{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 28.571428571428573,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 10.9375,
      "learning_rate": 1.8e-05,
      "loss": 2.708388328552246,
      "loss_d0": 4.110811305046082,
      "step": 10
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 6.375,
      "learning_rate": 3.8e-05,
      "loss": 1.2063720703125,
      "loss_d0": 1.8877601504325867,
      "step": 20
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 4.375,
      "learning_rate": 5.8e-05,
      "loss": 0.8983614921569825,
      "loss_d0": 1.4538785099983216,
      "step": 30
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 4.5,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.8970762252807617,
      "loss_d0": 1.425518774986267,
      "step": 40
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 4.4375,
      "learning_rate": 9.8e-05,
      "loss": 0.7297840118408203,
      "loss_d0": 1.340456622838974,
      "step": 50
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 3.390625,
      "learning_rate": 0.000118,
      "loss": 0.6489387512207031,
      "loss_d0": 1.0164853096008302,
      "step": 60
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.84375,
      "learning_rate": 0.000138,
      "loss": 0.6557669162750244,
      "loss_d0": 0.9850718915462494,
      "step": 70
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 3.640625,
      "learning_rate": 0.00015800000000000002,
      "loss": 0.5417943000793457,
      "loss_d0": 0.7510321617126465,
      "step": 80
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 3.171875,
      "learning_rate": 0.00017800000000000002,
      "loss": 0.5557333469390869,
      "loss_d0": 0.726891279220581,
      "step": 90
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 3.1875,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.5609831809997559,
      "loss_d0": 0.7925398766994476,
      "step": 100
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 4.625,
      "learning_rate": 0.0002,
      "loss": 0.49335689544677735,
      "loss_d0": 0.7104986011981964,
      "step": 110
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 2.921875,
      "learning_rate": 0.0002,
      "loss": 0.468343448638916,
      "loss_d0": 0.6982269883155823,
      "step": 120
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 2.46875,
      "learning_rate": 0.0002,
      "loss": 0.45113506317138674,
      "loss_d0": 0.6743539869785309,
      "step": 130
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.609375,
      "learning_rate": 0.0002,
      "loss": 0.4742081642150879,
      "loss_d0": 0.7392071902751922,
      "step": 140
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 2.640625,
      "learning_rate": 0.0002,
      "loss": 0.4260121822357178,
      "loss_d0": 0.6589436203241348,
      "step": 150
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 1.953125,
      "learning_rate": 0.0002,
      "loss": 0.39026267528533937,
      "loss_d0": 0.6157096266746521,
      "step": 160
    },
    {
      "epoch": 2.4285714285714284,
      "grad_norm": 2.09375,
      "learning_rate": 0.0002,
      "loss": 0.40253314971923826,
      "loss_d0": 0.6893241822719574,
      "step": 170
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 2.203125,
      "learning_rate": 0.0002,
      "loss": 0.3850980758666992,
      "loss_d0": 0.623932832479477,
      "step": 180
    },
    {
      "epoch": 2.7142857142857144,
      "grad_norm": 1.875,
      "learning_rate": 0.0002,
      "loss": 0.3460643768310547,
      "loss_d0": 0.6405902057886124,
      "step": 190
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002,
      "loss": 0.3911351203918457,
      "loss_d0": 0.6420928090810776,
      "step": 200
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002,
      "loss": 0.3302267074584961,
      "loss_d0": 0.5826890021562576,
      "step": 210
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 2.421875,
      "learning_rate": 0.0002,
      "loss": 0.3666944980621338,
      "loss_d0": 0.6297772079706192,
      "step": 220
    },
    {
      "epoch": 3.2857142857142856,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002,
      "loss": 0.3350435495376587,
      "loss_d0": 0.5840106904506683,
      "step": 230
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002,
      "loss": 0.31048002243041994,
      "loss_d0": 0.5492438584566116,
      "step": 240
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 2.234375,
      "learning_rate": 0.0002,
      "loss": 0.3417187690734863,
      "loss_d0": 0.6671169877052308,
      "step": 250
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0002,
      "loss": 0.31016125679016116,
      "loss_d0": 0.5638339132070541,
      "step": 260
    },
    {
      "epoch": 3.857142857142857,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0002,
      "loss": 0.3462181806564331,
