{
  "best_metric": 2.4190404415130615,
  "best_model_checkpoint": "./results/models/checkpoint-136458",
  "epoch": 18.0,
  "eval_steps": 500,
  "global_step": 136458,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06595435958316845,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0009986809128083366,
      "loss": 5.3104,
      "step": 500
    },
    {
      "epoch": 0.1319087191663369,
      "grad_norm": 0.640625,
      "learning_rate": 0.0009973618256166733,
      "loss": 4.3028,
      "step": 1000
    },
    {
      "epoch": 0.19786307874950534,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0009960427384250099,
      "loss": 3.9369,
      "step": 1500
    },
    {
      "epoch": 0.2638174383326738,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0009947236512333464,
      "loss": 3.7115,
      "step": 2000
    },
    {
      "epoch": 0.3297717979158422,
      "grad_norm": 0.59375,
      "learning_rate": 0.0009934045640416832,
      "loss": 3.5589,
      "step": 2500
    },
    {
      "epoch": 0.3957261574990107,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0009920854768500197,
      "loss": 3.4636,
      "step": 3000
    },
    {
      "epoch": 0.4616805170821791,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0009907663896583565,
      "loss": 3.3838,
      "step": 3500
    },
    {
      "epoch": 0.5276348766653476,
      "grad_norm": 0.5859375,
      "learning_rate": 0.000989447302466693,
      "loss": 3.3023,
      "step": 4000
    },
    {
      "epoch": 0.593589236248516,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0009881282152750298,
      "loss": 3.2445,
      "step": 4500
    },
    {
      "epoch": 0.6595435958316844,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0009868091280833663,
      "loss": 3.205,
      "step": 5000
    },
    {
      "epoch": 0.725497955414853,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0009854900408917029,
      "loss": 3.157,
      "step": 5500
    },
    {
      "epoch": 0.7914523149980214,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0009841709537000397,
      "loss": 3.126,
      "step": 6000
    },
    {
      "epoch": 0.8574066745811898,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0009828518665083762,
      "loss": 3.0838,
      "step": 6500
    },
    {
      "epoch": 0.9233610341643582,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0009815327793167127,
      "loss": 3.0584,
      "step": 7000
    },
    {
      "epoch": 0.9893153937475268,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0009802136921250495,
      "loss": 3.0325,
      "step": 7500
    },
    {
      "epoch": 1.0,
      "eval_loss": 3.0351455211639404,
      "eval_runtime": 4.9415,
      "eval_samples_per_second": 404.738,
      "eval_steps_per_second": 1.619,
      "step": 7581
    },
    {
      "epoch": 1.0552697533306952,
      "grad_norm": 0.55859375,
      "learning_rate": 0.000978894604933386,
      "loss": 2.9581,
      "step": 8000
    },
    {
      "epoch": 1.1212241129138636,
      "grad_norm": 0.625,
      "learning_rate": 0.0009775755177417228,
      "loss": 2.9426,
      "step": 8500
    },
    {
      "epoch": 1.187178472497032,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009762564305500594,
      "loss": 2.9122,
      "step": 9000
    },
    {
      "epoch": 1.2531328320802004,
      "grad_norm": 0.58203125,
      "learning_rate": 0.000974937343358396,
      "loss": 2.8982,
      "step": 9500
    },
    {
      "epoch": 1.3190871916633689,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0009736182561667326,
      "loss": 2.891,
      "step": 10000
    },
    {
      "epoch": 1.3850415512465375,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0009722991689750693,
      "loss": 2.8751,
      "step": 10500
    },
    {
      "epoch": 1.450995910829706,
      "grad_norm": 0.609375,
      "learning_rate": 0.0009709800817834059,
      "loss": 2.853,
      "step": 11000
    },
    {
      "epoch": 1.5169502704128743,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0009696609945917425,
      "loss": 2.842,
      "step": 11500
    },
    {
      "epoch": 1.5829046299960428,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0009683419074000792,
      "loss": 2.831,
      "step": 12000
    },
    {
      "epoch": 1.6488589895792112,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009670228202084158,
      "loss": 2.8257,
      "step": 12500
    },
    {
      "epoch": 1.7148133491623796,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0009657037330167524,
      "loss": 2.8069,
      "step": 13000
    },
    {
      "epoch": 1.780767708745548,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0009643846458250891,
      "loss": 2.7999,
      "step": 13500
    },
    {
      "epoch": 1.8467220683287167,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009630655586334257,
      "loss": 2.7877,
      "step": 14000
    },
    {
      "epoch": 1.9126764279118849,
      "grad_norm": 0.625,
      "learning_rate": 0.0009617464714417623,
      "loss": 2.7914,
      "step": 14500
    },
    {
      "epoch": 1.9786307874950535,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009604273842500989,
      "loss": 2.7667,
      "step": 15000
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.802887201309204,
      "eval_runtime": 5.0226,
      "eval_samples_per_second": 398.199,
      "eval_steps_per_second": 1.593,
      "step": 15162
    },
    {
      "epoch": 2.0445851470782217,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0009591082970584356,
      "loss": 2.6997,
      "step": 15500
    },
    {
      "epoch": 2.1105395066613903,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009577892098667722,
      "loss": 2.688,
      "step": 16000
    },
    {
      "epoch": 2.1764938662445585,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0009564701226751088,
      "loss": 2.6977,
      "step": 16500
    },
    {
      "epoch": 2.242448225827727,
      "grad_norm": 0.578125,
      "learning_rate": 0.0009551510354834455,
