{
  "best_metric": 0.9616548940464178,
  "best_model_checkpoint": "/mnt/data4_HDD_14TB/yang/voxceleb-checkpoints/ecapa-tdnn/voxceleb1/finetune/aam-len3-bs256-lr1e-3/checkpoint-2092",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 5230,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03824091778202677,
      "grad_norm": 5.8640031814575195,
      "learning_rate": 3.824091778202677e-05,
      "loss": 13.1781,
      "step": 20
    },
    {
      "epoch": 0.07648183556405354,
      "grad_norm": 4.678975582122803,
      "learning_rate": 7.648183556405354e-05,
      "loss": 12.9709,
      "step": 40
    },
    {
      "epoch": 0.1147227533460803,
      "grad_norm": 3.863873243331909,
      "learning_rate": 0.0001147227533460803,
      "loss": 12.7525,
      "step": 60
    },
    {
      "epoch": 0.15296367112810708,
      "grad_norm": 3.3697011470794678,
      "learning_rate": 0.00015296367112810707,
      "loss": 12.5664,
      "step": 80
    },
    {
      "epoch": 0.19120458891013384,
      "grad_norm": 3.032871961593628,
      "learning_rate": 0.00019120458891013384,
      "loss": 12.345,
      "step": 100
    },
    {
      "epoch": 0.2294455066921606,
      "grad_norm": 2.81247615814209,
      "learning_rate": 0.0002294455066921606,
      "loss": 12.1041,
      "step": 120
    },
    {
      "epoch": 0.2676864244741874,
      "grad_norm": 2.684664011001587,
      "learning_rate": 0.0002676864244741874,
      "loss": 11.8394,
      "step": 140
    },
    {
      "epoch": 0.30592734225621415,
      "grad_norm": 2.6299898624420166,
      "learning_rate": 0.00030592734225621415,
      "loss": 11.6226,
      "step": 160
    },
    {
      "epoch": 0.3441682600382409,
      "grad_norm": 2.584235429763794,
      "learning_rate": 0.00034416826003824094,
      "loss": 11.3617,
      "step": 180
    },
    {
      "epoch": 0.3824091778202677,
      "grad_norm": 2.6476144790649414,
      "learning_rate": 0.0003824091778202677,
      "loss": 11.0999,
      "step": 200
    },
    {
      "epoch": 0.42065009560229444,
      "grad_norm": 2.607042074203491,
      "learning_rate": 0.0004206500956022944,
      "loss": 10.9299,
      "step": 220
    },
    {
      "epoch": 0.4588910133843212,
      "grad_norm": 2.6292777061462402,
      "learning_rate": 0.0004588910133843212,
      "loss": 10.6686,
      "step": 240
    },
    {
      "epoch": 0.497131931166348,
      "grad_norm": 2.56091570854187,
      "learning_rate": 0.0004971319311663481,
      "loss": 10.4216,
      "step": 260
    },
    {
      "epoch": 0.5353728489483748,
      "grad_norm": 2.5721495151519775,
      "learning_rate": 0.0005353728489483748,
      "loss": 10.258,
      "step": 280
    },
    {
      "epoch": 0.5736137667304015,
      "grad_norm": 2.6123592853546143,
      "learning_rate": 0.0005736137667304016,
      "loss": 10.0441,
      "step": 300
    },
    {
      "epoch": 0.6118546845124283,
      "grad_norm": 2.5007669925689697,
      "learning_rate": 0.0006118546845124283,
      "loss": 9.7655,
      "step": 320
    },
    {
      "epoch": 0.6500956022944551,
      "grad_norm": 2.6687047481536865,
      "learning_rate": 0.000650095602294455,
      "loss": 9.6003,
      "step": 340
    },
    {
      "epoch": 0.6883365200764818,
      "grad_norm": 2.758013963699341,
      "learning_rate": 0.0006883365200764819,
      "loss": 9.3888,
      "step": 360
    },
    {
      "epoch": 0.7265774378585086,
      "grad_norm": 2.6198856830596924,
      "learning_rate": 0.0007265774378585086,
      "loss": 9.0976,
      "step": 380
    },
    {
      "epoch": 0.7648183556405354,
      "grad_norm": 2.6092536449432373,
      "learning_rate": 0.0007648183556405354,
      "loss": 8.9158,
      "step": 400
    },
    {
      "epoch": 0.8030592734225621,
      "grad_norm": 2.609232187271118,
      "learning_rate": 0.0008030592734225621,
      "loss": 8.7461,
      "step": 420
    },
    {
      "epoch": 0.8413001912045889,
      "grad_norm": 2.717844247817993,
      "learning_rate": 0.0008413001912045888,
      "loss": 8.5383,
      "step": 440
    },
    {
      "epoch": 0.8795411089866156,
      "grad_norm": 2.665982723236084,
      "learning_rate": 0.0008795411089866157,
      "loss": 8.243,
      "step": 460
    },
    {
      "epoch": 0.9177820267686424,
      "grad_norm": 2.6607024669647217,
      "learning_rate": 0.0009177820267686424,
      "loss": 8.143,
      "step": 480
    },
    {
      "epoch": 0.9560229445506692,
      "grad_norm": 2.6848645210266113,
      "learning_rate": 0.0009560229445506692,
      "loss": 7.9178,
      "step": 500
    },
    {
      "epoch": 0.994263862332696,
      "grad_norm": 2.72930908203125,
