{
  "best_metric": 0.7726819541375872,
  "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuned-lora-medmnistv2/checkpoint-1090",
  "epoch": 9.954337899543379,
  "eval_steps": 500,
  "global_step": 1090,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "grad_norm": 4.391805171966553,
      "learning_rate": 0.004954128440366973,
      "loss": 1.2554,
      "step": 10
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.9513263702392578,
      "learning_rate": 0.004908256880733945,
      "loss": 1.0074,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 3.8831961154937744,
      "learning_rate": 0.004862385321100918,
      "loss": 0.9068,
      "step": 30
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.7356910705566406,
      "learning_rate": 0.00481651376146789,
      "loss": 0.9577,
      "step": 40
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.5790963172912598,
      "learning_rate": 0.0047706422018348625,
      "loss": 0.9124,
      "step": 50
    },
    {
      "epoch": 0.55,
      "grad_norm": 2.810718297958374,
      "learning_rate": 0.004724770642201835,
      "loss": 0.8901,
      "step": 60
    },
    {
      "epoch": 0.64,
      "grad_norm": 2.3134875297546387,
      "learning_rate": 0.004678899082568808,
      "loss": 0.9111,
      "step": 70
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.8082305192947388,
      "learning_rate": 0.00463302752293578,
      "loss": 0.9059,
      "step": 80
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.4370285272598267,
      "learning_rate": 0.0045871559633027525,
      "loss": 0.9984,
      "step": 90
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.8687102794647217,
      "learning_rate": 0.004541284403669725,
      "loss": 0.9135,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7198404785643071,
      "eval_f1": 0.3049533355905189,
      "eval_loss": 0.7698342800140381,
      "eval_precision": 0.5179078489450006,
      "eval_recall": 0.3103063518754385,
      "eval_runtime": 5.826,
      "eval_samples_per_second": 172.159,
      "eval_steps_per_second": 10.814,
      "step": 109
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.885165810585022,
      "learning_rate": 0.004495412844036698,
      "loss": 0.8622,
      "step": 110
    },
    {
      "epoch": 1.1,
      "grad_norm": 2.4616353511810303,
      "learning_rate": 0.0044495412844036695,
      "loss": 0.9407,
      "step": 120
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.2185616493225098,
      "learning_rate": 0.004403669724770643,
      "loss": 0.8853,
      "step": 130
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.0212063789367676,
      "learning_rate": 0.004357798165137615,
      "loss": 0.8534,
      "step": 140
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.129385232925415,
      "learning_rate": 0.004311926605504587,
      "loss": 0.8613,
      "step": 150
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.726841926574707,
      "learning_rate": 0.0042660550458715595,
      "loss": 0.8104,
      "step": 160
    },
    {
      "epoch": 1.55,
      "grad_norm": 2.272818088531494,
      "learning_rate": 0.004220183486238533,
      "loss": 0.915,
      "step": 170
    },
    {
      "epoch": 1.64,
      "grad_norm": 1.117639422416687,
      "learning_rate": 0.004174311926605505,
      "loss": 0.8753,
      "step": 180
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.8981102108955383,
      "learning_rate": 0.004128440366972477,
      "loss": 0.8429,
      "step": 190
    },
    {
      "epoch": 1.83,
      "grad_norm": 1.282863974571228,
      "learning_rate": 0.00408256880733945,
      "loss": 0.8355,
      "step": 200
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.9718325734138489,
      "learning_rate": 0.004036697247706422,
      "loss": 0.8352,
      "step": 210
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7298105682951147,
      "eval_f1": 0.3883733866690214,
      "eval_loss": 0.7352049350738525,
      "eval_precision": 0.536225724752637,
      "eval_recall": 0.42308831430611293,
      "eval_runtime": 5.7891,
      "eval_samples_per_second": 173.258,
      "eval_steps_per_second": 10.883,
      "step": 219
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.3424675464630127,
