{
  "best_metric": 0.7676969092721835,
  "best_model_checkpoint": "vit-base-patch16-224-in21k-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": 1.0251981019973755,
      "learning_rate": 0.004954128440366973,
      "loss": 1.2326,
      "step": 10
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6157691478729248,
      "learning_rate": 0.004908256880733945,
      "loss": 0.9567,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.101731538772583,
      "learning_rate": 0.004862385321100918,
      "loss": 0.905,
      "step": 30
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.8456130623817444,
      "learning_rate": 0.00481651376146789,
      "loss": 0.9132,
      "step": 40
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.6145790219306946,
      "learning_rate": 0.0047706422018348625,
      "loss": 0.8798,
      "step": 50
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.1609084606170654,
      "learning_rate": 0.004724770642201835,
      "loss": 0.7849,
      "step": 60
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.7732934355735779,
      "learning_rate": 0.004678899082568808,
      "loss": 0.7901,
      "step": 70
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.7393956780433655,
      "learning_rate": 0.00463302752293578,
      "loss": 0.7932,
      "step": 80
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.667579174041748,
      "learning_rate": 0.0045871559633027525,
      "loss": 0.7811,
      "step": 90
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.8475561141967773,
      "learning_rate": 0.004541284403669725,
      "loss": 0.7579,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7427716849451645,
      "eval_f1": 0.3926656517433595,
      "eval_loss": 0.7045032978057861,
      "eval_precision": 0.5204175676877641,
      "eval_recall": 0.3710290953850673,
      "eval_runtime": 5.2715,
      "eval_samples_per_second": 190.268,
      "eval_steps_per_second": 11.951,
      "step": 109
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.7959792613983154,
      "learning_rate": 0.004495412844036698,
      "loss": 0.8206,
      "step": 110
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.823697566986084,
      "learning_rate": 0.0044495412844036695,
      "loss": 0.7631,
      "step": 120
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.7514823079109192,
      "learning_rate": 0.004403669724770643,
      "loss": 0.7404,
      "step": 130
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.4910705089569092,
      "learning_rate": 0.004357798165137615,
      "loss": 0.7533,
      "step": 140
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.1586723327636719,
      "learning_rate": 0.004311926605504587,
      "loss": 0.7781,
      "step": 150
    },
    {
      "epoch": 1.46,
      "grad_norm": 1.106838583946228,
      "learning_rate": 0.0042660550458715595,
      "loss": 0.7337,
      "step": 160
    },
    {
      "epoch": 1.55,
      "grad_norm": 1.4145853519439697,
      "learning_rate": 0.004220183486238533,
      "loss": 0.7871,
      "step": 170
    },
    {
      "epoch": 1.64,
      "grad_norm": 1.0210392475128174,
      "learning_rate": 0.004174311926605505,
      "loss": 0.827,
      "step": 180
    },
    {
      "epoch": 1.74,
      "grad_norm": 1.4612568616867065,
      "learning_rate": 0.004128440366972477,
      "loss": 0.8952,
      "step": 190
    },
    {
      "epoch": 1.83,
      "grad_norm": 2.333279848098755,
      "learning_rate": 0.00408256880733945,
      "loss": 0.8019,
      "step": 200
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.9485087394714355,
      "learning_rate": 0.004036697247706422,
      "loss": 0.7689,
      "step": 210
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7278165503489531,
      "eval_f1": 0.3573330762987477,
      "eval_loss": 0.75123530626297,
      "eval_precision": 0.3964031496153109,
      "eval_recall": 0.35266989798605725,
      "eval_runtime": 5.3783,
      "eval_samples_per_second": 186.49,
      "eval_steps_per_second": 11.714,
      "step": 219
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.6179887056350708,
      "learning_rate": 0.003990825688073394,
      "loss": 0.8527,
      "step": 220
    },
    {
