{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.22943673282092464,
  "eval_steps": 10,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001147183664104623,
      "eval_loss": 1.873344898223877,
      "eval_runtime": 12.7647,
      "eval_samples_per_second": 515.797,
      "eval_steps_per_second": 8.069,
      "step": 10
    },
    {
      "epoch": 0.002294367328209246,
      "eval_loss": 1.8726389408111572,
      "eval_runtime": 12.8667,
      "eval_samples_per_second": 511.709,
      "eval_steps_per_second": 8.005,
      "step": 20
    },
    {
      "epoch": 0.0034415509923138693,
      "eval_loss": 1.8714078664779663,
      "eval_runtime": 12.9103,
      "eval_samples_per_second": 509.979,
      "eval_steps_per_second": 7.978,
      "step": 30
    },
    {
      "epoch": 0.004588734656418492,
      "eval_loss": 1.8696790933609009,
      "eval_runtime": 12.947,
      "eval_samples_per_second": 508.534,
      "eval_steps_per_second": 7.955,
      "step": 40
    },
    {
      "epoch": 0.0057359183205231154,
      "eval_loss": 1.8675329685211182,
      "eval_runtime": 12.9458,
      "eval_samples_per_second": 508.582,
      "eval_steps_per_second": 7.956,
      "step": 50
    },
    {
      "epoch": 0.006883101984627739,
      "eval_loss": 1.8649154901504517,
      "eval_runtime": 13.0432,
      "eval_samples_per_second": 504.785,
      "eval_steps_per_second": 7.897,
      "step": 60
    },
    {
      "epoch": 0.008030285648732363,
      "eval_loss": 1.8619294166564941,
      "eval_runtime": 13.0638,
      "eval_samples_per_second": 503.988,
      "eval_steps_per_second": 7.884,
      "step": 70
    },
    {
      "epoch": 0.009177469312836984,
      "eval_loss": 1.8583979606628418,
      "eval_runtime": 13.0482,
      "eval_samples_per_second": 504.592,
      "eval_steps_per_second": 7.894,
      "step": 80
    },
    {
      "epoch": 0.010324652976941608,
      "eval_loss": 1.85438871383667,
      "eval_runtime": 13.0615,
      "eval_samples_per_second": 504.075,
      "eval_steps_per_second": 7.886,
      "step": 90
    },
    {
      "epoch": 0.011471836641046231,
      "grad_norm": 9.938580513000488,
      "learning_rate": 3.8226299694189603e-07,
      "loss": 3.1046,
      "step": 100
    },
    {
      "epoch": 0.011471836641046231,
      "eval_loss": 1.849947214126587,
      "eval_runtime": 13.0663,
      "eval_samples_per_second": 503.89,
      "eval_steps_per_second": 7.883,
      "step": 100
    },
    {
      "epoch": 0.012619020305150854,
      "eval_loss": 1.8451412916183472,
      "eval_runtime": 12.9771,
      "eval_samples_per_second": 507.357,
      "eval_steps_per_second": 7.937,
      "step": 110
    },
    {
      "epoch": 0.013766203969255477,
      "eval_loss": 1.8399487733840942,
      "eval_runtime": 13.0209,
      "eval_samples_per_second": 505.648,
      "eval_steps_per_second": 7.91,
      "step": 120
    },
    {
      "epoch": 0.0149133876333601,
      "eval_loss": 1.8342881202697754,
      "eval_runtime": 13.0369,
      "eval_samples_per_second": 505.028,
      "eval_steps_per_second": 7.901,
      "step": 130
    },
    {
      "epoch": 0.016060571297464726,
      "eval_loss": 1.8283486366271973,
      "eval_runtime": 13.0149,
      "eval_samples_per_second": 505.88,
      "eval_steps_per_second": 7.914,
      "step": 140
    },
    {
      "epoch": 0.017207754961569347,
      "eval_loss": 1.822334885597229,
      "eval_runtime": 13.0213,
      "eval_samples_per_second": 505.632,
      "eval_steps_per_second": 7.91,
      "step": 150
    },
    {
      "epoch": 0.01835493862567397,
      "eval_loss": 1.8158738613128662,
      "eval_runtime": 13.0599,
      "eval_samples_per_second": 504.14,
      "eval_steps_per_second": 7.887,
      "step": 160
    },
    {
      "epoch": 0.019502122289778594,
      "eval_loss": 1.8090614080429077,
      "eval_runtime": 13.034,
      "eval_samples_per_second": 505.14,
      "eval_steps_per_second": 7.902,
      "step": 170
    },
    {
      "epoch": 0.020649305953883215,
      "eval_loss": 1.8015782833099365,
      "eval_runtime": 13.0665,
      "eval_samples_per_second": 503.885,
      "eval_steps_per_second": 7.883,
      "step": 180
    },
    {
      "epoch": 0.02179648961798784,
      "eval_loss": 1.793796420097351,
      "eval_runtime": 13.0555,
      "eval_samples_per_second": 504.31,
      "eval_steps_per_second": 7.889,
      "step": 190
    },
    {
      "epoch": 0.022943673282092462,
      "grad_norm": 4.906337738037109,
      "learning_rate": 7.645259938837921e-07,
      "loss": 3.0303,
      "step": 200
    },
    {
      "epoch": 0.022943673282092462,
      "eval_loss": 1.785815715789795,
      "eval_runtime": 12.9925,
      "eval_samples_per_second": 506.754,
      "eval_steps_per_second": 7.928,
      "step": 200
    },
    {
      "epoch": 0.024090856946197087,
      "eval_loss": 1.7775053977966309,
      "eval_runtime": 13.0639,
      "eval_samples_per_second": 503.986,
      "eval_steps_per_second": 7.884,
      "step": 210
    },
    {
      "epoch": 0.025238040610301708,
      "eval_loss": 1.7692992687225342,
      "eval_runtime": 13.0129,
      "eval_samples_per_second": 505.96,
      "eval_steps_per_second": 7.915,
      "step": 220
    },
    {
      "epoch": 0.026385224274406333,
      "eval_loss": 1.760453224182129,
      "eval_runtime": 13.0078,
      "eval_samples_per_second": 506.158,
      "eval_steps_per_second": 7.918,
      "step": 230
    },
    {
      "epoch": 0.027532407938510955,
      "eval_loss": 1.751396656036377,
      "eval_runtime": 12.9957,
      "eval_samples_per_second": 506.628,
      "eval_steps_per_second": 7.926,
      "step": 240
    },
    {
      "epoch": 0.02867959160261558,
      "eval_loss": 1.7417218685150146,
      "eval_runtime": 12.9774,
      "eval_samples_per_second": 507.344,
      "eval_steps_per_second": 7.937,
      "step": 250
