{
  "best_metric": 0.9227471218737594,
  "best_model_checkpoint": "outputs/whisper-small-keyword-spotting-m-agv-grabo/checkpoint-723",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3615,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.7624309392265193e-05,
      "loss": 0.6646,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 5.524861878453039e-05,
      "loss": 0.5982,
      "step": 20
    },
    {
      "epoch": 0.04,
      "learning_rate": 8.287292817679558e-05,
      "loss": 0.5497,
      "step": 30
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00011049723756906077,
      "loss": 0.5489,
      "step": 40
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00013812154696132598,
      "loss": 0.5144,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00016574585635359117,
      "loss": 0.5082,
      "step": 60
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019337016574585636,
      "loss": 0.4936,
      "step": 70
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00022099447513812155,
      "loss": 0.4818,
      "step": 80
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00024861878453038676,
      "loss": 0.4708,
      "step": 90
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00027624309392265195,
      "loss": 0.4238,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00030386740331491714,
      "loss": 0.4179,
      "step": 110
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00033149171270718233,
      "loss": 0.4103,
      "step": 120
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0003591160220994475,
      "loss": 0.3913,
      "step": 130
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0003867403314917127,
      "loss": 0.3914,
      "step": 140
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004143646408839779,
      "loss": 0.3795,
      "step": 150
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004419889502762431,
      "loss": 0.3642,
      "step": 160
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00046961325966850834,
      "loss": 0.3802,
      "step": 170
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004972375690607735,
      "loss": 0.3794,
      "step": 180
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0005248618784530387,
      "loss": 0.3375,
      "step": 190
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0005524861878453039,
      "loss": 0.3635,
      "step": 200
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000580110497237569,
      "loss": 0.3237,
      "step": 210
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0006077348066298343,
      "loss": 0.326,
      "step": 220
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0006353591160220995,
      "loss": 0.3118,
      "step": 230
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0006629834254143647,
      "loss": 0.3553,
      "step": 240
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0006906077348066299,
      "loss": 0.2944,
      "step": 250
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.000718232044198895,
      "loss": 0.2763,
      "step": 260
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0007458563535911602,
      "loss": 0.3624,
      "step": 270
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0007734806629834254,
      "loss": 0.2961,
      "step": 280
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0008011049723756906,
      "loss": 0.3019,
      "step": 290
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0008287292817679558,
      "loss": 0.3244,
      "step": 300
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0008563535911602211,
      "loss": 0.2954,
      "step": 310
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0008839779005524862,
      "loss": 0.3259,
      "step": 320
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0009116022099447514,
      "loss": 0.2811,
      "step": 330
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0009392265193370167,
      "loss": 0.315,
      "step": 340
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0009668508287292818,
      "loss": 0.2881,
      "step": 350
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.000994475138121547,
      "loss": 0.3164,
      "step": 360
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0009975407316323394,
      "loss": 0.2444,
      "step": 370
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0009944666461727636,
      "loss": 0.2568,
      "step": 380
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000991392560713188,
      "loss": 0.2413,
      "step": 390
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.000988318475253612,
      "loss": 0.3132,
      "step": 400
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0009852443897940364,
      "loss": 0.3086,
      "step": 410
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0009821703043344605,
      "loss": 0.2726,
      "step": 420
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0009790962188748848,
      "loss": 0.2549,
      "step": 430
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.000976022133415309,
      "loss": 0.2724,
      "step": 440
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0009729480479557332,
      "loss": 0.2591,
      "step": 450
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0009698739624961574,
      "loss": 0.2414,
      "step": 460
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0009667998770365817,
      "loss": 0.2387,
      "step": 470
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0009637257915770059,
      "loss": 0.2447,
      "step": 480
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0009606517061174301,
      "loss": 0.2283,
      "step": 490
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0009575776206578543,
      "loss": 0.2752,
      "step": 500
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0009545035351982785,
      "loss": 0.2889,
      "step": 510
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0009514294497387027,