      "epoch": 2.1,
      "grad_norm": 1.0684826374053955,
      "learning_rate": 0.00394954128440367,
      "loss": 0.8174,
      "step": 230
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.6670346260070801,
      "learning_rate": 0.003903669724770642,
      "loss": 0.7894,
      "step": 240
    },
    {
      "epoch": 2.28,
      "grad_norm": 1.4650906324386597,
      "learning_rate": 0.003857798165137615,
      "loss": 0.7712,
      "step": 250
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.9125275611877441,
      "learning_rate": 0.003811926605504587,
      "loss": 0.8228,
      "step": 260
    },
    {
      "epoch": 2.47,
      "grad_norm": 2.175058603286743,
      "learning_rate": 0.0037706422018348625,
      "loss": 0.7592,
      "step": 270
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.2160431146621704,
      "learning_rate": 0.003724770642201835,
      "loss": 0.7986,
      "step": 280
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.9580293893814087,
      "learning_rate": 0.0036788990825688075,
      "loss": 0.7767,
      "step": 290
    },
    {
      "epoch": 2.74,
      "grad_norm": 1.544980525970459,
      "learning_rate": 0.0036330275229357802,
      "loss": 0.7882,
      "step": 300
    },
    {
      "epoch": 2.83,
      "grad_norm": 2.8008100986480713,
      "learning_rate": 0.003587155963302752,
      "loss": 0.8818,
      "step": 310
    },
    {
      "epoch": 2.92,
      "grad_norm": 3.061676025390625,
      "learning_rate": 0.003541284403669725,
      "loss": 0.7353,
      "step": 320
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7357926221335992,
      "eval_f1": 0.40020878247545494,
      "eval_loss": 0.7191066145896912,
      "eval_precision": 0.4630387632736769,
      "eval_recall": 0.4201909974743698,
      "eval_runtime": 5.4433,
      "eval_samples_per_second": 184.263,
      "eval_steps_per_second": 11.574,
      "step": 328
    },
    {
      "epoch": 3.01,
      "grad_norm": 6.1841206550598145,
      "learning_rate": 0.003495412844036697,
      "loss": 0.762,
      "step": 330
    },
    {
      "epoch": 3.11,
      "grad_norm": 2.103243589401245,
      "learning_rate": 0.00344954128440367,
      "loss": 0.798,
      "step": 340
    },
    {
      "epoch": 3.2,
      "grad_norm": 2.337041139602661,
      "learning_rate": 0.0034082568807339447,
      "loss": 0.7817,
      "step": 350
    },
    {
      "epoch": 3.29,
      "grad_norm": 3.2973079681396484,
      "learning_rate": 0.0033669724770642204,
      "loss": 0.7669,
      "step": 360
    },
    {
      "epoch": 3.38,
      "grad_norm": 1.7525138854980469,
      "learning_rate": 0.003321100917431193,
      "loss": 0.8788,
      "step": 370
    },
    {
      "epoch": 3.47,
      "grad_norm": 3.467615842819214,
      "learning_rate": 0.003275229357798165,
      "loss": 0.9385,
      "step": 380
    },
    {
      "epoch": 3.56,
      "grad_norm": 2.560269355773926,
      "learning_rate": 0.0032293577981651377,
      "loss": 0.9773,
      "step": 390
    },
    {
      "epoch": 3.65,
      "grad_norm": 0.9047974348068237,
      "learning_rate": 0.00318348623853211,
      "loss": 0.948,
      "step": 400
    },
    {
      "epoch": 3.74,
      "grad_norm": 0.7493894100189209,
      "learning_rate": 0.0031376146788990827,
      "loss": 0.9429,
      "step": 410
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.787327229976654,
      "learning_rate": 0.003091743119266055,
      "loss": 0.9193,
      "step": 420
    },
    {
      "epoch": 3.93,
      "grad_norm": 0.5133536458015442,
      "learning_rate": 0.0030458715596330278,
      "loss": 0.8429,
      "step": 430
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6809571286141576,
      "eval_f1": 0.18506746606597574,
      "eval_loss": 0.7857978940010071,
      "eval_precision": 0.4280081257540274,
      "eval_recall": 0.1812976208760752,
      "eval_runtime": 5.4671,
      "eval_samples_per_second": 183.46,
      "eval_steps_per_second": 11.523,
      "step": 438
    },
    {
      "epoch": 4.02,
      "grad_norm": 1.170052170753479,
      "learning_rate": 0.003,
      "loss": 0.7773,
      "step": 440
    },
    {
      "epoch": 4.11,
      "grad_norm": 1.006057858467102,