    },
    {
      "epoch": 0.0298267752667202,
      "eval_loss": 1.7319914102554321,
      "eval_runtime": 13.0219,
      "eval_samples_per_second": 505.611,
      "eval_steps_per_second": 7.91,
      "step": 260
    },
    {
      "epoch": 0.030973958930824826,
      "eval_loss": 1.7227253913879395,
      "eval_runtime": 13.0026,
      "eval_samples_per_second": 506.361,
      "eval_steps_per_second": 7.922,
      "step": 270
    },
    {
      "epoch": 0.03212114259492945,
      "eval_loss": 1.7133797407150269,
      "eval_runtime": 12.9757,
      "eval_samples_per_second": 507.409,
      "eval_steps_per_second": 7.938,
      "step": 280
    },
    {
      "epoch": 0.03326832625903407,
      "eval_loss": 1.704041600227356,
      "eval_runtime": 12.9845,
      "eval_samples_per_second": 507.065,
      "eval_steps_per_second": 7.933,
      "step": 290
    },
    {
      "epoch": 0.034415509923138694,
      "grad_norm": 4.665822505950928,
      "learning_rate": 1.1467889908256882e-06,
      "loss": 2.9459,
      "step": 300
    },
    {
      "epoch": 0.034415509923138694,
      "eval_loss": 1.6940686702728271,
      "eval_runtime": 13.0019,
      "eval_samples_per_second": 506.387,
      "eval_steps_per_second": 7.922,
      "step": 300
    },
    {
      "epoch": 0.035562693587243316,
      "eval_loss": 1.683342695236206,
      "eval_runtime": 13.0065,
      "eval_samples_per_second": 506.209,
      "eval_steps_per_second": 7.919,
      "step": 310
    },
    {
      "epoch": 0.03670987725134794,
      "eval_loss": 1.6724653244018555,
      "eval_runtime": 13.0129,
      "eval_samples_per_second": 505.96,
      "eval_steps_per_second": 7.915,
      "step": 320
    },
    {
      "epoch": 0.037857060915452566,
      "eval_loss": 1.6614341735839844,
      "eval_runtime": 12.9921,
      "eval_samples_per_second": 506.769,
      "eval_steps_per_second": 7.928,
      "step": 330
    },
    {
      "epoch": 0.03900424457955719,
      "eval_loss": 1.6510112285614014,
      "eval_runtime": 13.0242,
      "eval_samples_per_second": 505.52,
      "eval_steps_per_second": 7.908,
      "step": 340
    },
    {
      "epoch": 0.04015142824366181,
      "eval_loss": 1.6401513814926147,
      "eval_runtime": 12.9214,
      "eval_samples_per_second": 509.542,
      "eval_steps_per_second": 7.971,
      "step": 350
    },
    {
      "epoch": 0.04129861190776643,
      "eval_loss": 1.6295816898345947,
      "eval_runtime": 12.9563,
      "eval_samples_per_second": 508.171,
      "eval_steps_per_second": 7.95,
      "step": 360
    },
    {
      "epoch": 0.04244579557187106,
      "eval_loss": 1.6187150478363037,
      "eval_runtime": 12.9758,
      "eval_samples_per_second": 507.405,
      "eval_steps_per_second": 7.938,
      "step": 370
    },
    {
      "epoch": 0.04359297923597568,
      "eval_loss": 1.607272982597351,
      "eval_runtime": 12.9876,
      "eval_samples_per_second": 506.947,
      "eval_steps_per_second": 7.931,
      "step": 380
    },
    {
      "epoch": 0.0447401629000803,
      "eval_loss": 1.5961676836013794,
      "eval_runtime": 12.9782,
      "eval_samples_per_second": 507.313,
      "eval_steps_per_second": 7.936,
      "step": 390
    },
    {
      "epoch": 0.045887346564184923,
      "grad_norm": 4.870114326477051,
      "learning_rate": 1.5290519877675841e-06,
      "loss": 2.7813,
      "step": 400
    },
    {
      "epoch": 0.045887346564184923,
      "eval_loss": 1.5848218202590942,
      "eval_runtime": 12.9783,
      "eval_samples_per_second": 507.309,
      "eval_steps_per_second": 7.936,
      "step": 400
    },
    {
      "epoch": 0.04703453022828955,
      "eval_loss": 1.5734797716140747,
      "eval_runtime": 12.9739,
      "eval_samples_per_second": 507.482,
      "eval_steps_per_second": 7.939,
      "step": 410
    },
    {
      "epoch": 0.04818171389239417,
      "eval_loss": 1.562021255493164,
      "eval_runtime": 12.9388,
      "eval_samples_per_second": 508.855,
      "eval_steps_per_second": 7.961,
      "step": 420
    },
    {
      "epoch": 0.049328897556498795,
      "eval_loss": 1.5495364665985107,
      "eval_runtime": 12.9412,
      "eval_samples_per_second": 508.764,
      "eval_steps_per_second": 7.959,
      "step": 430
    },
    {
      "epoch": 0.050476081220603417,
      "eval_loss": 1.5375314950942993,
      "eval_runtime": 12.9686,
      "eval_samples_per_second": 507.687,
      "eval_steps_per_second": 7.942,
      "step": 440
    },
    {
      "epoch": 0.051623264884708045,
      "eval_loss": 1.525598168373108,
      "eval_runtime": 12.9695,
      "eval_samples_per_second": 507.651,
      "eval_steps_per_second": 7.942,
      "step": 450
    },
    {
      "epoch": 0.052770448548812667,
      "eval_loss": 1.5132672786712646,
      "eval_runtime": 12.8961,
      "eval_samples_per_second": 510.543,
      "eval_steps_per_second": 7.987,
      "step": 460
    },
    {
      "epoch": 0.05391763221291729,
      "eval_loss": 1.5012215375900269,
      "eval_runtime": 12.9428,
      "eval_samples_per_second": 508.7,
      "eval_steps_per_second": 7.958,
      "step": 470
    },
    {
      "epoch": 0.05506481587702191,
      "eval_loss": 1.4892219305038452,
      "eval_runtime": 12.9208,
      "eval_samples_per_second": 509.567,
      "eval_steps_per_second": 7.972,
      "step": 480
    },
    {
      "epoch": 0.05621199954112653,
      "eval_loss": 1.4768636226654053,
      "eval_runtime": 12.9423,
      "eval_samples_per_second": 508.721,
      "eval_steps_per_second": 7.958,
      "step": 490
    },
    {
      "epoch": 0.05735918320523116,
      "grad_norm": 4.155641555786133,
      "learning_rate": 1.9113149847094803e-06,
      "loss": 2.6308,
      "step": 500
    },
    {
      "epoch": 0.05735918320523116,
      "eval_loss": 1.4640088081359863,
      "eval_runtime": 12.8729,
      "eval_samples_per_second": 511.462,
