{
  "best_metric": 0.9638336347197106,
  "best_model_checkpoint": "resnet-50-finetuned-resnet50/checkpoint-1705",
  "epoch": 11.996784565916398,
  "global_step": 1860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 2.688172043010753e-06,
      "loss": 1.0948,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 5.376344086021506e-06,
      "loss": 1.0931,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 8.064516129032258e-06,
      "loss": 1.0889,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 1.0861,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 1.0796,
      "step": 50
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 1.0725,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.881720430107527e-05,
      "loss": 1.0602,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 1.0512,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 1.0386,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 1.0228,
      "step": 100
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 1.0067,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.981,
      "step": 120
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.9705,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.942,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.9229,
      "step": 150
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6424050632911392,
      "eval_loss": 0.8797757625579834,
      "eval_runtime": 237.5636,
      "eval_samples_per_second": 9.311,
      "eval_steps_per_second": 0.295,
      "step": 155
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.9396,
      "step": 160
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.8696,
      "step": 170
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.8369,
      "step": 180
    },
    {
      "epoch": 1.23,
      "learning_rate": 4.98805256869773e-05,
      "loss": 0.8137,
      "step": 190
    },
    {
      "epoch": 1.29,
      "learning_rate": 4.9581839904420555e-05,
      "loss": 0.7895,
      "step": 200
    },
    {
      "epoch": 1.35,
      "learning_rate": 4.92831541218638e-05,
      "loss": 0.7494,
      "step": 210
    },
    {
      "epoch": 1.42,
      "learning_rate": 4.898446833930705e-05,
      "loss": 0.7313,
      "step": 220
    },
    {
      "epoch": 1.48,
      "learning_rate": 4.86857825567503e-05,
      "loss": 0.6833,
      "step": 230
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.8387096774193554e-05,
      "loss": 0.6755,
      "step": 240
    },
    {
      "epoch": 1.61,
      "learning_rate": 4.80884109916368e-05,
      "loss": 0.6313,
      "step": 250
    },
    {
      "epoch": 1.68,
      "learning_rate": 4.778972520908005e-05,
      "loss": 0.6044,
      "step": 260
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.74910394265233e-05,
      "loss": 0.5762,
      "step": 270
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.7192353643966546e-05,
      "loss": 0.5494,
      "step": 280
    },
    {
      "epoch": 1.87,
      "learning_rate": 4.6893667861409805e-05,
      "loss": 0.5436,
      "step": 290
    },
    {
      "epoch": 1.93,
      "learning_rate": 4.659498207885305e-05,
      "loss": 0.5149,
      "step": 300
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4797,
      "step": 310
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8747739602169982,
      "eval_loss": 0.37069129943847656,
      "eval_runtime": 244.0414,
      "eval_samples_per_second": 9.064,
      "eval_steps_per_second": 0.287,
      "step": 310
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.5997610513739546e-05,
      "loss": 0.4886,
      "step": 320
    },
    {
      "epoch": 2.13,
      "learning_rate": 4.56989247311828e-05,
      "loss": 0.45,
      "step": 330
    },
    {
      "epoch": 2.19,
      "learning_rate": 4.540023894862604e-05,
      "loss": 0.4523,
      "step": 340
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.51015531660693e-05,
      "loss": 0.4588,
      "step": 350
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.4802867383512545e-05,
      "loss": 0.4219,
      "step": 360