      "loss": 2.6954,
      "step": 17000
    },
    {
      "epoch": 2.308402585410896,
      "grad_norm": 0.53125,
      "learning_rate": 0.000953831948291782,
      "loss": 2.6841,
      "step": 17500
    },
    {
      "epoch": 2.374356944994064,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0009525128611001187,
      "loss": 2.6851,
      "step": 18000
    },
    {
      "epoch": 2.4403113045772327,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009511937739084554,
      "loss": 2.6696,
      "step": 18500
    },
    {
      "epoch": 2.506265664160401,
      "grad_norm": 0.54296875,
      "learning_rate": 0.000949874686716792,
      "loss": 2.6717,
      "step": 19000
    },
    {
      "epoch": 2.5722200237435695,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009485555995251286,
      "loss": 2.6728,
      "step": 19500
    },
    {
      "epoch": 2.6381743833267377,
      "grad_norm": 0.578125,
      "learning_rate": 0.0009472365123334653,
      "loss": 2.6487,
      "step": 20000
    },
    {
      "epoch": 2.7041287429099063,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0009459174251418019,
      "loss": 2.6549,
      "step": 20500
    },
    {
      "epoch": 2.770083102493075,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0009445983379501385,
      "loss": 2.6441,
      "step": 21000
    },
    {
      "epoch": 2.836037462076243,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0009432792507584752,
      "loss": 2.6374,
      "step": 21500
    },
    {
      "epoch": 2.901991821659412,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009419601635668118,
      "loss": 2.6291,
      "step": 22000
    },
    {
      "epoch": 2.96794618124258,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0009406410763751484,
      "loss": 2.6326,
      "step": 22500
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.696772575378418,
      "eval_runtime": 5.0551,
      "eval_samples_per_second": 395.639,
      "eval_steps_per_second": 1.583,
      "step": 22743
    },
    {
      "epoch": 3.0339005408257487,
      "grad_norm": 0.68359375,
      "learning_rate": 0.000939321989183485,
      "loss": 2.5858,
      "step": 23000
    },
    {
      "epoch": 3.099854900408917,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0009380029019918217,
      "loss": 2.5525,
      "step": 23500
    },
    {
      "epoch": 3.1658092599920855,
      "grad_norm": 0.515625,
      "learning_rate": 0.0009366838148001583,
      "loss": 2.559,
      "step": 24000
    },
    {
      "epoch": 3.231763619575254,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009353647276084949,
      "loss": 2.5623,
      "step": 24500
    },
    {
      "epoch": 3.2977179791584224,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009340456404168316,
      "loss": 2.5558,
      "step": 25000
    },
    {
      "epoch": 3.363672338741591,
      "grad_norm": 0.72265625,
      "learning_rate": 0.0009327265532251682,
      "loss": 2.5571,
      "step": 25500
    },
    {
      "epoch": 3.429626698324759,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0009314074660335047,
      "loss": 2.561,
      "step": 26000
    },
    {
      "epoch": 3.495581057907928,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0009300883788418415,
      "loss": 2.5557,
      "step": 26500
    },
    {
      "epoch": 3.561535417491096,
      "grad_norm": 0.58203125,
      "learning_rate": 0.000928769291650178,
      "loss": 2.5473,
      "step": 27000
    },
    {
      "epoch": 3.6274897770742647,
      "grad_norm": 0.578125,
      "learning_rate": 0.0009274502044585147,
      "loss": 2.5441,
      "step": 27500
    },
    {
      "epoch": 3.6934441366574333,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0009261311172668514,
      "loss": 2.5392,
      "step": 28000
    },
    {
      "epoch": 3.7593984962406015,
      "grad_norm": 0.53515625,
      "learning_rate": 0.000924812030075188,
      "loss": 2.5421,
      "step": 28500
    },
    {
      "epoch": 3.8253528558237697,
      "grad_norm": 0.59375,
      "learning_rate": 0.0009234929428835246,
      "loss": 2.5356,
      "step": 29000
    },
    {
      "epoch": 3.8913072154069384,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0009221738556918613,
      "loss": 2.5418,
      "step": 29500
    },
    {
      "epoch": 3.957261574990107,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0009208547685001979,
      "loss": 2.5229,
      "step": 30000
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.629683017730713,
      "eval_runtime": 4.5326,
      "eval_samples_per_second": 441.249,
      "eval_steps_per_second": 1.765,
      "step": 30324
    },
    {
      "epoch": 4.023215934573275,
      "grad_norm": 0.6953125,
      "learning_rate": 0.0009195356813085345,
      "loss": 2.4939,
      "step": 30500
    },
    {
      "epoch": 4.089170294156443,
      "grad_norm": 0.546875,
      "learning_rate": 0.0009182165941168712,
      "loss": 2.4514,
      "step": 31000
    },
    {
      "epoch": 4.1551246537396125,
      "grad_norm": 0.578125,
      "learning_rate": 0.0009168975069252078,
      "loss": 2.4666,
      "step": 31500
    },
    {
      "epoch": 4.221079013322781,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009155784197335444,
      "loss": 2.4649,
      "step": 32000
    },
    {
      "epoch": 4.287033372905949,
      "grad_norm": 0.57421875,
      "learning_rate": 0.000914259332541881,
      "loss": 2.4642,
      "step": 32500
    },
    {
      "epoch": 4.352987732489117,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0009129402453502177,
      "loss": 2.4669,
      "step": 33000
    },
    {
      "epoch": 4.418942092072286,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0009116211581585543,
      "loss": 2.4686,
      "step": 33500
    },
    {
      "epoch": 4.484896451655454,
      "grad_norm": 0.53125,
      "learning_rate": 0.0009103020709668909,
      "loss": 2.4809,
      "step": 34000
    },
    {
