{
  "best_metric": 0.7956131605184447,
  "best_model_checkpoint": "deit-base-patch16-224-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": 3.768280267715454,
      "learning_rate": 0.004954128440366973,
      "loss": 1.2551,
      "step": 10
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.6224374771118164,
      "learning_rate": 0.004908256880733945,
      "loss": 1.1194,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.8902679085731506,
      "learning_rate": 0.004862385321100918,
      "loss": 0.9226,
      "step": 30
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.336371898651123,
      "learning_rate": 0.00481651376146789,
      "loss": 0.9343,
      "step": 40
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.9396265745162964,
      "learning_rate": 0.0047706422018348625,
      "loss": 0.849,
      "step": 50
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.6278876066207886,
      "learning_rate": 0.004724770642201835,
      "loss": 0.8533,
      "step": 60
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.5454286336898804,
      "learning_rate": 0.004678899082568808,
      "loss": 0.8229,
      "step": 70
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.064020037651062,
      "learning_rate": 0.00463302752293578,
      "loss": 0.8622,
      "step": 80
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.5100964307785034,
      "learning_rate": 0.0045871559633027525,
      "loss": 0.8633,
      "step": 90
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.3673182725906372,
      "learning_rate": 0.004541284403669725,
      "loss": 0.9012,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7228315054835494,
      "eval_f1": 0.34621811450823214,
      "eval_loss": 0.7629963159561157,
      "eval_precision": 0.42629140706004787,
      "eval_recall": 0.32692877938779574,
      "eval_runtime": 6.2499,
      "eval_samples_per_second": 160.484,
      "eval_steps_per_second": 10.08,
      "step": 109
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9423499703407288,
      "learning_rate": 0.004495412844036698,
      "loss": 0.8922,
      "step": 110
    },
    {
      "epoch": 1.1,
      "grad_norm": 2.50795841217041,
      "learning_rate": 0.0044495412844036695,
      "loss": 0.8791,
      "step": 120
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.4502021074295044,
      "learning_rate": 0.004403669724770643,
      "loss": 0.8973,
      "step": 130
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.703600287437439,
      "learning_rate": 0.004357798165137615,
      "loss": 0.8246,
      "step": 140
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.7422580122947693,
      "learning_rate": 0.004311926605504587,
      "loss": 0.7913,
      "step": 150
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.146699070930481,
      "learning_rate": 0.0042660550458715595,
      "loss": 0.8028,
      "step": 160
    },
    {
      "epoch": 1.55,
      "grad_norm": 2.6000311374664307,
      "learning_rate": 0.004220183486238533,
      "loss": 0.9162,
      "step": 170
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.8894189596176147,
      "learning_rate": 0.004174311926605505,
      "loss": 0.9022,
      "step": 180
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.9741529226303101,
      "learning_rate": 0.004128440366972477,
      "loss": 0.8879,
      "step": 190
    },
    {
      "epoch": 1.83,
      "grad_norm": 2.898531198501587,
      "learning_rate": 0.00408256880733945,
      "loss": 0.7974,
      "step": 200
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.5416871309280396,
      "learning_rate": 0.004036697247706422,
      "loss": 0.7636,
      "step": 210
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7288135593220338,
      "eval_f1": 0.3788829143385214,
      "eval_loss": 0.7212122082710266,
      "eval_precision": 0.591156357572359,
      "eval_recall": 0.3630725573348524,
      "eval_runtime": 5.4131,
      "eval_samples_per_second": 185.292,
      "eval_steps_per_second": 11.638,
      "step": 219
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.9239124059677124,
      "learning_rate": 0.003990825688073394,
      "loss": 0.8466,
      "step": 220
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.9725480675697327,