      "loss_d0": 0.6273518323898315,
      "step": 270
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0002,
      "loss": 0.37842116355895994,
      "loss_d0": 0.6331767499446869,
      "step": 280
    },
    {
      "epoch": 4.142857142857143,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002,
      "loss": 0.2758188247680664,
      "loss_d0": 0.5233808279037475,
      "step": 290
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002,
      "loss": 0.2679539680480957,
      "loss_d0": 0.512147718667984,
      "step": 300
    },
    {
      "epoch": 4.428571428571429,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002,
      "loss": 0.2771256446838379,
      "loss_d0": 0.5529826283454895,
      "step": 310
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002,
      "loss": 0.2626373767852783,
      "loss_d0": 0.4893135607242584,
      "step": 320
    },
    {
      "epoch": 4.714285714285714,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002,
      "loss": 0.29569981098175047,
      "loss_d0": 0.5720477998256683,
      "step": 330
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002,
      "loss": 0.2952061653137207,
      "loss_d0": 0.5526792675256729,
      "step": 340
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.78125,
      "learning_rate": 0.0002,
      "loss": 0.3209908723831177,
      "loss_d0": 0.6238142669200897,
      "step": 350
    },
    {
      "epoch": 5.142857142857143,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002,
      "loss": 0.2546632528305054,
      "loss_d0": 0.497738191485405,
      "step": 360
    },
    {
      "epoch": 5.285714285714286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002,
      "loss": 0.2360783815383911,
      "loss_d0": 0.43793114721775056,
      "step": 370
    },
    {
      "epoch": 5.428571428571429,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002,
      "loss": 0.23914880752563478,
      "loss_d0": 0.46601290702819825,
      "step": 380
    },
    {
      "epoch": 5.571428571428571,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002,
      "loss": 0.24625580310821532,
      "loss_d0": 0.5180164843797683,
      "step": 390
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002,
      "loss": 0.30511114597320554,
      "loss_d0": 0.570301228761673,
      "step": 400
    },
    {
      "epoch": 5.857142857142857,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002,
      "loss": 0.26278665065765383,
      "loss_d0": 0.5415922611951828,
      "step": 410
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002,
      "loss": 0.2830559015274048,
      "loss_d0": 0.5763564020395279,
      "step": 420
    },
    {
      "epoch": 6.142857142857143,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002,
      "loss": 0.23301992416381836,
      "loss_d0": 0.47966580390930175,
      "step": 430
    },
    {
      "epoch": 6.285714285714286,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002,
      "loss": 0.2130380630493164,
      "loss_d0": 0.4446234554052353,
      "step": 440
    },
    {
      "epoch": 6.428571428571429,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002,
      "loss": 0.2604910612106323,
      "loss_d0": 0.5528126150369644,
      "step": 450
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002,
      "loss": 0.2402413845062256,
      "loss_d0": 0.4986030846834183,
      "step": 460
    },
    {
      "epoch": 6.714285714285714,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0002,
      "loss": 0.24624488353729249,
      "loss_d0": 0.5280796885490417,
      "step": 470
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002,
      "loss": 0.2268777847290039,
      "loss_d0": 0.4914736434817314,
      "step": 480
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002,
      "loss": 0.25430734157562257,
      "loss_d0": 0.5139662355184555,
      "step": 490
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002,
      "loss": 0.19951030015945434,
      "loss_d0": 0.45199823677539824,
      "step": 500
    },
    {
      "epoch": 7.142857142857143,
      "eval_loss": 8.546619415283203,
      "eval_runtime": 0.3482,
      "eval_samples_per_second": 718.039,
      "eval_steps_per_second": 71.804,
      "step": 500
    },
    {
      "epoch": 7.285714285714286,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0002,
      "loss": 0.19481804370880126,
      "loss_d0": 0.45815878808498384,
      "step": 510
    },
    {
      "epoch": 7.428571428571429,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002,
      "loss": 0.2255556106567383,
      "loss_d0": 0.5097730100154877,
      "step": 520
    },
    {
      "epoch": 7.571428571428571,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002,
      "loss": 0.22788920402526855,
      "loss_d0": 0.44859754145145414,
      "step": 530
    },
    {
      "epoch": 7.714285714285714,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002,
