{
  "best_metric": 0.7926221335992024,
  "best_model_checkpoint": "swin-large-patch4-window7-224-in22k-finetuned-lora-medmnistv2/checkpoint-985",
  "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": 3.442108154296875,
      "learning_rate": 0.004954128440366973,
      "loss": 1.1496,
      "step": 10
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.3745663166046143,
      "learning_rate": 0.004908256880733945,
      "loss": 0.9658,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.6516236066818237,
      "learning_rate": 0.004862385321100918,
      "loss": 0.9225,
      "step": 30
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.4029324054718018,
      "learning_rate": 0.00481651376146789,
      "loss": 0.8968,
      "step": 40
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.2768313884735107,
      "learning_rate": 0.0047706422018348625,
      "loss": 0.8582,
      "step": 50
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.250333309173584,
      "learning_rate": 0.004724770642201835,
      "loss": 0.8362,
      "step": 60
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.2696894407272339,
      "learning_rate": 0.004678899082568808,
      "loss": 0.8948,
      "step": 70
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.02017080783844,
      "learning_rate": 0.00463302752293578,
      "loss": 0.8448,
      "step": 80
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.1102752685546875,
      "learning_rate": 0.0045871559633027525,
      "loss": 0.8288,
      "step": 90
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.8990870714187622,
      "learning_rate": 0.004541284403669725,
      "loss": 0.8516,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7547357926221336,
      "eval_f1": 0.41878801809381894,
      "eval_loss": 0.7470152974128723,
      "eval_precision": 0.5455908887866956,
      "eval_recall": 0.3913939658670338,
      "eval_runtime": 11.5403,
      "eval_samples_per_second": 86.913,
      "eval_steps_per_second": 5.459,
      "step": 109
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.1006461381912231,
      "learning_rate": 0.004495412844036698,
      "loss": 0.8742,
      "step": 110
    },
    {
      "epoch": 1.1,
      "grad_norm": 1.271594524383545,
      "learning_rate": 0.0044495412844036695,
      "loss": 0.8544,
      "step": 120
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.24767005443573,
      "learning_rate": 0.004403669724770643,
      "loss": 0.792,
      "step": 130
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.5129131078720093,
      "learning_rate": 0.004357798165137615,
      "loss": 0.8107,
      "step": 140
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.728676974773407,
      "learning_rate": 0.004311926605504587,
      "loss": 0.7921,
      "step": 150
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.073712944984436,
      "learning_rate": 0.0042660550458715595,
      "loss": 0.7453,
      "step": 160
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.1818609237670898,
      "learning_rate": 0.004220183486238533,
      "loss": 0.8023,
      "step": 170
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.6278601884841919,
      "learning_rate": 0.004174311926605505,
      "loss": 0.8196,
      "step": 180
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.796804666519165,
      "learning_rate": 0.004128440366972477,
      "loss": 0.8063,
      "step": 190
    },
    {
      "epoch": 1.83,
      "grad_norm": 1.4323725700378418,
      "learning_rate": 0.00408256880733945,
      "loss": 0.7128,
      "step": 200
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.5659187436103821,
      "learning_rate": 0.004036697247706422,
      "loss": 0.7738,
      "step": 210
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7168494516450648,
      "eval_f1": 0.3577314775338443,
      "eval_loss": 0.8952543139457703,
      "eval_precision": 0.42246935817657555,
      "eval_recall": 0.4458814441250039,
      "eval_runtime": 11.2901,
      "eval_samples_per_second": 88.839,
      "eval_steps_per_second": 5.58,
      "step": 219
    },
    {
      "epoch": 2.01,
      "grad_norm": 2.2657077312469482,
      "learning_rate": 0.003990825688073394,
      "loss": 0.8065,
      "step": 220