      "learning_rate": 0.003944954128440367,
      "loss": 0.8851,
      "step": 230
    },
    {
      "epoch": 2.19,
      "grad_norm": 1.1037930250167847,
      "learning_rate": 0.0038990825688073397,
      "loss": 0.7814,
      "step": 240
    },
    {
      "epoch": 2.28,
      "grad_norm": 1.5679117441177368,
      "learning_rate": 0.0038532110091743124,
      "loss": 0.7781,
      "step": 250
    },
    {
      "epoch": 2.37,
      "grad_norm": 0.7649789452552795,
      "learning_rate": 0.0038073394495412843,
      "loss": 0.8029,
      "step": 260
    },
    {
      "epoch": 2.47,
      "grad_norm": 1.4153032302856445,
      "learning_rate": 0.003761467889908257,
      "loss": 0.7439,
      "step": 270
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.177064299583435,
      "learning_rate": 0.0037155963302752293,
      "loss": 0.8781,
      "step": 280
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.9871839284896851,
      "learning_rate": 0.003669724770642202,
      "loss": 0.7439,
      "step": 290
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.7878830432891846,
      "learning_rate": 0.0036238532110091743,
      "loss": 0.8001,
      "step": 300
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.0698702335357666,
      "learning_rate": 0.003577981651376147,
      "loss": 0.8568,
      "step": 310
    },
    {
      "epoch": 2.92,
      "grad_norm": 0.9107087850570679,
      "learning_rate": 0.0035321100917431194,
      "loss": 0.7189,
      "step": 320
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7258225324027916,
      "eval_f1": 0.393460298491453,
      "eval_loss": 0.7621892690658569,
      "eval_precision": 0.4464810481413866,
      "eval_recall": 0.422982191132074,
      "eval_runtime": 5.2671,
      "eval_samples_per_second": 190.428,
      "eval_steps_per_second": 11.961,
      "step": 328
    },
    {
      "epoch": 3.01,
      "grad_norm": 1.1653732061386108,
      "learning_rate": 0.003486238532110092,
      "loss": 0.7245,
      "step": 330
    },
    {
      "epoch": 3.11,
      "grad_norm": 0.693824291229248,
      "learning_rate": 0.0034403669724770644,
      "loss": 0.7883,
      "step": 340
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.7583943009376526,
      "learning_rate": 0.003394495412844037,
      "loss": 0.7618,
      "step": 350
    },
    {
      "epoch": 3.29,
      "grad_norm": 1.0829322338104248,
      "learning_rate": 0.003348623853211009,
      "loss": 0.7534,
      "step": 360
    },
    {
      "epoch": 3.38,
      "grad_norm": 1.2232468128204346,
      "learning_rate": 0.0033027522935779817,
      "loss": 0.6827,
      "step": 370
    },
    {
      "epoch": 3.47,
      "grad_norm": 1.0095463991165161,
      "learning_rate": 0.003256880733944954,
      "loss": 0.7615,
      "step": 380
    },
    {
      "epoch": 3.56,
      "grad_norm": 1.059535026550293,
      "learning_rate": 0.003211009174311927,
      "loss": 0.8094,
      "step": 390
    },
    {
      "epoch": 3.65,
      "grad_norm": 0.9983842372894287,
      "learning_rate": 0.003165137614678899,
      "loss": 0.7649,
      "step": 400
    },
    {
      "epoch": 3.74,
      "grad_norm": 0.6007803082466125,
      "learning_rate": 0.003119266055045872,
      "loss": 0.7825,
      "step": 410
    },
    {
      "epoch": 3.84,
      "grad_norm": 1.1800183057785034,
      "learning_rate": 0.003073394495412844,
      "loss": 0.7296,
      "step": 420
    },
    {
      "epoch": 3.93,
      "grad_norm": 1.049116611480713,
      "learning_rate": 0.003027522935779817,
      "loss": 0.6904,
      "step": 430
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7437686939182453,
      "eval_f1": 0.4115031629499951,
      "eval_loss": 0.7280949950218201,
      "eval_precision": 0.48879535404583313,
      "eval_recall": 0.4484100461852218,
      "eval_runtime": 5.2319,
      "eval_samples_per_second": 191.708,
      "eval_steps_per_second": 12.041,
      "step": 438
    },
    {
      "epoch": 4.02,
      "grad_norm": 0.9003276824951172,
      "learning_rate": 0.002981651376146789,
      "loss": 0.6584,
      "step": 440
    },
    {
      "epoch": 4.11,
      "grad_norm": 1.3586938381195068,
      "learning_rate": 0.002935779816513762,