      "learning_rate": 0.003990825688073394,
      "loss": 0.8943,
      "step": 220
    },
    {
      "epoch": 2.1,
      "grad_norm": 1.352690577507019,
      "learning_rate": 0.003944954128440367,
      "loss": 0.8426,
      "step": 230
    },
    {
      "epoch": 2.19,
      "grad_norm": 1.743643879890442,
      "learning_rate": 0.0038990825688073397,
      "loss": 0.7838,
      "step": 240
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.9074971675872803,
      "learning_rate": 0.0038532110091743124,
      "loss": 0.8546,
      "step": 250
    },
    {
      "epoch": 2.37,
      "grad_norm": 0.6970580816268921,
      "learning_rate": 0.0038073394495412843,
      "loss": 0.8481,
      "step": 260
    },
    {
      "epoch": 2.47,
      "grad_norm": 1.3498753309249878,
      "learning_rate": 0.003761467889908257,
      "loss": 0.766,
      "step": 270
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.2216682434082031,
      "learning_rate": 0.0037155963302752293,
      "loss": 0.8654,
      "step": 280
    },
    {
      "epoch": 2.65,
      "grad_norm": 1.1758558750152588,
      "learning_rate": 0.003669724770642202,
      "loss": 0.8438,
      "step": 290
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.8012980222702026,
      "learning_rate": 0.0036238532110091743,
      "loss": 0.8363,
      "step": 300
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.5546942949295044,
      "learning_rate": 0.003577981651376147,
      "loss": 0.8998,
      "step": 310
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.9878047704696655,
      "learning_rate": 0.0035321100917431194,
      "loss": 0.7891,
      "step": 320
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7178464606181456,
      "eval_f1": 0.36668660138047887,
      "eval_loss": 0.7574967741966248,
      "eval_precision": 0.3953602798961963,
      "eval_recall": 0.40001083689608274,
      "eval_runtime": 5.9834,
      "eval_samples_per_second": 167.63,
      "eval_steps_per_second": 10.529,
      "step": 328
    },
    {
      "epoch": 3.01,
      "grad_norm": 1.59013032913208,
      "learning_rate": 0.003486238532110092,
      "loss": 0.8096,
      "step": 330
    },
    {
      "epoch": 3.11,
      "grad_norm": 0.7891358733177185,
      "learning_rate": 0.0034403669724770644,
      "loss": 0.8351,
      "step": 340
    },
    {
      "epoch": 3.2,
      "grad_norm": 1.1285407543182373,
      "learning_rate": 0.003394495412844037,
      "loss": 0.8189,
      "step": 350
    },
    {
      "epoch": 3.29,
      "grad_norm": 1.166321873664856,
      "learning_rate": 0.003348623853211009,
      "loss": 0.7379,
      "step": 360
    },
    {
      "epoch": 3.38,
      "grad_norm": 0.8681693077087402,
      "learning_rate": 0.0033027522935779817,
      "loss": 0.7388,
      "step": 370
    },
    {
      "epoch": 3.47,
      "grad_norm": 1.2175371646881104,
      "learning_rate": 0.003256880733944954,
      "loss": 0.8183,
      "step": 380
    },
    {
      "epoch": 3.56,
      "grad_norm": 1.4015443325042725,
      "learning_rate": 0.003211009174311927,
      "loss": 0.8545,
      "step": 390
    },
    {
      "epoch": 3.65,
      "grad_norm": 1.3555314540863037,
      "learning_rate": 0.003165137614678899,
      "loss": 0.8038,
      "step": 400
    },
    {
      "epoch": 3.74,
      "grad_norm": 1.3774343729019165,
      "learning_rate": 0.003119266055045872,
      "loss": 0.8346,
      "step": 410
    },
    {
      "epoch": 3.84,
      "grad_norm": 1.0004535913467407,
      "learning_rate": 0.003073394495412844,
      "loss": 0.7919,
      "step": 420
    },
    {
      "epoch": 3.93,
      "grad_norm": 0.7557776570320129,
      "learning_rate": 0.003027522935779817,
      "loss": 0.7649,
      "step": 430
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7417746759720838,
      "eval_f1": 0.41456723287460584,
      "eval_loss": 0.6878895163536072,
      "eval_precision": 0.5009001421261136,
      "eval_recall": 0.3971855759911965,
      "eval_runtime": 5.8982,
      "eval_samples_per_second": 170.052,
      "eval_steps_per_second": 10.681,
      "step": 438
    },
    {
      "epoch": 4.02,
      "grad_norm": 0.8360877633094788,
      "learning_rate": 0.002981651376146789,
      "loss": 0.7322,
      "step": 440
    },
    {
      "epoch": 4.11,