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.6983966827392578,
      "learning_rate": 0.003944954128440367,
      "loss": 0.8244,
      "step": 230
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.9509605765342712,
      "learning_rate": 0.0038990825688073397,
      "loss": 0.7429,
      "step": 240
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.8736411333084106,
      "learning_rate": 0.0038532110091743124,
      "loss": 0.7479,
      "step": 250
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.0205260515213013,
      "learning_rate": 0.0038073394495412843,
      "loss": 0.7733,
      "step": 260
    },
    {
      "epoch": 2.47,
      "grad_norm": 0.9213468432426453,
      "learning_rate": 0.003761467889908257,
      "loss": 0.6993,
      "step": 270
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.7223290801048279,
      "learning_rate": 0.0037155963302752293,
      "loss": 0.7582,
      "step": 280
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.8736669421195984,
      "learning_rate": 0.003669724770642202,
      "loss": 0.7483,
      "step": 290
    },
    {
      "epoch": 2.74,
      "grad_norm": 1.2014636993408203,
      "learning_rate": 0.0036238532110091743,
      "loss": 0.7592,
      "step": 300
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.290935754776001,
      "learning_rate": 0.003577981651376147,
      "loss": 0.819,
      "step": 310
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.751524806022644,
      "learning_rate": 0.0035321100917431194,
      "loss": 0.6994,
      "step": 320
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7607178464606181,
      "eval_f1": 0.5104809166351773,
      "eval_loss": 0.6592820286750793,
      "eval_precision": 0.6256931769842607,
      "eval_recall": 0.5058522052667251,
      "eval_runtime": 11.3494,
      "eval_samples_per_second": 88.375,
      "eval_steps_per_second": 5.551,
      "step": 328
    },
    {
      "epoch": 3.01,
      "grad_norm": 1.0127859115600586,
      "learning_rate": 0.003486238532110092,
      "loss": 0.723,
      "step": 330
    },
    {
      "epoch": 3.11,
      "grad_norm": 0.8557056784629822,
      "learning_rate": 0.0034403669724770644,
      "loss": 0.762,
      "step": 340
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.7101117968559265,
      "learning_rate": 0.003394495412844037,
      "loss": 0.746,
      "step": 350
    },
    {
      "epoch": 3.29,
      "grad_norm": 0.732359766960144,
      "learning_rate": 0.003348623853211009,
      "loss": 0.7152,
      "step": 360
    },
    {
      "epoch": 3.38,
      "grad_norm": 0.7687917947769165,
      "learning_rate": 0.0033027522935779817,
      "loss": 0.6849,
      "step": 370
    },
    {
      "epoch": 3.47,
      "grad_norm": 0.9754200577735901,
      "learning_rate": 0.003256880733944954,
      "loss": 0.721,
      "step": 380
    },
    {
      "epoch": 3.56,
      "grad_norm": 0.9926655888557434,
      "learning_rate": 0.003211009174311927,
      "loss": 0.7457,
      "step": 390
    },
    {
      "epoch": 3.65,
      "grad_norm": 1.4650967121124268,
      "learning_rate": 0.003165137614678899,
      "loss": 0.7209,
      "step": 400
    },
    {
      "epoch": 3.74,
      "grad_norm": 0.7933741211891174,
      "learning_rate": 0.003119266055045872,
      "loss": 0.7785,
      "step": 410
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.7204951047897339,
      "learning_rate": 0.003073394495412844,
      "loss": 0.7276,
      "step": 420
    },
    {
      "epoch": 3.93,
      "grad_norm": 1.0182371139526367,
      "learning_rate": 0.003027522935779817,
      "loss": 0.6731,
      "step": 430
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7716849451645065,
      "eval_f1": 0.5382789492033663,
      "eval_loss": 0.6144729852676392,
      "eval_precision": 0.6321625904242083,
      "eval_recall": 0.500050058598068,
      "eval_runtime": 11.305,
      "eval_samples_per_second": 88.722,
      "eval_steps_per_second": 5.573,
      "step": 438
    },
    {
      "epoch": 4.02,
      "grad_norm": 0.5864942669868469,
      "learning_rate": 0.002981651376146789,
      "loss": 0.685,
      "step": 440
    },
    {
      "epoch": 4.11,
      "grad_norm": 0.816683292388916,