      "learning_rate": 0.002954128440366973,
      "loss": 0.8049,
      "step": 450
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.5809192657470703,
      "learning_rate": 0.002908256880733945,
      "loss": 0.81,
      "step": 460
    },
    {
      "epoch": 4.29,
      "grad_norm": 0.7377546429634094,
      "learning_rate": 0.002862385321100918,
      "loss": 0.7943,
      "step": 470
    },
    {
      "epoch": 4.38,
      "grad_norm": 0.5214153528213501,
      "learning_rate": 0.0028165137614678897,
      "loss": 0.7664,
      "step": 480
    },
    {
      "epoch": 4.47,
      "grad_norm": 0.615777313709259,
      "learning_rate": 0.0027706422018348624,
      "loss": 0.7769,
      "step": 490
    },
    {
      "epoch": 4.57,
      "grad_norm": 0.5365859866142273,
      "learning_rate": 0.0027247706422018347,
      "loss": 0.8232,
      "step": 500
    },
    {
      "epoch": 4.66,
      "grad_norm": 0.7110781669616699,
      "learning_rate": 0.0026788990825688075,
      "loss": 0.7755,
      "step": 510
    },
    {
      "epoch": 4.75,
      "grad_norm": 0.8415727615356445,
      "learning_rate": 0.0026330275229357798,
      "loss": 0.8154,
      "step": 520
    },
    {
      "epoch": 4.84,
      "grad_norm": 0.7670438289642334,
      "learning_rate": 0.0025871559633027525,
      "loss": 0.7386,
      "step": 530
    },
    {
      "epoch": 4.93,
      "grad_norm": 0.8495368957519531,
      "learning_rate": 0.002541284403669725,
      "loss": 0.7929,
      "step": 540
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7218344965104686,
      "eval_f1": 0.35225874228381837,
      "eval_loss": 0.7013418674468994,
      "eval_precision": 0.5157907633577941,
      "eval_recall": 0.397101205649215,
      "eval_runtime": 5.3335,
      "eval_samples_per_second": 188.058,
      "eval_steps_per_second": 11.812,
      "step": 547
    },
    {
      "epoch": 5.02,
      "grad_norm": 0.9542437791824341,
      "learning_rate": 0.0024954128440366975,
      "loss": 0.75,
      "step": 550
    },
    {
      "epoch": 5.11,
      "grad_norm": 0.7318850755691528,
      "learning_rate": 0.00244954128440367,
      "loss": 0.7665,
      "step": 560
    },
    {
      "epoch": 5.21,
      "grad_norm": 1.0321910381317139,
      "learning_rate": 0.0024036697247706426,
      "loss": 0.7015,
      "step": 570
    },
    {
      "epoch": 5.3,
      "grad_norm": 0.786632239818573,
      "learning_rate": 0.002357798165137615,
      "loss": 0.7756,
      "step": 580
    },
    {
      "epoch": 5.39,
      "grad_norm": 0.8995092511177063,
      "learning_rate": 0.002311926605504587,
      "loss": 0.79,
      "step": 590
    },
    {
      "epoch": 5.48,
      "grad_norm": 0.7932329773902893,
      "learning_rate": 0.00226605504587156,
      "loss": 0.7069,
      "step": 600
    },
    {
      "epoch": 5.57,
      "grad_norm": 0.7889924645423889,
      "learning_rate": 0.002220183486238532,
      "loss": 0.7507,
      "step": 610
    },
    {
      "epoch": 5.66,
      "grad_norm": 0.8880289793014526,
      "learning_rate": 0.002174311926605505,
      "loss": 0.7341,
      "step": 620
    },
    {
      "epoch": 5.75,
      "grad_norm": 0.5282242298126221,
      "learning_rate": 0.0021284403669724773,
      "loss": 0.685,
      "step": 630
    },
    {
      "epoch": 5.84,
      "grad_norm": 1.16853666305542,
      "learning_rate": 0.0020825688073394496,
      "loss": 0.7149,
      "step": 640
    },
    {
      "epoch": 5.94,
      "grad_norm": 0.6140002012252808,
      "learning_rate": 0.0020366972477064223,
      "loss": 0.6804,
      "step": 650
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7607178464606181,
      "eval_f1": 0.43914392933644575,
      "eval_loss": 0.6822256445884705,
      "eval_precision": 0.5010781993513719,
      "eval_recall": 0.4239506721942319,
      "eval_runtime": 5.2475,
      "eval_samples_per_second": 191.137,
      "eval_steps_per_second": 12.006,
      "step": 657
    },
    {
      "epoch": 6.03,
      "grad_norm": 0.7350448369979858,
      "learning_rate": 0.0019908256880733946,
      "loss": 0.7111,
      "step": 660
    },
    {
      "epoch": 6.12,
      "grad_norm": 0.6007147431373596,
      "learning_rate": 0.001944954128440367,