      "grad_norm": 1.399274468421936,
      "learning_rate": 0.002935779816513762,
      "loss": 0.7425,
      "step": 450
    },
    {
      "epoch": 4.2,
      "grad_norm": 1.0945734977722168,
      "learning_rate": 0.0028899082568807338,
      "loss": 0.7295,
      "step": 460
    },
    {
      "epoch": 4.29,
      "grad_norm": 1.5122932195663452,
      "learning_rate": 0.0028440366972477065,
      "loss": 0.7892,
      "step": 470
    },
    {
      "epoch": 4.38,
      "grad_norm": 0.7687821388244629,
      "learning_rate": 0.002798165137614679,
      "loss": 0.7614,
      "step": 480
    },
    {
      "epoch": 4.47,
      "grad_norm": 1.25822913646698,
      "learning_rate": 0.0027522935779816515,
      "loss": 0.7811,
      "step": 490
    },
    {
      "epoch": 4.57,
      "grad_norm": 0.7886181473731995,
      "learning_rate": 0.002706422018348624,
      "loss": 0.8093,
      "step": 500
    },
    {
      "epoch": 4.66,
      "grad_norm": 0.7840601801872253,
      "learning_rate": 0.0026605504587155966,
      "loss": 0.738,
      "step": 510
    },
    {
      "epoch": 4.75,
      "grad_norm": 1.1541500091552734,
      "learning_rate": 0.002614678899082569,
      "loss": 0.8208,
      "step": 520
    },
    {
      "epoch": 4.84,
      "grad_norm": 1.354428768157959,
      "learning_rate": 0.0025688073394495416,
      "loss": 0.746,
      "step": 530
    },
    {
      "epoch": 4.93,
      "grad_norm": 0.8335555195808411,
      "learning_rate": 0.0025229357798165135,
      "loss": 0.8146,
      "step": 540
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7178464606181456,
      "eval_f1": 0.36409764256305227,
      "eval_loss": 0.7471081614494324,
      "eval_precision": 0.4490225302647663,
      "eval_recall": 0.4141043546195771,
      "eval_runtime": 5.7321,
      "eval_samples_per_second": 174.978,
      "eval_steps_per_second": 10.991,
      "step": 547
    },
    {
      "epoch": 5.02,
      "grad_norm": 1.1929486989974976,
      "learning_rate": 0.0024770642201834866,
      "loss": 0.7308,
      "step": 550
    },
    {
      "epoch": 5.11,
      "grad_norm": 1.202407717704773,
      "learning_rate": 0.002431192660550459,
      "loss": 0.7956,
      "step": 560
    },
    {
      "epoch": 5.21,
      "grad_norm": 1.1130154132843018,
      "learning_rate": 0.0023853211009174312,
      "loss": 0.7042,
      "step": 570
    },
    {
      "epoch": 5.3,
      "grad_norm": 1.1920500993728638,
      "learning_rate": 0.002339449541284404,
      "loss": 0.7945,
      "step": 580
    },
    {
      "epoch": 5.39,
      "grad_norm": 1.4452427625656128,
      "learning_rate": 0.0022935779816513763,
      "loss": 0.7757,
      "step": 590
    },
    {
      "epoch": 5.48,
      "grad_norm": 1.1632148027420044,
      "learning_rate": 0.002247706422018349,
      "loss": 0.7595,
      "step": 600
    },
    {
      "epoch": 5.57,
      "grad_norm": 0.8493008017539978,
      "learning_rate": 0.0022018348623853213,
      "loss": 0.7757,
      "step": 610
    },
    {
      "epoch": 5.66,
      "grad_norm": 1.3676623106002808,
      "learning_rate": 0.0021559633027522936,
      "loss": 0.7664,
      "step": 620
    },
    {
      "epoch": 5.75,
      "grad_norm": 0.9162562489509583,
      "learning_rate": 0.0021100917431192663,
      "loss": 0.7303,
      "step": 630
    },
    {
      "epoch": 5.84,
      "grad_norm": 0.7179450392723083,
      "learning_rate": 0.0020642201834862386,
      "loss": 0.7088,
      "step": 640
    },
    {
      "epoch": 5.94,
      "grad_norm": 0.7027204632759094,
      "learning_rate": 0.002018348623853211,
      "loss": 0.6831,
      "step": 650
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7367896311066799,
      "eval_f1": 0.42520915568023343,
      "eval_loss": 0.7007002830505371,
      "eval_precision": 0.47770601644698146,
      "eval_recall": 0.4148257042870626,
      "eval_runtime": 5.9496,
      "eval_samples_per_second": 168.582,
      "eval_steps_per_second": 10.589,
      "step": 657
    },
    {
      "epoch": 6.03,
      "grad_norm": 0.7743313312530518,
      "learning_rate": 0.0019724770642201837,
      "loss": 0.7261,
      "step": 660
    },
    {
      "epoch": 6.12,
      "grad_norm": 0.6318005323410034,
      "learning_rate": 0.0019266055045871562,