      "eval_steps_per_second": 8.001,
      "step": 500
    },
    {
      "epoch": 0.05850636686933578,
      "eval_loss": 1.4513096809387207,
      "eval_runtime": 12.9653,
      "eval_samples_per_second": 507.817,
      "eval_steps_per_second": 7.944,
      "step": 510
    },
    {
      "epoch": 0.0596535505334404,
      "eval_loss": 1.439149260520935,
      "eval_runtime": 12.9443,
      "eval_samples_per_second": 508.639,
      "eval_steps_per_second": 7.957,
      "step": 520
    },
    {
      "epoch": 0.060800734197545024,
      "eval_loss": 1.426237940788269,
      "eval_runtime": 12.9496,
      "eval_samples_per_second": 508.433,
      "eval_steps_per_second": 7.954,
      "step": 530
    },
    {
      "epoch": 0.06194791786164965,
      "eval_loss": 1.4129557609558105,
      "eval_runtime": 12.9822,
      "eval_samples_per_second": 507.157,
      "eval_steps_per_second": 7.934,
      "step": 540
    },
    {
      "epoch": 0.06309510152575427,
      "eval_loss": 1.3997886180877686,
      "eval_runtime": 12.9979,
      "eval_samples_per_second": 506.542,
      "eval_steps_per_second": 7.924,
      "step": 550
    },
    {
      "epoch": 0.0642422851898589,
      "eval_loss": 1.3873906135559082,
      "eval_runtime": 12.9378,
      "eval_samples_per_second": 508.895,
      "eval_steps_per_second": 7.961,
      "step": 560
    },
    {
      "epoch": 0.06538946885396352,
      "eval_loss": 1.3751789331436157,
      "eval_runtime": 12.9624,
      "eval_samples_per_second": 507.932,
      "eval_steps_per_second": 7.946,
      "step": 570
    },
    {
      "epoch": 0.06653665251806815,
      "eval_loss": 1.3620370626449585,
      "eval_runtime": 12.9498,
      "eval_samples_per_second": 508.426,
      "eval_steps_per_second": 7.954,
      "step": 580
    },
    {
      "epoch": 0.06768383618217276,
      "eval_loss": 1.3485124111175537,
      "eval_runtime": 12.9759,
      "eval_samples_per_second": 507.403,
      "eval_steps_per_second": 7.938,
      "step": 590
    },
    {
      "epoch": 0.06883101984627739,
      "grad_norm": 5.262124061584473,
      "learning_rate": 2.2935779816513764e-06,
      "loss": 2.4452,
      "step": 600
    },
    {
      "epoch": 0.06883101984627739,
      "eval_loss": 1.3349775075912476,
      "eval_runtime": 12.9631,
      "eval_samples_per_second": 507.902,
      "eval_steps_per_second": 7.946,
      "step": 600
    },
    {
      "epoch": 0.06997820351038202,
      "eval_loss": 1.3213400840759277,
      "eval_runtime": 12.9619,
      "eval_samples_per_second": 507.951,
      "eval_steps_per_second": 7.946,
      "step": 610
    },
    {
      "epoch": 0.07112538717448663,
      "eval_loss": 1.308822512626648,
      "eval_runtime": 12.9652,
      "eval_samples_per_second": 507.82,
      "eval_steps_per_second": 7.944,
      "step": 620
    },
    {
      "epoch": 0.07227257083859126,
      "eval_loss": 1.296485185623169,
      "eval_runtime": 13.0441,
      "eval_samples_per_second": 504.75,
      "eval_steps_per_second": 7.896,
      "step": 630
    },
    {
      "epoch": 0.07341975450269587,
      "eval_loss": 1.283867597579956,
      "eval_runtime": 12.9822,
      "eval_samples_per_second": 507.154,
      "eval_steps_per_second": 7.934,
      "step": 640
    },
    {
      "epoch": 0.0745669381668005,
      "eval_loss": 1.2713148593902588,
      "eval_runtime": 12.9775,
      "eval_samples_per_second": 507.338,
      "eval_steps_per_second": 7.937,
      "step": 650
    },
    {
      "epoch": 0.07571412183090513,
      "eval_loss": 1.2591922283172607,
      "eval_runtime": 13.0227,
      "eval_samples_per_second": 505.578,
      "eval_steps_per_second": 7.909,
      "step": 660
    },
    {
      "epoch": 0.07686130549500975,
      "eval_loss": 1.246610164642334,
      "eval_runtime": 13.014,
      "eval_samples_per_second": 505.917,
      "eval_steps_per_second": 7.915,
      "step": 670
    },
    {
      "epoch": 0.07800848915911437,
      "eval_loss": 1.2331972122192383,
      "eval_runtime": 12.9634,
      "eval_samples_per_second": 507.891,
      "eval_steps_per_second": 7.945,
      "step": 680
    },
    {
      "epoch": 0.079155672823219,
      "eval_loss": 1.2203081846237183,
      "eval_runtime": 12.9664,
      "eval_samples_per_second": 507.775,
      "eval_steps_per_second": 7.944,
      "step": 690
    },
    {
      "epoch": 0.08030285648732362,
      "grad_norm": 3.824066400527954,
      "learning_rate": 2.6758409785932725e-06,
      "loss": 2.2626,
      "step": 700
    },
    {
      "epoch": 0.08030285648732362,
      "eval_loss": 1.207729697227478,
      "eval_runtime": 13.037,
      "eval_samples_per_second": 505.025,
      "eval_steps_per_second": 7.901,
      "step": 700
    },
    {
      "epoch": 0.08145004015142825,
      "eval_loss": 1.195885419845581,
      "eval_runtime": 12.9625,
      "eval_samples_per_second": 507.928,
      "eval_steps_per_second": 7.946,
      "step": 710
    },
    {
      "epoch": 0.08259722381553286,
      "eval_loss": 1.1840639114379883,
      "eval_runtime": 13.0124,
      "eval_samples_per_second": 505.98,
      "eval_steps_per_second": 7.916,
      "step": 720
    },
    {
      "epoch": 0.08374440747963749,
      "eval_loss": 1.1725258827209473,
      "eval_runtime": 12.9851,
      "eval_samples_per_second": 507.043,
      "eval_steps_per_second": 7.932,
      "step": 730
    },
    {
      "epoch": 0.08489159114374212,
      "eval_loss": 1.1619255542755127,
      "eval_runtime": 13.1041,
      "eval_samples_per_second": 502.438,
      "eval_steps_per_second": 7.86,
      "step": 740
    },
    {
      "epoch": 0.08603877480784673,
      "eval_loss": 1.1515777111053467,
      "eval_runtime": 12.9648,
      "eval_samples_per_second": 507.837,
      "eval_steps_per_second": 7.945,
      "step": 750
    },
    {
      "epoch": 0.08718595847195136,
      "eval_loss": 1.1416434049606323,