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.4504181600955796e-05,
      "loss": 0.4036,
      "step": 370
    },
    {
      "epoch": 2.45,
      "learning_rate": 4.420549581839905e-05,
      "loss": 0.3875,
      "step": 380
    },
    {
      "epoch": 2.51,
      "learning_rate": 4.390681003584229e-05,
      "loss": 0.3735,
      "step": 390
    },
    {
      "epoch": 2.58,
      "learning_rate": 4.360812425328555e-05,
      "loss": 0.3719,
      "step": 400
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.3309438470728796e-05,
      "loss": 0.3958,
      "step": 410
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.301075268817205e-05,
      "loss": 0.3896,
      "step": 420
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.271206690561529e-05,
      "loss": 0.38,
      "step": 430
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.241338112305854e-05,
      "loss": 0.3629,
      "step": 440
    },
    {
      "epoch": 2.9,
      "learning_rate": 4.2114695340501795e-05,
      "loss": 0.3358,
      "step": 450
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.1816009557945046e-05,
      "loss": 0.3201,
      "step": 460
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9231464737793852,
      "eval_loss": 0.21964775025844574,
      "eval_runtime": 228.0728,
      "eval_samples_per_second": 9.699,
      "eval_steps_per_second": 0.307,
      "step": 465
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.15173237753883e-05,
      "loss": 0.367,
      "step": 470
    },
    {
      "epoch": 3.1,
      "learning_rate": 4.121863799283154e-05,
      "loss": 0.3273,
      "step": 480
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.0919952210274794e-05,
      "loss": 0.3418,
      "step": 490
    },
    {
      "epoch": 3.23,
      "learning_rate": 4.062126642771804e-05,
      "loss": 0.3462,
      "step": 500
    },
    {
      "epoch": 3.29,
      "learning_rate": 4.032258064516129e-05,
      "loss": 0.3211,
      "step": 510
    },
    {
      "epoch": 3.35,
      "learning_rate": 4.002389486260454e-05,
      "loss": 0.3173,
      "step": 520
    },
    {
      "epoch": 3.42,
      "learning_rate": 3.972520908004779e-05,
      "loss": 0.3148,
      "step": 530
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.9426523297491045e-05,
      "loss": 0.3089,
      "step": 540
    },
    {
      "epoch": 3.55,
      "learning_rate": 3.912783751493429e-05,
      "loss": 0.3132,
      "step": 550
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.882915173237754e-05,
      "loss": 0.3126,
      "step": 560
    },
    {
      "epoch": 3.68,
      "learning_rate": 3.8530465949820786e-05,
      "loss": 0.3335,
      "step": 570
    },
    {
      "epoch": 3.74,
      "learning_rate": 3.8231780167264044e-05,
      "loss": 0.3107,
      "step": 580
    },
    {
      "epoch": 3.8,
      "learning_rate": 3.793309438470729e-05,
      "loss": 0.3289,
      "step": 590
    },
    {
      "epoch": 3.87,
      "learning_rate": 3.763440860215054e-05,
      "loss": 0.3225,
      "step": 600
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.7335722819593785e-05,
      "loss": 0.3188,
      "step": 610
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2978,
      "step": 620
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9452983725135624,
      "eval_loss": 0.16960099339485168,
      "eval_runtime": 228.3891,
      "eval_samples_per_second": 9.685,
      "eval_steps_per_second": 0.306,
      "step": 620
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.673835125448029e-05,
      "loss": 0.3221,
      "step": 630
    },
    {
      "epoch": 4.13,
      "learning_rate": 3.643966547192354e-05,
      "loss": 0.3077,
      "step": 640
    },
    {
      "epoch": 4.19,
      "learning_rate": 3.614097968936679e-05,
      "loss": 0.2895,
      "step": 650
    },
    {
      "epoch": 4.26,
      "learning_rate": 3.5842293906810036e-05,
      "loss": 0.2822,
      "step": 660
    },
    {
      "epoch": 4.32,
      "learning_rate": 3.554360812425329e-05,
      "loss": 0.2864,
      "step": 670
    },
    {
