{
  "best_metric": 0.8728268768516966,
  "best_model_checkpoint": "stool-condition-classification/checkpoint-500",
  "epoch": 10.0,
  "eval_steps": 100,
  "global_step": 1020,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 7.584359169006348,
      "learning_rate": 0.00019803921568627454,
      "loss": 0.6059,
      "step": 10
    },
    {
      "epoch": 0.2,
      "grad_norm": 3.984663486480713,
      "learning_rate": 0.000196078431372549,
      "loss": 0.69,
      "step": 20
    },
    {
      "epoch": 0.29,
      "grad_norm": 5.750972270965576,
      "learning_rate": 0.00019411764705882354,
      "loss": 0.652,
      "step": 30
    },
    {
      "epoch": 0.39,
      "grad_norm": 5.540163993835449,
      "learning_rate": 0.00019215686274509807,
      "loss": 0.567,
      "step": 40
    },
    {
      "epoch": 0.49,
      "grad_norm": 12.890495300292969,
      "learning_rate": 0.00019019607843137254,
      "loss": 0.5255,
      "step": 50
    },
    {
      "epoch": 0.59,
      "grad_norm": 11.914591789245605,
      "learning_rate": 0.00018823529411764707,
      "loss": 0.5776,
      "step": 60
    },
    {
      "epoch": 0.69,
      "grad_norm": 8.995306015014648,
      "learning_rate": 0.00018627450980392157,
      "loss": 0.4868,
      "step": 70
    },
    {
      "epoch": 0.78,
      "grad_norm": 7.732036590576172,
      "learning_rate": 0.00018431372549019607,
      "loss": 0.6819,
      "step": 80
    },
    {
      "epoch": 0.88,
      "grad_norm": 9.876835823059082,
      "learning_rate": 0.0001823529411764706,
      "loss": 0.5809,
      "step": 90
    },
    {
      "epoch": 0.98,
      "grad_norm": 6.603707790374756,
      "learning_rate": 0.0001803921568627451,
      "loss": 0.5076,
      "step": 100
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.7730870712401056,
      "eval_auroc": 0.8537928335848846,
      "eval_f1": 0.6906474820143884,
      "eval_loss": 0.5360854268074036,
      "eval_model_selection": 0.559226340209067,
      "eval_npv": 0.7060931899641577,
      "eval_ppv": 0.96,
      "eval_runtime": 21.3071,
      "eval_samples_per_second": 17.787,
      "eval_sensitivity": 0.5393258426966292,
      "eval_specificty": 0.9800995024875622,
      "eval_steps_per_second": 2.253,
      "step": 100
    },
    {
      "epoch": 1.08,
      "grad_norm": 2.871717929840088,
      "learning_rate": 0.00017843137254901963,
      "loss": 0.6088,
      "step": 110
    },
    {
      "epoch": 1.18,
      "grad_norm": 7.064380645751953,
      "learning_rate": 0.00017647058823529413,
      "loss": 0.5249,
      "step": 120
    },
    {
      "epoch": 1.27,
      "grad_norm": 4.083934307098389,
      "learning_rate": 0.00017450980392156863,
      "loss": 0.5367,
      "step": 130
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.9436535835266113,
      "learning_rate": 0.00017254901960784316,
      "loss": 0.4674,
      "step": 140
    },
    {
      "epoch": 1.47,
      "grad_norm": 2.449876308441162,
      "learning_rate": 0.00017058823529411766,
      "loss": 0.5506,
      "step": 150
    },
    {
      "epoch": 1.57,
      "grad_norm": 2.2175521850585938,
      "learning_rate": 0.00016862745098039216,
      "loss": 0.4266,
      "step": 160
    },
    {
      "epoch": 1.67,
      "grad_norm": 1.2798587083816528,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.483,
      "step": 170
    },
    {
      "epoch": 1.76,
      "grad_norm": 4.6635847091674805,
      "learning_rate": 0.0001647058823529412,
      "loss": 0.5022,
      "step": 180
    },
    {
      "epoch": 1.86,
      "grad_norm": 10.006440162658691,
      "learning_rate": 0.0001627450980392157,
      "loss": 0.5173,