      "eval_runtime": 13.0349,
      "eval_samples_per_second": 505.104,
      "eval_steps_per_second": 7.902,
      "step": 760
    },
    {
      "epoch": 0.08833314213605599,
      "eval_loss": 1.1320044994354248,
      "eval_runtime": 13.0667,
      "eval_samples_per_second": 503.875,
      "eval_steps_per_second": 7.883,
      "step": 770
    },
    {
      "epoch": 0.0894803258001606,
      "eval_loss": 1.1226693391799927,
      "eval_runtime": 13.3034,
      "eval_samples_per_second": 494.912,
      "eval_steps_per_second": 7.742,
      "step": 780
    },
    {
      "epoch": 0.09062750946426523,
      "eval_loss": 1.113792061805725,
      "eval_runtime": 12.9793,
      "eval_samples_per_second": 507.271,
      "eval_steps_per_second": 7.936,
      "step": 790
    },
    {
      "epoch": 0.09177469312836985,
      "grad_norm": 3.358140707015991,
      "learning_rate": 3.0581039755351682e-06,
      "loss": 2.0044,
      "step": 800
    },
    {
      "epoch": 0.09177469312836985,
      "eval_loss": 1.1052675247192383,
      "eval_runtime": 13.0193,
      "eval_samples_per_second": 505.709,
      "eval_steps_per_second": 7.911,
      "step": 800
    },
    {
      "epoch": 0.09292187679247448,
      "eval_loss": 1.0964922904968262,
      "eval_runtime": 13.1289,
      "eval_samples_per_second": 501.489,
      "eval_steps_per_second": 7.845,
      "step": 810
    },
    {
      "epoch": 0.0940690604565791,
      "eval_loss": 1.0879426002502441,
      "eval_runtime": 12.9727,
      "eval_samples_per_second": 507.526,
      "eval_steps_per_second": 7.94,
      "step": 820
    },
    {
      "epoch": 0.09521624412068372,
      "eval_loss": 1.0795575380325317,
      "eval_runtime": 13.1138,
      "eval_samples_per_second": 502.068,
      "eval_steps_per_second": 7.854,
      "step": 830
    },
    {
      "epoch": 0.09636342778478835,
      "eval_loss": 1.0718028545379639,
      "eval_runtime": 12.9813,
      "eval_samples_per_second": 507.191,
      "eval_steps_per_second": 7.934,
      "step": 840
    },
    {
      "epoch": 0.09751061144889296,
      "eval_loss": 1.0643888711929321,
      "eval_runtime": 13.0171,
      "eval_samples_per_second": 505.795,
      "eval_steps_per_second": 7.913,
      "step": 850
    },
    {
      "epoch": 0.09865779511299759,
      "eval_loss": 1.0564391613006592,
      "eval_runtime": 13.0169,
      "eval_samples_per_second": 505.806,
      "eval_steps_per_second": 7.913,
      "step": 860
    },
    {
      "epoch": 0.09980497877710222,
      "eval_loss": 1.0490267276763916,
      "eval_runtime": 13.0703,
      "eval_samples_per_second": 503.738,
      "eval_steps_per_second": 7.88,
      "step": 870
    },
    {
      "epoch": 0.10095216244120683,
      "eval_loss": 1.0417358875274658,
      "eval_runtime": 13.0303,
      "eval_samples_per_second": 505.284,
      "eval_steps_per_second": 7.905,
      "step": 880
    },
    {
      "epoch": 0.10209934610531146,
      "eval_loss": 1.0353840589523315,
      "eval_runtime": 12.9765,
      "eval_samples_per_second": 507.381,
      "eval_steps_per_second": 7.937,
      "step": 890
    },
    {
      "epoch": 0.10324652976941609,
      "grad_norm": 3.705247402191162,
      "learning_rate": 3.4403669724770644e-06,
      "loss": 1.8763,
      "step": 900
    },
    {
      "epoch": 0.10324652976941609,
      "eval_loss": 1.0295778512954712,
      "eval_runtime": 13.0396,
      "eval_samples_per_second": 504.924,
      "eval_steps_per_second": 7.899,
      "step": 900
    },
    {
      "epoch": 0.1043937134335207,
      "eval_loss": 1.0238651037216187,
      "eval_runtime": 12.973,
      "eval_samples_per_second": 507.515,
      "eval_steps_per_second": 7.94,
      "step": 910
    },
    {
      "epoch": 0.10554089709762533,
      "eval_loss": 1.0180109739303589,
      "eval_runtime": 13.0196,
      "eval_samples_per_second": 505.7,
      "eval_steps_per_second": 7.911,
      "step": 920
    },
    {
      "epoch": 0.10668808076172995,
      "eval_loss": 1.0122654438018799,
      "eval_runtime": 12.9877,
      "eval_samples_per_second": 506.942,
      "eval_steps_per_second": 7.931,
      "step": 930
    },
    {
      "epoch": 0.10783526442583458,
      "eval_loss": 1.0065423250198364,
      "eval_runtime": 13.0039,
      "eval_samples_per_second": 506.312,
      "eval_steps_per_second": 7.921,
      "step": 940
    },
    {
      "epoch": 0.1089824480899392,
      "eval_loss": 1.0008338689804077,
      "eval_runtime": 13.1129,
      "eval_samples_per_second": 502.101,
      "eval_steps_per_second": 7.855,
      "step": 950
    },
    {
      "epoch": 0.11012963175404382,
      "eval_loss": 0.9950375556945801,
      "eval_runtime": 13.0191,
      "eval_samples_per_second": 505.718,
      "eval_steps_per_second": 7.911,
      "step": 960
    },
    {
      "epoch": 0.11127681541814845,
      "eval_loss": 0.9893631935119629,
      "eval_runtime": 12.9794,
      "eval_samples_per_second": 507.267,
      "eval_steps_per_second": 7.936,
      "step": 970
    },
    {
      "epoch": 0.11242399908225306,
      "eval_loss": 0.9840025305747986,
      "eval_runtime": 12.9799,
      "eval_samples_per_second": 507.247,
      "eval_steps_per_second": 7.935,
      "step": 980
    },
    {
      "epoch": 0.11357118274635769,
      "eval_loss": 0.9792994856834412,
      "eval_runtime": 12.9866,
      "eval_samples_per_second": 506.983,
      "eval_steps_per_second": 7.931,
      "step": 990
    },
    {
      "epoch": 0.11471836641046232,
      "grad_norm": 4.240755558013916,
      "learning_rate": 3.8226299694189605e-06,
      "loss": 1.7287,
      "step": 1000
    },
    {
      "epoch": 0.11471836641046232,
      "eval_loss": 0.9751574993133545,
      "eval_runtime": 13.0083,
      "eval_samples_per_second": 506.136,
      "eval_steps_per_second": 7.918,
      "step": 1000
    },
    {