      "loss": 0.702,
      "step": 670
    },
    {
      "epoch": 6.21,
      "grad_norm": 0.696121096611023,
      "learning_rate": 0.0018807339449541285,
      "loss": 0.6966,
      "step": 680
    },
    {
      "epoch": 6.3,
      "grad_norm": 0.5956413149833679,
      "learning_rate": 0.001834862385321101,
      "loss": 0.7176,
      "step": 690
    },
    {
      "epoch": 6.39,
      "grad_norm": 1.6218867301940918,
      "learning_rate": 0.0017889908256880735,
      "loss": 0.7223,
      "step": 700
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.1353025436401367,
      "learning_rate": 0.001743119266055046,
      "loss": 0.7098,
      "step": 710
    },
    {
      "epoch": 6.58,
      "grad_norm": 0.7476430535316467,
      "learning_rate": 0.0016972477064220186,
      "loss": 0.6411,
      "step": 720
    },
    {
      "epoch": 6.67,
      "grad_norm": 1.3945239782333374,
      "learning_rate": 0.0016513761467889909,
      "loss": 0.7219,
      "step": 730
    },
    {
      "epoch": 6.76,
      "grad_norm": 0.9898785948753357,
      "learning_rate": 0.0016055045871559634,
      "loss": 0.7456,
      "step": 740
    },
    {
      "epoch": 6.85,
      "grad_norm": 0.7352449893951416,
      "learning_rate": 0.001559633027522936,
      "loss": 0.7057,
      "step": 750
    },
    {
      "epoch": 6.94,
      "grad_norm": 1.0120513439178467,
      "learning_rate": 0.0015137614678899084,
      "loss": 0.695,
      "step": 760
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7427716849451645,
      "eval_f1": 0.48411655206463156,
      "eval_loss": 0.6797036528587341,
      "eval_precision": 0.4638223826629623,
      "eval_recall": 0.5333725550353186,
      "eval_runtime": 5.7522,
      "eval_samples_per_second": 174.367,
      "eval_steps_per_second": 10.952,
      "step": 766
    },
    {
      "epoch": 7.03,
      "grad_norm": 1.297319769859314,
      "learning_rate": 0.001467889908256881,
      "loss": 0.7541,
      "step": 770
    },
    {
      "epoch": 7.12,
      "grad_norm": 1.1115490198135376,
      "learning_rate": 0.0014220183486238532,
      "loss": 0.6944,
      "step": 780
    },
    {
      "epoch": 7.21,
      "grad_norm": 0.881907045841217,
      "learning_rate": 0.0013761467889908258,
      "loss": 0.7047,
      "step": 790
    },
    {
      "epoch": 7.31,
      "grad_norm": 0.9110414981842041,
      "learning_rate": 0.0013302752293577983,
      "loss": 0.7227,
      "step": 800
    },
    {
      "epoch": 7.4,
      "grad_norm": 0.7190865874290466,
      "learning_rate": 0.0012844036697247708,
      "loss": 0.7263,
      "step": 810
    },
    {
      "epoch": 7.49,
      "grad_norm": 0.9148305654525757,
      "learning_rate": 0.0012385321100917433,
      "loss": 0.6336,
      "step": 820
    },
    {
      "epoch": 7.58,
      "grad_norm": 0.7972878813743591,
      "learning_rate": 0.0011926605504587156,
      "loss": 0.6886,
      "step": 830
    },
    {
      "epoch": 7.67,
      "grad_norm": 0.9237717986106873,
      "learning_rate": 0.0011467889908256881,
      "loss": 0.6212,
      "step": 840
    },
    {
      "epoch": 7.76,
      "grad_norm": 1.1942154169082642,
      "learning_rate": 0.0011009174311926607,
      "loss": 0.6202,
      "step": 850
    },
    {
      "epoch": 7.85,
      "grad_norm": 1.2370058298110962,
      "learning_rate": 0.0010550458715596332,
      "loss": 0.703,
      "step": 860
    },
    {
      "epoch": 7.95,
      "grad_norm": 1.2201330661773682,
      "learning_rate": 0.0010091743119266055,
      "loss": 0.6646,
      "step": 870
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7537387836490529,
      "eval_f1": 0.4932964122925023,
      "eval_loss": 0.6534218192100525,
      "eval_precision": 0.6130395728356862,
      "eval_recall": 0.5077119285550199,
      "eval_runtime": 6.1104,
      "eval_samples_per_second": 164.148,
      "eval_steps_per_second": 10.31,
      "step": 876
    },
    {
      "epoch": 8.04,
      "grad_norm": 0.904547393321991,
      "learning_rate": 0.0009633027522935781,
      "loss": 0.6991,
      "step": 880
    },
    {
      "epoch": 8.13,
      "grad_norm": 0.7196776270866394,
      "learning_rate": 0.0009174311926605505,
      "loss": 0.612,
      "step": 890
    },
    {
      "epoch": 8.22,