      "step": 190
    },
    {
      "epoch": 1.96,
      "grad_norm": 43.053401947021484,
      "learning_rate": 0.00016078431372549022,
      "loss": 0.4086,
      "step": 200
    },
    {
      "epoch": 1.96,
      "eval_accuracy": 0.783641160949868,
      "eval_auroc": 0.8727989267147409,
      "eval_f1": 0.722972972972973,
      "eval_loss": 0.48566874861717224,
      "eval_model_selection": 0.655849963664822,
      "eval_npv": 0.7279693486590039,
      "eval_ppv": 0.9067796610169492,
      "eval_runtime": 20.6546,
      "eval_samples_per_second": 18.349,
      "eval_sensitivity": 0.601123595505618,
      "eval_specificty": 0.945273631840796,
      "eval_steps_per_second": 2.324,
      "step": 200
    },
    {
      "epoch": 2.06,
      "grad_norm": 6.759799957275391,
      "learning_rate": 0.0001588235294117647,
      "loss": 0.4347,
      "step": 210
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.868139386177063,
      "learning_rate": 0.00015686274509803922,
      "loss": 0.4856,
      "step": 220
    },
    {
      "epoch": 2.25,
      "grad_norm": 6.924500465393066,
      "learning_rate": 0.00015490196078431375,
      "loss": 0.6559,
      "step": 230
    },
    {
      "epoch": 2.35,
      "grad_norm": 3.674975872039795,
      "learning_rate": 0.00015294117647058822,
      "loss": 0.4086,
      "step": 240
    },
    {
      "epoch": 2.45,
      "grad_norm": 3.160930633544922,
      "learning_rate": 0.00015098039215686275,
      "loss": 0.4172,
      "step": 250
    },
    {
      "epoch": 2.55,
      "grad_norm": 5.0961785316467285,
      "learning_rate": 0.00014901960784313728,
      "loss": 0.3948,
      "step": 260
    },
    {
      "epoch": 2.65,
      "grad_norm": 3.1396780014038086,
      "learning_rate": 0.00014705882352941178,
      "loss": 0.3086,
      "step": 270
    },
    {
      "epoch": 2.75,
      "grad_norm": 8.720426559448242,
      "learning_rate": 0.00014509803921568628,
      "loss": 0.5149,
      "step": 280
    },
    {
      "epoch": 2.84,
      "grad_norm": 2.7797322273254395,
      "learning_rate": 0.00014313725490196078,
      "loss": 0.6524,
      "step": 290
    },
    {
      "epoch": 2.94,
      "grad_norm": 8.662543296813965,
      "learning_rate": 0.0001411764705882353,
      "loss": 0.5208,
      "step": 300
    },
    {
      "epoch": 2.94,
      "eval_accuracy": 0.7598944591029023,
      "eval_auroc": 0.8058862988428643,
      "eval_f1": 0.7055016181229774,
      "eval_loss": 0.5108699798583984,
      "eval_model_selection": 0.7218122868802057,
      "eval_npv": 0.7217741935483871,
      "eval_ppv": 0.8320610687022901,
      "eval_runtime": 20.396,
      "eval_samples_per_second": 18.582,
      "eval_sensitivity": 0.6123595505617978,
      "eval_specificty": 0.8905472636815921,
      "eval_steps_per_second": 2.353,
      "step": 300
    },
    {
      "epoch": 3.04,
      "grad_norm": 5.801065444946289,
      "learning_rate": 0.0001392156862745098,
      "loss": 0.4872,
      "step": 310
    },
    {
      "epoch": 3.14,
      "grad_norm": 4.54748010635376,
      "learning_rate": 0.0001372549019607843,
      "loss": 0.4804,
      "step": 320
    },
    {
      "epoch": 3.24,
      "grad_norm": 6.328420639038086,
      "learning_rate": 0.00013529411764705884,
      "loss": 0.5523,
      "step": 330
    },
    {
      "epoch": 3.33,
      "grad_norm": 2.860734462738037,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.4061,
      "step": 340
    },
    {
      "epoch": 3.43,
      "grad_norm": 5.824751377105713,
      "learning_rate": 0.00013137254901960784,
      "loss": 0.4895,
      "step": 350
    },
    {
      "epoch": 3.53,