      "epoch": 0.11586555007456693,
      "eval_loss": 0.9705988764762878,
      "eval_runtime": 12.9835,
      "eval_samples_per_second": 507.107,
      "eval_steps_per_second": 7.933,
      "step": 1010
    },
    {
      "epoch": 0.11701273373867156,
      "eval_loss": 0.9658572673797607,
      "eval_runtime": 13.0627,
      "eval_samples_per_second": 504.031,
      "eval_steps_per_second": 7.885,
      "step": 1020
    },
    {
      "epoch": 0.11815991740277619,
      "eval_loss": 0.9614543318748474,
      "eval_runtime": 13.0169,
      "eval_samples_per_second": 505.803,
      "eval_steps_per_second": 7.913,
      "step": 1030
    },
    {
      "epoch": 0.1193071010668808,
      "eval_loss": 0.9571945667266846,
      "eval_runtime": 12.9762,
      "eval_samples_per_second": 507.389,
      "eval_steps_per_second": 7.938,
      "step": 1040
    },
    {
      "epoch": 0.12045428473098543,
      "eval_loss": 0.9531480073928833,
      "eval_runtime": 13.0036,
      "eval_samples_per_second": 506.323,
      "eval_steps_per_second": 7.921,
      "step": 1050
    },
    {
      "epoch": 0.12160146839509005,
      "eval_loss": 0.9493557810783386,
      "eval_runtime": 13.0046,
      "eval_samples_per_second": 506.282,
      "eval_steps_per_second": 7.92,
      "step": 1060
    },
    {
      "epoch": 0.12274865205919468,
      "eval_loss": 0.9455849528312683,
      "eval_runtime": 13.0676,
      "eval_samples_per_second": 503.843,
      "eval_steps_per_second": 7.882,
      "step": 1070
    },
    {
      "epoch": 0.1238958357232993,
      "eval_loss": 0.9415374398231506,
      "eval_runtime": 13.0698,
      "eval_samples_per_second": 503.755,
      "eval_steps_per_second": 7.881,
      "step": 1080
    },
    {
      "epoch": 0.12504301938740392,
      "eval_loss": 0.9376588463783264,
      "eval_runtime": 12.9843,
      "eval_samples_per_second": 507.072,
      "eval_steps_per_second": 7.933,
      "step": 1090
    },
    {
      "epoch": 0.12619020305150855,
      "grad_norm": 3.3081679344177246,
      "learning_rate": 4.204892966360857e-06,
      "loss": 1.6312,
      "step": 1100
    },
    {
      "epoch": 0.12619020305150855,
      "eval_loss": 0.9338813424110413,
      "eval_runtime": 13.0365,
      "eval_samples_per_second": 505.043,
      "eval_steps_per_second": 7.901,
      "step": 1100
    },
    {
      "epoch": 0.12733738671561318,
      "eval_loss": 0.9302825927734375,
      "eval_runtime": 13.0692,
      "eval_samples_per_second": 503.779,
      "eval_steps_per_second": 7.881,
      "step": 1110
    },
    {
      "epoch": 0.1284845703797178,
      "eval_loss": 0.9267016053199768,
      "eval_runtime": 13.0455,
      "eval_samples_per_second": 504.694,
      "eval_steps_per_second": 7.895,
      "step": 1120
    },
    {
      "epoch": 0.1296317540438224,
      "eval_loss": 0.9232119917869568,
      "eval_runtime": 13.0051,
      "eval_samples_per_second": 506.262,
      "eval_steps_per_second": 7.92,
      "step": 1130
    },
    {
      "epoch": 0.13077893770792703,
      "eval_loss": 0.9197220206260681,
      "eval_runtime": 13.0069,
      "eval_samples_per_second": 506.192,
      "eval_steps_per_second": 7.919,
      "step": 1140
    },
    {
      "epoch": 0.13192612137203166,
      "eval_loss": 0.9162309765815735,
      "eval_runtime": 12.9851,
      "eval_samples_per_second": 507.044,
      "eval_steps_per_second": 7.932,
      "step": 1150
    },
    {
      "epoch": 0.1330733050361363,
      "eval_loss": 0.9127652049064636,
      "eval_runtime": 12.9768,
      "eval_samples_per_second": 507.366,
      "eval_steps_per_second": 7.937,
      "step": 1160
    },
    {
      "epoch": 0.13422048870024092,
      "eval_loss": 0.9096914529800415,
      "eval_runtime": 12.9953,
      "eval_samples_per_second": 506.645,
      "eval_steps_per_second": 7.926,
      "step": 1170
    },
    {
      "epoch": 0.13536767236434552,
      "eval_loss": 0.9069137573242188,
      "eval_runtime": 13.0109,
      "eval_samples_per_second": 506.038,
      "eval_steps_per_second": 7.916,
      "step": 1180
    },
    {
      "epoch": 0.13651485602845015,
      "eval_loss": 0.9039744734764099,
      "eval_runtime": 12.9772,
      "eval_samples_per_second": 507.352,
      "eval_steps_per_second": 7.937,
      "step": 1190
    },
    {
      "epoch": 0.13766203969255478,
      "grad_norm": 3.420077085494995,
      "learning_rate": 4.587155963302753e-06,
      "loss": 1.5316,
      "step": 1200
    },
    {
      "epoch": 0.13766203969255478,
      "eval_loss": 0.9009912610054016,
      "eval_runtime": 12.9848,
      "eval_samples_per_second": 507.056,
      "eval_steps_per_second": 7.932,
      "step": 1200
    },
    {
      "epoch": 0.1388092233566594,
      "eval_loss": 0.8979319930076599,
      "eval_runtime": 13.0295,
      "eval_samples_per_second": 505.313,
      "eval_steps_per_second": 7.905,
      "step": 1210
    },
    {
      "epoch": 0.13995640702076403,
      "eval_loss": 0.8946565985679626,
      "eval_runtime": 13.0119,
      "eval_samples_per_second": 505.997,
      "eval_steps_per_second": 7.916,
      "step": 1220
    },
    {
      "epoch": 0.14110359068486863,
      "eval_loss": 0.8915460705757141,
      "eval_runtime": 12.9979,
      "eval_samples_per_second": 506.545,
      "eval_steps_per_second": 7.924,
      "step": 1230
    },
    {
      "epoch": 0.14225077434897326,
      "eval_loss": 0.8887993097305298,
      "eval_runtime": 12.9633,
      "eval_samples_per_second": 507.895,
      "eval_steps_per_second": 7.945,
      "step": 1240
    },
    {
      "epoch": 0.1433979580130779,
      "eval_loss": 0.886054277420044,
      "eval_runtime": 12.9996,
      "eval_samples_per_second": 506.477,
      "eval_steps_per_second": 7.923,
      "step": 1250
    },
    {
      "epoch": 0.14454514167718252,
      "eval_loss": 0.8833284974098206,
      "eval_runtime": 13.1133,