      "learning_rate": 0.0009942638623326961,
      "loss": 7.6771,
      "step": 520
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6003363605785402,
      "eval_loss": 6.701026916503906,
      "eval_runtime": 210.9446,
      "eval_samples_per_second": 70.469,
      "eval_steps_per_second": 70.469,
      "step": 523
    },
    {
      "epoch": 1.0325047801147227,
      "grad_norm": 2.6760189533233643,
      "learning_rate": 0.0009963883577650309,
      "loss": 7.2977,
      "step": 540
    },
    {
      "epoch": 1.0707456978967496,
      "grad_norm": 2.761121988296509,
      "learning_rate": 0.0009921393669003612,
      "loss": 7.1159,
      "step": 560
    },
    {
      "epoch": 1.1089866156787762,
      "grad_norm": 2.6698880195617676,
      "learning_rate": 0.0009878903760356915,
      "loss": 6.8399,
      "step": 580
    },
    {
      "epoch": 1.147227533460803,
      "grad_norm": 2.64094614982605,
      "learning_rate": 0.0009836413851710218,
      "loss": 6.6834,
      "step": 600
    },
    {
      "epoch": 1.1854684512428297,
      "grad_norm": 2.7290735244750977,
      "learning_rate": 0.0009793923943063523,
      "loss": 6.5621,
      "step": 620
    },
    {
      "epoch": 1.2237093690248566,
      "grad_norm": 2.7097644805908203,
      "learning_rate": 0.0009751434034416827,
      "loss": 6.217,
      "step": 640
    },
    {
      "epoch": 1.2619502868068833,
      "grad_norm": 2.703202962875366,
      "learning_rate": 0.000970894412577013,
      "loss": 6.0556,
      "step": 660
    },
    {
      "epoch": 1.3001912045889101,
      "grad_norm": 2.7082841396331787,
      "learning_rate": 0.0009666454217123433,
      "loss": 5.9178,
      "step": 680
    },
    {
      "epoch": 1.338432122370937,
      "grad_norm": 2.6659677028656006,
      "learning_rate": 0.0009623964308476737,
      "loss": 5.6926,
      "step": 700
    },
    {
      "epoch": 1.3766730401529637,
      "grad_norm": 2.656085252761841,
      "learning_rate": 0.000958147439983004,
      "loss": 5.4895,
      "step": 720
    },
    {
      "epoch": 1.4149139579349903,
      "grad_norm": 2.692253351211548,
      "learning_rate": 0.0009538984491183344,
      "loss": 5.3583,
      "step": 740
    },
    {
      "epoch": 1.4531548757170172,
      "grad_norm": 2.6969852447509766,
      "learning_rate": 0.0009496494582536647,
      "loss": 5.207,
      "step": 760
    },
    {
      "epoch": 1.491395793499044,
      "grad_norm": 2.675426483154297,
      "learning_rate": 0.0009454004673889951,
      "loss": 5.0834,
      "step": 780
    },
    {
      "epoch": 1.5296367112810707,
      "grad_norm": 2.750749111175537,
      "learning_rate": 0.0009411514765243255,
      "loss": 4.9146,
      "step": 800
    },
    {
      "epoch": 1.5678776290630974,
      "grad_norm": 2.668574333190918,
      "learning_rate": 0.0009369024856596558,
      "loss": 4.7151,
      "step": 820
    },
    {
      "epoch": 1.6061185468451242,
      "grad_norm": 2.688476324081421,
      "learning_rate": 0.0009326534947949862,
      "loss": 4.6307,
      "step": 840
    },
    {
      "epoch": 1.644359464627151,
      "grad_norm": 2.8131325244903564,
      "learning_rate": 0.0009284045039303166,
      "loss": 4.5571,
      "step": 860
    },
    {
      "epoch": 1.682600382409178,
      "grad_norm": 2.68613338470459,
      "learning_rate": 0.0009241555130656469,
      "loss": 4.3895,
      "step": 880
    },
    {
      "epoch": 1.7208413001912046,
      "grad_norm": 2.662827730178833,
      "learning_rate": 0.0009199065222009773,
      "loss": 4.2343,
      "step": 900
    },
    {
      "epoch": 1.7590822179732313,
      "grad_norm": 2.5778844356536865,
      "learning_rate": 0.0009156575313363077,
      "loss": 4.1223,
      "step": 920
    },
    {
      "epoch": 1.7973231357552581,
      "grad_norm": 2.7444920539855957,
      "learning_rate": 0.000911408540471638,
      "loss": 4.0716,
      "step": 940
    },
    {
      "epoch": 1.835564053537285,
      "grad_norm": 2.7660489082336426,
      "learning_rate": 0.0009071595496069684,
      "loss": 3.9948,
      "step": 960
    },
    {
      "epoch": 1.8738049713193117,
      "grad_norm": 2.6412644386291504,
      "learning_rate": 0.0009029105587422988,
      "loss": 3.9169,
      "step": 980
    },
    {
      "epoch": 1.9120458891013383,
      "grad_norm": 2.571195602416992,
      "learning_rate": 0.0008986615678776291,
      "loss": 3.7099,
      "step": 1000
    },
    {
      "epoch": 1.9502868068833652,
      "grad_norm": 2.6130528450012207,
      "learning_rate": 0.0008944125770129595,
      "loss": 3.5425,
      "step": 1020
    },
    {
      "epoch": 1.988527724665392,