      "loss": 0.2295910358428955,
      "loss_d0": 0.469280743598938,
      "step": 540
    },
    {
      "epoch": 7.857142857142857,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002,
      "loss": 0.22432618141174315,
      "loss_d0": 0.4931183695793152,
      "step": 550
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.34375,
      "learning_rate": 0.0002,
      "loss": 0.22481327056884765,
      "loss_d0": 0.4790141642093658,
      "step": 560
    },
    {
      "epoch": 8.142857142857142,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002,
      "loss": 0.1898207187652588,
      "loss_d0": 0.45880764722824097,
      "step": 570
    },
    {
      "epoch": 8.285714285714286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002,
      "loss": 0.16722708940505981,
      "loss_d0": 0.39154875874519346,
      "step": 580
    },
    {
      "epoch": 8.428571428571429,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002,
      "loss": 0.20029706954956056,
      "loss_d0": 0.4694953829050064,
      "step": 590
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002,
      "loss": 0.1980130195617676,
      "loss_d0": 0.429129758477211,
      "step": 600
    },
    {
      "epoch": 8.714285714285714,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002,
      "loss": 0.21289968490600586,
      "loss_d0": 0.5128728121519088,
      "step": 610
    },
    {
      "epoch": 8.857142857142858,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002,
      "loss": 0.20417430400848388,
      "loss_d0": 0.4555791884660721,
      "step": 620
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002,
      "loss": 0.2080092191696167,
      "loss_d0": 0.48112956881523133,
      "step": 630
    },
    {
      "epoch": 9.142857142857142,
      "grad_norm": 1.125,
      "learning_rate": 0.0002,
      "loss": 0.17301057577133178,
      "loss_d0": 0.3914658397436142,
      "step": 640
    },
    {
      "epoch": 9.285714285714286,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002,
      "loss": 0.16326009035110473,
      "loss_d0": 0.3957242280244827,
      "step": 650
    },
    {
      "epoch": 9.428571428571429,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 0.23928072452545165,
      "loss_d0": 0.47887236177921294,
      "step": 660
    },
    {
      "epoch": 9.571428571428571,
      "grad_norm": 1.125,
      "learning_rate": 0.0002,
      "loss": 0.21262478828430176,
      "loss_d0": 0.49389356970787046,
      "step": 670
    },
    {
      "epoch": 9.714285714285714,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002,
      "loss": 0.21038594245910644,
      "loss_d0": 0.4909708023071289,
      "step": 680
    },
    {
      "epoch": 9.857142857142858,
      "grad_norm": 0.84375,
      "learning_rate": 0.0002,
      "loss": 0.19068275690078734,
      "loss_d0": 0.43951269090175626,
      "step": 690
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002,
      "loss": 0.20882303714752198,
      "loss_d0": 0.4652249127626419,
      "step": 700
    },
    {
      "epoch": 10.142857142857142,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0002,
      "loss": 0.16138590574264527,
      "loss_d0": 0.40691248178482053,
      "step": 710
    },
    {
      "epoch": 10.285714285714286,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002,
      "loss": 0.1708214044570923,
      "loss_d0": 0.44721868336200715,
      "step": 720
    },
    {
      "epoch": 10.428571428571429,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002,
      "loss": 0.16305239200592042,
      "loss_d0": 0.411194321513176,
      "step": 730
    },
    {
      "epoch": 10.571428571428571,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002,
      "loss": 0.18613686561584472,
      "loss_d0": 0.46029442846775054,
      "step": 740
    },
    {
      "epoch": 10.714285714285714,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002,
      "loss": 0.17803105115890502,
      "loss_d0": 0.4383477747440338,
      "step": 750
    },
    {
      "epoch": 10.857142857142858,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002,
      "loss": 0.2087890625,
      "loss_d0": 0.4998590737581253,
      "step": 760
    },
    {
      "epoch": 11.0,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0002,
      "loss": 0.19215229749679566,
      "loss_d0": 0.4216267496347427,
      "step": 770
    },
    {
      "epoch": 11.142857142857142,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 0.161735200881958,
      "loss_d0": 0.40338513553142546,
      "step": 780
    },
    {
      "epoch": 11.285714285714286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002,
      "loss": 0.16771552562713624,
      "loss_d0": 0.42275542616844175,
      "step": 790
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002,
      "loss": 0.15335218906402587,