      "epoch": 4.550850811238623,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0009089829837752276,
      "loss": 2.4708,
      "step": 34500
    },
    {
      "epoch": 4.616805170821792,
      "grad_norm": 0.5625,
      "learning_rate": 0.0009076638965835642,
      "loss": 2.4701,
      "step": 35000
    },
    {
      "epoch": 4.68275953040496,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0009063448093919007,
      "loss": 2.4714,
      "step": 35500
    },
    {
      "epoch": 4.748713889988128,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0009050257222002375,
      "loss": 2.4626,
      "step": 36000
    },
    {
      "epoch": 4.814668249571296,
      "grad_norm": 0.57421875,
      "learning_rate": 0.000903706635008574,
      "loss": 2.4696,
      "step": 36500
    },
    {
      "epoch": 4.880622609154465,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0009023875478169107,
      "loss": 2.4614,
      "step": 37000
    },
    {
      "epoch": 4.9465769687376335,
      "grad_norm": 0.5625,
      "learning_rate": 0.0009010684606252473,
      "loss": 2.4544,
      "step": 37500
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.593965768814087,
      "eval_runtime": 4.0976,
      "eval_samples_per_second": 488.092,
      "eval_steps_per_second": 1.952,
      "step": 37905
    },
    {
      "epoch": 5.012531328320802,
      "grad_norm": 0.55078125,
      "learning_rate": 0.000899749373433584,
      "loss": 2.4455,
      "step": 38000
    },
    {
      "epoch": 5.078485687903971,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0008984302862419205,
      "loss": 2.3874,
      "step": 38500
    },
    {
      "epoch": 5.144440047487139,
      "grad_norm": 0.609375,
      "learning_rate": 0.0008971111990502573,
      "loss": 2.3907,
      "step": 39000
    },
    {
      "epoch": 5.210394407070307,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0008957921118585939,
      "loss": 2.3965,
      "step": 39500
    },
    {
      "epoch": 5.276348766653475,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0008944730246669305,
      "loss": 2.4055,
      "step": 40000
    },
    {
      "epoch": 5.3423031262366445,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0008931539374752672,
      "loss": 2.4024,
      "step": 40500
    },
    {
      "epoch": 5.408257485819813,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0008918348502836038,
      "loss": 2.402,
      "step": 41000
    },
    {
      "epoch": 5.474211845402981,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0008905157630919404,
      "loss": 2.4073,
      "step": 41500
    },
    {
      "epoch": 5.54016620498615,
      "grad_norm": 0.58203125,
      "learning_rate": 0.000889196675900277,
      "loss": 2.4046,
      "step": 42000
    },
    {
      "epoch": 5.606120564569318,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0008878775887086137,
      "loss": 2.4127,
      "step": 42500
    },
    {
      "epoch": 5.672074924152486,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0008865585015169503,
      "loss": 2.4102,
      "step": 43000
    },
    {
      "epoch": 5.738029283735655,
      "grad_norm": 0.546875,
      "learning_rate": 0.0008852394143252869,
      "loss": 2.4021,
      "step": 43500
    },
    {
      "epoch": 5.803983643318824,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0008839203271336235,
      "loss": 2.4041,
      "step": 44000
    },
    {
      "epoch": 5.869938002901992,
      "grad_norm": 0.59375,
      "learning_rate": 0.0008826012399419602,
      "loss": 2.4034,
      "step": 44500
    },
    {
      "epoch": 5.93589236248516,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0008812821527502967,
      "loss": 2.4059,
      "step": 45000
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.5455453395843506,
      "eval_runtime": 4.3171,
      "eval_samples_per_second": 463.276,
      "eval_steps_per_second": 1.853,
      "step": 45486
    },
    {
      "epoch": 6.001846722068329,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0008799630655586335,
      "loss": 2.3998,
      "step": 45500
    },
    {
      "epoch": 6.067801081651497,
      "grad_norm": 0.546875,
      "learning_rate": 0.00087864397836697,
      "loss": 2.3343,
      "step": 46000
    },
    {
      "epoch": 6.1337554412346655,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0008773248911753067,
      "loss": 2.3341,
      "step": 46500
    },
    {
      "epoch": 6.199709800817834,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0008760058039836433,
      "loss": 2.3527,
      "step": 47000
    },
    {
      "epoch": 6.265664160401003,
      "grad_norm": 0.58203125,
      "learning_rate": 0.00087468671679198,
      "loss": 2.3633,
      "step": 47500
    },
    {
      "epoch": 6.331618519984171,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0008733676296003165,
      "loss": 2.3615,
      "step": 48000
    },
    {
      "epoch": 6.397572879567339,
      "grad_norm": 0.546875,
      "learning_rate": 0.0008720485424086533,
      "loss": 2.3654,
      "step": 48500
    },
    {
      "epoch": 6.463527239150508,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0008707294552169899,
      "loss": 2.3695,
      "step": 49000
    },
    {
      "epoch": 6.5294815987336765,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0008694103680253265,
      "loss": 2.3698,
      "step": 49500
    },
    {
      "epoch": 6.595435958316845,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0008680912808336632,
      "loss": 2.3671,
      "step": 50000
    },
    {
      "epoch": 6.661390317900013,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0008667721936419998,
      "loss": 2.3611,
      "step": 50500
    },
    {
      "epoch": 6.727344677483182,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0008654531064503364,
      "loss": 2.371,
      "step": 51000
    },
    {
      "epoch": 6.79329903706635,
      "grad_norm": 0.6484375,
      "learning_rate": 0.000864134019258673,
      "loss": 2.3693,
      "step": 51500