      "epoch": 4.39,
      "learning_rate": 3.524492234169653e-05,
      "loss": 0.2868,
      "step": 680
    },
    {
      "epoch": 4.45,
      "learning_rate": 3.494623655913979e-05,
      "loss": 0.2721,
      "step": 690
    },
    {
      "epoch": 4.51,
      "learning_rate": 3.4647550776583035e-05,
      "loss": 0.2455,
      "step": 700
    },
    {
      "epoch": 4.58,
      "learning_rate": 3.4348864994026287e-05,
      "loss": 0.2657,
      "step": 710
    },
    {
      "epoch": 4.64,
      "learning_rate": 3.405017921146954e-05,
      "loss": 0.2797,
      "step": 720
    },
    {
      "epoch": 4.71,
      "learning_rate": 3.375149342891278e-05,
      "loss": 0.3008,
      "step": 730
    },
    {
      "epoch": 4.77,
      "learning_rate": 3.3452807646356034e-05,
      "loss": 0.2696,
      "step": 740
    },
    {
      "epoch": 4.84,
      "learning_rate": 3.3154121863799286e-05,
      "loss": 0.2978,
      "step": 750
    },
    {
      "epoch": 4.9,
      "learning_rate": 3.285543608124254e-05,
      "loss": 0.2775,
      "step": 760
    },
    {
      "epoch": 4.96,
      "learning_rate": 3.255675029868578e-05,
      "loss": 0.3208,
      "step": 770
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9525316455696202,
      "eval_loss": 0.13934803009033203,
      "eval_runtime": 228.8309,
      "eval_samples_per_second": 9.667,
      "eval_steps_per_second": 0.306,
      "step": 775
    },
    {
      "epoch": 5.03,
      "learning_rate": 3.2258064516129034e-05,
      "loss": 0.3063,
      "step": 780
    },
    {
      "epoch": 5.1,
      "learning_rate": 3.195937873357228e-05,
      "loss": 0.2632,
      "step": 790
    },
    {
      "epoch": 5.16,
      "learning_rate": 3.1660692951015537e-05,
      "loss": 0.2518,
      "step": 800
    },
    {
      "epoch": 5.23,
      "learning_rate": 3.136200716845878e-05,
      "loss": 0.253,
      "step": 810
    },
    {
      "epoch": 5.29,
      "learning_rate": 3.106332138590203e-05,
      "loss": 0.2748,
      "step": 820
    },
    {
      "epoch": 5.35,
      "learning_rate": 3.0764635603345284e-05,
      "loss": 0.2349,
      "step": 830
    },
    {
      "epoch": 5.42,
      "learning_rate": 3.046594982078853e-05,
      "loss": 0.2857,
      "step": 840
    },
    {
      "epoch": 5.48,
      "learning_rate": 3.016726403823178e-05,
      "loss": 0.2364,
      "step": 850
    },
    {
      "epoch": 5.55,
      "learning_rate": 2.9868578255675032e-05,
      "loss": 0.2772,
      "step": 860
    },
    {
      "epoch": 5.61,
      "learning_rate": 2.9569892473118284e-05,
      "loss": 0.2522,
      "step": 870
    },
    {
      "epoch": 5.68,
      "learning_rate": 2.9271206690561532e-05,
      "loss": 0.2404,
      "step": 880
    },
    {
      "epoch": 5.74,
      "learning_rate": 2.897252090800478e-05,
      "loss": 0.2913,
      "step": 890
    },
    {
      "epoch": 5.8,
      "learning_rate": 2.8673835125448028e-05,
      "loss": 0.2843,
      "step": 900
    },
    {
      "epoch": 5.87,
      "learning_rate": 2.8375149342891276e-05,
      "loss": 0.2918,
      "step": 910
    },
    {
      "epoch": 5.93,
      "learning_rate": 2.807646356033453e-05,
      "loss": 0.2479,
      "step": 920
    },
    {
      "epoch": 6.0,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.2599,
      "step": 930
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9561482820976492,
      "eval_loss": 0.1326771080493927,
      "eval_runtime": 235.7718,
      "eval_samples_per_second": 9.382,
      "eval_steps_per_second": 0.297,
      "step": 930
    },
    {
      "epoch": 6.06,
      "learning_rate": 2.747909199522103e-05,
      "loss": 0.2479,
      "step": 940
    },
    {
      "epoch": 6.13,
      "learning_rate": 2.718040621266428e-05,
      "loss": 0.2865,
      "step": 950
    },
    {
      "epoch": 6.19,
      "learning_rate": 2.6881720430107527e-05,
      "loss": 0.2508,
      "step": 960
    },
    {
      "epoch": 6.26,
      "learning_rate": 2.6583034647550775e-05,
      "loss": 0.279,
      "step": 970
    },
    {
      "epoch": 6.32,