      "loss": 0.7175,
      "step": 450
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.5924133062362671,
      "learning_rate": 0.0028899082568807338,
      "loss": 0.67,
      "step": 460
    },
    {
      "epoch": 4.29,
      "grad_norm": 0.9187817573547363,
      "learning_rate": 0.0028440366972477065,
      "loss": 0.697,
      "step": 470
    },
    {
      "epoch": 4.38,
      "grad_norm": 0.7464513778686523,
      "learning_rate": 0.002798165137614679,
      "loss": 0.6691,
      "step": 480
    },
    {
      "epoch": 4.47,
      "grad_norm": 0.6532416343688965,
      "learning_rate": 0.0027522935779816515,
      "loss": 0.7217,
      "step": 490
    },
    {
      "epoch": 4.57,
      "grad_norm": 0.7422506809234619,
      "learning_rate": 0.002706422018348624,
      "loss": 0.7619,
      "step": 500
    },
    {
      "epoch": 4.66,
      "grad_norm": 0.8745353817939758,
      "learning_rate": 0.0026605504587155966,
      "loss": 0.7264,
      "step": 510
    },
    {
      "epoch": 4.75,
      "grad_norm": 0.7916358113288879,
      "learning_rate": 0.002614678899082569,
      "loss": 0.7592,
      "step": 520
    },
    {
      "epoch": 4.84,
      "grad_norm": 1.0833995342254639,
      "learning_rate": 0.0025688073394495416,
      "loss": 0.7263,
      "step": 530
    },
    {
      "epoch": 4.93,
      "grad_norm": 1.0176126956939697,
      "learning_rate": 0.0025229357798165135,
      "loss": 0.7658,
      "step": 540
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7397806580259222,
      "eval_f1": 0.3752685826577452,
      "eval_loss": 0.7215412259101868,
      "eval_precision": 0.4855492779862528,
      "eval_recall": 0.42520631993699903,
      "eval_runtime": 5.5429,
      "eval_samples_per_second": 180.953,
      "eval_steps_per_second": 11.366,
      "step": 547
    },
    {
      "epoch": 5.02,
      "grad_norm": 1.45496666431427,
      "learning_rate": 0.0024770642201834866,
      "loss": 0.7024,
      "step": 550
    },
    {
      "epoch": 5.11,
      "grad_norm": 0.8751833438873291,
      "learning_rate": 0.002431192660550459,
      "loss": 0.741,
      "step": 560
    },
    {
      "epoch": 5.21,
      "grad_norm": 0.7062868475914001,
      "learning_rate": 0.0023853211009174312,
      "loss": 0.6733,
      "step": 570
    },
    {
      "epoch": 5.3,
      "grad_norm": 0.9199981689453125,
      "learning_rate": 0.002339449541284404,
      "loss": 0.6988,
      "step": 580
    },
    {
      "epoch": 5.39,
      "grad_norm": 0.8544884324073792,
      "learning_rate": 0.0022935779816513763,
      "loss": 0.7073,
      "step": 590
    },
    {
      "epoch": 5.48,
      "grad_norm": 1.0523695945739746,
      "learning_rate": 0.002247706422018349,
      "loss": 0.6822,
      "step": 600
    },
    {
      "epoch": 5.57,
      "grad_norm": 0.866678774356842,
      "learning_rate": 0.0022018348623853213,
      "loss": 0.6784,
      "step": 610
    },
    {
      "epoch": 5.66,
      "grad_norm": 0.6989203095436096,
      "learning_rate": 0.0021559633027522936,
      "loss": 0.6796,
      "step": 620
    },
    {
      "epoch": 5.75,
      "grad_norm": 0.4832663834095001,
      "learning_rate": 0.0021100917431192663,
      "loss": 0.6472,
      "step": 630
    },
    {
      "epoch": 5.84,
      "grad_norm": 1.1230007410049438,
      "learning_rate": 0.0020642201834862386,
      "loss": 0.6693,
      "step": 640
    },
    {
      "epoch": 5.94,
      "grad_norm": 0.6341413259506226,
      "learning_rate": 0.002018348623853211,
      "loss": 0.6363,
      "step": 650
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7676969092721835,
      "eval_f1": 0.5121263559819323,
      "eval_loss": 0.6328557133674622,
      "eval_precision": 0.6350277126119822,
      "eval_recall": 0.49284577902844884,
      "eval_runtime": 5.4702,
      "eval_samples_per_second": 183.357,
      "eval_steps_per_second": 11.517,
      "step": 657
    },
    {
      "epoch": 6.03,
      "grad_norm": 0.7254393696784973,
      "learning_rate": 0.0019724770642201837,
      "loss": 0.6398,
      "step": 660
    },
    {
      "epoch": 6.12,
      "grad_norm": 0.6462991237640381,
      "learning_rate": 0.0019266055045871562,
      "loss": 0.6703,
      "step": 670
