    },
    {
      "epoch": 0.176,
      "grad_norm": 3.1588313579559326,
      "learning_rate": 9.822731197970998e-06,
      "loss": 4.7214,
      "step": 1375
    },
    {
      "epoch": 0.17664,
      "grad_norm": 3.431478261947632,
      "learning_rate": 9.821397825046387e-06,
      "loss": 4.8892,
      "step": 1380
    },
    {
      "epoch": 0.17728,
      "grad_norm": 3.7104616165161133,
      "learning_rate": 9.820059547422952e-06,
      "loss": 4.8027,
      "step": 1385
    },
    {
      "epoch": 0.17792,
      "grad_norm": 3.189239263534546,
      "learning_rate": 9.818716366462098e-06,
      "loss": 4.8692,
      "step": 1390
    },
    {
      "epoch": 0.17856,
      "grad_norm": 3.3543105125427246,
      "learning_rate": 9.81736828353021e-06,
      "loss": 4.9076,
      "step": 1395
    },
    {
      "epoch": 0.1792,
      "grad_norm": 3.2962117195129395,
      "learning_rate": 9.816015299998663e-06,
      "loss": 4.93,
      "step": 1400
    },
    {
      "epoch": 0.1792,
      "eval_loss": 1.2212570905685425,
      "eval_runtime": 7.0462,
      "eval_samples_per_second": 141.92,
      "eval_steps_per_second": 17.74,
      "step": 1400
    },
    {
      "epoch": 0.17984,
      "grad_norm": 3.2857654094696045,
      "learning_rate": 9.814657417243814e-06,
      "loss": 4.7544,
      "step": 1405
    },
    {
      "epoch": 0.18048,
      "grad_norm": 3.31211256980896,
      "learning_rate": 9.813294636647009e-06,
      "loss": 4.9007,
      "step": 1410
    },
    {
      "epoch": 0.18112,
      "grad_norm": 3.3026342391967773,
      "learning_rate": 9.81192695959457e-06,
      "loss": 4.8136,
      "step": 1415
    },
    {
      "epoch": 0.18176,
      "grad_norm": 3.6015031337738037,
      "learning_rate": 9.810554387477812e-06,
      "loss": 4.8296,
      "step": 1420
    },
    {
      "epoch": 0.1824,
      "grad_norm": 3.5558950901031494,
      "learning_rate": 9.809176921693013e-06,
      "loss": 4.9049,
      "step": 1425
    },
    {
      "epoch": 0.18304,
      "grad_norm": 3.272860288619995,
      "learning_rate": 9.807794563641442e-06,
      "loss": 4.868,
      "step": 1430
    },
    {
      "epoch": 0.18368,
      "grad_norm": 3.427809715270996,
      "learning_rate": 9.806407314729341e-06,
      "loss": 4.7899,
      "step": 1435
    },
    {
      "epoch": 0.18432,
      "grad_norm": 3.545553207397461,
      "learning_rate": 9.805015176367924e-06,
      "loss": 4.9774,
      "step": 1440
    },
    {
      "epoch": 0.18496,
      "grad_norm": 3.5434036254882812,
      "learning_rate": 9.803618149973383e-06,
      "loss": 4.8174,
      "step": 1445
    },
    {
      "epoch": 0.1856,
      "grad_norm": 3.5401341915130615,
      "learning_rate": 9.802216236966882e-06,
      "loss": 4.8138,
      "step": 1450
    },
    {
      "epoch": 0.18624,
      "grad_norm": 3.339459180831909,
      "learning_rate": 9.800809438774557e-06,
      "loss": 4.9385,
      "step": 1455
    },
    {
      "epoch": 0.18688,
      "grad_norm": 3.541703224182129,
      "learning_rate": 9.799397756827508e-06,
      "loss": 4.8764,
      "step": 1460
    },
    {
      "epoch": 0.18752,
      "grad_norm": 3.3053269386291504,
      "learning_rate": 9.79798119256181e-06,
      "loss": 4.5765,
      "step": 1465
    },
    {
      "epoch": 0.18816,
      "grad_norm": 3.461660146713257,
      "learning_rate": 9.7965597474185e-06,
      "loss": 4.6125,
      "step": 1470
    },
    {
      "epoch": 0.1888,
      "grad_norm": 3.564030885696411,
      "learning_rate": 9.795133422843583e-06,
      "loss": 4.8758,
      "step": 1475
    },
    {
      "epoch": 0.18944,
      "grad_norm": 3.635293483734131,
      "learning_rate": 9.793702220288028e-06,
      "loss": 4.7954,
      "step": 1480
    },
    {
      "epoch": 0.19008,
      "grad_norm": 3.4663326740264893,
      "learning_rate": 9.792266141207763e-06,
      "loss": 4.8442,
      "step": 1485
    },
    {
      "epoch": 0.19072,
      "grad_norm": 3.556608200073242,
      "learning_rate": 9.790825187063677e-06,
      "loss": 4.8431,
      "step": 1490
    },
    {
      "epoch": 0.19136,