    },
    {
      "epoch": 6.859253396649518,
      "grad_norm": 0.59375,
      "learning_rate": 0.0008628149320670097,
      "loss": 2.3575,
      "step": 52000
    },
    {
      "epoch": 6.9252077562326875,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0008614958448753462,
      "loss": 2.3638,
      "step": 52500
    },
    {
      "epoch": 6.991162115815856,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0008601767576836829,
      "loss": 2.3601,
      "step": 53000
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.5313472747802734,
      "eval_runtime": 4.3987,
      "eval_samples_per_second": 454.683,
      "eval_steps_per_second": 1.819,
      "step": 53067
    },
    {
      "epoch": 7.057116475399024,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0008588576704920195,
      "loss": 2.3004,
      "step": 53500
    },
    {
      "epoch": 7.123070834982192,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0008575385833003562,
      "loss": 2.2926,
      "step": 54000
    },
    {
      "epoch": 7.189025194565361,
      "grad_norm": 0.734375,
      "learning_rate": 0.0008562194961086927,
      "loss": 2.2966,
      "step": 54500
    },
    {
      "epoch": 7.254979554148529,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0008549004089170295,
      "loss": 2.3004,
      "step": 55000
    },
    {
      "epoch": 7.3209339137316976,
      "grad_norm": 0.5859375,
      "learning_rate": 0.000853581321725366,
      "loss": 2.3168,
      "step": 55500
    },
    {
      "epoch": 7.386888273314866,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0008522622345337027,
      "loss": 2.3264,
      "step": 56000
    },
    {
      "epoch": 7.452842632898035,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0008509431473420393,
      "loss": 2.3132,
      "step": 56500
    },
    {
      "epoch": 7.518796992481203,
      "grad_norm": 0.625,
      "learning_rate": 0.000849624060150376,
      "loss": 2.3263,
      "step": 57000
    },
    {
      "epoch": 7.584751352064371,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0008483049729587125,
      "loss": 2.3318,
      "step": 57500
    },
    {
      "epoch": 7.650705711647539,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0008469858857670493,
      "loss": 2.3313,
      "step": 58000
    },
    {
      "epoch": 7.7166600712307085,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0008456667985753858,
      "loss": 2.3265,
      "step": 58500
    },
    {
      "epoch": 7.782614430813877,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0008443477113837225,
      "loss": 2.3401,
      "step": 59000
    },
    {
      "epoch": 7.848568790397045,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0008430286241920592,
      "loss": 2.3385,
      "step": 59500
    },
    {
      "epoch": 7.914523149980214,
      "grad_norm": 0.578125,
      "learning_rate": 0.0008417095370003958,
      "loss": 2.3224,
      "step": 60000
    },
    {
      "epoch": 7.980477509563382,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0008403904498087324,
      "loss": 2.3394,
      "step": 60500
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.508885383605957,
      "eval_runtime": 4.5718,
      "eval_samples_per_second": 437.467,
      "eval_steps_per_second": 1.75,
      "step": 60648
    },
    {
      "epoch": 8.04643186914655,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0008390713626170689,
      "loss": 2.2768,
      "step": 61000
    },
    {
      "epoch": 8.112386228729719,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0008377522754254057,
      "loss": 2.2642,
      "step": 61500
    },
    {
      "epoch": 8.178340588312887,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0008364331882337422,
      "loss": 2.2747,
      "step": 62000
    },
    {
      "epoch": 8.244294947896057,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0008351141010420789,
      "loss": 2.2778,
      "step": 62500
    },
    {
      "epoch": 8.310249307479225,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0008337950138504155,
      "loss": 2.2876,
      "step": 63000
    },
    {
      "epoch": 8.376203667062393,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0008324759266587522,
      "loss": 2.2983,
      "step": 63500
    },
    {
      "epoch": 8.442158026645561,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0008311568394670887,
      "loss": 2.3014,
      "step": 64000
    },
    {
      "epoch": 8.50811238622873,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0008298377522754255,
      "loss": 2.2997,
      "step": 64500
    },
    {
      "epoch": 8.574066745811898,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000828518665083762,
      "loss": 2.2962,
      "step": 65000
    },
    {
      "epoch": 8.640021105395066,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0008271995778920987,
      "loss": 2.2889,
      "step": 65500
    },
    {
      "epoch": 8.705975464978234,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0008258804907004353,
      "loss": 2.2984,
      "step": 66000
    },
    {
      "epoch": 8.771929824561404,
      "grad_norm": 0.59375,
      "learning_rate": 0.000824561403508772,
      "loss": 2.3166,
      "step": 66500
    },
    {
      "epoch": 8.837884184144572,
      "grad_norm": 0.73828125,
      "learning_rate": 0.0008232423163171085,
      "loss": 2.3091,
      "step": 67000
    },
    {
      "epoch": 8.90383854372774,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0008219232291254453,
      "loss": 2.2962,
      "step": 67500
    },
    {
      "epoch": 8.969792903310909,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0008206041419337818,
      "loss": 2.306,
      "step": 68000
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.5000669956207275,
      "eval_runtime": 4.9297,
      "eval_samples_per_second": 405.707,
      "eval_steps_per_second": 1.623,
      "step": 68229
    },
    {
      "epoch": 9.035747262894077,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0008192850547421185,