      "loss": 0.2681,
      "step": 520
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.000948355364279127,
      "loss": 0.2498,
      "step": 530
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0009452812788195512,
      "loss": 0.2686,
      "step": 540
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0009422071933599754,
      "loss": 0.2308,
      "step": 550
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0009391331079003996,
      "loss": 0.2405,
      "step": 560
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0009360590224408239,
      "loss": 0.2557,
      "step": 570
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0009329849369812481,
      "loss": 0.4182,
      "step": 580
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0009299108515216723,
      "loss": 0.2541,
      "step": 590
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0009268367660620966,
      "loss": 0.2168,
      "step": 600
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0009237626806025208,
      "loss": 0.2416,
      "step": 610
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.000920688595142945,
      "loss": 0.2436,
      "step": 620
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0009176145096833693,
      "loss": 0.245,
      "step": 630
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0009145404242237935,
      "loss": 0.2471,
      "step": 640
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0009114663387642177,
      "loss": 0.2252,
      "step": 650
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.000908392253304642,
      "loss": 0.2087,
      "step": 660
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0009053181678450662,
      "loss": 0.2288,
      "step": 670
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0009022440823854904,
      "loss": 0.258,
      "step": 680
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0008991699969259147,
      "loss": 0.2467,
      "step": 690
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0008960959114663388,
      "loss": 0.2665,
      "step": 700
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0008930218260067629,
      "loss": 0.2265,
      "step": 710
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0008899477405471871,
      "loss": 0.2841,
      "step": 720
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9227471218737594,
      "eval_loss": 0.16996660828590393,
      "eval_runtime": 475.1508,
      "eval_samples_per_second": 26.507,
      "eval_steps_per_second": 0.829,
      "step": 723
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008868736550876114,
      "loss": 0.2341,
      "step": 730
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0008837995696280356,
      "loss": 0.1977,
      "step": 740
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0008807254841684598,
      "loss": 0.2085,
      "step": 750
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0008776513987088841,
      "loss": 0.2595,
      "step": 760
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0008745773132493083,
      "loss": 0.2888,
      "step": 770
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0008715032277897325,
      "loss": 0.1983,
      "step": 780
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0008684291423301568,
      "loss": 0.2042,
      "step": 790
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.000865355056870581,
      "loss": 0.2073,
      "step": 800
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0008622809714110052,
      "loss": 0.2405,
      "step": 810
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0008592068859514295,
      "loss": 0.1793,
      "step": 820
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0008561328004918537,
      "loss": 0.1951,
      "step": 830
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0008530587150322779,
      "loss": 0.209,
      "step": 840
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0008499846295727022,
      "loss": 0.2235,
      "step": 850
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0008469105441131264,
      "loss": 0.2534,
      "step": 860
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0008438364586535506,
      "loss": 0.2315,
      "step": 870
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0008407623731939749,
      "loss": 0.3287,
      "step": 880
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.000837688287734399,
      "loss": 0.1967,
      "step": 890
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0008346142022748232,
      "loss": 0.1835,
      "step": 900
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0008315401168152474,
      "loss": 0.2297,
      "step": 910
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0008284660313556717,
      "loss": 0.2412,
      "step": 920
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0008253919458960959,
      "loss": 0.2503,
      "step": 930
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0008223178604365201,
      "loss": 0.2083,
      "step": 940
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0008192437749769444,
      "loss": 0.1994,
      "step": 950
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0008161696895173686,
      "loss": 0.1757,
      "step": 960
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0008130956040577928,
      "loss": 0.214,
      "step": 970
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0008100215185982171,
      "loss": 0.2483,
      "step": 980
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0008069474331386413,
      "loss": 0.2136,
      "step": 990
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0008038733476790655,
      "loss": 0.2111,
      "step": 1000
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0008007992622194898,
      "loss": 0.1751,
      "step": 1010
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.000797725176759914,
      "loss": 0.2262,
      "step": 1020