      "loss_d0": 0.3944630563259125,
      "step": 800
    },
    {
      "epoch": 11.571428571428571,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002,
      "loss": 0.16975181102752684,
      "loss_d0": 0.42211559414863586,
      "step": 810
    },
    {
      "epoch": 11.714285714285714,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0002,
      "loss": 0.18913717269897462,
      "loss_d0": 0.43886401802301406,
      "step": 820
    },
    {
      "epoch": 11.857142857142858,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002,
      "loss": 0.22327446937561035,
      "loss_d0": 0.5353886485099792,
      "step": 830
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002,
      "loss": 0.15874338150024414,
      "loss_d0": 0.38432416021823884,
      "step": 840
    },
    {
      "epoch": 12.142857142857142,
      "grad_norm": 0.8125,
      "learning_rate": 0.0002,
      "loss": 0.1377707004547119,
      "loss_d0": 0.3911532461643219,
      "step": 850
    },
    {
      "epoch": 12.285714285714286,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 0.16991720199584961,
      "loss_d0": 0.4460686922073364,
      "step": 860
    },
    {
      "epoch": 12.428571428571429,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002,
      "loss": 0.16523797512054444,
      "loss_d0": 0.43365546464920046,
      "step": 870
    },
    {
      "epoch": 12.571428571428571,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002,
      "loss": 0.15238118171691895,
      "loss_d0": 0.42346865832805636,
      "step": 880
    },
    {
      "epoch": 12.714285714285714,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0002,
      "loss": 0.14495233297348023,
      "loss_d0": 0.35719306766986847,
      "step": 890
    },
    {
      "epoch": 12.857142857142858,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002,
      "loss": 0.17073333263397217,
      "loss_d0": 0.43814408481121064,
      "step": 900
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002,
      "loss": 0.1850340723991394,
      "loss_d0": 0.482748007774353,
      "step": 910
    },
    {
      "epoch": 13.142857142857142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002,
      "loss": 0.13537302017211914,
      "loss_d0": 0.34865332692861556,
      "step": 920
    },
    {
      "epoch": 13.285714285714286,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0002,
      "loss": 0.1343899965286255,
      "loss_d0": 0.3968281477689743,
      "step": 930
    },
    {
      "epoch": 13.428571428571429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002,
      "loss": 0.1584508538246155,
      "loss_d0": 0.4577269285917282,
      "step": 940
    },
    {
      "epoch": 13.571428571428571,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002,
      "loss": 0.14411535263061523,
      "loss_d0": 0.3959138482809067,
      "step": 950
    },
    {
      "epoch": 13.714285714285714,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002,
      "loss": 0.17094615697860718,
      "loss_d0": 0.46275514960289,
      "step": 960
    },
    {
      "epoch": 13.857142857142858,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002,
      "loss": 0.15647470951080322,
      "loss_d0": 0.43458292484283445,
      "step": 970
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0002,
      "loss": 0.16071619987487792,
      "loss_d0": 0.4134894758462906,
      "step": 980
    },
    {
      "epoch": 14.142857142857142,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002,
      "loss": 0.14269460439682008,
      "loss_d0": 0.4106316104531288,
      "step": 990
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0002,
      "loss": 0.15636165142059327,
      "loss_d0": 0.42688301801681516,
      "step": 1000
    },
    {
      "epoch": 14.285714285714286,
      "eval_loss": 9.74100399017334,
      "eval_runtime": 0.3462,
      "eval_samples_per_second": 722.046,
      "eval_steps_per_second": 72.205,
      "step": 1000
    },
    {
      "epoch": 14.428571428571429,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0002,
      "loss": 0.1325446605682373,
      "loss_d0": 0.38657407015562056,
      "step": 1010
    },
    {
      "epoch": 14.571428571428571,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0002,
      "loss": 0.14686695337295533,
      "loss_d0": 0.4054566860198975,
      "step": 1020
    },
    {
      "epoch": 14.714285714285714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002,
      "loss": 0.13976905345916749,
      "loss_d0": 0.38684512078762057,
      "step": 1030
    },
    {
      "epoch": 14.857142857142858,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002,
      "loss": 0.1563257694244385,
      "loss_d0": 0.4456890881061554,
      "step": 1040
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002,
      "loss": 0.1601293206214905,
      "loss_d0": 0.4465105265378952,
      "step": 1050