      "loss": 2.2644,
      "step": 68500
    },
    {
      "epoch": 9.101701622477245,
      "grad_norm": 0.6015625,
      "learning_rate": 0.000817965967550455,
      "loss": 2.2261,
      "step": 69000
    },
    {
      "epoch": 9.167655982060413,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0008166468803587918,
      "loss": 2.2395,
      "step": 69500
    },
    {
      "epoch": 9.233610341643583,
      "grad_norm": 0.625,
      "learning_rate": 0.0008153277931671284,
      "loss": 2.2461,
      "step": 70000
    },
    {
      "epoch": 9.299564701226751,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0008140087059754649,
      "loss": 2.256,
      "step": 70500
    },
    {
      "epoch": 9.36551906080992,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0008126896187838017,
      "loss": 2.2607,
      "step": 71000
    },
    {
      "epoch": 9.431473420393088,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0008113705315921382,
      "loss": 2.2576,
      "step": 71500
    },
    {
      "epoch": 9.497427779976256,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0008100514444004749,
      "loss": 2.2616,
      "step": 72000
    },
    {
      "epoch": 9.563382139559424,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0008087323572088115,
      "loss": 2.2711,
      "step": 72500
    },
    {
      "epoch": 9.629336499142592,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0008074132700171482,
      "loss": 2.2697,
      "step": 73000
    },
    {
      "epoch": 9.695290858725762,
      "grad_norm": 0.609375,
      "learning_rate": 0.0008060941828254847,
      "loss": 2.2548,
      "step": 73500
    },
    {
      "epoch": 9.76124521830893,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0008047750956338215,
      "loss": 2.2583,
      "step": 74000
    },
    {
      "epoch": 9.827199577892099,
      "grad_norm": 0.62109375,
      "learning_rate": 0.000803456008442158,
      "loss": 2.2598,
      "step": 74500
    },
    {
      "epoch": 9.893153937475267,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0008021369212504947,
      "loss": 2.2696,
      "step": 75000
    },
    {
      "epoch": 9.959108297058435,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0008008178340588313,
      "loss": 2.2692,
      "step": 75500
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.4777557849884033,
      "eval_runtime": 3.8042,
      "eval_samples_per_second": 525.734,
      "eval_steps_per_second": 2.103,
      "step": 75810
    },
    {
      "epoch": 10.025062656641603,
      "grad_norm": 0.625,
      "learning_rate": 0.000799498746867168,
      "loss": 2.2365,
      "step": 76000
    },
    {
      "epoch": 10.091017016224772,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0007981796596755045,
      "loss": 2.1978,
      "step": 76500
    },
    {
      "epoch": 10.156971375807942,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0007968605724838413,
      "loss": 2.2063,
      "step": 77000
    },
    {
      "epoch": 10.22292573539111,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0007955414852921778,
      "loss": 2.215,
      "step": 77500
    },
    {
      "epoch": 10.288880094974278,
      "grad_norm": 0.70703125,
      "learning_rate": 0.0007942223981005145,
      "loss": 2.222,
      "step": 78000
    },
    {
      "epoch": 10.354834454557446,
      "grad_norm": 0.75,
      "learning_rate": 0.000792903310908851,
      "loss": 2.2226,
      "step": 78500
    },
    {
      "epoch": 10.420788814140614,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0007915842237171877,
      "loss": 2.2292,
      "step": 79000
    },
    {
      "epoch": 10.486743173723783,
      "grad_norm": 0.609375,
      "learning_rate": 0.0007902651365255243,
      "loss": 2.2305,
      "step": 79500
    },
    {
      "epoch": 10.55269753330695,
      "grad_norm": 0.625,
      "learning_rate": 0.0007889460493338609,
      "loss": 2.2233,
      "step": 80000
    },
    {
      "epoch": 10.61865189289012,
      "grad_norm": 0.65625,
      "learning_rate": 0.0007876269621421977,
      "loss": 2.2378,
      "step": 80500
    },
    {
      "epoch": 10.684606252473289,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0007863078749505342,
      "loss": 2.2273,
      "step": 81000
    },
    {
      "epoch": 10.750560612056457,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0007849887877588709,
      "loss": 2.2363,
      "step": 81500
    },
    {
      "epoch": 10.816514971639625,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0007836697005672075,
      "loss": 2.2386,
      "step": 82000
    },
    {
      "epoch": 10.882469331222794,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0007823506133755442,
      "loss": 2.2348,
      "step": 82500
    },
    {
      "epoch": 10.948423690805962,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0007810315261838807,
      "loss": 2.2359,
      "step": 83000
    },
    {
      "epoch": 11.0,
      "eval_loss": 2.4718410968780518,
      "eval_runtime": 4.0713,
      "eval_samples_per_second": 491.242,
      "eval_steps_per_second": 1.965,
      "step": 83391
    },
    {
      "epoch": 11.01437805038913,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0007797124389922175,
      "loss": 2.2184,
      "step": 83500
    },
    {
      "epoch": 11.0803324099723,
      "grad_norm": 0.6015625,
      "learning_rate": 0.000778393351800554,
      "loss": 2.1719,
      "step": 84000
    },
    {
      "epoch": 11.146286769555468,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0007770742646088907,
      "loss": 2.1817,
      "step": 84500
    },
    {
      "epoch": 11.212241129138636,
      "grad_norm": 0.8125,
      "learning_rate": 0.0007757551774172273,
      "loss": 2.185,
      "step": 85000
    },
    {
      "epoch": 11.278195488721805,
      "grad_norm": 0.62890625,
      "learning_rate": 0.000774436090225564,
      "loss": 2.1905,
      "step": 85500
    },
    {