      "grad_norm": 1.802435278892517,
      "learning_rate": 0.00012941176470588237,
      "loss": 0.4011,
      "step": 360
    },
    {
      "epoch": 3.63,
      "grad_norm": 2.1423304080963135,
      "learning_rate": 0.00012745098039215687,
      "loss": 0.3881,
      "step": 370
    },
    {
      "epoch": 3.73,
      "grad_norm": 2.6464078426361084,
      "learning_rate": 0.00012549019607843137,
      "loss": 0.4371,
      "step": 380
    },
    {
      "epoch": 3.82,
      "grad_norm": 1.5666183233261108,
      "learning_rate": 0.0001235294117647059,
      "loss": 0.4256,
      "step": 390
    },
    {
      "epoch": 3.92,
      "grad_norm": 7.859738349914551,
      "learning_rate": 0.00012156862745098039,
      "loss": 0.474,
      "step": 400
    },
    {
      "epoch": 3.92,
      "eval_accuracy": 0.7994722955145118,
      "eval_auroc": 0.8600536642629548,
      "eval_f1": 0.7432432432432432,
      "eval_loss": 0.5211557149887085,
      "eval_model_selection": 0.6577785231147633,
      "eval_npv": 0.7394636015325671,
      "eval_ppv": 0.9322033898305084,
      "eval_runtime": 20.1111,
      "eval_samples_per_second": 18.845,
      "eval_sensitivity": 0.6179775280898876,
      "eval_specificty": 0.9601990049751243,
      "eval_steps_per_second": 2.387,
      "step": 400
    },
    {
      "epoch": 4.02,
      "grad_norm": 5.231843948364258,
      "learning_rate": 0.0001196078431372549,
      "loss": 0.3969,
      "step": 410
    },
    {
      "epoch": 4.12,
      "grad_norm": 1.5469104051589966,
      "learning_rate": 0.00011764705882352942,
      "loss": 0.4781,
      "step": 420
    },
    {
      "epoch": 4.22,
      "grad_norm": 5.2077765464782715,
      "learning_rate": 0.00011568627450980394,
      "loss": 0.4261,
      "step": 430
    },
    {
      "epoch": 4.31,
      "grad_norm": 0.9990509748458862,
      "learning_rate": 0.00011372549019607843,
      "loss": 0.3773,
      "step": 440
    },
    {
      "epoch": 4.41,
      "grad_norm": 5.277149677276611,
      "learning_rate": 0.00011176470588235294,
      "loss": 0.4551,
      "step": 450
    },
    {
      "epoch": 4.51,
      "grad_norm": 1.3224270343780518,
      "learning_rate": 0.00010980392156862746,
      "loss": 0.3533,
      "step": 460
    },
    {
      "epoch": 4.61,
      "grad_norm": 2.360970973968506,
      "learning_rate": 0.00010784313725490196,
      "loss": 0.4317,
      "step": 470
    },
    {
      "epoch": 4.71,
      "grad_norm": 2.176283836364746,
      "learning_rate": 0.00010588235294117647,
      "loss": 0.4718,
      "step": 480
    },
    {
      "epoch": 4.8,
      "grad_norm": 1.7329970598220825,
      "learning_rate": 0.00010392156862745099,
      "loss": 0.426,
      "step": 490
    },
    {
      "epoch": 4.9,
      "grad_norm": 7.07577657699585,
      "learning_rate": 0.00010196078431372549,
      "loss": 0.4285,
      "step": 500
    },
    {
      "epoch": 4.9,
      "eval_accuracy": 0.7757255936675461,
      "eval_auroc": 0.8728268768516966,
      "eval_f1": 0.7578347578347578,
      "eval_loss": 0.4510786235332489,
      "eval_model_selection": 0.9461959863603332,
      "eval_npv": 0.7815533980582524,
      "eval_ppv": 0.7687861271676301,
      "eval_runtime": 20.6194,
      "eval_samples_per_second": 18.381,
      "eval_sensitivity": 0.7471910112359551,
      "eval_specificty": 0.8009950248756219,
      "eval_steps_per_second": 2.328,
      "step": 500
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.841150760650635,
      "learning_rate": 0.0001,
      "loss": 0.3839,
      "step": 510
    },
    {
      "epoch": 5.1,
      "grad_norm": 2.587709903717041,