      "learning_rate": 2.628434886499403e-05,
      "loss": 0.2534,
      "step": 980
    },
    {
      "epoch": 6.39,
      "learning_rate": 2.5985663082437278e-05,
      "loss": 0.2779,
      "step": 990
    },
    {
      "epoch": 6.45,
      "learning_rate": 2.5686977299880526e-05,
      "loss": 0.2627,
      "step": 1000
    },
    {
      "epoch": 6.51,
      "learning_rate": 2.5388291517323774e-05,
      "loss": 0.2588,
      "step": 1010
    },
    {
      "epoch": 6.58,
      "learning_rate": 2.5089605734767026e-05,
      "loss": 0.2611,
      "step": 1020
    },
    {
      "epoch": 6.64,
      "learning_rate": 2.4790919952210277e-05,
      "loss": 0.2677,
      "step": 1030
    },
    {
      "epoch": 6.71,
      "learning_rate": 2.4492234169653525e-05,
      "loss": 0.2389,
      "step": 1040
    },
    {
      "epoch": 6.77,
      "learning_rate": 2.4193548387096777e-05,
      "loss": 0.252,
      "step": 1050
    },
    {
      "epoch": 6.84,
      "learning_rate": 2.3894862604540025e-05,
      "loss": 0.2363,
      "step": 1060
    },
    {
      "epoch": 6.9,
      "learning_rate": 2.3596176821983273e-05,
      "loss": 0.2602,
      "step": 1070
    },
    {
      "epoch": 6.96,
      "learning_rate": 2.3297491039426525e-05,
      "loss": 0.2407,
      "step": 1080
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9593128390596745,
      "eval_loss": 0.12799127399921417,
      "eval_runtime": 237.4274,
      "eval_samples_per_second": 9.317,
      "eval_steps_per_second": 0.295,
      "step": 1085
    },
    {
      "epoch": 7.03,
      "learning_rate": 2.2998805256869773e-05,
      "loss": 0.2479,
      "step": 1090
    },
    {
      "epoch": 7.1,
      "learning_rate": 2.270011947431302e-05,
      "loss": 0.2604,
      "step": 1100
    },
    {
      "epoch": 7.16,
      "learning_rate": 2.2401433691756272e-05,
      "loss": 0.2578,
      "step": 1110
    },
    {
      "epoch": 7.23,
      "learning_rate": 2.2102747909199524e-05,
      "loss": 0.2373,
      "step": 1120
    },
    {
      "epoch": 7.29,
      "learning_rate": 2.1804062126642775e-05,
      "loss": 0.2429,
      "step": 1130
    },
    {
      "epoch": 7.35,
      "learning_rate": 2.1505376344086024e-05,
      "loss": 0.2534,
      "step": 1140
    },
    {
      "epoch": 7.42,
      "learning_rate": 2.120669056152927e-05,
      "loss": 0.2726,
      "step": 1150
    },
    {
      "epoch": 7.48,
      "learning_rate": 2.0908004778972523e-05,
      "loss": 0.2441,
      "step": 1160
    },
    {
      "epoch": 7.55,
      "learning_rate": 2.060931899641577e-05,
      "loss": 0.2613,
      "step": 1170
    },
    {
      "epoch": 7.61,
      "learning_rate": 2.031063321385902e-05,
      "loss": 0.254,
      "step": 1180
    },
    {
      "epoch": 7.68,
      "learning_rate": 2.001194743130227e-05,
      "loss": 0.2415,
      "step": 1190
    },
    {
      "epoch": 7.74,
      "learning_rate": 1.9713261648745522e-05,
      "loss": 0.2539,
      "step": 1200
    },
    {
      "epoch": 7.8,
      "learning_rate": 1.941457586618877e-05,
      "loss": 0.2473,
      "step": 1210
    },
    {
      "epoch": 7.87,
      "learning_rate": 1.9115890083632022e-05,
      "loss": 0.2419,
      "step": 1220
    },
    {
      "epoch": 7.93,
      "learning_rate": 1.881720430107527e-05,
      "loss": 0.2377,
      "step": 1230
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.2364,
      "step": 1240
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9593128390596745,
      "eval_loss": 0.11666559427976608,
      "eval_runtime": 231.3229,
      "eval_samples_per_second": 9.562,
      "eval_steps_per_second": 0.303,
      "step": 1240
    },
    {
      "epoch": 8.06,
      "learning_rate": 1.821983273596177e-05,
      "loss": 0.2481,
      "step": 1250
    },
    {
      "epoch": 8.13,
      "learning_rate": 1.7921146953405018e-05,
      "loss": 0.2723,
      "step": 1260
    },
    {
      "epoch": 8.19,
      "learning_rate": 1.7622461170848266e-05,
      "loss": 0.2392,
      "step": 1270
    },
    {
      "epoch": 8.26,