      "epoch": 11.344149848304973,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0007731170030339005,
      "loss": 2.1953,
      "step": 86000
    },
    {
      "epoch": 11.410104207888141,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0007717979158422373,
      "loss": 2.1873,
      "step": 86500
    },
    {
      "epoch": 11.47605856747131,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0007704788286505738,
      "loss": 2.2014,
      "step": 87000
    },
    {
      "epoch": 11.54201292705448,
      "grad_norm": 0.703125,
      "learning_rate": 0.0007691597414589104,
      "loss": 2.2029,
      "step": 87500
    },
    {
      "epoch": 11.607967286637647,
      "grad_norm": 0.69140625,
      "learning_rate": 0.000767840654267247,
      "loss": 2.2041,
      "step": 88000
    },
    {
      "epoch": 11.673921646220816,
      "grad_norm": 0.59375,
      "learning_rate": 0.0007665215670755837,
      "loss": 2.2031,
      "step": 88500
    },
    {
      "epoch": 11.739876005803984,
      "grad_norm": 0.71875,
      "learning_rate": 0.0007652024798839203,
      "loss": 2.2059,
      "step": 89000
    },
    {
      "epoch": 11.805830365387152,
      "grad_norm": 0.65625,
      "learning_rate": 0.0007638833926922569,
      "loss": 2.21,
      "step": 89500
    },
    {
      "epoch": 11.87178472497032,
      "grad_norm": 0.6875,
      "learning_rate": 0.0007625643055005937,
      "loss": 2.2105,
      "step": 90000
    },
    {
      "epoch": 11.937739084553488,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0007612452183089302,
      "loss": 2.2089,
      "step": 90500
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.453855514526367,
      "eval_runtime": 3.8561,
      "eval_samples_per_second": 518.662,
      "eval_steps_per_second": 2.075,
      "step": 90972
    },
    {
      "epoch": 12.003693444136658,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0007599261311172669,
      "loss": 2.2131,
      "step": 91000
    },
    {
      "epoch": 12.069647803719826,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0007586070439256035,
      "loss": 2.1413,
      "step": 91500
    },
    {
      "epoch": 12.135602163302995,
      "grad_norm": 0.640625,
      "learning_rate": 0.0007572879567339402,
      "loss": 2.1529,
      "step": 92000
    },
    {
      "epoch": 12.201556522886163,
      "grad_norm": 0.609375,
      "learning_rate": 0.0007559688695422767,
      "loss": 2.1531,
      "step": 92500
    },
    {
      "epoch": 12.267510882469331,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0007546497823506135,
      "loss": 2.1672,
      "step": 93000
    },
    {
      "epoch": 12.3334652420525,
      "grad_norm": 0.6484375,
      "learning_rate": 0.00075333069515895,
      "loss": 2.1653,
      "step": 93500
    },
    {
      "epoch": 12.399419601635667,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0007520116079672867,
      "loss": 2.1711,
      "step": 94000
    },
    {
      "epoch": 12.465373961218837,
      "grad_norm": 0.59375,
      "learning_rate": 0.0007506925207756233,
      "loss": 2.1787,
      "step": 94500
    },
    {
      "epoch": 12.531328320802006,
      "grad_norm": 0.60546875,
      "learning_rate": 0.00074937343358396,
      "loss": 2.1737,
      "step": 95000
    },
    {
      "epoch": 12.597282680385174,
      "grad_norm": 0.65625,
      "learning_rate": 0.0007480543463922965,
      "loss": 2.1787,
      "step": 95500
    },
    {
      "epoch": 12.663237039968342,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0007467352592006333,
      "loss": 2.1764,
      "step": 96000
    },
    {
      "epoch": 12.72919139955151,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0007454161720089698,
      "loss": 2.1902,
      "step": 96500
    },
    {
      "epoch": 12.795145759134678,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0007440970848173064,
      "loss": 2.1778,
      "step": 97000
    },
    {
      "epoch": 12.861100118717847,
      "grad_norm": 0.61328125,
      "learning_rate": 0.000742777997625643,
      "loss": 2.1791,
      "step": 97500
    },
    {
      "epoch": 12.927054478301017,
      "grad_norm": 0.875,
      "learning_rate": 0.0007414589104339797,
      "loss": 2.194,
      "step": 98000
    },
    {
      "epoch": 12.993008837884185,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0007401398232423163,
      "loss": 2.1869,
      "step": 98500
    },
    {
      "epoch": 13.0,
      "eval_loss": 2.4518699645996094,
      "eval_runtime": 4.8206,
      "eval_samples_per_second": 414.883,
      "eval_steps_per_second": 1.66,
      "step": 98553
    },
    {
      "epoch": 13.058963197467353,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0007388207360506529,
      "loss": 2.1188,
      "step": 99000
    },
    {
      "epoch": 13.124917557050521,
      "grad_norm": 0.640625,
      "learning_rate": 0.0007375016488589896,
      "loss": 2.1277,
      "step": 99500
    },
    {
      "epoch": 13.19087191663369,
      "grad_norm": 0.734375,
      "learning_rate": 0.0007361825616673262,
      "loss": 2.139,
      "step": 100000
    },
    {
      "epoch": 13.256826276216858,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0007348634744756628,
      "loss": 2.1469,
      "step": 100500
    },
    {
      "epoch": 13.322780635800026,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0007335443872839995,
      "loss": 2.1406,
      "step": 101000
    },
    {
      "epoch": 13.388734995383194,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0007322253000923362,
      "loss": 2.1448,
      "step": 101500
    },
    {
      "epoch": 13.454689354966364,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0007309062129006727,
      "loss": 2.1424,
      "step": 102000
    },
    {
      "epoch": 13.520643714549532,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0007295871257090095,
      "loss": 2.1505,
      "step": 102500
    },
    {
      "epoch": 13.5865980741327,
      "grad_norm": 0.7578125,
      "learning_rate": 0.000728268038517346,