      "learning_rate": 0.002935779816513762,
      "loss": 0.6859,
      "step": 450
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.6105664968490601,
      "learning_rate": 0.0028899082568807338,
      "loss": 0.6785,
      "step": 460
    },
    {
      "epoch": 4.29,
      "grad_norm": 1.0114610195159912,
      "learning_rate": 0.0028440366972477065,
      "loss": 0.6661,
      "step": 470
    },
    {
      "epoch": 4.38,
      "grad_norm": 1.1204463243484497,
      "learning_rate": 0.002798165137614679,
      "loss": 0.664,
      "step": 480
    },
    {
      "epoch": 4.47,
      "grad_norm": 1.0051954984664917,
      "learning_rate": 0.0027522935779816515,
      "loss": 0.7172,
      "step": 490
    },
    {
      "epoch": 4.57,
      "grad_norm": 0.6869407892227173,
      "learning_rate": 0.002706422018348624,
      "loss": 0.7341,
      "step": 500
    },
    {
      "epoch": 4.66,
      "grad_norm": 0.720870316028595,
      "learning_rate": 0.0026605504587155966,
      "loss": 0.7028,
      "step": 510
    },
    {
      "epoch": 4.75,
      "grad_norm": 0.9448994994163513,
      "learning_rate": 0.002614678899082569,
      "loss": 0.7383,
      "step": 520
    },
    {
      "epoch": 4.84,
      "grad_norm": 0.9630647897720337,
      "learning_rate": 0.0025688073394495416,
      "loss": 0.6727,
      "step": 530
    },
    {
      "epoch": 4.93,
      "grad_norm": 0.8725408315658569,
      "learning_rate": 0.0025229357798165135,
      "loss": 0.7266,
      "step": 540
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7397806580259222,
      "eval_f1": 0.4934894169090878,
      "eval_loss": 0.6838864088058472,
      "eval_precision": 0.5519819307791096,
      "eval_recall": 0.5344054167239178,
      "eval_runtime": 11.2776,
      "eval_samples_per_second": 88.937,
      "eval_steps_per_second": 5.586,
      "step": 547
    },
    {
      "epoch": 5.02,
      "grad_norm": 1.1792114973068237,
      "learning_rate": 0.0024770642201834866,
      "loss": 0.6676,
      "step": 550
    },
    {
      "epoch": 5.11,
      "grad_norm": 0.7483982443809509,
      "learning_rate": 0.002431192660550459,
      "loss": 0.7229,
      "step": 560
    },
    {
      "epoch": 5.21,
      "grad_norm": 0.8732199668884277,
      "learning_rate": 0.0023853211009174312,
      "loss": 0.6366,
      "step": 570
    },
    {
      "epoch": 5.3,
      "grad_norm": 1.1174126863479614,
      "learning_rate": 0.002339449541284404,
      "loss": 0.6855,
      "step": 580
    },
    {
      "epoch": 5.39,
      "grad_norm": 0.7935824990272522,
      "learning_rate": 0.0022935779816513763,
      "loss": 0.7225,
      "step": 590
    },
    {
      "epoch": 5.48,
      "grad_norm": 0.9188850522041321,
      "learning_rate": 0.002247706422018349,
      "loss": 0.6657,
      "step": 600
    },
    {
      "epoch": 5.57,
      "grad_norm": 0.7955396771430969,
      "learning_rate": 0.0022018348623853213,
      "loss": 0.6797,
      "step": 610
    },
    {
      "epoch": 5.66,
      "grad_norm": 1.0317906141281128,
      "learning_rate": 0.0021559633027522936,
      "loss": 0.6589,
      "step": 620
    },
    {
      "epoch": 5.75,
      "grad_norm": 0.6561917662620544,
      "learning_rate": 0.0021100917431192663,
      "loss": 0.6435,
      "step": 630
    },
    {
      "epoch": 5.84,
      "grad_norm": 0.8070980906486511,
      "learning_rate": 0.0020642201834862386,
      "loss": 0.6303,
      "step": 640
    },
    {
      "epoch": 5.94,
      "grad_norm": 0.6900261044502258,
      "learning_rate": 0.002018348623853211,
      "loss": 0.6388,
      "step": 650
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7666999002991027,
      "eval_f1": 0.5338252978902394,
      "eval_loss": 0.6242878437042236,
      "eval_precision": 0.6116740127093266,
      "eval_recall": 0.5062826891866704,
      "eval_runtime": 11.256,
      "eval_samples_per_second": 89.108,
      "eval_steps_per_second": 5.597,
      "step": 657
    },
    {
      "epoch": 6.03,
      "grad_norm": 0.8859825134277344,
      "learning_rate": 0.0019724770642201837,
      "loss": 0.6252,
      "step": 660
    },
    {
      "epoch": 6.12,
      "grad_norm": 0.7855720520019531,
      "learning_rate": 0.0019266055045871562,
      "loss": 0.6026,