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0007946510913003382,
      "loss": 0.2485,
      "step": 1030
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0007915770058407625,
      "loss": 0.2275,
      "step": 1040
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0007885029203811867,
      "loss": 0.2771,
      "step": 1050
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0007854288349216109,
      "loss": 0.2777,
      "step": 1060
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0007823547494620352,
      "loss": 0.2614,
      "step": 1070
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0007792806640024594,
      "loss": 0.2189,
      "step": 1080
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0007762065785428835,
      "loss": 0.1682,
      "step": 1090
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0007731324930833076,
      "loss": 0.2051,
      "step": 1100
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0007700584076237319,
      "loss": 0.2981,
      "step": 1110
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0007669843221641561,
      "loss": 0.2469,
      "step": 1120
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0007639102367045803,
      "loss": 0.2232,
      "step": 1130
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0007608361512450046,
      "loss": 0.2312,
      "step": 1140
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0007577620657854288,
      "loss": 0.2055,
      "step": 1150
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.000754687980325853,
      "loss": 0.1844,
      "step": 1160
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0007516138948662773,
      "loss": 0.185,
      "step": 1170
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0007485398094067015,
      "loss": 0.2394,
      "step": 1180
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0007454657239471257,
      "loss": 0.2908,
      "step": 1190
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00074239163848755,
      "loss": 0.2708,
      "step": 1200
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0007393175530279742,
      "loss": 0.1798,
      "step": 1210
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0007362434675683984,
      "loss": 0.1992,
      "step": 1220
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0007331693821088227,
      "loss": 0.2794,
      "step": 1230
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0007300952966492469,
      "loss": 0.2765,
      "step": 1240
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0007270212111896711,
      "loss": 0.2008,
      "step": 1250
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0007239471257300953,
      "loss": 0.1814,
      "step": 1260
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0007208730402705196,
      "loss": 0.1971,
      "step": 1270
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0007177989548109437,
      "loss": 0.179,
      "step": 1280
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0007147248693513679,
      "loss": 0.2237,
      "step": 1290
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0007116507838917922,
      "loss": 0.1961,
      "step": 1300
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0007085766984322164,
      "loss": 0.2523,
      "step": 1310
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0007055026129726406,
      "loss": 0.2659,
      "step": 1320
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0007024285275130649,
      "loss": 0.2138,
      "step": 1330
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0006993544420534891,
      "loss": 0.1813,
      "step": 1340
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0006962803565939133,
      "loss": 0.1657,
      "step": 1350
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0006932062711343376,
      "loss": 0.2167,
      "step": 1360
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0006901321856747618,
      "loss": 0.2037,
      "step": 1370
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.000687058100215186,
      "loss": 0.1702,
      "step": 1380
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0006839840147556103,
      "loss": 0.1792,
      "step": 1390
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0006809099292960345,
      "loss": 0.2097,
      "step": 1400
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0006778358438364587,
      "loss": 0.2015,
      "step": 1410
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.000674761758376883,
      "loss": 0.1708,
      "step": 1420
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0006716876729173072,
      "loss": 0.1777,
      "step": 1430
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0006686135874577314,
      "loss": 0.1826,
      "step": 1440
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9147280666931322,
      "eval_loss": 0.21437880396842957,
      "eval_runtime": 474.1279,
      "eval_samples_per_second": 26.565,
      "eval_steps_per_second": 0.831,
      "step": 1446
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006655395019981556,
      "loss": 0.1741,
      "step": 1450
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006624654165385799,
      "loss": 0.1686,
      "step": 1460
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006593913310790041,
      "loss": 0.2035,
      "step": 1470
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006563172456194282,
      "loss": 0.2041,
      "step": 1480
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006532431601598524,
      "loss": 0.179,
      "step": 1490
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006501690747002766,
      "loss": 0.2548,
      "step": 1500
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006470949892407008,
      "loss": 0.1814,
      "step": 1510
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006440209037811251,
      "loss": 0.2271,
      "step": 1520
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006409468183215493,