      "loss": 0.6981,
      "step": 670
    },
    {
      "epoch": 6.21,
      "grad_norm": 0.8717228770256042,
      "learning_rate": 0.0018990825688073396,
      "loss": 0.666,
      "step": 680
    },
    {
      "epoch": 6.3,
      "grad_norm": 0.5795207023620605,
      "learning_rate": 0.0018532110091743121,
      "loss": 0.7206,
      "step": 690
    },
    {
      "epoch": 6.39,
      "grad_norm": 0.9550521373748779,
      "learning_rate": 0.0018073394495412844,
      "loss": 0.7388,
      "step": 700
    },
    {
      "epoch": 6.48,
      "grad_norm": 1.932844877243042,
      "learning_rate": 0.001761467889908257,
      "loss": 0.7198,
      "step": 710
    },
    {
      "epoch": 6.58,
      "grad_norm": 0.8446419835090637,
      "learning_rate": 0.0017155963302752295,
      "loss": 0.6446,
      "step": 720
    },
    {
      "epoch": 6.67,
      "grad_norm": 1.2699838876724243,
      "learning_rate": 0.001669724770642202,
      "loss": 0.683,
      "step": 730
    },
    {
      "epoch": 6.76,
      "grad_norm": 0.8056779503822327,
      "learning_rate": 0.0016238532110091745,
      "loss": 0.7272,
      "step": 740
    },
    {
      "epoch": 6.85,
      "grad_norm": 0.8416026830673218,
      "learning_rate": 0.0015779816513761468,
      "loss": 0.7436,
      "step": 750
    },
    {
      "epoch": 6.94,
      "grad_norm": 0.8119596242904663,
      "learning_rate": 0.0015321100917431193,
      "loss": 0.6922,
      "step": 760
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7666999002991027,
      "eval_f1": 0.5226745508907684,
      "eval_loss": 0.653312087059021,
      "eval_precision": 0.6761986767002692,
      "eval_recall": 0.5105511763591389,
      "eval_runtime": 5.3357,
      "eval_samples_per_second": 187.978,
      "eval_steps_per_second": 11.807,
      "step": 766
    },
    {
      "epoch": 7.03,
      "grad_norm": 1.218758225440979,
      "learning_rate": 0.0014862385321100919,
      "loss": 0.7674,
      "step": 770
    },
    {
      "epoch": 7.12,
      "grad_norm": 1.0076816082000732,
      "learning_rate": 0.0014403669724770644,
      "loss": 0.6783,
      "step": 780
    },
    {
      "epoch": 7.21,
      "grad_norm": 1.2119933366775513,
      "learning_rate": 0.0013944954128440369,
      "loss": 0.6743,
      "step": 790
    },
    {
      "epoch": 7.31,
      "grad_norm": 0.7832847237586975,
      "learning_rate": 0.0013486238532110092,
      "loss": 0.751,
      "step": 800
    },
    {
      "epoch": 7.4,
      "grad_norm": 0.7728474140167236,
      "learning_rate": 0.0013027522935779817,
      "loss": 0.7252,
      "step": 810
    },
    {
      "epoch": 7.49,
      "grad_norm": 1.1566990613937378,
      "learning_rate": 0.0012568807339449542,
      "loss": 0.6176,
      "step": 820
    },
    {
      "epoch": 7.58,
      "grad_norm": 0.8185287117958069,
      "learning_rate": 0.0012110091743119267,
      "loss": 0.6928,
      "step": 830
    },
    {
      "epoch": 7.67,
      "grad_norm": 1.3969968557357788,
      "learning_rate": 0.0011651376146788993,
      "loss": 0.6475,
      "step": 840
    },
    {
      "epoch": 7.76,
      "grad_norm": 1.7943429946899414,
      "learning_rate": 0.0011192660550458716,
      "loss": 0.6414,
      "step": 850
    },
    {
      "epoch": 7.85,
      "grad_norm": 1.6549348831176758,
      "learning_rate": 0.001073394495412844,
      "loss": 0.717,
      "step": 860
    },
    {
      "epoch": 7.95,
      "grad_norm": 1.016310691833496,
      "learning_rate": 0.0010275229357798166,
      "loss": 0.6563,
      "step": 870
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7467597208374875,
      "eval_f1": 0.4496255647360247,
      "eval_loss": 0.6757794618606567,
      "eval_precision": 0.4547903051227771,
      "eval_recall": 0.4588581671719845,
      "eval_runtime": 5.4168,
      "eval_samples_per_second": 185.164,
      "eval_steps_per_second": 11.63,
      "step": 876
    },
    {
      "epoch": 8.04,
      "grad_norm": 1.1468579769134521,
      "learning_rate": 0.0009816513761467891,
      "loss": 0.7141,
      "step": 880
    },
    {
      "epoch": 8.13,
      "grad_norm": 3.9916632175445557,
      "learning_rate": 0.0009357798165137615,
      "loss": 0.6384,