    },
    {
      "epoch": 15.142857142857142,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0002,
      "loss": 0.15074552297592164,
      "loss_d0": 0.41963100135326387,
      "step": 1060
    },
    {
      "epoch": 15.285714285714286,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002,
      "loss": 0.15185041427612306,
      "loss_d0": 0.41460293233394624,
      "step": 1070
    },
    {
      "epoch": 15.428571428571429,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0002,
      "loss": 0.13449288606643678,
      "loss_d0": 0.40143118500709535,
      "step": 1080
    },
    {
      "epoch": 15.571428571428571,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002,
      "loss": 0.16524744033813477,
      "loss_d0": 0.44636636078357694,
      "step": 1090
    },
    {
      "epoch": 15.714285714285714,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002,
      "loss": 0.1261012315750122,
      "loss_d0": 0.35532553046941756,
      "step": 1100
    },
    {
      "epoch": 15.857142857142858,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0002,
      "loss": 0.16132652759552002,
      "loss_d0": 0.45412015467882155,
      "step": 1110
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.875,
      "learning_rate": 0.0002,
      "loss": 0.1498228430747986,
      "loss_d0": 0.3984955310821533,
      "step": 1120
    },
    {
      "epoch": 16.142857142857142,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002,
      "loss": 0.11573998928070069,
      "loss_d0": 0.3429853171110153,
      "step": 1130
    },
    {
      "epoch": 16.285714285714285,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0002,
      "loss": 0.13562896251678466,
      "loss_d0": 0.3974081426858902,
      "step": 1140
    },
    {
      "epoch": 16.428571428571427,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 0.14604614973068236,
      "loss_d0": 0.43393263816833494,
      "step": 1150
    },
    {
      "epoch": 16.571428571428573,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002,
      "loss": 0.12731246948242186,
      "loss_d0": 0.36327935606241224,
      "step": 1160
    },
    {
      "epoch": 16.714285714285715,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002,
      "loss": 0.14498001337051392,
      "loss_d0": 0.4154079735279083,
      "step": 1170
    },
    {
      "epoch": 16.857142857142858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002,
      "loss": 0.16311801671981813,
      "loss_d0": 0.45202054679393766,
      "step": 1180
    },
    {
      "epoch": 17.0,
      "grad_norm": 1.125,
      "learning_rate": 0.0002,
      "loss": 0.15291047096252441,
      "loss_d0": 0.4574140697717667,
      "step": 1190
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002,
      "loss": 0.11962804794311524,
      "loss_d0": 0.35834813863039017,
      "step": 1200
    },
    {
      "epoch": 17.285714285714285,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002,
      "loss": 0.14257519245147704,
      "loss_d0": 0.40999800264835357,
      "step": 1210
    },
    {
      "epoch": 17.428571428571427,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002,
      "loss": 0.12993460893630981,
      "loss_d0": 0.3714532881975174,
      "step": 1220
    },
    {
      "epoch": 17.571428571428573,
      "grad_norm": 0.984375,
      "learning_rate": 0.0002,
      "loss": 0.13073623180389404,
      "loss_d0": 0.3804581880569458,
      "step": 1230
    },
    {
      "epoch": 17.714285714285715,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002,
      "loss": 0.13893918991088866,
      "loss_d0": 0.40951029658317567,
      "step": 1240
    },
    {
      "epoch": 17.857142857142858,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002,
      "loss": 0.13479981422424317,
      "loss_d0": 0.4055067300796509,
      "step": 1250
    },
    {
      "epoch": 18.0,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002,
      "loss": 0.14387743473052977,
      "loss_d0": 0.45950326323509216,
      "step": 1260
    },
    {
      "epoch": 18.142857142857142,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0002,
      "loss": 0.12405726909637452,
      "loss_d0": 0.37750509530305865,
      "step": 1270
    },
    {
      "epoch": 18.285714285714285,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0002,
      "loss": 0.12020995616912841,
      "loss_d0": 0.3599585384130478,
      "step": 1280
    },
    {
      "epoch": 18.428571428571427,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0002,
      "loss": 0.1472527027130127,
      "loss_d0": 0.3992807060480118,
      "step": 1290
    },
    {
      "epoch": 18.571428571428573,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0002,
      "loss": 0.15526074171066284,
      "loss_d0": 0.457242825627327,
      "step": 1300
    },
    {
      "epoch": 18.714285714285715,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002,