      "grad_norm": 3.726987838745117,
      "learning_rate": 9.789379359321624e-06,
      "loss": 4.8309,
      "step": 1495
    },
    {
      "epoch": 0.192,
      "grad_norm": 3.535627603530884,
      "learning_rate": 9.78792865945241e-06,
      "loss": 4.8779,
      "step": 1500
    },
    {
      "epoch": 0.192,
      "eval_loss": 1.2059489488601685,
      "eval_runtime": 6.8452,
      "eval_samples_per_second": 146.087,
      "eval_steps_per_second": 18.261,
      "step": 1500
    },
    {
      "epoch": 0.19264,
      "grad_norm": 3.3409645557403564,
      "learning_rate": 9.7864730889318e-06,
      "loss": 4.8398,
      "step": 1505
    },
    {
      "epoch": 0.19328,
      "grad_norm": 3.240247964859009,
      "learning_rate": 9.78501264924051e-06,
      "loss": 4.689,
      "step": 1510
    },
    {
      "epoch": 0.19392,
      "grad_norm": 3.6355326175689697,
      "learning_rate": 9.783547341864216e-06,
      "loss": 4.7737,
      "step": 1515
    },
    {
      "epoch": 0.19456,
      "grad_norm": 3.4650771617889404,
      "learning_rate": 9.78207716829354e-06,
      "loss": 4.7844,
      "step": 1520
    },
    {
      "epoch": 0.1952,
      "grad_norm": 3.281463146209717,
      "learning_rate": 9.780602130024055e-06,
      "loss": 4.6872,
      "step": 1525
    },
    {
      "epoch": 0.19584,
      "grad_norm": 3.264622926712036,
      "learning_rate": 9.779122228556289e-06,
      "loss": 4.7438,
      "step": 1530
    },
    {
      "epoch": 0.19648,
      "grad_norm": 3.598848342895508,
      "learning_rate": 9.777637465395706e-06,
      "loss": 4.6983,
      "step": 1535
    },
    {
      "epoch": 0.19712,
      "grad_norm": 3.3951942920684814,
      "learning_rate": 9.776147842052725e-06,
      "loss": 4.8429,
      "step": 1540
    },
    {
      "epoch": 0.19776,
      "grad_norm": 3.088014841079712,
      "learning_rate": 9.774653360042706e-06,
      "loss": 4.8207,
      "step": 1545
    },
    {
      "epoch": 0.1984,
      "grad_norm": 3.4452457427978516,
      "learning_rate": 9.773154020885953e-06,
      "loss": 4.8426,
      "step": 1550
    },
    {
      "epoch": 0.19904,
      "grad_norm": 3.3782291412353516,
      "learning_rate": 9.771649826107707e-06,
      "loss": 4.9081,
      "step": 1555
    },
    {
      "epoch": 0.19968,
      "grad_norm": 3.420620918273926,
      "learning_rate": 9.770140777238153e-06,
      "loss": 4.8296,
      "step": 1560
    },
    {
      "epoch": 0.20032,
      "grad_norm": 3.3439509868621826,
      "learning_rate": 9.76862687581241e-06,
      "loss": 4.626,
      "step": 1565
    },
    {
      "epoch": 0.20096,
      "grad_norm": 3.2657105922698975,
      "learning_rate": 9.76710812337054e-06,
      "loss": 4.8035,
      "step": 1570
    },
    {
      "epoch": 0.2016,
      "grad_norm": 3.4240477085113525,
      "learning_rate": 9.765584521457533e-06,
      "loss": 4.776,
      "step": 1575
    },
    {
      "epoch": 0.20224,
      "grad_norm": 3.7116453647613525,
      "learning_rate": 9.764056071623314e-06,
      "loss": 4.8099,
      "step": 1580
    },
    {
      "epoch": 0.20288,
      "grad_norm": 3.3470919132232666,
      "learning_rate": 9.762522775422741e-06,
      "loss": 4.6686,
      "step": 1585
    },
    {
      "epoch": 0.20352,
      "grad_norm": 3.552156925201416,
      "learning_rate": 9.760984634415602e-06,
      "loss": 4.7256,
      "step": 1590
    },
    {
      "epoch": 0.20416,
      "grad_norm": 3.144547939300537,
      "learning_rate": 9.759441650166612e-06,
      "loss": 4.6914,
      "step": 1595
    },
    {
      "epoch": 0.2048,
      "grad_norm": 3.3078038692474365,
      "learning_rate": 9.757893824245414e-06,
      "loss": 4.7828,
      "step": 1600
    },
    {
      "epoch": 0.2048,
      "eval_loss": 1.1752163171768188,
      "eval_runtime": 6.6642,
      "eval_samples_per_second": 150.056,
      "eval_steps_per_second": 18.757,
      "step": 1600
    }
  ],
  "logging_steps": 5,
  "max_steps": 15624,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.585434243497984e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}