      "learning_rate": 9.80392156862745e-05,
      "loss": 0.396,
      "step": 520
    },
    {
      "epoch": 5.2,
      "grad_norm": 2.148972272872925,
      "learning_rate": 9.607843137254903e-05,
      "loss": 0.3475,
      "step": 530
    },
    {
      "epoch": 5.29,
      "grad_norm": 6.0918803215026855,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.3433,
      "step": 540
    },
    {
      "epoch": 5.39,
      "grad_norm": 1.9835032224655151,
      "learning_rate": 9.215686274509804e-05,
      "loss": 0.3289,
      "step": 550
    },
    {
      "epoch": 5.49,
      "grad_norm": 1.912084698677063,
      "learning_rate": 9.019607843137255e-05,
      "loss": 0.3887,
      "step": 560
    },
    {
      "epoch": 5.59,
      "grad_norm": 2.5347495079040527,
      "learning_rate": 8.823529411764706e-05,
      "loss": 0.4516,
      "step": 570
    },
    {
      "epoch": 5.69,
      "grad_norm": 3.123342990875244,
      "learning_rate": 8.627450980392158e-05,
      "loss": 0.3973,
      "step": 580
    },
    {
      "epoch": 5.78,
      "grad_norm": 6.096330165863037,
      "learning_rate": 8.431372549019608e-05,
      "loss": 0.3895,
      "step": 590
    },
    {
      "epoch": 5.88,
      "grad_norm": 1.6143405437469482,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.3506,
      "step": 600
    },
    {
      "epoch": 5.88,
      "eval_accuracy": 0.8047493403693932,
      "eval_auroc": 0.8691095086365922,
      "eval_f1": 0.7658227848101267,
      "eval_loss": 0.47164368629455566,
      "eval_model_selection": 0.764352395326737,
      "eval_npv": 0.7634854771784232,
      "eval_ppv": 0.8768115942028986,
      "eval_runtime": 20.5548,
      "eval_samples_per_second": 18.438,
      "eval_sensitivity": 0.6797752808988764,
      "eval_specificty": 0.9154228855721394,
      "eval_steps_per_second": 2.335,
      "step": 600
    },
    {
      "epoch": 5.98,
      "grad_norm": 2.3068525791168213,
      "learning_rate": 8.039215686274511e-05,
      "loss": 0.3566,
      "step": 610
    },
    {
      "epoch": 6.08,
      "grad_norm": 3.027449369430542,
      "learning_rate": 7.843137254901961e-05,
      "loss": 0.3662,
      "step": 620
    },
    {
      "epoch": 6.18,
      "grad_norm": 3.0861737728118896,
      "learning_rate": 7.647058823529411e-05,
      "loss": 0.4272,
      "step": 630
    },
    {
      "epoch": 6.27,
      "grad_norm": 2.3498377799987793,
      "learning_rate": 7.450980392156864e-05,
      "loss": 0.2825,
      "step": 640
    },
    {
      "epoch": 6.37,
      "grad_norm": 2.3852789402008057,
      "learning_rate": 7.254901960784314e-05,
      "loss": 0.361,
      "step": 650
    },
    {
      "epoch": 6.47,
      "grad_norm": 2.655799627304077,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.291,
      "step": 660
    },
    {
      "epoch": 6.57,
      "grad_norm": 4.058769226074219,
      "learning_rate": 6.862745098039216e-05,
      "loss": 0.3045,
      "step": 670
    },
    {
      "epoch": 6.67,
      "grad_norm": 4.20621919631958,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.3931,
      "step": 680
    },
    {
      "epoch": 6.76,
      "grad_norm": 4.316690444946289,
      "learning_rate": 6.470588235294118e-05,
      "loss": 0.2858,
      "step": 690
    },
    {
      "epoch": 6.86,
      "grad_norm": 3.3390917778015137,
      "learning_rate": 6.274509803921569e-05,
      "loss": 0.4239,
      "step": 700
    },
    {
      "epoch": 6.86,
      "eval_accuracy": 0.8100263852242744,
      "eval_auroc": 0.8517245234501649,
      "eval_f1": 0.7677419354838709,
      "eval_loss": 0.504310131072998,
      "eval_model_selection": 0.7332159427581195,