    },
    {
      "epoch": 6.21,
      "grad_norm": 0.7446946501731873,
      "learning_rate": 0.0018807339449541285,
      "loss": 0.6184,
      "step": 680
    },
    {
      "epoch": 6.3,
      "grad_norm": 0.5196910500526428,
      "learning_rate": 0.001834862385321101,
      "loss": 0.6308,
      "step": 690
    },
    {
      "epoch": 6.39,
      "grad_norm": 1.019028902053833,
      "learning_rate": 0.0017889908256880735,
      "loss": 0.664,
      "step": 700
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.3352869749069214,
      "learning_rate": 0.001743119266055046,
      "loss": 0.638,
      "step": 710
    },
    {
      "epoch": 6.58,
      "grad_norm": 0.8105664253234863,
      "learning_rate": 0.0016972477064220186,
      "loss": 0.5801,
      "step": 720
    },
    {
      "epoch": 6.67,
      "grad_norm": 1.0205323696136475,
      "learning_rate": 0.0016513761467889909,
      "loss": 0.6923,
      "step": 730
    },
    {
      "epoch": 6.76,
      "grad_norm": 0.48183074593544006,
      "learning_rate": 0.0016055045871559634,
      "loss": 0.6583,
      "step": 740
    },
    {
      "epoch": 6.85,
      "grad_norm": 0.7392696738243103,
      "learning_rate": 0.001559633027522936,
      "loss": 0.6784,
      "step": 750
    },
    {
      "epoch": 6.94,
      "grad_norm": 0.7750418782234192,
      "learning_rate": 0.0015137614678899084,
      "loss": 0.6299,
      "step": 760
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7716849451645065,
      "eval_f1": 0.5713473581488847,
      "eval_loss": 0.6116508841514587,
      "eval_precision": 0.5962341473849692,
      "eval_recall": 0.5781145727281558,
      "eval_runtime": 5.4791,
      "eval_samples_per_second": 183.06,
      "eval_steps_per_second": 11.498,
      "step": 766
    },
    {
      "epoch": 7.03,
      "grad_norm": 1.6914126873016357,
      "learning_rate": 0.001467889908256881,
      "loss": 0.7058,
      "step": 770
    },
    {
      "epoch": 7.12,
      "grad_norm": 1.1266179084777832,
      "learning_rate": 0.0014220183486238532,
      "loss": 0.5969,
      "step": 780
    },
    {
      "epoch": 7.21,
      "grad_norm": 0.8198122978210449,
      "learning_rate": 0.0013761467889908258,
      "loss": 0.6296,
      "step": 790
    },
    {
      "epoch": 7.31,
      "grad_norm": 0.8668613433837891,
      "learning_rate": 0.0013302752293577983,
      "loss": 0.6779,
      "step": 800
    },
    {
      "epoch": 7.4,
      "grad_norm": 0.757351279258728,
      "learning_rate": 0.0012844036697247708,
      "loss": 0.6589,
      "step": 810
    },
    {
      "epoch": 7.49,
      "grad_norm": 1.231268286705017,
      "learning_rate": 0.0012385321100917433,
      "loss": 0.5909,
      "step": 820
    },
    {
      "epoch": 7.58,
      "grad_norm": 0.6386240720748901,
      "learning_rate": 0.0011926605504587156,
      "loss": 0.6166,
      "step": 830
    },
    {
      "epoch": 7.67,
      "grad_norm": 0.7646284103393555,
      "learning_rate": 0.0011467889908256881,
      "loss": 0.5815,
      "step": 840
    },
    {
      "epoch": 7.76,
      "grad_norm": 1.1417447328567505,
      "learning_rate": 0.0011009174311926607,
      "loss": 0.5825,
      "step": 850
    },
    {
      "epoch": 7.85,
      "grad_norm": 0.8213087916374207,
      "learning_rate": 0.0010550458715596332,
      "loss": 0.652,
      "step": 860
    },
    {
      "epoch": 7.95,
      "grad_norm": 0.8747499585151672,
      "learning_rate": 0.0010091743119266055,
      "loss": 0.6011,
      "step": 870
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7796610169491526,
      "eval_f1": 0.5902452724888688,
      "eval_loss": 0.5919255018234253,
      "eval_precision": 0.6162323714867404,
      "eval_recall": 0.575685794889542,
      "eval_runtime": 5.3294,
      "eval_samples_per_second": 188.202,
      "eval_steps_per_second": 11.821,
      "step": 876
    },
    {
      "epoch": 8.04,
      "grad_norm": 0.9114809632301331,
      "learning_rate": 0.0009633027522935781,
      "loss": 0.6365,
      "step": 880
    },
    {
      "epoch": 8.13,
      "grad_norm": 0.7782912254333496,
      "learning_rate": 0.0009174311926605505,
      "loss": 0.5558,
      "step": 890
    },
    {
      "epoch": 8.22,
      "grad_norm": 1.2907453775405884,