      "eval_samples_per_second": 502.086,
      "eval_steps_per_second": 7.855,
      "step": 1260
    },
    {
      "epoch": 0.14569232534128715,
      "eval_loss": 0.8806452751159668,
      "eval_runtime": 13.0208,
      "eval_samples_per_second": 505.654,
      "eval_steps_per_second": 7.91,
      "step": 1270
    },
    {
      "epoch": 0.14683950900539175,
      "eval_loss": 0.8778645992279053,
      "eval_runtime": 13.1071,
      "eval_samples_per_second": 502.322,
      "eval_steps_per_second": 7.858,
      "step": 1280
    },
    {
      "epoch": 0.14798669266949638,
      "eval_loss": 0.8747658133506775,
      "eval_runtime": 13.0362,
      "eval_samples_per_second": 505.055,
      "eval_steps_per_second": 7.901,
      "step": 1290
    },
    {
      "epoch": 0.149133876333601,
      "grad_norm": 3.3882789611816406,
      "learning_rate": 4.969418960244649e-06,
      "loss": 1.4961,
      "step": 1300
    },
    {
      "epoch": 0.149133876333601,
      "eval_loss": 0.8717615008354187,
      "eval_runtime": 13.0104,
      "eval_samples_per_second": 506.058,
      "eval_steps_per_second": 7.917,
      "step": 1300
    },
    {
      "epoch": 0.15028105999770563,
      "eval_loss": 0.8690453767776489,
      "eval_runtime": 13.0189,
      "eval_samples_per_second": 505.727,
      "eval_steps_per_second": 7.912,
      "step": 1310
    },
    {
      "epoch": 0.15142824366181026,
      "eval_loss": 0.8664180040359497,
      "eval_runtime": 12.988,
      "eval_samples_per_second": 506.931,
      "eval_steps_per_second": 7.93,
      "step": 1320
    },
    {
      "epoch": 0.1525754273259149,
      "eval_loss": 0.8634527921676636,
      "eval_runtime": 12.9897,
      "eval_samples_per_second": 506.861,
      "eval_steps_per_second": 7.929,
      "step": 1330
    },
    {
      "epoch": 0.1537226109900195,
      "eval_loss": 0.8603318929672241,
      "eval_runtime": 13.0269,
      "eval_samples_per_second": 505.417,
      "eval_steps_per_second": 7.907,
      "step": 1340
    },
    {
      "epoch": 0.15486979465412412,
      "eval_loss": 0.8573653697967529,
      "eval_runtime": 12.9945,
      "eval_samples_per_second": 506.677,
      "eval_steps_per_second": 7.926,
      "step": 1350
    },
    {
      "epoch": 0.15601697831822875,
      "eval_loss": 0.8545491695404053,
      "eval_runtime": 12.9981,
      "eval_samples_per_second": 506.536,
      "eval_steps_per_second": 7.924,
      "step": 1360
    },
    {
      "epoch": 0.15716416198233338,
      "eval_loss": 0.85209721326828,
      "eval_runtime": 13.0301,
      "eval_samples_per_second": 505.293,
      "eval_steps_per_second": 7.905,
      "step": 1370
    },
    {
      "epoch": 0.158311345646438,
      "eval_loss": 0.8497452139854431,
      "eval_runtime": 13.1229,
      "eval_samples_per_second": 501.717,
      "eval_steps_per_second": 7.849,
      "step": 1380
    },
    {
      "epoch": 0.1594585293105426,
      "eval_loss": 0.8474416136741638,
      "eval_runtime": 13.0624,
      "eval_samples_per_second": 504.043,
      "eval_steps_per_second": 7.885,
      "step": 1390
    },
    {
      "epoch": 0.16060571297464724,
      "grad_norm": 3.7449283599853516,
      "learning_rate": 5.351681957186545e-06,
      "loss": 1.451,
      "step": 1400
    },
    {
      "epoch": 0.16060571297464724,
      "eval_loss": 0.8452581763267517,
      "eval_runtime": 13.0191,
      "eval_samples_per_second": 505.72,
      "eval_steps_per_second": 7.911,
      "step": 1400
    },
    {
      "epoch": 0.16175289663875186,
      "eval_loss": 0.8428558111190796,
      "eval_runtime": 13.0038,
      "eval_samples_per_second": 506.314,
      "eval_steps_per_second": 7.921,
      "step": 1410
    },
    {
      "epoch": 0.1629000803028565,
      "eval_loss": 0.8404139876365662,
      "eval_runtime": 12.9904,
      "eval_samples_per_second": 506.837,
      "eval_steps_per_second": 7.929,
      "step": 1420
    },
    {
      "epoch": 0.16404726396696112,
      "eval_loss": 0.8380302786827087,
      "eval_runtime": 12.9812,
      "eval_samples_per_second": 507.195,
      "eval_steps_per_second": 7.935,
      "step": 1430
    },
    {
      "epoch": 0.16519444763106572,
      "eval_loss": 0.8357470631599426,
      "eval_runtime": 13.0241,
      "eval_samples_per_second": 505.526,
      "eval_steps_per_second": 7.908,
      "step": 1440
    },
    {
      "epoch": 0.16634163129517035,
      "eval_loss": 0.8336036205291748,
      "eval_runtime": 13.142,
      "eval_samples_per_second": 500.99,
      "eval_steps_per_second": 7.837,
      "step": 1450
    },
    {
      "epoch": 0.16748881495927498,
      "eval_loss": 0.8312162756919861,
      "eval_runtime": 12.9935,
      "eval_samples_per_second": 506.714,
      "eval_steps_per_second": 7.927,
      "step": 1460
    },
    {
      "epoch": 0.1686359986233796,
      "eval_loss": 0.8288681507110596,
      "eval_runtime": 13.1018,
      "eval_samples_per_second": 502.527,
      "eval_steps_per_second": 7.862,
      "step": 1470
    },
    {
      "epoch": 0.16978318228748424,
      "eval_loss": 0.8261700868606567,
      "eval_runtime": 13.0327,
      "eval_samples_per_second": 505.19,
      "eval_steps_per_second": 7.903,
      "step": 1480
    },
    {
      "epoch": 0.17093036595158884,
      "eval_loss": 0.8235771656036377,
      "eval_runtime": 12.9895,
      "eval_samples_per_second": 506.873,
      "eval_steps_per_second": 7.93,
      "step": 1490
    },
    {
      "epoch": 0.17207754961569346,
      "grad_norm": 2.5695531368255615,
      "learning_rate": 5.733944954128441e-06,
      "loss": 1.4177,
      "step": 1500
    },
    {
      "epoch": 0.17207754961569346,
      "eval_loss": 0.8213275074958801,
      "eval_runtime": 13.0468,
      "eval_samples_per_second": 504.647,
      "eval_steps_per_second": 7.895,
      "step": 1500
    },
    {
      "epoch": 0.1732247332797981,