      "grad_norm": 2.7550861835479736,
      "learning_rate": 0.0008901635861482899,
      "loss": 3.5879,
      "step": 1040
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9140935082408341,
      "eval_loss": 1.9992759227752686,
      "eval_runtime": 203.0806,
      "eval_samples_per_second": 73.198,
      "eval_steps_per_second": 73.198,
      "step": 1046
    },
    {
      "epoch": 2.026768642447419,
      "grad_norm": 2.4564621448516846,
      "learning_rate": 0.0008859145952836202,
      "loss": 3.2225,
      "step": 1060
    },
    {
      "epoch": 2.0650095602294454,
      "grad_norm": 2.484492063522339,
      "learning_rate": 0.0008816656044189504,
      "loss": 2.9864,
      "step": 1080
    },
    {
      "epoch": 2.1032504780114722,
      "grad_norm": 2.467353582382202,
      "learning_rate": 0.000877416613554281,
      "loss": 3.0552,
      "step": 1100
    },
    {
      "epoch": 2.141491395793499,
      "grad_norm": 2.594728469848633,
      "learning_rate": 0.0008731676226896112,
      "loss": 2.9168,
      "step": 1120
    },
    {
      "epoch": 2.179732313575526,
      "grad_norm": 2.4294815063476562,
      "learning_rate": 0.0008689186318249415,
      "loss": 2.8856,
      "step": 1140
    },
    {
      "epoch": 2.2179732313575524,
      "grad_norm": 2.521456480026245,
      "learning_rate": 0.000864669640960272,
      "loss": 2.8125,
      "step": 1160
    },
    {
      "epoch": 2.2562141491395793,
      "grad_norm": 2.5333268642425537,
      "learning_rate": 0.0008604206500956023,
      "loss": 2.8216,
      "step": 1180
    },
    {
      "epoch": 2.294455066921606,
      "grad_norm": 2.449925422668457,
      "learning_rate": 0.0008561716592309326,
      "loss": 2.7326,
      "step": 1200
    },
    {
      "epoch": 2.332695984703633,
      "grad_norm": 2.5326759815216064,
      "learning_rate": 0.000851922668366263,
      "loss": 2.5496,
      "step": 1220
    },
    {
      "epoch": 2.3709369024856595,
      "grad_norm": 2.4263978004455566,
      "learning_rate": 0.0008478861270448269,
      "loss": 2.6125,
      "step": 1240
    },
    {
      "epoch": 2.4091778202676863,
      "grad_norm": 2.4790937900543213,
      "learning_rate": 0.0008436371361801573,
      "loss": 2.6434,
      "step": 1260
    },
    {
      "epoch": 2.447418738049713,
      "grad_norm": 2.52268385887146,
      "learning_rate": 0.0008393881453154876,
      "loss": 2.5629,
      "step": 1280
    },
    {
      "epoch": 2.48565965583174,
      "grad_norm": 2.4463202953338623,
      "learning_rate": 0.0008353516039940514,
      "loss": 2.4858,
      "step": 1300
    },
    {
      "epoch": 2.5239005736137665,
      "grad_norm": 2.455935001373291,
      "learning_rate": 0.0008311026131293817,
      "loss": 2.431,
      "step": 1320
    },
    {
      "epoch": 2.5621414913957934,
      "grad_norm": 2.3563170433044434,
      "learning_rate": 0.0008268536222647122,
      "loss": 2.327,
      "step": 1340
    },
    {
      "epoch": 2.6003824091778203,
      "grad_norm": 2.4477434158325195,
      "learning_rate": 0.0008226046314000425,
      "loss": 2.4019,
      "step": 1360
    },
    {
      "epoch": 2.638623326959847,
      "grad_norm": 2.2741150856018066,
      "learning_rate": 0.0008183556405353728,
      "loss": 2.2696,
      "step": 1380
    },
    {
      "epoch": 2.676864244741874,
      "grad_norm": 2.2840123176574707,
      "learning_rate": 0.0008141066496707033,
      "loss": 2.2923,
      "step": 1400
    },
    {
      "epoch": 2.7151051625239004,
      "grad_norm": 2.3673081398010254,
      "learning_rate": 0.0008098576588060336,
      "loss": 2.2051,
      "step": 1420
    },
    {
      "epoch": 2.7533460803059273,
      "grad_norm": 2.2644526958465576,
      "learning_rate": 0.0008056086679413639,
      "loss": 2.1845,
      "step": 1440
    },
    {
      "epoch": 2.791586998087954,
      "grad_norm": 2.392956256866455,
      "learning_rate": 0.0008013596770766944,
      "loss": 2.1307,
      "step": 1460
    },
    {
      "epoch": 2.8298279158699806,
      "grad_norm": 2.361642837524414,
      "learning_rate": 0.0007973231357552581,
      "loss": 2.0802,
      "step": 1480
    },
    {
      "epoch": 2.8680688336520075,
      "grad_norm": 2.3466880321502686,
      "learning_rate": 0.000793286594433822,
      "loss": 1.9928,
      "step": 1500
    },
    {
      "epoch": 2.9063097514340344,
      "grad_norm": 2.3408572673797607,
      "learning_rate": 0.0007894625026556193,
      "loss": 1.9699,
      "step": 1520
    },
    {
      "epoch": 2.9445506692160612,
      "grad_norm": 2.3541271686553955,
      "learning_rate": 0.0007854259613341832,
      "loss": 1.9908,