      "loss": 0.1752,
      "step": 1530
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006378727328619735,
      "loss": 0.1917,
      "step": 1540
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006347986474023978,
      "loss": 0.1763,
      "step": 1550
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.000631724561942822,
      "loss": 0.2002,
      "step": 1560
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006286504764832462,
      "loss": 0.214,
      "step": 1570
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006255763910236704,
      "loss": 0.1708,
      "step": 1580
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006225023055640947,
      "loss": 0.1559,
      "step": 1590
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006194282201045189,
      "loss": 0.1594,
      "step": 1600
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006163541346449431,
      "loss": 0.1601,
      "step": 1610
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006132800491853674,
      "loss": 0.1779,
      "step": 1620
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006102059637257916,
      "loss": 0.1746,
      "step": 1630
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006071318782662158,
      "loss": 0.2524,
      "step": 1640
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006040577928066401,
      "loss": 0.2179,
      "step": 1650
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006009837073470643,
      "loss": 0.1713,
      "step": 1660
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0005979096218874884,
      "loss": 0.1803,
      "step": 1670
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0005948355364279127,
      "loss": 0.189,
      "step": 1680
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0005917614509683369,
      "loss": 0.1712,
      "step": 1690
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0005886873655087611,
      "loss": 0.1769,
      "step": 1700
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0005856132800491854,
      "loss": 0.1795,
      "step": 1710
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0005825391945896096,
      "loss": 0.1535,
      "step": 1720
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0005794651091300338,
      "loss": 0.1562,
      "step": 1730
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0005763910236704581,
      "loss": 0.2128,
      "step": 1740
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0005733169382108823,
      "loss": 0.1707,
      "step": 1750
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0005702428527513065,
      "loss": 0.1792,
      "step": 1760
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0005671687672917308,
      "loss": 0.1504,
      "step": 1770
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.000564094681832155,
      "loss": 0.1598,
      "step": 1780
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0005610205963725792,
      "loss": 0.1737,
      "step": 1790
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0005579465109130034,
      "loss": 0.2276,
      "step": 1800
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0005548724254534277,
      "loss": 0.2368,
      "step": 1810
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0005517983399938519,
      "loss": 0.2196,
      "step": 1820
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0005487242545342761,
      "loss": 0.2351,
      "step": 1830
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0005456501690747004,
      "loss": 0.198,
      "step": 1840
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0005425760836151246,
      "loss": 0.1952,
      "step": 1850
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0005395019981555488,
      "loss": 0.1659,
      "step": 1860
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0005364279126959729,
      "loss": 0.1596,
      "step": 1870
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0005333538272363971,
      "loss": 0.1564,
      "step": 1880
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0005302797417768213,
      "loss": 0.2239,
      "step": 1890
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0005272056563172456,
      "loss": 0.1639,
      "step": 1900
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0005241315708576698,
      "loss": 0.1504,
      "step": 1910
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.000521057485398094,
      "loss": 0.1506,
      "step": 1920
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0005179833999385182,
      "loss": 0.154,
      "step": 1930
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0005149093144789425,
      "loss": 0.2018,
      "step": 1940
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0005118352290193667,
      "loss": 0.1539,
      "step": 1950
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0005087611435597909,
      "loss": 0.1614,
      "step": 1960
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0005056870581002152,
      "loss": 0.1625,
      "step": 1970
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0005026129726406394,
      "loss": 0.1542,
      "step": 1980
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0004995388871810636,
      "loss": 0.1852,
      "step": 1990
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0004964648017214879,
      "loss": 0.183,
      "step": 2000
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0004933907162619121,
      "loss": 0.1372,
      "step": 2010
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0004903166308023363,
      "loss": 0.1798,
      "step": 2020
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.0004872425453427605,
      "loss": 0.1427,
      "step": 2030
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.00048416845988318475,
      "loss": 0.1477,
      "step": 2040
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.000481094374423609,
      "loss": 0.1675,