      "step": 670
    },
    {
      "epoch": 6.21,
      "grad_norm": 0.7128086090087891,
      "learning_rate": 0.0018807339449541285,
      "loss": 0.5799,
      "step": 680
    },
    {
      "epoch": 6.3,
      "grad_norm": 0.6269203424453735,
      "learning_rate": 0.001834862385321101,
      "loss": 0.6358,
      "step": 690
    },
    {
      "epoch": 6.39,
      "grad_norm": 1.1196659803390503,
      "learning_rate": 0.0017889908256880735,
      "loss": 0.6282,
      "step": 700
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.253045678138733,
      "learning_rate": 0.001743119266055046,
      "loss": 0.6109,
      "step": 710
    },
    {
      "epoch": 6.58,
      "grad_norm": 0.946592390537262,
      "learning_rate": 0.0016972477064220186,
      "loss": 0.58,
      "step": 720
    },
    {
      "epoch": 6.67,
      "grad_norm": 0.9241579174995422,
      "learning_rate": 0.0016513761467889909,
      "loss": 0.6492,
      "step": 730
    },
    {
      "epoch": 6.76,
      "grad_norm": 0.9794093370437622,
      "learning_rate": 0.0016055045871559634,
      "loss": 0.6333,
      "step": 740
    },
    {
      "epoch": 6.85,
      "grad_norm": 0.9750506281852722,
      "learning_rate": 0.001559633027522936,
      "loss": 0.6883,
      "step": 750
    },
    {
      "epoch": 6.94,
      "grad_norm": 1.1783692836761475,
      "learning_rate": 0.0015137614678899084,
      "loss": 0.6495,
      "step": 760
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7826520438683948,
      "eval_f1": 0.6162957297860807,
      "eval_loss": 0.6161015033721924,
      "eval_precision": 0.6357243111624864,
      "eval_recall": 0.6152698884549003,
      "eval_runtime": 11.1928,
      "eval_samples_per_second": 89.611,
      "eval_steps_per_second": 5.629,
      "step": 766
    },
    {
      "epoch": 7.03,
      "grad_norm": 1.0667985677719116,
      "learning_rate": 0.001467889908256881,
      "loss": 0.6615,
      "step": 770
    },
    {
      "epoch": 7.12,
      "grad_norm": 1.2581771612167358,
      "learning_rate": 0.0014220183486238532,
      "loss": 0.5515,
      "step": 780
    },
    {
      "epoch": 7.21,
      "grad_norm": 1.012791395187378,
      "learning_rate": 0.0013761467889908258,
      "loss": 0.607,
      "step": 790
    },
    {
      "epoch": 7.31,
      "grad_norm": 0.9969860911369324,
      "learning_rate": 0.0013302752293577983,
      "loss": 0.6396,
      "step": 800
    },
    {
      "epoch": 7.4,
      "grad_norm": 0.9849869608879089,
      "learning_rate": 0.0012844036697247708,
      "loss": 0.6563,
      "step": 810
    },
    {
      "epoch": 7.49,
      "grad_norm": 1.0733628273010254,
      "learning_rate": 0.0012385321100917433,
      "loss": 0.5833,
      "step": 820
    },
    {
      "epoch": 7.58,
      "grad_norm": 1.044959306716919,
      "learning_rate": 0.0011926605504587156,
      "loss": 0.5745,
      "step": 830
    },
    {
      "epoch": 7.67,
      "grad_norm": 0.8886712193489075,
      "learning_rate": 0.0011467889908256881,
      "loss": 0.5437,
      "step": 840
    },
    {
      "epoch": 7.76,
      "grad_norm": 1.2174370288848877,
      "learning_rate": 0.0011009174311926607,
      "loss": 0.5654,
      "step": 850
    },
    {
      "epoch": 7.85,
      "grad_norm": 0.8696100115776062,
      "learning_rate": 0.0010550458715596332,
      "loss": 0.6165,
      "step": 860
    },
    {
      "epoch": 7.95,
      "grad_norm": 1.2088637351989746,
      "learning_rate": 0.0010091743119266055,
      "loss": 0.5639,
      "step": 870
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7836490528414756,
      "eval_f1": 0.5930825856311631,
      "eval_loss": 0.5751714706420898,
      "eval_precision": 0.601839864113542,
      "eval_recall": 0.59116309877434,
      "eval_runtime": 11.2311,
      "eval_samples_per_second": 89.305,
      "eval_steps_per_second": 5.609,
      "step": 876
    },
    {
      "epoch": 8.04,
      "grad_norm": 0.6919325590133667,
      "learning_rate": 0.0009633027522935781,
      "loss": 0.6158,
      "step": 880
    },
    {
      "epoch": 8.13,
      "grad_norm": 1.1032708883285522,
      "learning_rate": 0.0009174311926605505,
      "loss": 0.4922,
      "step": 890
    },
    {
      "epoch": 8.22,