      "loss": 2.1533,
      "step": 103000
    },
    {
      "epoch": 13.652552433715869,
      "grad_norm": 0.671875,
      "learning_rate": 0.0007269489513256827,
      "loss": 2.1655,
      "step": 103500
    },
    {
      "epoch": 13.718506793299037,
      "grad_norm": 0.609375,
      "learning_rate": 0.0007256298641340193,
      "loss": 2.1599,
      "step": 104000
    },
    {
      "epoch": 13.784461152882205,
      "grad_norm": 0.59765625,
      "learning_rate": 0.000724310776942356,
      "loss": 2.1582,
      "step": 104500
    },
    {
      "epoch": 13.850415512465373,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0007229916897506925,
      "loss": 2.1664,
      "step": 105000
    },
    {
      "epoch": 13.916369872048543,
      "grad_norm": 0.65625,
      "learning_rate": 0.0007216726025590291,
      "loss": 2.1638,
      "step": 105500
    },
    {
      "epoch": 13.982324231631711,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0007203535153673658,
      "loss": 2.1643,
      "step": 106000
    },
    {
      "epoch": 14.0,
      "eval_loss": 2.442619562149048,
      "eval_runtime": 5.7628,
      "eval_samples_per_second": 347.051,
      "eval_steps_per_second": 1.388,
      "step": 106134
    },
    {
      "epoch": 14.04827859121488,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0007190344281757024,
      "loss": 2.1108,
      "step": 106500
    },
    {
      "epoch": 14.114232950798048,
      "grad_norm": 0.67578125,
      "learning_rate": 0.000717715340984039,
      "loss": 2.1076,
      "step": 107000
    },
    {
      "epoch": 14.180187310381216,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0007163962537923757,
      "loss": 2.1202,
      "step": 107500
    },
    {
      "epoch": 14.246141669964384,
      "grad_norm": 0.625,
      "learning_rate": 0.0007150771666007123,
      "loss": 2.1221,
      "step": 108000
    },
    {
      "epoch": 14.312096029547552,
      "grad_norm": 0.625,
      "learning_rate": 0.0007137580794090489,
      "loss": 2.1262,
      "step": 108500
    },
    {
      "epoch": 14.378050389130722,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0007124389922173856,
      "loss": 2.1376,
      "step": 109000
    },
    {
      "epoch": 14.44400474871389,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0007111199050257222,
      "loss": 2.1344,
      "step": 109500
    },
    {
      "epoch": 14.509959108297059,
      "grad_norm": 0.65625,
      "learning_rate": 0.0007098008178340588,
      "loss": 2.1379,
      "step": 110000
    },
    {
      "epoch": 14.575913467880227,
      "grad_norm": 0.671875,
      "learning_rate": 0.0007084817306423955,
      "loss": 2.1354,
      "step": 110500
    },
    {
      "epoch": 14.641867827463395,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0007071626434507322,
      "loss": 2.1339,
      "step": 111000
    },
    {
      "epoch": 14.707822187046563,
      "grad_norm": 0.65625,
      "learning_rate": 0.0007058435562590687,
      "loss": 2.1297,
      "step": 111500
    },
    {
      "epoch": 14.773776546629731,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0007045244690674055,
      "loss": 2.1355,
      "step": 112000
    },
    {
      "epoch": 14.839730906212901,
      "grad_norm": 0.6640625,
      "learning_rate": 0.000703205381875742,
      "loss": 2.1385,
      "step": 112500
    },
    {
      "epoch": 14.90568526579607,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0007018862946840787,
      "loss": 2.1399,
      "step": 113000
    },
    {
      "epoch": 14.971639625379238,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0007005672074924153,
      "loss": 2.149,
      "step": 113500
    },
    {
      "epoch": 15.0,
      "eval_loss": 2.43693208694458,
      "eval_runtime": 4.8208,
      "eval_samples_per_second": 414.869,
      "eval_steps_per_second": 1.659,
      "step": 113715
    },
    {
      "epoch": 15.037593984962406,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0006992481203007519,
      "loss": 2.1078,
      "step": 114000
    },
    {
      "epoch": 15.103548344545574,
      "grad_norm": 0.6875,
      "learning_rate": 0.0006979290331090885,
      "loss": 2.0965,
      "step": 114500
    },
    {
      "epoch": 15.169502704128742,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0006966099459174251,
      "loss": 2.1037,
      "step": 115000
    },
    {
      "epoch": 15.23545706371191,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0006952908587257618,
      "loss": 2.1074,
      "step": 115500
    },
    {
      "epoch": 15.30141142329508,
      "grad_norm": 0.734375,
      "learning_rate": 0.0006939717715340984,
      "loss": 2.1103,
      "step": 116000
    },
    {
      "epoch": 15.367365782878249,
      "grad_norm": 0.69140625,
      "learning_rate": 0.000692652684342435,
      "loss": 2.1089,
      "step": 116500
    },
    {
      "epoch": 15.433320142461417,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0006913335971507717,
      "loss": 2.1088,
      "step": 117000
    },
    {
      "epoch": 15.499274502044585,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0006900145099591083,
      "loss": 2.1112,
      "step": 117500
    },
    {
      "epoch": 15.565228861627753,
      "grad_norm": 0.609375,
      "learning_rate": 0.0006886954227674449,
      "loss": 2.1153,
      "step": 118000
    },
    {
      "epoch": 15.631183221210922,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0006873763355757816,
      "loss": 2.1321,
      "step": 118500
    },
    {
      "epoch": 15.69713758079409,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0006860572483841182,
      "loss": 2.1233,
      "step": 119000
    },
    {
      "epoch": 15.763091940377258,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0006847381611924548,
      "loss": 2.1361,
      "step": 119500
    },
    {
      "epoch": 15.829046299960428,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0006834190740007915,
      "loss": 2.1296,
      "step": 120000
    },
    {