      "step": 2050
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.0004780202889640332,
      "loss": 0.1769,
      "step": 2060
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.000475253612050415,
      "loss": 0.2386,
      "step": 2070
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0004721795265908392,
      "loss": 0.1842,
      "step": 2080
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.00046910544113126345,
      "loss": 0.1855,
      "step": 2090
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0004660313556716877,
      "loss": 0.1855,
      "step": 2100
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0004629572702121119,
      "loss": 0.1694,
      "step": 2110
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.00045988318475253615,
      "loss": 0.1774,
      "step": 2120
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.0004568090992929603,
      "loss": 0.1704,
      "step": 2130
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.00045373501383338456,
      "loss": 0.1604,
      "step": 2140
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0004506609283738088,
      "loss": 0.1807,
      "step": 2150
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.000447586842914233,
      "loss": 0.2425,
      "step": 2160
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.914410480349345,
      "eval_loss": 0.19619111716747284,
      "eval_runtime": 474.4554,
      "eval_samples_per_second": 26.546,
      "eval_steps_per_second": 0.83,
      "step": 2169
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00044451275745465725,
      "loss": 0.1888,
      "step": 2170
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0004414386719950815,
      "loss": 0.171,
      "step": 2180
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0004383645865355057,
      "loss": 0.1572,
      "step": 2190
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.00043529050107592995,
      "loss": 0.1851,
      "step": 2200
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0004322164156163542,
      "loss": 0.1688,
      "step": 2210
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0004291423301567784,
      "loss": 0.2044,
      "step": 2220
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0004260682446972026,
      "loss": 0.1637,
      "step": 2230
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00042299415923762677,
      "loss": 0.1795,
      "step": 2240
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.000419920073778051,
      "loss": 0.2163,
      "step": 2250
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.00041684598831847523,
      "loss": 0.2959,
      "step": 2260
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.00041377190285889947,
      "loss": 0.1883,
      "step": 2270
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0004106978173993237,
      "loss": 0.1615,
      "step": 2280
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.00040762373193974793,
      "loss": 0.1716,
      "step": 2290
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.00040454964648017216,
      "loss": 0.1548,
      "step": 2300
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0004014755610205964,
      "loss": 0.1689,
      "step": 2310
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0003984014755610206,
      "loss": 0.1307,
      "step": 2320
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0003953273901014448,
      "loss": 0.171,
      "step": 2330
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.00039225330464186904,
      "loss": 0.1799,
      "step": 2340
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00038917921918229327,
      "loss": 0.1637,
      "step": 2350
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0003861051337227175,
      "loss": 0.1673,
      "step": 2360
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.00038303104826314173,
      "loss": 0.1606,
      "step": 2370
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.00037995696280356596,
      "loss": 0.1588,
      "step": 2380
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0003768828773439902,
      "loss": 0.1895,
      "step": 2390
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.00037380879188441443,
      "loss": 0.1782,
      "step": 2400
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.00037073470642483866,
      "loss": 0.1842,
      "step": 2410
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.00036766062096526284,
      "loss": 0.138,
      "step": 2420
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.00036458653550568707,
      "loss": 0.1522,
      "step": 2430
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.00036151245004611125,
      "loss": 0.1433,
      "step": 2440
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0003584383645865355,
      "loss": 0.1681,
      "step": 2450
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0003553642791269597,
      "loss": 0.1797,
      "step": 2460
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.00035229019366738395,
      "loss": 0.1606,
      "step": 2470
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0003492161082078082,
      "loss": 0.1424,
      "step": 2480
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0003461420227482324,
      "loss": 0.1439,
      "step": 2490
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.00034306793728865664,
      "loss": 0.1623,
      "step": 2500
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0003399938518290809,
      "loss": 0.1482,
      "step": 2510
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.00033691976636950505,
      "loss": 0.1742,
      "step": 2520
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0003338456809099293,
      "loss": 0.1851,
      "step": 2530
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0003307715954503535,
      "loss": 0.1319,
      "step": 2540
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.00032769750999077775,