      "grad_norm": 1.5484873056411743,
      "learning_rate": 0.000871559633027523,
      "loss": 0.5453,
      "step": 900
    },
    {
      "epoch": 8.31,
      "grad_norm": 1.1881980895996094,
      "learning_rate": 0.0008256880733944954,
      "loss": 0.6307,
      "step": 910
    },
    {
      "epoch": 8.4,
      "grad_norm": 0.80064457654953,
      "learning_rate": 0.000779816513761468,
      "loss": 0.5129,
      "step": 920
    },
    {
      "epoch": 8.49,
      "grad_norm": 0.7097495794296265,
      "learning_rate": 0.0007339449541284405,
      "loss": 0.6111,
      "step": 930
    },
    {
      "epoch": 8.58,
      "grad_norm": 1.0539182424545288,
      "learning_rate": 0.0006880733944954129,
      "loss": 0.5803,
      "step": 940
    },
    {
      "epoch": 8.68,
      "grad_norm": 1.1059874296188354,
      "learning_rate": 0.0006422018348623854,
      "loss": 0.6122,
      "step": 950
    },
    {
      "epoch": 8.77,
      "grad_norm": 0.969431459903717,
      "learning_rate": 0.0005963302752293578,
      "loss": 0.5662,
      "step": 960
    },
    {
      "epoch": 8.86,
      "grad_norm": 0.9566834568977356,
      "learning_rate": 0.0005504587155963303,
      "loss": 0.5317,
      "step": 970
    },
    {
      "epoch": 8.95,
      "grad_norm": 1.040589690208435,
      "learning_rate": 0.0005045871559633027,
      "loss": 0.6012,
      "step": 980
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7926221335992024,
      "eval_f1": 0.6175857153722378,
      "eval_loss": 0.5508496165275574,
      "eval_precision": 0.6303209720740276,
      "eval_recall": 0.6194980716058,
      "eval_runtime": 11.1597,
      "eval_samples_per_second": 89.877,
      "eval_steps_per_second": 5.645,
      "step": 985
    },
    {
      "epoch": 9.04,
      "grad_norm": 0.8962224721908569,
      "learning_rate": 0.00045871559633027525,
      "loss": 0.5106,
      "step": 990
    },
    {
      "epoch": 9.13,
      "grad_norm": 0.9521830677986145,
      "learning_rate": 0.0004128440366972477,
      "loss": 0.5768,
      "step": 1000
    },
    {
      "epoch": 9.22,
      "grad_norm": 1.102037787437439,
      "learning_rate": 0.00036697247706422024,
      "loss": 0.5709,
      "step": 1010
    },
    {
      "epoch": 9.32,
      "grad_norm": 1.3637118339538574,
      "learning_rate": 0.0003211009174311927,
      "loss": 0.4719,
      "step": 1020
    },
    {
      "epoch": 9.41,
      "grad_norm": 0.8370681405067444,
      "learning_rate": 0.00027522935779816516,
      "loss": 0.5717,
      "step": 1030
    },
    {
      "epoch": 9.5,
      "grad_norm": 1.0408118963241577,
      "learning_rate": 0.00022935779816513763,
      "loss": 0.5216,
      "step": 1040
    },
    {
      "epoch": 9.59,
      "grad_norm": 1.2250603437423706,
      "learning_rate": 0.00018348623853211012,
      "loss": 0.5648,
      "step": 1050
    },
    {
      "epoch": 9.68,
      "grad_norm": 0.9561938047409058,
      "learning_rate": 0.00013761467889908258,
      "loss": 0.5316,
      "step": 1060
    },
    {
      "epoch": 9.77,
      "grad_norm": 1.2182472944259644,
      "learning_rate": 9.174311926605506e-05,
      "loss": 0.5773,
      "step": 1070
    },
    {
      "epoch": 9.86,
      "grad_norm": 1.1085007190704346,
      "learning_rate": 4.587155963302753e-05,
      "loss": 0.5777,
      "step": 1080
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.9443553686141968,
      "learning_rate": 0.0,
      "loss": 0.5468,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "eval_accuracy": 0.7856430707876371,
      "eval_f1": 0.6355311291077588,
      "eval_loss": 0.566506028175354,
      "eval_precision": 0.6470430666433826,
      "eval_recall": 0.6287911299620902,
      "eval_runtime": 11.1527,
      "eval_samples_per_second": 89.934,
      "eval_steps_per_second": 5.649,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "step": 1090,
      "total_flos": 1.2358226368562393e+19,
      "train_loss": 0.6880708248243419,
      "train_runtime": 1791.2534,
      "train_samples_per_second": 39.118,
      "train_steps_per_second": 0.609
    }
  ],
  "logging_steps": 10,
  "max_steps": 1090,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.2358226368562393e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}