      "step": 1540
    },
    {
      "epoch": 2.982791586998088,
      "grad_norm": 2.4044065475463867,
      "learning_rate": 0.0007813894200127471,
      "loss": 1.9536,
      "step": 1560
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9607130844265052,
      "eval_loss": 0.8233553767204285,
      "eval_runtime": 419.4359,
      "eval_samples_per_second": 35.44,
      "eval_steps_per_second": 35.44,
      "step": 1569
    },
    {
      "epoch": 3.0210325047801145,
      "grad_norm": 2.1922998428344727,
      "learning_rate": 0.0007771404291480774,
      "loss": 1.8218,
      "step": 1580
    },
    {
      "epoch": 3.0592734225621414,
      "grad_norm": NaN,
      "learning_rate": 0.0007731038878266412,
      "loss": 1.5913,
      "step": 1600
    },
    {
      "epoch": 3.0975143403441683,
      "grad_norm": NaN,
      "learning_rate": 0.000769492245591672,
      "loss": 1.6189,
      "step": 1620
    },
    {
      "epoch": 3.135755258126195,
      "grad_norm": 2.179269313812256,
      "learning_rate": 0.0007658806033567028,
      "loss": 1.5457,
      "step": 1640
    },
    {
      "epoch": 3.173996175908222,
      "grad_norm": 2.30602765083313,
      "learning_rate": 0.0007618440620352667,
      "loss": 1.5588,
      "step": 1660
    },
    {
      "epoch": 3.2122370936902485,
      "grad_norm": 2.100261688232422,
      "learning_rate": 0.0007578075207138306,
      "loss": 1.5639,
      "step": 1680
    },
    {
      "epoch": 3.2504780114722753,
      "grad_norm": 2.185692548751831,
      "learning_rate": 0.0007544083280220948,
      "loss": 1.5383,
      "step": 1700
    },
    {
      "epoch": 3.288718929254302,
      "grad_norm": 2.014509439468384,
      "learning_rate": 0.0007503717867006586,
      "loss": 1.5617,
      "step": 1720
    },
    {
      "epoch": 3.3269598470363286,
      "grad_norm": 1.757117509841919,
      "learning_rate": 0.0007467601444656894,
      "loss": 1.6206,
      "step": 1740
    },
    {
      "epoch": 3.3652007648183555,
      "grad_norm": 1.4908952713012695,
      "learning_rate": 0.0007435734013171872,
      "loss": 1.6565,
      "step": 1760
    },
    {
      "epoch": 3.4034416826003824,
      "grad_norm": 1.310102939605713,
      "learning_rate": 0.000739961759082218,
      "loss": 1.6018,
      "step": 1780
    },
    {
      "epoch": 3.4416826003824093,
      "grad_norm": 1.3842021226882935,
      "learning_rate": 0.0007359252177607819,
      "loss": 1.5565,
      "step": 1800
    },
    {
      "epoch": 3.479923518164436,
      "grad_norm": 0.23385359346866608,
      "learning_rate": 0.0007321011259825792,
      "loss": 1.6196,
      "step": 1820
    },
    {
      "epoch": 3.5181644359464626,
      "grad_norm": 0.0,
      "learning_rate": 0.0007291268323773104,
      "loss": 1.5137,
      "step": 1840
    },
    {
      "epoch": 3.5564053537284894,
      "grad_norm": NaN,
      "learning_rate": 0.0007253027405991077,
      "loss": 1.5645,
      "step": 1860
    },
    {
      "epoch": 3.5946462715105163,
      "grad_norm": NaN,
      "learning_rate": 0.0007216910983641384,
      "loss": 1.5224,
      "step": 1880
    },
    {
      "epoch": 3.632887189292543,
      "grad_norm": 0.0,
      "learning_rate": 0.0007185043552156363,
      "loss": 1.4859,
      "step": 1900
    },
    {
      "epoch": 3.67112810707457,
      "grad_norm": 0.0,
      "learning_rate": 0.0007155300616103676,
      "loss": 1.5356,
      "step": 1920
    },
    {
      "epoch": 3.7093690248565965,
      "grad_norm": 0.0,
      "learning_rate": 0.0007125557680050988,
      "loss": 1.5673,
      "step": 1940
    },
    {
      "epoch": 3.7476099426386233,
      "grad_norm": 0.0,
      "learning_rate": 0.00070958147439983,
      "loss": 1.4688,
      "step": 1960
    },
    {
      "epoch": 3.78585086042065,
      "grad_norm": 0.0,
      "learning_rate": 0.0007061822817080943,
      "loss": 1.5848,
      "step": 1980
    },
    {
      "epoch": 3.8240917782026767,
      "grad_norm": 0.0,
      "learning_rate": 0.0007036328871892926,
      "loss": 1.5157,
      "step": 2000
    },
    {
      "epoch": 3.8623326959847035,
      "grad_norm": NaN,
      "learning_rate": 0.0007006585935840239,
      "loss": 1.5408,
      "step": 2020
    },
    {
      "epoch": 3.9005736137667304,
      "grad_norm": 0.0,
      "learning_rate": 0.0006978967495219886,
      "loss": 1.4978,
      "step": 2040
    },
    {
      "epoch": 3.9388145315487573,
      "grad_norm": 0.0,
      "learning_rate": 0.0006949224559167198,
      "loss": 1.557,
      "step": 2060
    },
    {
      "epoch": 3.977055449330784,
      "grad_norm": 0.0,
      "learning_rate": 0.0006921606118546845,
      "loss": 1.5008,