      "learning_rate": 1.7323775388291518e-05,
      "loss": 0.2257,
      "step": 1280
    },
    {
      "epoch": 8.32,
      "learning_rate": 1.702508960573477e-05,
      "loss": 0.23,
      "step": 1290
    },
    {
      "epoch": 8.39,
      "learning_rate": 1.6726403823178017e-05,
      "loss": 0.2237,
      "step": 1300
    },
    {
      "epoch": 8.45,
      "learning_rate": 1.642771804062127e-05,
      "loss": 0.2384,
      "step": 1310
    },
    {
      "epoch": 8.51,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.2336,
      "step": 1320
    },
    {
      "epoch": 8.58,
      "learning_rate": 1.5830346475507768e-05,
      "loss": 0.2745,
      "step": 1330
    },
    {
      "epoch": 8.64,
      "learning_rate": 1.5531660692951016e-05,
      "loss": 0.2079,
      "step": 1340
    },
    {
      "epoch": 8.71,
      "learning_rate": 1.5232974910394265e-05,
      "loss": 0.2189,
      "step": 1350
    },
    {
      "epoch": 8.77,
      "learning_rate": 1.4934289127837516e-05,
      "loss": 0.2357,
      "step": 1360
    },
    {
      "epoch": 8.84,
      "learning_rate": 1.4635603345280766e-05,
      "loss": 0.2575,
      "step": 1370
    },
    {
      "epoch": 8.9,
      "learning_rate": 1.4336917562724014e-05,
      "loss": 0.2583,
      "step": 1380
    },
    {
      "epoch": 8.96,
      "learning_rate": 1.4038231780167265e-05,
      "loss": 0.2147,
      "step": 1390
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9597649186256781,
      "eval_loss": 0.11601197719573975,
      "eval_runtime": 233.1414,
      "eval_samples_per_second": 9.488,
      "eval_steps_per_second": 0.3,
      "step": 1395
    },
    {
      "epoch": 9.03,
      "learning_rate": 1.3739545997610515e-05,
      "loss": 0.2431,
      "step": 1400
    },
    {
      "epoch": 9.1,
      "learning_rate": 1.3440860215053763e-05,
      "loss": 0.2511,
      "step": 1410
    },
    {
      "epoch": 9.16,
      "learning_rate": 1.3142174432497015e-05,
      "loss": 0.207,
      "step": 1420
    },
    {
      "epoch": 9.23,
      "learning_rate": 1.2843488649940263e-05,
      "loss": 0.2803,
      "step": 1430
    },
    {
      "epoch": 9.29,
      "learning_rate": 1.2544802867383513e-05,
      "loss": 0.2462,
      "step": 1440
    },
    {
      "epoch": 9.35,
      "learning_rate": 1.2246117084826763e-05,
      "loss": 0.2465,
      "step": 1450
    },
    {
      "epoch": 9.42,
      "learning_rate": 1.1947431302270013e-05,
      "loss": 0.2425,
      "step": 1460
    },
    {
      "epoch": 9.48,
      "learning_rate": 1.1648745519713262e-05,
      "loss": 0.2511,
      "step": 1470
    },
    {
      "epoch": 9.55,
      "learning_rate": 1.135005973715651e-05,
      "loss": 0.2333,
      "step": 1480
    },
    {
      "epoch": 9.61,
      "learning_rate": 1.1051373954599762e-05,
      "loss": 0.2188,
      "step": 1490
    },
    {
      "epoch": 9.68,
      "learning_rate": 1.0752688172043012e-05,
      "loss": 0.2549,
      "step": 1500
    },
    {
      "epoch": 9.74,
      "learning_rate": 1.0454002389486262e-05,
      "loss": 0.2827,
      "step": 1510
    },
    {
      "epoch": 9.8,
      "learning_rate": 1.015531660692951e-05,
      "loss": 0.2465,
      "step": 1520
    },
    {
      "epoch": 9.87,
      "learning_rate": 9.856630824372761e-06,
      "loss": 0.2231,
      "step": 1530
    },
    {
      "epoch": 9.93,
      "learning_rate": 9.557945041816011e-06,
      "loss": 0.2576,
      "step": 1540
    },
    {
      "epoch": 10.0,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.2634,
      "step": 1550
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9624773960216998,
      "eval_loss": 0.10948529094457626,
      "eval_runtime": 227.8347,
      "eval_samples_per_second": 9.709,
      "eval_steps_per_second": 0.307,
      "step": 1550
    },
    {
      "epoch": 10.06,
      "learning_rate": 8.960573476702509e-06,
      "loss": 0.2642,
      "step": 1560
    },
    {
      "epoch": 10.13,
      "learning_rate": 8.661887694145759e-06,
      "loss": 0.2714,
      "step": 1570
    },
    {
      "epoch": 10.19,