      "epoch": 15.895000659543596,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0006820999868091281,
      "loss": 2.1236,
      "step": 120500
    },
    {
      "epoch": 15.960955019126764,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0006807808996174647,
      "loss": 2.1354,
      "step": 121000
    },
    {
      "epoch": 16.0,
      "eval_loss": 2.431525945663452,
      "eval_runtime": 5.3462,
      "eval_samples_per_second": 374.095,
      "eval_steps_per_second": 1.496,
      "step": 121296
    },
    {
      "epoch": 16.026909378709934,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0006794618124258015,
      "loss": 2.1031,
      "step": 121500
    },
    {
      "epoch": 16.0928637382931,
      "grad_norm": 0.66015625,
      "learning_rate": 0.000678142725234138,
      "loss": 2.0653,
      "step": 122000
    },
    {
      "epoch": 16.15881809787627,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0006768236380424745,
      "loss": 2.0796,
      "step": 122500
    },
    {
      "epoch": 16.224772457459437,
      "grad_norm": 0.796875,
      "learning_rate": 0.0006755045508508113,
      "loss": 2.0916,
      "step": 123000
    },
    {
      "epoch": 16.290726817042607,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0006741854636591479,
      "loss": 2.0888,
      "step": 123500
    },
    {
      "epoch": 16.356681176625774,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0006728663764674845,
      "loss": 2.0981,
      "step": 124000
    },
    {
      "epoch": 16.422635536208944,
      "grad_norm": 0.65625,
      "learning_rate": 0.0006715472892758211,
      "loss": 2.1025,
      "step": 124500
    },
    {
      "epoch": 16.488589895792114,
      "grad_norm": 0.640625,
      "learning_rate": 0.0006702282020841578,
      "loss": 2.0996,
      "step": 125000
    },
    {
      "epoch": 16.55454425537528,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0006689091148924944,
      "loss": 2.103,
      "step": 125500
    },
    {
      "epoch": 16.62049861495845,
      "grad_norm": 0.73828125,
      "learning_rate": 0.000667590027700831,
      "loss": 2.1105,
      "step": 126000
    },
    {
      "epoch": 16.686452974541616,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0006662709405091677,
      "loss": 2.1138,
      "step": 126500
    },
    {
      "epoch": 16.752407334124786,
      "grad_norm": 0.65625,
      "learning_rate": 0.0006649518533175043,
      "loss": 2.1095,
      "step": 127000
    },
    {
      "epoch": 16.818361693707953,
      "grad_norm": 0.640625,
      "learning_rate": 0.0006636327661258409,
      "loss": 2.1114,
      "step": 127500
    },
    {
      "epoch": 16.884316053291123,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0006623136789341776,
      "loss": 2.1012,
      "step": 128000
    },
    {
      "epoch": 16.950270412874293,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0006609945917425142,
      "loss": 2.109,
      "step": 128500
    },
    {
      "epoch": 17.0,
      "eval_loss": 2.4235010147094727,
      "eval_runtime": 5.2236,
      "eval_samples_per_second": 382.874,
      "eval_steps_per_second": 1.531,
      "step": 128877
    },
    {
      "epoch": 17.01622477245746,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0006596755045508508,
      "loss": 2.0946,
      "step": 129000
    },
    {
      "epoch": 17.08217913204063,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0006583564173591875,
      "loss": 2.0571,
      "step": 129500
    },
    {
      "epoch": 17.148133491623796,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0006570373301675241,
      "loss": 2.0727,
      "step": 130000
    },
    {
      "epoch": 17.214087851206965,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0006557182429758607,
      "loss": 2.0699,
      "step": 130500
    },
    {
      "epoch": 17.280042210790132,
      "grad_norm": 0.6953125,
      "learning_rate": 0.0006543991557841973,
      "loss": 2.0763,
      "step": 131000
    },
    {
      "epoch": 17.345996570373302,
      "grad_norm": 0.7265625,
      "learning_rate": 0.000653080068592534,
      "loss": 2.0804,
      "step": 131500
    },
    {
      "epoch": 17.411950929956472,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0006517609814008705,
      "loss": 2.0832,
      "step": 132000
    },
    {
      "epoch": 17.47790528953964,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0006504418942092073,
      "loss": 2.0821,
      "step": 132500
    },
    {
      "epoch": 17.54385964912281,
      "grad_norm": 0.6875,
      "learning_rate": 0.0006491228070175439,
      "loss": 2.0821,
      "step": 133000
    },
    {
      "epoch": 17.609814008705975,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0006478037198258805,
      "loss": 2.0886,
      "step": 133500
    },
    {
      "epoch": 17.675768368289145,
      "grad_norm": 0.66015625,
      "learning_rate": 0.000646484632634217,
      "loss": 2.0945,
      "step": 134000
    },
    {
      "epoch": 17.74172272787231,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0006451655454425538,
      "loss": 2.0918,
      "step": 134500
    },
    {
      "epoch": 17.80767708745548,
      "grad_norm": 0.72265625,
      "learning_rate": 0.0006438464582508904,
      "loss": 2.099,
      "step": 135000
    },
    {
      "epoch": 17.873631447038647,
      "grad_norm": 0.68359375,
      "learning_rate": 0.000642527371059227,
      "loss": 2.097,
      "step": 135500
    },
    {
      "epoch": 17.939585806621817,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0006412082838675637,
      "loss": 2.0928,
      "step": 136000
    },
    {
      "epoch": 18.0,
      "eval_loss": 2.4190404415130615,
      "eval_runtime": 4.6336,
      "eval_samples_per_second": 431.626,
      "eval_steps_per_second": 1.727,
      "step": 136458
    }
  ],
  "logging_steps": 500,
  "max_steps": 379050,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8932082563541893e+18,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}