      "step": 2080
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9616548940464178,
      "eval_loss": 0.6434142589569092,
      "eval_runtime": 182.6336,
      "eval_samples_per_second": 81.392,
      "eval_steps_per_second": 81.392,
      "step": 2092
    },
    {
      "epoch": 4.015296367112811,
      "grad_norm": NaN,
      "learning_rate": 0.0006887614191629488,
      "loss": 1.4216,
      "step": 2100
    },
    {
      "epoch": 4.053537284894838,
      "grad_norm": 0.0,
      "learning_rate": 0.0006859995751009136,
      "loss": 1.3752,
      "step": 2120
    },
    {
      "epoch": 4.091778202676864,
      "grad_norm": 0.0,
      "learning_rate": 0.0006828128319524112,
      "loss": 1.3751,
      "step": 2140
    },
    {
      "epoch": 4.130019120458891,
      "grad_norm": NaN,
      "learning_rate": 0.0006794136392606756,
      "loss": 1.3608,
      "step": 2160
    },
    {
      "epoch": 4.168260038240918,
      "grad_norm": NaN,
      "learning_rate": 0.0006766517951986402,
      "loss": 1.3152,
      "step": 2180
    },
    {
      "epoch": 4.2065009560229445,
      "grad_norm": 0.0,
      "learning_rate": 0.0006736775015933716,
      "loss": 1.3721,
      "step": 2200
    },
    {
      "epoch": 4.244741873804971,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 1.3709,
      "step": 2220
    },
    {
      "epoch": 4.282982791586998,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2240
    },
    {
      "epoch": 4.321223709369025,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2260
    },
    {
      "epoch": 4.359464627151052,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2280
    },
    {
      "epoch": 4.397705544933078,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2300
    },
    {
      "epoch": 4.435946462715105,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2320
    },
    {
      "epoch": 4.474187380497132,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2340
    },
    {
      "epoch": 4.512428298279159,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2360
    },
    {
      "epoch": 4.550669216061186,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2380
    },
    {
      "epoch": 4.588910133843212,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2400
    },
    {
      "epoch": 4.627151051625239,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2420
    },
    {
      "epoch": 4.665391969407266,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2440
    },
    {
      "epoch": 4.7036328871892925,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2460
    },
    {
      "epoch": 4.741873804971319,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2480
    },
    {
      "epoch": 4.780114722753346,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2500
    },
    {
      "epoch": 4.818355640535373,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2520
    },
    {
      "epoch": 4.8565965583174,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2540
    },
    {
      "epoch": 4.894837476099426,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2560
    },
    {
      "epoch": 4.933078393881453,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2580
    },
    {
      "epoch": 4.97131931166348,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2600
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.0005381769256643121,
      "eval_loss": NaN,
      "eval_runtime": 180.7668,
      "eval_samples_per_second": 82.233,
      "eval_steps_per_second": 82.233,
      "step": 2615
    },
    {
      "epoch": 5.009560229445507,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2620
    },
    {
      "epoch": 5.047801147227533,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2640
    },
    {
      "epoch": 5.08604206500956,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2660
    },
    {
      "epoch": 5.124282982791587,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2680
    },
    {
      "epoch": 5.162523900573614,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2700
    },
    {
      "epoch": 5.2007648183556405,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2720
    },
    {
      "epoch": 5.239005736137667,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2740
    },
    {
      "epoch": 5.277246653919694,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2760
    },
    {
      "epoch": 5.315487571701721,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2780
    },
    {
      "epoch": 5.353728489483748,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2800
    },
    {
      "epoch": 5.3919694072657744,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2820