      "grad_norm": 1.1908409595489502,
      "learning_rate": 0.000871559633027523,
      "loss": 0.6583,
      "step": 900
    },
    {
      "epoch": 8.31,
      "grad_norm": 1.1398776769638062,
      "learning_rate": 0.0008256880733944954,
      "loss": 0.7221,
      "step": 910
    },
    {
      "epoch": 8.4,
      "grad_norm": 0.867472767829895,
      "learning_rate": 0.000779816513761468,
      "loss": 0.6012,
      "step": 920
    },
    {
      "epoch": 8.49,
      "grad_norm": 0.7770695686340332,
      "learning_rate": 0.0007339449541284405,
      "loss": 0.6763,
      "step": 930
    },
    {
      "epoch": 8.58,
      "grad_norm": 1.4209235906600952,
      "learning_rate": 0.0006880733944954129,
      "loss": 0.6451,
      "step": 940
    },
    {
      "epoch": 8.68,
      "grad_norm": 1.1582767963409424,
      "learning_rate": 0.0006422018348623854,
      "loss": 0.6745,
      "step": 950
    },
    {
      "epoch": 8.77,
      "grad_norm": 1.2608932256698608,
      "learning_rate": 0.0005963302752293578,
      "loss": 0.6464,
      "step": 960
    },
    {
      "epoch": 8.86,
      "grad_norm": 0.7971704006195068,
      "learning_rate": 0.0005504587155963303,
      "loss": 0.6251,
      "step": 970
    },
    {
      "epoch": 8.95,
      "grad_norm": 1.318163275718689,
      "learning_rate": 0.0005045871559633027,
      "loss": 0.675,
      "step": 980
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7666999002991027,
      "eval_f1": 0.5308101410608023,
      "eval_loss": 0.6237577795982361,
      "eval_precision": 0.6518028517787527,
      "eval_recall": 0.5430883155707511,
      "eval_runtime": 5.972,
      "eval_samples_per_second": 167.95,
      "eval_steps_per_second": 10.549,
      "step": 985
    },
    {
      "epoch": 9.04,
      "grad_norm": 0.9488193392753601,
      "learning_rate": 0.00045871559633027525,
      "loss": 0.6078,
      "step": 990
    },
    {
      "epoch": 9.13,
      "grad_norm": 0.8466370105743408,
      "learning_rate": 0.0004128440366972477,
      "loss": 0.6368,
      "step": 1000
    },
    {
      "epoch": 9.22,
      "grad_norm": 1.245497465133667,
      "learning_rate": 0.00036697247706422024,
      "loss": 0.6639,
      "step": 1010
    },
    {
      "epoch": 9.32,
      "grad_norm": 1.0719431638717651,
      "learning_rate": 0.0003211009174311927,
      "loss": 0.5846,
      "step": 1020
    },
    {
      "epoch": 9.41,
      "grad_norm": 0.8983196020126343,
      "learning_rate": 0.00027522935779816516,
      "loss": 0.663,
      "step": 1030
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.8495015501976013,
      "learning_rate": 0.00022935779816513763,
      "loss": 0.5945,
      "step": 1040
    },
    {
      "epoch": 9.59,
      "grad_norm": 1.3694357872009277,
      "learning_rate": 0.00018348623853211012,
      "loss": 0.6575,
      "step": 1050
    },
    {
      "epoch": 9.68,
      "grad_norm": 1.014473557472229,
      "learning_rate": 0.00013761467889908258,
      "loss": 0.6433,
      "step": 1060
    },
    {
      "epoch": 9.77,
      "grad_norm": 1.0054584741592407,
      "learning_rate": 9.174311926605506e-05,
      "loss": 0.6797,
      "step": 1070
    },
    {
      "epoch": 9.86,
      "grad_norm": 1.37436044216156,
      "learning_rate": 4.587155963302753e-05,
      "loss": 0.6579,
      "step": 1080
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.9838040471076965,
      "learning_rate": 0.0,
      "loss": 0.6145,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "eval_accuracy": 0.7726819541375872,
      "eval_f1": 0.5282805571871881,
      "eval_loss": 0.6095667481422424,
      "eval_precision": 0.6426600715638663,
      "eval_recall": 0.5346360087577886,
      "eval_runtime": 6.0311,
      "eval_samples_per_second": 166.305,
      "eval_steps_per_second": 10.446,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "step": 1090,
      "total_flos": 5.440571948014866e+18,
      "train_loss": 0.7656277652180523,
      "train_runtime": 925.5702,
      "train_samples_per_second": 75.705,
      "train_steps_per_second": 1.178
    }
  ],
  "logging_steps": 10,
  "max_steps": 1090,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 5.440571948014866e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}