      "learning_rate": 8.363201911589009e-06,
      "loss": 0.2354,
      "step": 1580
    },
    {
      "epoch": 10.26,
      "learning_rate": 8.064516129032258e-06,
      "loss": 0.2322,
      "step": 1590
    },
    {
      "epoch": 10.32,
      "learning_rate": 7.765830346475508e-06,
      "loss": 0.2227,
      "step": 1600
    },
    {
      "epoch": 10.39,
      "learning_rate": 7.467144563918758e-06,
      "loss": 0.2399,
      "step": 1610
    },
    {
      "epoch": 10.45,
      "learning_rate": 7.168458781362007e-06,
      "loss": 0.233,
      "step": 1620
    },
    {
      "epoch": 10.51,
      "learning_rate": 6.869772998805258e-06,
      "loss": 0.2572,
      "step": 1630
    },
    {
      "epoch": 10.58,
      "learning_rate": 6.5710872162485075e-06,
      "loss": 0.2219,
      "step": 1640
    },
    {
      "epoch": 10.64,
      "learning_rate": 6.2724014336917564e-06,
      "loss": 0.2517,
      "step": 1650
    },
    {
      "epoch": 10.71,
      "learning_rate": 5.973715651135006e-06,
      "loss": 0.2471,
      "step": 1660
    },
    {
      "epoch": 10.77,
      "learning_rate": 5.675029868578255e-06,
      "loss": 0.2378,
      "step": 1670
    },
    {
      "epoch": 10.84,
      "learning_rate": 5.376344086021506e-06,
      "loss": 0.2384,
      "step": 1680
    },
    {
      "epoch": 10.9,
      "learning_rate": 5.077658303464755e-06,
      "loss": 0.2426,
      "step": 1690
    },
    {
      "epoch": 10.96,
      "learning_rate": 4.7789725209080055e-06,
      "loss": 0.2159,
      "step": 1700
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9638336347197106,
      "eval_loss": 0.10932076722383499,
      "eval_runtime": 252.327,
      "eval_samples_per_second": 8.766,
      "eval_steps_per_second": 0.277,
      "step": 1705
    },
    {
      "epoch": 11.03,
      "learning_rate": 4.4802867383512545e-06,
      "loss": 0.2404,
      "step": 1710
    },
    {
      "epoch": 11.1,
      "learning_rate": 4.181600955794504e-06,
      "loss": 0.2461,
      "step": 1720
    },
    {
      "epoch": 11.16,
      "learning_rate": 3.882915173237754e-06,
      "loss": 0.239,
      "step": 1730
    },
    {
      "epoch": 11.23,
      "learning_rate": 3.5842293906810035e-06,
      "loss": 0.2171,
      "step": 1740
    },
    {
      "epoch": 11.29,
      "learning_rate": 3.2855436081242537e-06,
      "loss": 0.2376,
      "step": 1750
    },
    {
      "epoch": 11.35,
      "learning_rate": 2.986857825567503e-06,
      "loss": 0.2301,
      "step": 1760
    },
    {
      "epoch": 11.42,
      "learning_rate": 2.688172043010753e-06,
      "loss": 0.2307,
      "step": 1770
    },
    {
      "epoch": 11.48,
      "learning_rate": 2.3894862604540028e-06,
      "loss": 0.2216,
      "step": 1780
    },
    {
      "epoch": 11.55,
      "learning_rate": 2.090800477897252e-06,
      "loss": 0.2246,
      "step": 1790
    },
    {
      "epoch": 11.61,
      "learning_rate": 1.7921146953405017e-06,
      "loss": 0.2639,
      "step": 1800
    },
    {
      "epoch": 11.68,
      "learning_rate": 1.4934289127837516e-06,
      "loss": 0.2243,
      "step": 1810
    },
    {
      "epoch": 11.74,
      "learning_rate": 1.1947431302270014e-06,
      "loss": 0.2217,
      "step": 1820
    },
    {
      "epoch": 11.8,
      "learning_rate": 8.960573476702509e-07,
      "loss": 0.2324,
      "step": 1830
    },
    {
      "epoch": 11.87,
      "learning_rate": 5.973715651135007e-07,
      "loss": 0.2496,
      "step": 1840
    },
    {
      "epoch": 11.93,
      "learning_rate": 2.9868578255675034e-07,
      "loss": 0.2464,
      "step": 1850
    },
    {
      "epoch": 12.0,
      "learning_rate": 0.0,
      "loss": 0.2355,
      "step": 1860
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9638336347197106,
      "eval_loss": 0.11173169314861298,
      "eval_runtime": 228.1044,
      "eval_samples_per_second": 9.697,
      "eval_steps_per_second": 0.307,
      "step": 1860
    }
  ],
  "max_steps": 1860,
  "num_train_epochs": 12,
  "total_flos": 5.070182024513249e+18,
  "trial_name": null,
  "trial_params": null
}