      "loss": 0.14091994762420654,
      "loss_d0": 0.4314684599637985,
      "step": 1310
    },
    {
      "epoch": 18.857142857142858,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0002,
      "loss": 0.1281859755516052,
      "loss_d0": 0.36913466453552246,
      "step": 1320
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.828125,
      "learning_rate": 0.0002,
      "loss": 0.1549570679664612,
      "loss_d0": 0.43077765703201293,
      "step": 1330
    },
    {
      "epoch": 19.142857142857142,
      "grad_norm": 0.73828125,
      "learning_rate": 0.0002,
      "loss": 0.11442582607269287,
      "loss_d0": 0.374704709649086,
      "step": 1340
    },
    {
      "epoch": 19.285714285714285,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002,
      "loss": 0.1251933455467224,
      "loss_d0": 0.37866236865520475,
      "step": 1350
    },
    {
      "epoch": 19.428571428571427,
      "grad_norm": 0.6953125,
      "learning_rate": 0.0002,
      "loss": 0.12533226013183593,
      "loss_d0": 0.38566732704639434,
      "step": 1360
    },
    {
      "epoch": 19.571428571428573,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002,
      "loss": 0.13898290395736695,
      "loss_d0": 0.41185433566570284,
      "step": 1370
    },
    {
      "epoch": 19.714285714285715,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002,
      "loss": 0.14139026403427124,
      "loss_d0": 0.3994438648223877,
      "step": 1380
    },
    {
      "epoch": 19.857142857142858,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002,
      "loss": 0.14629528522491456,
      "loss_d0": 0.45764580070972444,
      "step": 1390
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0002,
      "loss": 0.13958772420883178,
      "loss_d0": 0.41138762086629865,
      "step": 1400
    },
    {
      "epoch": 20.142857142857142,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0002,
      "loss": 0.10666416883468628,
      "loss_d0": 0.35182701647281645,
      "step": 1410
    },
    {
      "epoch": 20.285714285714285,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0002,
      "loss": 0.12388570308685302,
      "loss_d0": 0.3976111799478531,
      "step": 1420
    },
    {
      "epoch": 20.428571428571427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002,
      "loss": 0.12193892002105713,
      "loss_d0": 0.3856766760349274,
      "step": 1430
    },
    {
      "epoch": 20.571428571428573,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0002,
      "loss": 0.12145572900772095,
      "loss_d0": 0.3665471762418747,
      "step": 1440
    },
    {
      "epoch": 20.714285714285715,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0002,
      "loss": 0.13988080024719238,
      "loss_d0": 0.39173959791660307,
      "step": 1450
    },
    {
      "epoch": 20.857142857142858,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0002,
      "loss": 0.1655336618423462,
      "loss_d0": 0.4761664211750031,
      "step": 1460
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.8125,
      "learning_rate": 0.0002,
      "loss": 0.13238286972045898,
      "loss_d0": 0.42673201858997345,
      "step": 1470
    },
    {
      "epoch": 21.142857142857142,
      "grad_norm": 0.9375,
      "learning_rate": 0.0002,
      "loss": 0.11268144845962524,
      "loss_d0": 0.38569384068250656,
      "step": 1480
    },
    {
      "epoch": 21.285714285714285,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002,
      "loss": 0.1253903865814209,
      "loss_d0": 0.4076909214258194,
      "step": 1490
    },
    {
      "epoch": 21.428571428571427,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0002,
      "loss": 0.13616844415664672,
      "loss_d0": 0.43757488429546354,
      "step": 1500
    },
    {
      "epoch": 21.428571428571427,
      "eval_loss": 8.753684043884277,
      "eval_runtime": 0.3463,
      "eval_samples_per_second": 721.916,
      "eval_steps_per_second": 72.192,
      "step": 1500
    },
    {
      "epoch": 21.571428571428573,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0002,
      "loss": 0.13126761913299562,
      "loss_d0": 0.3969725713133812,
      "step": 1510
    },
    {
      "epoch": 21.714285714285715,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0002,
      "loss": 0.12211315631866455,
      "loss_d0": 0.394762022793293,
      "step": 1520
    },
    {
      "epoch": 21.857142857142858,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002,
      "loss": 0.13277941942214966,
      "loss_d0": 0.37962048798799514,
      "step": 1530
    },
    {
      "epoch": 22.0,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002,
      "loss": 0.11672606468200683,
      "loss_d0": 0.36649424582719803,
      "step": 1540
    },
    {
      "epoch": 22.142857142857142,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002,