      "learning_rate": 0.000871559633027523,
      "loss": 0.5756,
      "step": 900
    },
    {
      "epoch": 8.31,
      "grad_norm": 1.1318057775497437,
      "learning_rate": 0.0008256880733944954,
      "loss": 0.6574,
      "step": 910
    },
    {
      "epoch": 8.4,
      "grad_norm": 0.5756754875183105,
      "learning_rate": 0.000779816513761468,
      "loss": 0.5494,
      "step": 920
    },
    {
      "epoch": 8.49,
      "grad_norm": 0.7023201584815979,
      "learning_rate": 0.0007339449541284405,
      "loss": 0.6101,
      "step": 930
    },
    {
      "epoch": 8.58,
      "grad_norm": 0.9913588166236877,
      "learning_rate": 0.0006880733944954129,
      "loss": 0.6034,
      "step": 940
    },
    {
      "epoch": 8.68,
      "grad_norm": 0.982480525970459,
      "learning_rate": 0.0006422018348623854,
      "loss": 0.5981,
      "step": 950
    },
    {
      "epoch": 8.77,
      "grad_norm": 1.1374191045761108,
      "learning_rate": 0.0005963302752293578,
      "loss": 0.6134,
      "step": 960
    },
    {
      "epoch": 8.86,
      "grad_norm": 0.7602503299713135,
      "learning_rate": 0.0005504587155963303,
      "loss": 0.5561,
      "step": 970
    },
    {
      "epoch": 8.95,
      "grad_norm": 0.8967719674110413,
      "learning_rate": 0.0005045871559633027,
      "loss": 0.6043,
      "step": 980
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7946161515453639,
      "eval_f1": 0.5983005873522603,
      "eval_loss": 0.5475942492485046,
      "eval_precision": 0.6295002331514908,
      "eval_recall": 0.5812943349243114,
      "eval_runtime": 5.2749,
      "eval_samples_per_second": 190.145,
      "eval_steps_per_second": 11.943,
      "step": 985
    },
    {
      "epoch": 9.04,
      "grad_norm": 0.9292692542076111,
      "learning_rate": 0.00045871559633027525,
      "loss": 0.5415,
      "step": 990
    },
    {
      "epoch": 9.13,
      "grad_norm": 0.8102772235870361,
      "learning_rate": 0.0004128440366972477,
      "loss": 0.5611,
      "step": 1000
    },
    {
      "epoch": 9.22,
      "grad_norm": 1.1832762956619263,
      "learning_rate": 0.00036697247706422024,
      "loss": 0.5876,
      "step": 1010
    },
    {
      "epoch": 9.32,
      "grad_norm": 1.1258927583694458,
      "learning_rate": 0.0003211009174311927,
      "loss": 0.4946,
      "step": 1020
    },
    {
      "epoch": 9.41,
      "grad_norm": 0.8979453444480896,
      "learning_rate": 0.00027522935779816516,
      "loss": 0.5668,
      "step": 1030
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.9506226778030396,
      "learning_rate": 0.00022935779816513763,
      "loss": 0.5384,
      "step": 1040
    },
    {
      "epoch": 9.59,
      "grad_norm": 1.1283026933670044,
      "learning_rate": 0.00018348623853211012,
      "loss": 0.5772,
      "step": 1050
    },
    {
      "epoch": 9.68,
      "grad_norm": 0.9244889616966248,
      "learning_rate": 0.00013761467889908258,
      "loss": 0.5662,
      "step": 1060
    },
    {
      "epoch": 9.77,
      "grad_norm": 0.8923238515853882,
      "learning_rate": 9.174311926605506e-05,
      "loss": 0.5921,
      "step": 1070
    },
    {
      "epoch": 9.86,
      "grad_norm": 1.1653807163238525,
      "learning_rate": 4.587155963302753e-05,
      "loss": 0.5846,
      "step": 1080
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.8477634191513062,
      "learning_rate": 0.0,
      "loss": 0.5671,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "eval_accuracy": 0.7956131605184447,
      "eval_f1": 0.5997799904348683,
      "eval_loss": 0.5543830990791321,
      "eval_precision": 0.627341257239154,
      "eval_recall": 0.5810367281093276,
      "eval_runtime": 5.4717,
      "eval_samples_per_second": 183.308,
      "eval_steps_per_second": 11.514,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "step": 1090,
      "total_flos": 5.442882169274339e+18,
      "train_loss": 0.7150778630457887,
      "train_runtime": 818.9501,
      "train_samples_per_second": 85.561,
      "train_steps_per_second": 1.331
    }
  ],
  "logging_steps": 10,
  "max_steps": 1090,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 5.442882169274339e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}