      "eval_loss": 0.8189388513565063,
      "eval_runtime": 12.9877,
      "eval_samples_per_second": 506.943,
      "eval_steps_per_second": 7.931,
      "step": 1510
    },
    {
      "epoch": 0.17437191694390272,
      "eval_loss": 0.8168380856513977,
      "eval_runtime": 12.9915,
      "eval_samples_per_second": 506.793,
      "eval_steps_per_second": 7.928,
      "step": 1520
    },
    {
      "epoch": 0.17551910060800735,
      "eval_loss": 0.814733624458313,
      "eval_runtime": 13.1726,
      "eval_samples_per_second": 499.824,
      "eval_steps_per_second": 7.819,
      "step": 1530
    },
    {
      "epoch": 0.17666628427211198,
      "eval_loss": 0.8127268552780151,
      "eval_runtime": 13.0207,
      "eval_samples_per_second": 505.657,
      "eval_steps_per_second": 7.91,
      "step": 1540
    },
    {
      "epoch": 0.17781346793621658,
      "eval_loss": 0.8106749057769775,
      "eval_runtime": 13.0973,
      "eval_samples_per_second": 502.697,
      "eval_steps_per_second": 7.864,
      "step": 1550
    },
    {
      "epoch": 0.1789606516003212,
      "eval_loss": 0.808172345161438,
      "eval_runtime": 13.2116,
      "eval_samples_per_second": 498.349,
      "eval_steps_per_second": 7.796,
      "step": 1560
    },
    {
      "epoch": 0.18010783526442584,
      "eval_loss": 0.8059112429618835,
      "eval_runtime": 12.9871,
      "eval_samples_per_second": 506.963,
      "eval_steps_per_second": 7.931,
      "step": 1570
    },
    {
      "epoch": 0.18125501892853046,
      "eval_loss": 0.8035888671875,
      "eval_runtime": 13.0963,
      "eval_samples_per_second": 502.737,
      "eval_steps_per_second": 7.865,
      "step": 1580
    },
    {
      "epoch": 0.1824022025926351,
      "eval_loss": 0.8014644384384155,
      "eval_runtime": 13.0028,
      "eval_samples_per_second": 506.351,
      "eval_steps_per_second": 7.921,
      "step": 1590
    },
    {
      "epoch": 0.1835493862567397,
      "grad_norm": 2.681454658508301,
      "learning_rate": 6.1162079510703365e-06,
      "loss": 1.3734,
      "step": 1600
    },
    {
      "epoch": 0.1835493862567397,
      "eval_loss": 0.7993160486221313,
      "eval_runtime": 12.9893,
      "eval_samples_per_second": 506.879,
      "eval_steps_per_second": 7.93,
      "step": 1600
    },
    {
      "epoch": 0.18469656992084432,
      "eval_loss": 0.7969831228256226,
      "eval_runtime": 13.0137,
      "eval_samples_per_second": 505.929,
      "eval_steps_per_second": 7.915,
      "step": 1610
    },
    {
      "epoch": 0.18584375358494895,
      "eval_loss": 0.7947937250137329,
      "eval_runtime": 13.0467,
      "eval_samples_per_second": 504.648,
      "eval_steps_per_second": 7.895,
      "step": 1620
    },
    {
      "epoch": 0.18699093724905358,
      "eval_loss": 0.7922271490097046,
      "eval_runtime": 13.1133,
      "eval_samples_per_second": 502.086,
      "eval_steps_per_second": 7.855,
      "step": 1630
    },
    {
      "epoch": 0.1881381209131582,
      "eval_loss": 0.790046751499176,
      "eval_runtime": 13.0687,
      "eval_samples_per_second": 503.801,
      "eval_steps_per_second": 7.881,
      "step": 1640
    },
    {
      "epoch": 0.1892853045772628,
      "eval_loss": 0.7877430319786072,
      "eval_runtime": 13.1712,
      "eval_samples_per_second": 499.877,
      "eval_steps_per_second": 7.82,
      "step": 1650
    },
    {
      "epoch": 0.19043248824136744,
      "eval_loss": 0.7852274179458618,
      "eval_runtime": 13.0721,
      "eval_samples_per_second": 503.669,
      "eval_steps_per_second": 7.879,
      "step": 1660
    },
    {
      "epoch": 0.19157967190547207,
      "eval_loss": 0.782863438129425,
      "eval_runtime": 12.9741,
      "eval_samples_per_second": 507.471,
      "eval_steps_per_second": 7.939,
      "step": 1670
    },
    {
      "epoch": 0.1927268555695767,
      "eval_loss": 0.7804363965988159,
      "eval_runtime": 12.9826,
      "eval_samples_per_second": 507.142,
      "eval_steps_per_second": 7.934,
      "step": 1680
    },
    {
      "epoch": 0.19387403923368132,
      "eval_loss": 0.7779432535171509,
      "eval_runtime": 12.9731,
      "eval_samples_per_second": 507.514,
      "eval_steps_per_second": 7.94,
      "step": 1690
    },
    {
      "epoch": 0.19502122289778592,
      "grad_norm": 3.1423346996307373,
      "learning_rate": 6.4984709480122335e-06,
      "loss": 1.3327,
      "step": 1700
    },
    {
      "epoch": 0.19502122289778592,
      "eval_loss": 0.7756665945053101,
      "eval_runtime": 12.9767,
      "eval_samples_per_second": 507.371,
      "eval_steps_per_second": 7.937,
      "step": 1700
    },
    {
      "epoch": 0.19616840656189055,
      "eval_loss": 0.7737661004066467,
      "eval_runtime": 12.9633,
      "eval_samples_per_second": 507.896,
      "eval_steps_per_second": 7.946,
      "step": 1710
    },
    {
      "epoch": 0.19731559022599518,
      "eval_loss": 0.7719039916992188,
      "eval_runtime": 12.9798,
      "eval_samples_per_second": 507.248,
      "eval_steps_per_second": 7.935,
      "step": 1720
    },
    {
      "epoch": 0.1984627738900998,
      "eval_loss": 0.7699739933013916,
      "eval_runtime": 12.9817,
      "eval_samples_per_second": 507.177,
      "eval_steps_per_second": 7.934,
      "step": 1730
    },
    {
      "epoch": 0.19960995755420444,
      "eval_loss": 0.7679038643836975,
      "eval_runtime": 13.0593,
      "eval_samples_per_second": 504.164,
      "eval_steps_per_second": 7.887,
      "step": 1740
    },
    {
      "epoch": 0.20075714121830904,
      "eval_loss": 0.765801191329956,
      "eval_runtime": 12.9827,
      "eval_samples_per_second": 507.136,
      "eval_steps_per_second": 7.934,
      "step": 1750
    },
    {
      "epoch": 0.20190432488241367,
      "eval_loss": 0.7640795707702637,
      "eval_runtime": 12.9885,
      "eval_samples_per_second": 506.908,