      "loss": 0.10678048133850097,
      "loss_d0": 0.3356593608856201,
      "step": 1550
    },
    {
      "epoch": 22.285714285714285,
      "grad_norm": 0.734375,
      "learning_rate": 0.0002,
      "loss": 0.10783629417419434,
      "loss_d0": 0.35811730176210405,
      "step": 1560
    },
    {
      "epoch": 22.428571428571427,
      "grad_norm": 0.71875,
      "learning_rate": 0.0002,
      "loss": 0.12049505710601807,
      "loss_d0": 0.37791029214859007,
      "step": 1570
    },
    {
      "epoch": 22.571428571428573,
      "grad_norm": 0.70703125,
      "learning_rate": 0.0002,
      "loss": 0.17957345247268677,
      "loss_d0": 0.39931109845638274,
      "step": 1580
    },
    {
      "epoch": 22.714285714285715,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0002,
      "loss": 0.13404223918914795,
      "loss_d0": 0.41865653693675997,
      "step": 1590
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002,
      "loss": 0.1373995065689087,
      "loss_d0": 0.41133707463741304,
      "step": 1600
    },
    {
      "epoch": 23.0,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001999900348908142,
      "loss": 0.1371894121170044,
      "loss_d0": 0.43064315617084503,
      "step": 1610
    },
    {
      "epoch": 23.142857142857142,
      "grad_norm": 0.76171875,
      "learning_rate": 0.00019995559043291586,
      "loss": 0.09948145151138306,
      "loss_d0": 0.3413606286048889,
      "step": 1620
    },
    {
      "epoch": 23.285714285714285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001998965530147941,
      "loss": 0.12103539705276489,
      "loss_d0": 0.4073850393295288,
      "step": 1630
    },
    {
      "epoch": 23.428571428571427,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00019981293877668964,
      "loss": 0.13946337699890138,
      "loss_d0": 0.44305639564990995,
      "step": 1640
    },
    {
      "epoch": 23.571428571428573,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00019970477057790025,
      "loss": 0.12069251537322997,
      "loss_d0": 0.3936402976512909,
      "step": 1650
    },
    {
      "epoch": 23.714285714285715,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0001995720779905314,
      "loss": 0.11377420425415039,
      "loss_d0": 0.3999328136444092,
      "step": 1660
    },
    {
      "epoch": 23.857142857142858,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0001994148972914116,
      "loss": 0.12190061807632446,
      "loss_d0": 0.3969143331050873,
      "step": 1670
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0001992332714521745,
      "loss": 0.13412773609161377,
      "loss_d0": 0.3998068705201149,
      "step": 1680
    },
    {
      "epoch": 24.142857142857142,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0001990272501275111,
      "loss": 0.11265419721603394,
      "loss_d0": 0.3656838908791542,
      "step": 1690
    },
    {
      "epoch": 24.285714285714285,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00019879688964159436,
      "loss": 0.10430265665054321,
      "loss_d0": 0.34578651189804077,
      "step": 1700
    },
    {
      "epoch": 24.428571428571427,
      "grad_norm": 0.546875,
      "learning_rate": 0.00019854225297268106,
      "loss": 0.1179135799407959,
      "loss_d0": 0.37750212103128433,
      "step": 1710
    },
    {
      "epoch": 24.571428571428573,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0001982634097358939,
      "loss": 0.1166454792022705,
      "loss_d0": 0.39225142896175386,
      "step": 1720
    },
    {
      "epoch": 24.714285714285715,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00019796043616418963,
      "loss": 0.11532022953033447,
      "loss_d0": 0.3840861052274704,
      "step": 1730
    },
    {
      "epoch": 24.857142857142858,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00019763341508751762,
      "loss": 0.14431958198547362,
      "loss_d0": 0.42472186088562014,
      "step": 1740
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.859375,
      "learning_rate": 0.00019728243591017483,
      "loss": 0.14231926202774048,
      "loss_d0": 0.4452060103416443,
      "step": 1750
    },
    {
      "epoch": 25.142857142857142,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001969075945863639,
      "loss": 0.12176300287246704,
      "loss_d0": 0.3826014012098312,
      "step": 1760
    },
    {
      "epoch": 25.285714285714285,
      "grad_norm": 0.71484375,
      "learning_rate": 0.00019650899359396003,
      "loss": 0.10076014995574951,
      "loss_d0": 0.3468409776687622,
      "step": 1770
    },
    {
      "epoch": 25.428571428571427,
      "grad_norm": 0.73046875,
      "learning_rate": 0.00019608674190649446,
      "loss": 0.1326352596282959,
      "loss_d0": 0.41551732271909714,
      "step": 1780
    },
    {
      "epoch": 25.571428571428573,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00019564095496336245,