    },
    {
      "epoch": 5.430210325047801,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2840
    },
    {
      "epoch": 5.468451242829828,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2860
    },
    {
      "epoch": 5.506692160611855,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2880
    },
    {
      "epoch": 5.544933078393882,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2900
    },
    {
      "epoch": 5.583173996175908,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2920
    },
    {
      "epoch": 5.621414913957935,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2940
    },
    {
      "epoch": 5.659655831739962,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2960
    },
    {
      "epoch": 5.6978967495219885,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 2980
    },
    {
      "epoch": 5.736137667304015,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3000
    },
    {
      "epoch": 5.774378585086042,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3020
    },
    {
      "epoch": 5.812619502868069,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3040
    },
    {
      "epoch": 5.850860420650095,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3060
    },
    {
      "epoch": 5.8891013384321225,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3080
    },
    {
      "epoch": 5.927342256214149,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3100
    },
    {
      "epoch": 5.965583173996176,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3120
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.0005381769256643121,
      "eval_loss": NaN,
      "eval_runtime": 182.9382,
      "eval_samples_per_second": 81.257,
      "eval_steps_per_second": 81.257,
      "step": 3138
    },
    {
      "epoch": 6.003824091778203,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3140
    },
    {
      "epoch": 6.042065009560229,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3160
    },
    {
      "epoch": 6.080305927342256,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3180
    },
    {
      "epoch": 6.118546845124283,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3200
    },
    {
      "epoch": 6.15678776290631,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3220
    },
    {
      "epoch": 6.195028680688337,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3240
    },
    {
      "epoch": 6.233269598470363,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3260
    },
    {
      "epoch": 6.27151051625239,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3280
    },
    {
      "epoch": 6.309751434034417,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3300
    },
    {
      "epoch": 6.347992351816444,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3320
    },
    {
      "epoch": 6.3862332695984705,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3340
    },
    {
      "epoch": 6.424474187380497,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3360
    },
    {
      "epoch": 6.462715105162524,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3380
    },
    {
      "epoch": 6.500956022944551,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3400
    },
    {
      "epoch": 6.539196940726577,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3420
    },
    {
      "epoch": 6.577437858508604,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3440
    },
    {
      "epoch": 6.615678776290631,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3460
    },
    {
      "epoch": 6.653919694072657,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3480
    },
    {
      "epoch": 6.692160611854685,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 6.730401529636711,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3520
    },
    {
      "epoch": 6.768642447418738,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3540
    },
    {
      "epoch": 6.806883365200765,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3560
    },
    {
      "epoch": 6.845124282982791,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3580
    },
    {
      "epoch": 6.8833652007648185,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3600
    },
    {
      "epoch": 6.921606118546845,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3620
    },
    {
      "epoch": 6.959847036328872,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3640
    },
    {
      "epoch": 6.998087954110899,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3660
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.0005381769256643121,