      "eval_steps_per_second": 7.93,
      "step": 1760
    },
    {
      "epoch": 0.2030515085465183,
      "eval_loss": 0.7621497511863708,
      "eval_runtime": 12.9872,
      "eval_samples_per_second": 506.96,
      "eval_steps_per_second": 7.931,
      "step": 1770
    },
    {
      "epoch": 0.20419869221062292,
      "eval_loss": 0.7601312398910522,
      "eval_runtime": 12.9805,
      "eval_samples_per_second": 507.223,
      "eval_steps_per_second": 7.935,
      "step": 1780
    },
    {
      "epoch": 0.20534587587472755,
      "eval_loss": 0.7580214738845825,
      "eval_runtime": 12.9849,
      "eval_samples_per_second": 507.05,
      "eval_steps_per_second": 7.932,
      "step": 1790
    },
    {
      "epoch": 0.20649305953883218,
      "grad_norm": 3.4953625202178955,
      "learning_rate": 6.880733944954129e-06,
      "loss": 1.2804,
      "step": 1800
    },
    {
      "epoch": 0.20649305953883218,
      "eval_loss": 0.7558028697967529,
      "eval_runtime": 12.989,
      "eval_samples_per_second": 506.889,
      "eval_steps_per_second": 7.93,
      "step": 1800
    },
    {
      "epoch": 0.20764024320293678,
      "eval_loss": 0.7535884380340576,
      "eval_runtime": 12.9795,
      "eval_samples_per_second": 507.262,
      "eval_steps_per_second": 7.936,
      "step": 1810
    },
    {
      "epoch": 0.2087874268670414,
      "eval_loss": 0.7514472603797913,
      "eval_runtime": 12.9901,
      "eval_samples_per_second": 506.846,
      "eval_steps_per_second": 7.929,
      "step": 1820
    },
    {
      "epoch": 0.20993461053114604,
      "eval_loss": 0.7493338584899902,
      "eval_runtime": 13.0011,
      "eval_samples_per_second": 506.418,
      "eval_steps_per_second": 7.922,
      "step": 1830
    },
    {
      "epoch": 0.21108179419525067,
      "eval_loss": 0.7473368644714355,
      "eval_runtime": 13.0424,
      "eval_samples_per_second": 504.815,
      "eval_steps_per_second": 7.897,
      "step": 1840
    },
    {
      "epoch": 0.2122289778593553,
      "eval_loss": 0.7450571060180664,
      "eval_runtime": 13.1033,
      "eval_samples_per_second": 502.468,
      "eval_steps_per_second": 7.861,
      "step": 1850
    },
    {
      "epoch": 0.2133761615234599,
      "eval_loss": 0.7429091930389404,
      "eval_runtime": 12.9903,
      "eval_samples_per_second": 506.841,
      "eval_steps_per_second": 7.929,
      "step": 1860
    },
    {
      "epoch": 0.21452334518756452,
      "eval_loss": 0.7408153414726257,
      "eval_runtime": 12.983,
      "eval_samples_per_second": 507.126,
      "eval_steps_per_second": 7.933,
      "step": 1870
    },
    {
      "epoch": 0.21567052885166915,
      "eval_loss": 0.7389461994171143,
      "eval_runtime": 12.9922,
      "eval_samples_per_second": 506.767,
      "eval_steps_per_second": 7.928,
      "step": 1880
    },
    {
      "epoch": 0.21681771251577378,
      "eval_loss": 0.7368388772010803,
      "eval_runtime": 13.0283,
      "eval_samples_per_second": 505.36,
      "eval_steps_per_second": 7.906,
      "step": 1890
    },
    {
      "epoch": 0.2179648961798784,
      "grad_norm": 2.869717597961426,
      "learning_rate": 7.262996941896026e-06,
      "loss": 1.2255,
      "step": 1900
    },
    {
      "epoch": 0.2179648961798784,
      "eval_loss": 0.73488450050354,
      "eval_runtime": 12.9924,
      "eval_samples_per_second": 506.759,
      "eval_steps_per_second": 7.928,
      "step": 1900
    },
    {
      "epoch": 0.219112079843983,
      "eval_loss": 0.7328305244445801,
      "eval_runtime": 12.9892,
      "eval_samples_per_second": 506.881,
      "eval_steps_per_second": 7.93,
      "step": 1910
    },
    {
      "epoch": 0.22025926350808764,
      "eval_loss": 0.7309767007827759,
      "eval_runtime": 13.1016,
      "eval_samples_per_second": 502.534,
      "eval_steps_per_second": 7.862,
      "step": 1920
    },
    {
      "epoch": 0.22140644717219227,
      "eval_loss": 0.7293325662612915,
      "eval_runtime": 12.9943,
      "eval_samples_per_second": 506.684,
      "eval_steps_per_second": 7.927,
      "step": 1930
    },
    {
      "epoch": 0.2225536308362969,
      "eval_loss": 0.7277292013168335,
      "eval_runtime": 13.0122,
      "eval_samples_per_second": 505.986,
      "eval_steps_per_second": 7.916,
      "step": 1940
    },
    {
      "epoch": 0.22370081450040152,
      "eval_loss": 0.7258683443069458,
      "eval_runtime": 13.0042,
      "eval_samples_per_second": 506.298,
      "eval_steps_per_second": 7.921,
      "step": 1950
    },
    {
      "epoch": 0.22484799816450612,
      "eval_loss": 0.7239750027656555,
      "eval_runtime": 13.0619,
      "eval_samples_per_second": 504.061,
      "eval_steps_per_second": 7.886,
      "step": 1960
    },
    {
      "epoch": 0.22599518182861075,
      "eval_loss": 0.7221319079399109,
      "eval_runtime": 12.9944,
      "eval_samples_per_second": 506.679,
      "eval_steps_per_second": 7.926,
      "step": 1970
    },
    {
      "epoch": 0.22714236549271538,
      "eval_loss": 0.720324695110321,
      "eval_runtime": 12.982,
      "eval_samples_per_second": 507.165,
      "eval_steps_per_second": 7.934,
      "step": 1980
    },
    {
      "epoch": 0.22828954915682,
      "eval_loss": 0.7184233069419861,
      "eval_runtime": 12.9988,
      "eval_samples_per_second": 506.507,
      "eval_steps_per_second": 7.924,
      "step": 1990
    },
    {
      "epoch": 0.22943673282092464,
      "grad_norm": 9.275369644165039,
      "learning_rate": 7.645259938837921e-06,
      "loss": 1.2635,
      "step": 2000
    },
    {
      "epoch": 0.22943673282092464,
      "eval_loss": 0.7165008187294006,
      "eval_runtime": 12.9966,
      "eval_samples_per_second": 506.594,
      "eval_steps_per_second": 7.925,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 26151,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}