      "eval_npv": 0.7611336032388664,
      "eval_ppv": 0.9015151515151515,
      "eval_runtime": 20.9343,
      "eval_samples_per_second": 18.104,
      "eval_sensitivity": 0.6685393258426966,
      "eval_specificty": 0.9353233830845771,
      "eval_steps_per_second": 2.293,
      "step": 700
    },
    {
      "epoch": 6.96,
      "grad_norm": 3.034562587738037,
      "learning_rate": 6.078431372549019e-05,
      "loss": 0.3033,
      "step": 710
    },
    {
      "epoch": 7.06,
      "grad_norm": 0.7752771973609924,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.2963,
      "step": 720
    },
    {
      "epoch": 7.16,
      "grad_norm": 1.2533752918243408,
      "learning_rate": 5.6862745098039215e-05,
      "loss": 0.2974,
      "step": 730
    },
    {
      "epoch": 7.25,
      "grad_norm": 1.748043179512024,
      "learning_rate": 5.490196078431373e-05,
      "loss": 0.3549,
      "step": 740
    },
    {
      "epoch": 7.35,
      "grad_norm": 2.4874391555786133,
      "learning_rate": 5.294117647058824e-05,
      "loss": 0.413,
      "step": 750
    },
    {
      "epoch": 7.45,
      "grad_norm": 3.017951726913452,
      "learning_rate": 5.0980392156862745e-05,
      "loss": 0.2809,
      "step": 760
    },
    {
      "epoch": 7.55,
      "grad_norm": 5.023625373840332,
      "learning_rate": 4.901960784313725e-05,
      "loss": 0.3603,
      "step": 770
    },
    {
      "epoch": 7.65,
      "grad_norm": 2.754984140396118,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.2837,
      "step": 780
    },
    {
      "epoch": 7.75,
      "grad_norm": 1.0377743244171143,
      "learning_rate": 4.5098039215686275e-05,
      "loss": 0.3169,
      "step": 790
    },
    {
      "epoch": 7.84,
      "grad_norm": 4.163206100463867,
      "learning_rate": 4.313725490196079e-05,
      "loss": 0.2447,
      "step": 800
    },
    {
      "epoch": 7.84,
      "eval_accuracy": 0.8073878627968337,
      "eval_auroc": 0.8591872100173291,
      "eval_f1": 0.7711598746081505,
      "eval_loss": 0.580413818359375,
      "eval_model_selection": 0.7805634747610264,
      "eval_npv": 0.7689075630252101,
      "eval_ppv": 0.8723404255319149,
      "eval_runtime": 20.1671,
      "eval_samples_per_second": 18.793,
      "eval_sensitivity": 0.6910112359550562,
      "eval_specificty": 0.9104477611940298,
      "eval_steps_per_second": 2.38,
      "step": 800
    },
    {
      "epoch": 7.94,
      "grad_norm": 3.30311918258667,
      "learning_rate": 4.11764705882353e-05,
      "loss": 0.3555,
      "step": 810
    },
    {
      "epoch": 8.04,
      "grad_norm": 4.967993259429932,
      "learning_rate": 3.9215686274509805e-05,
      "loss": 0.2679,
      "step": 820
    },
    {
      "epoch": 8.14,
      "grad_norm": 4.838258266448975,
      "learning_rate": 3.725490196078432e-05,
      "loss": 0.2416,
      "step": 830
    },
    {
      "epoch": 8.24,
      "grad_norm": 4.202385425567627,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.1758,
      "step": 840
    },
    {
      "epoch": 8.33,
      "grad_norm": 3.0037460327148438,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.3192,
      "step": 850
    },
    {
      "epoch": 8.43,
      "grad_norm": 4.337649822235107,
      "learning_rate": 3.137254901960784e-05,
      "loss": 0.2669,
      "step": 860
    },
    {
      "epoch": 8.53,
      "grad_norm": 6.011264801025391,
      "learning_rate": 2.9411764705882354e-05,
      "loss": 0.228,
      "step": 870
    },
    {
      "epoch": 8.63,
      "grad_norm": 0.5027784109115601,
      "learning_rate": 2.7450980392156865e-05,
      "loss": 0.3023,
      "step": 880
    },
    {