      "eval_loss": NaN,
      "eval_runtime": 188.9144,
      "eval_samples_per_second": 78.686,
      "eval_steps_per_second": 78.686,
      "step": 3661
    },
    {
      "epoch": 7.036328871892925,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3680
    },
    {
      "epoch": 7.074569789674952,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3700
    },
    {
      "epoch": 7.112810707456979,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3720
    },
    {
      "epoch": 7.151051625239006,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3740
    },
    {
      "epoch": 7.189292543021033,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3760
    },
    {
      "epoch": 7.227533460803059,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3780
    },
    {
      "epoch": 7.265774378585086,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3800
    },
    {
      "epoch": 7.304015296367113,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3820
    },
    {
      "epoch": 7.342256214149139,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3840
    },
    {
      "epoch": 7.3804971319311665,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3860
    },
    {
      "epoch": 7.418738049713193,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3880
    },
    {
      "epoch": 7.45697896749522,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3900
    },
    {
      "epoch": 7.495219885277247,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3920
    },
    {
      "epoch": 7.533460803059273,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3940
    },
    {
      "epoch": 7.5717017208413,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3960
    },
    {
      "epoch": 7.609942638623327,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 3980
    },
    {
      "epoch": 7.648183556405353,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 7.686424474187381,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4020
    },
    {
      "epoch": 7.724665391969407,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4040
    },
    {
      "epoch": 7.762906309751434,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4060
    },
    {
      "epoch": 7.801147227533461,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4080
    },
    {
      "epoch": 7.839388145315487,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4100
    },
    {
      "epoch": 7.8776290630975145,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4120
    },
    {
      "epoch": 7.915869980879541,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4140
    },
    {
      "epoch": 7.954110898661568,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4160
    },
    {
      "epoch": 7.992351816443595,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4180
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.0005381769256643121,
      "eval_loss": NaN,
      "eval_runtime": 167.1036,
      "eval_samples_per_second": 88.957,
      "eval_steps_per_second": 88.957,
      "step": 4184
    },
    {
      "epoch": 8.030592734225621,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4200
    },
    {
      "epoch": 8.068833652007648,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4220
    },
    {
      "epoch": 8.107074569789676,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4240
    },
    {
      "epoch": 8.145315487571702,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4260
    },
    {
      "epoch": 8.183556405353729,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4280
    },
    {
      "epoch": 8.221797323135755,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4300
    },
    {
      "epoch": 8.260038240917781,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4320
    },
    {
      "epoch": 8.29827915869981,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4340
    },
    {
      "epoch": 8.336520076481836,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4360
    },
    {
      "epoch": 8.374760994263863,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4380
    },
    {
      "epoch": 8.413001912045889,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4400
    },
    {
      "epoch": 8.451242829827915,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4420
    },
    {
      "epoch": 8.489483747609942,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4440
    },
    {
      "epoch": 8.52772466539197,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4460
    },
    {
      "epoch": 8.565965583173996,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4480
    },
    {