      "loss": 0.1203911304473877,
      "loss_d0": 0.3934634208679199,
      "step": 1790
    },
    {
      "epoch": 25.714285714285715,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001951717546382631,
      "loss": 0.12641458511352538,
      "loss_d0": 0.3750816985964775,
      "step": 1800
    },
    {
      "epoch": 25.857142857142858,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0001946792692058803,
      "loss": 0.13473284244537354,
      "loss_d0": 0.4147344782948494,
      "step": 1810
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.671875,
      "learning_rate": 0.00019416363330681388,
      "loss": 0.12597525119781494,
      "loss_d0": 0.41172686219215393,
      "step": 1820
    },
    {
      "epoch": 26.142857142857142,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00019362498791077002,
      "loss": 0.13244423866271973,
      "loss_d0": 0.4096812799572945,
      "step": 1830
    },
    {
      "epoch": 26.285714285714285,
      "grad_norm": 0.640625,
      "learning_rate": 0.00019306348027802166,
      "loss": 0.10660384893417359,
      "loss_d0": 0.3657086730003357,
      "step": 1840
    },
    {
      "epoch": 26.428571428571427,
      "grad_norm": 0.9375,
      "learning_rate": 0.000192479263919149,
      "loss": 0.10279349088668824,
      "loss_d0": 0.3708105951547623,
      "step": 1850
    },
    {
      "epoch": 26.571428571428573,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00019187249855307114,
      "loss": 0.11480600833892822,
      "loss_d0": 0.3930380791425705,
      "step": 1860
    },
    {
      "epoch": 26.714285714285715,
      "grad_norm": 0.65625,
      "learning_rate": 0.00019124335006338072,
      "loss": 0.11895076036453248,
      "loss_d0": 0.39364840239286425,
      "step": 1870
    },
    {
      "epoch": 26.857142857142858,
      "grad_norm": 0.625,
      "learning_rate": 0.00019059199045299278,
      "loss": 0.1063655138015747,
      "loss_d0": 0.36443009227514267,
      "step": 1880
    },
    {
      "epoch": 27.0,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018991859779712102,
      "loss": 0.12195394039154053,
      "loss_d0": 0.42418347895145414,
      "step": 1890
    },
    {
      "epoch": 27.142857142857142,
      "grad_norm": 0.66796875,
      "learning_rate": 0.00018922335619459373,
      "loss": 0.09252588152885437,
      "loss_d0": 0.3217478543519974,
      "step": 1900
    },
    {
      "epoch": 27.285714285714285,
      "grad_norm": 0.50390625,
      "learning_rate": 0.00018850645571752317,
      "loss": 0.10999109745025634,
      "loss_d0": 0.36287291944026945,
      "step": 1910
    },
    {
      "epoch": 27.428571428571427,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0001877680923593415,
      "loss": 0.10641486644744873,
      "loss_d0": 0.3734027534723282,
      "step": 1920
    },
    {
      "epoch": 27.571428571428573,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00018700846798121838,
      "loss": 0.11276865005493164,
      "loss_d0": 0.392790412902832,
      "step": 1930
    },
    {
      "epoch": 27.714285714285715,
      "grad_norm": 0.6640625,
      "learning_rate": 0.00018622779025687395,
      "loss": 0.11621142625808716,
      "loss_d0": 0.4068701401352882,
      "step": 1940
    },
    {
      "epoch": 27.857142857142858,
      "grad_norm": 0.84375,
      "learning_rate": 0.00018542627261580302,
      "loss": 0.10486786365509033,
      "loss_d0": 0.3720921754837036,
      "step": 1950
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.9375,
      "learning_rate": 0.00018460413418492565,
      "loss": 0.125832998752594,
      "loss_d0": 0.4209293022751808,
      "step": 1960
    },
    {
      "epoch": 28.142857142857142,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0001837615997286797,
      "loss": 0.09495269060134888,
      "loss_d0": 0.32309970781207087,
      "step": 1970
    },
    {
      "epoch": 28.285714285714285,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00018289889958757294,
      "loss": 0.1047176718711853,
      "loss_d0": 0.35154553800821303,
      "step": 1980
    },
    {
      "epoch": 28.428571428571427,
      "grad_norm": 0.484375,
      "learning_rate": 0.00018201626961520997,
      "loss": 0.10913280248641968,
      "loss_d0": 0.3813262224197388,
      "step": 1990
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00018111395111381214,
      "loss": 0.11296517848968506,
      "loss_d0": 0.4021457523107529,
      "step": 2000
    },
    {
      "epoch": 28.571428571428573,
      "eval_loss": 9.778786659240723,
      "eval_runtime": 0.3465,
      "eval_samples_per_second": 721.509,
      "eval_steps_per_second": 72.151,
      "step": 2000
    }
  ],
  "logging_steps": 10,
  "max_steps": 3500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.577720168579072e+16,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}