      "step": 890
    },
    {
      "epoch": 8.22,
      "grad_norm": 1.3290263414382935,
      "learning_rate": 0.000889908256880734,
      "loss": 0.6598,
      "step": 900
    },
    {
      "epoch": 8.31,
      "grad_norm": 2.3329484462738037,
      "learning_rate": 0.0008440366972477065,
      "loss": 0.744,
      "step": 910
    },
    {
      "epoch": 8.4,
      "grad_norm": 1.5532264709472656,
      "learning_rate": 0.000798165137614679,
      "loss": 0.64,
      "step": 920
    },
    {
      "epoch": 8.49,
      "grad_norm": 0.8567171692848206,
      "learning_rate": 0.0007522935779816515,
      "loss": 0.6803,
      "step": 930
    },
    {
      "epoch": 8.58,
      "grad_norm": 0.8911744356155396,
      "learning_rate": 0.0007064220183486239,
      "loss": 0.6808,
      "step": 940
    },
    {
      "epoch": 8.68,
      "grad_norm": 0.8479019999504089,
      "learning_rate": 0.0006605504587155964,
      "loss": 0.6848,
      "step": 950
    },
    {
      "epoch": 8.77,
      "grad_norm": 1.6710033416748047,
      "learning_rate": 0.0006146788990825688,
      "loss": 0.6663,
      "step": 960
    },
    {
      "epoch": 8.86,
      "grad_norm": 0.7447875142097473,
      "learning_rate": 0.0005688073394495413,
      "loss": 0.6389,
      "step": 970
    },
    {
      "epoch": 8.95,
      "grad_norm": 1.6779547929763794,
      "learning_rate": 0.0005229357798165138,
      "loss": 0.6985,
      "step": 980
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7647058823529411,
      "eval_f1": 0.49146885132144036,
      "eval_loss": 0.6264088749885559,
      "eval_precision": 0.6450867892155376,
      "eval_recall": 0.4691799634422585,
      "eval_runtime": 5.2889,
      "eval_samples_per_second": 189.641,
      "eval_steps_per_second": 11.912,
      "step": 985
    },
    {
      "epoch": 9.04,
      "grad_norm": 0.9829887747764587,
      "learning_rate": 0.00047706422018348627,
      "loss": 0.608,
      "step": 990
    },
    {
      "epoch": 9.13,
      "grad_norm": 0.8272444009780884,
      "learning_rate": 0.00043119266055045873,
      "loss": 0.6661,
      "step": 1000
    },
    {
      "epoch": 9.22,
      "grad_norm": 1.056658387184143,
      "learning_rate": 0.0003853211009174312,
      "loss": 0.6982,
      "step": 1010
    },
    {
      "epoch": 9.32,
      "grad_norm": 0.8954740166664124,
      "learning_rate": 0.0003394495412844037,
      "loss": 0.5824,
      "step": 1020
    },
    {
      "epoch": 9.41,
      "grad_norm": 1.664057731628418,
      "learning_rate": 0.0002935779816513762,
      "loss": 0.657,
      "step": 1030
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.9140777587890625,
      "learning_rate": 0.00024770642201834864,
      "loss": 0.5975,
      "step": 1040
    },
    {
      "epoch": 9.59,
      "grad_norm": 1.4085102081298828,
      "learning_rate": 0.0002018348623853211,
      "loss": 0.6377,
      "step": 1050
    },
    {
      "epoch": 9.68,
      "grad_norm": 0.9538210034370422,
      "learning_rate": 0.0001559633027522936,
      "loss": 0.6508,
      "step": 1060
    },
    {
      "epoch": 9.77,
      "grad_norm": 1.5182217359542847,
      "learning_rate": 0.00011009174311926606,
      "loss": 0.6776,
      "step": 1070
    },
    {
      "epoch": 9.86,
      "grad_norm": 2.307227849960327,
      "learning_rate": 6.422018348623854e-05,
      "loss": 0.6758,
      "step": 1080
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.9896465539932251,
      "learning_rate": 1.834862385321101e-05,
      "loss": 0.6283,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "eval_accuracy": 0.7676969092721835,
      "eval_f1": 0.50879281522548,
      "eval_loss": 0.6179494857788086,
      "eval_precision": 0.5889156507510112,
      "eval_recall": 0.4796382542284182,
      "eval_runtime": 5.4809,
      "eval_samples_per_second": 182.998,
      "eval_steps_per_second": 11.494,
      "step": 1090
    },
    {
      "epoch": 9.95,
      "step": 1090,
      "total_flos": 5.442882169274339e+18,
      "train_loss": 0.7558124568484245,
      "train_runtime": 825.7866,
      "train_samples_per_second": 84.852,
      "train_steps_per_second": 1.32
    }
  ],
  "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
}