      "epoch": 8.73,
      "grad_norm": 1.8822238445281982,
      "learning_rate": 2.5490196078431373e-05,
      "loss": 0.3109,
      "step": 890
    },
    {
      "epoch": 8.82,
      "grad_norm": 3.6861536502838135,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 0.1739,
      "step": 900
    },
    {
      "epoch": 8.82,
      "eval_accuracy": 0.8073878627968337,
      "eval_auroc": 0.856224495500028,
      "eval_f1": 0.7767584097859327,
      "eval_loss": 0.6224877238273621,
      "eval_model_selection": 0.8229358823858236,
      "eval_npv": 0.7782608695652173,
      "eval_ppv": 0.8523489932885906,
      "eval_runtime": 20.4294,
      "eval_samples_per_second": 18.552,
      "eval_sensitivity": 0.7134831460674157,
      "eval_specificty": 0.8905472636815921,
      "eval_steps_per_second": 2.35,
      "step": 900
    },
    {
      "epoch": 8.92,
      "grad_norm": 5.0469770431518555,
      "learning_rate": 2.1568627450980395e-05,
      "loss": 0.4354,
      "step": 910
    },
    {
      "epoch": 9.02,
      "grad_norm": 4.443832874298096,
      "learning_rate": 1.9607843137254903e-05,
      "loss": 0.334,
      "step": 920
    },
    {
      "epoch": 9.12,
      "grad_norm": 2.0471107959747314,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 0.2051,
      "step": 930
    },
    {
      "epoch": 9.22,
      "grad_norm": 3.442962169647217,
      "learning_rate": 1.568627450980392e-05,
      "loss": 0.2112,
      "step": 940
    },
    {
      "epoch": 9.31,
      "grad_norm": 2.31436824798584,
      "learning_rate": 1.3725490196078432e-05,
      "loss": 0.3268,
      "step": 950
    },
    {
      "epoch": 9.41,
      "grad_norm": 2.2342071533203125,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 0.3107,
      "step": 960
    },
    {
      "epoch": 9.51,
      "grad_norm": 3.2926645278930664,
      "learning_rate": 9.803921568627451e-06,
      "loss": 0.2284,
      "step": 970
    },
    {
      "epoch": 9.61,
      "grad_norm": 2.639843702316284,
      "learning_rate": 7.84313725490196e-06,
      "loss": 0.2269,
      "step": 980
    },
    {
      "epoch": 9.71,
      "grad_norm": 2.7015938758850098,
      "learning_rate": 5.882352941176471e-06,
      "loss": 0.1987,
      "step": 990
    },
    {
      "epoch": 9.8,
      "grad_norm": 8.358205795288086,
      "learning_rate": 3.92156862745098e-06,
      "loss": 0.2888,
      "step": 1000
    },
    {
      "epoch": 9.8,
      "eval_accuracy": 0.8047493403693932,
      "eval_auroc": 0.856979149197831,
      "eval_f1": 0.783625730994152,
      "eval_loss": 0.5807113647460938,
      "eval_model_selection": 0.9020627201073285,
      "eval_npv": 0.7953488372093023,
      "eval_ppv": 0.8170731707317073,
      "eval_runtime": 20.5189,
      "eval_samples_per_second": 18.471,
      "eval_sensitivity": 0.7528089887640449,
      "eval_specificty": 0.8507462686567164,
      "eval_steps_per_second": 2.339,
      "step": 1000
    },
    {
      "epoch": 9.9,
      "grad_norm": 3.527721881866455,
      "learning_rate": 1.96078431372549e-06,
      "loss": 0.3236,
      "step": 1010
    },
    {
      "epoch": 10.0,
      "grad_norm": 6.274048805236816,
      "learning_rate": 0.0,
      "loss": 0.26,
      "step": 1020
    },
    {
      "epoch": 10.0,
      "step": 1020,
      "total_flos": 1.25924483123712e+18,
      "train_loss": 0.40091259911948557,
      "train_runtime": 1906.0636,
      "train_samples_per_second": 8.525,
      "train_steps_per_second": 0.535
    }
  ],
  "logging_steps": 10,
  "max_steps": 1020,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 1.25924483123712e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}