      "epoch": 8.604206500956023,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 8.64244741873805,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4520
    },
    {
      "epoch": 8.680688336520076,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4540
    },
    {
      "epoch": 8.718929254302104,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4560
    },
    {
      "epoch": 8.75717017208413,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4580
    },
    {
      "epoch": 8.795411089866157,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4600
    },
    {
      "epoch": 8.833652007648183,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4620
    },
    {
      "epoch": 8.87189292543021,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4640
    },
    {
      "epoch": 8.910133843212238,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4660
    },
    {
      "epoch": 8.948374760994264,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4680
    },
    {
      "epoch": 8.98661567877629,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4700
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.0005381769256643121,
      "eval_loss": NaN,
      "eval_runtime": 176.3855,
      "eval_samples_per_second": 84.276,
      "eval_steps_per_second": 84.276,
      "step": 4707
    },
    {
      "epoch": 9.024856596558317,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4720
    },
    {
      "epoch": 9.063097514340344,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4740
    },
    {
      "epoch": 9.101338432122372,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4760
    },
    {
      "epoch": 9.139579349904398,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4780
    },
    {
      "epoch": 9.177820267686425,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4800
    },
    {
      "epoch": 9.216061185468451,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4820
    },
    {
      "epoch": 9.254302103250478,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4840
    },
    {
      "epoch": 9.292543021032504,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4860
    },
    {
      "epoch": 9.330783938814532,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4880
    },
    {
      "epoch": 9.369024856596559,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4900
    },
    {
      "epoch": 9.407265774378585,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4920
    },
    {
      "epoch": 9.445506692160611,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4940
    },
    {
      "epoch": 9.483747609942638,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4960
    },
    {
      "epoch": 9.521988527724666,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 4980
    },
    {
      "epoch": 9.560229445506693,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5000
    },
    {
      "epoch": 9.598470363288719,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5020
    },
    {
      "epoch": 9.636711281070745,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5040
    },
    {
      "epoch": 9.674952198852772,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5060
    },
    {
      "epoch": 9.7131931166348,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5080
    },
    {
      "epoch": 9.751434034416826,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5100
    },
    {
      "epoch": 9.789674952198853,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5120
    },
    {
      "epoch": 9.82791586998088,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5140
    },
    {
      "epoch": 9.866156787762906,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5160
    },
    {
      "epoch": 9.904397705544934,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5180
    },
    {
      "epoch": 9.94263862332696,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5200
    },
    {
      "epoch": 9.980879541108987,
      "grad_norm": NaN,
      "learning_rate": 0.0006717654557042703,
      "loss": 0.0,
      "step": 5220
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.0005381769256643121,
      "eval_loss": NaN,
      "eval_runtime": 183.845,
      "eval_samples_per_second": 80.856,
      "eval_steps_per_second": 80.856,
      "step": 5230
    },
    {
      "epoch": 10.0,
      "step": 5230,
      "total_flos": 8.1484088684544e+18,
      "train_loss": 1.9844060945237343,
      "train_runtime": 19376.6921,
      "train_samples_per_second": 69.04,
      "train_steps_per_second": 0.27
    }
  ],
  "logging_steps": 20,
  "max_steps": 5230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.1484088684544e+18,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}