      "loss": 0.1445,
      "step": 2550
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.000324623424531202,
      "loss": 0.1687,
      "step": 2560
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0003215493390716262,
      "loss": 0.1547,
      "step": 2570
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.00031847525361205044,
      "loss": 0.1399,
      "step": 2580
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.0003154011681524747,
      "loss": 0.1409,
      "step": 2590
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0003123270826928989,
      "loss": 0.1422,
      "step": 2600
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.00030925299723332314,
      "loss": 0.1663,
      "step": 2610
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0003061789117737473,
      "loss": 0.1772,
      "step": 2620
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0003031048263141715,
      "loss": 0.1479,
      "step": 2630
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.00030003074085459573,
      "loss": 0.2015,
      "step": 2640
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.00029695665539501996,
      "loss": 0.1665,
      "step": 2650
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0002938825699354442,
      "loss": 0.151,
      "step": 2660
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0002908084844758684,
      "loss": 0.1354,
      "step": 2670
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.00028773439901629266,
      "loss": 0.1601,
      "step": 2680
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0002846603135567169,
      "loss": 0.1495,
      "step": 2690
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0002815862280971411,
      "loss": 0.1288,
      "step": 2700
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.00027851214263756535,
      "loss": 0.1375,
      "step": 2710
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.00027543805717798953,
      "loss": 0.1395,
      "step": 2720
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.00027236397171841376,
      "loss": 0.1605,
      "step": 2730
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.000269289886258838,
      "loss": 0.1627,
      "step": 2740
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0002662158007992622,
      "loss": 0.1511,
      "step": 2750
    },
    {
      "epoch": 3.82,
      "learning_rate": 0.00026314171533968646,
      "loss": 0.153,
      "step": 2760
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.0002600676298801107,
      "loss": 0.1617,
      "step": 2770
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0002569935444205349,
      "loss": 0.1371,
      "step": 2780
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.00025391945896095916,
      "loss": 0.1389,
      "step": 2790
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0002508453735013834,
      "loss": 0.1336,
      "step": 2800
    },
    {
      "epoch": 3.89,
      "learning_rate": 0.00024777128804180757,
      "loss": 0.1533,
      "step": 2810
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.0002446972025822318,
      "loss": 0.1717,
      "step": 2820
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.00024162311712265603,
      "loss": 0.1487,
      "step": 2830
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.00023854903166308023,
      "loss": 0.1646,
      "step": 2840
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.00023547494620350447,
      "loss": 0.1433,
      "step": 2850
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0002324008607439287,
      "loss": 0.17,
      "step": 2860
    },
    {
      "epoch": 3.97,
      "learning_rate": 0.00022932677528435293,
      "loss": 0.1905,
      "step": 2870
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.00022625268982477714,
      "loss": 0.1478,
      "step": 2880
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00022317860436520134,
      "loss": 0.164,
      "step": 2890
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8864628820960698,
      "eval_loss": 0.3099741041660309,
      "eval_runtime": 481.6212,
      "eval_samples_per_second": 26.151,
      "eval_steps_per_second": 0.818,
      "step": 2892
    },
    {
      "epoch": 4.01,
      "learning_rate": 0.00022010451890562557,
      "loss": 0.1351,
      "step": 2900
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.0002170304334460498,
      "loss": 0.1478,
      "step": 2910
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.00021395634798647404,
      "loss": 0.1262,
      "step": 2920
    },
    {
      "epoch": 4.05,
      "learning_rate": 0.00021088226252689824,
      "loss": 0.1532,
      "step": 2930
    },
    {
      "epoch": 4.07,
      "learning_rate": 0.00020780817706732247,
      "loss": 0.1281,
      "step": 2940
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0002047340916077467,
      "loss": 0.1589,
      "step": 2950
    },
    {
      "epoch": 4.09,
      "learning_rate": 0.00020166000614817094,
      "loss": 0.1528,
      "step": 2960
    },
    {
      "epoch": 4.11,
      "learning_rate": 0.00019858592068859514,
      "loss": 0.1503,
      "step": 2970
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.00019551183522901935,
      "loss": 0.1493,
      "step": 2980
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.00019243774976944358,
      "loss": 0.1585,
      "step": 2990
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0001893636643098678,
      "loss": 0.1414,
      "step": 3000
    },
    {
      "epoch": 4.16,
      "learning_rate": 0.00018628957885029204,
      "loss": 0.1394,
      "step": 3010
    },
    {
      "epoch": 4.18,
      "learning_rate": 0.00018321549339071628,
      "loss": 0.1633,
      "step": 3020
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.00018014140793114048,
      "loss": 0.1323,
      "step": 3030
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.00017706732247156471,
      "loss": 0.1531,
      "step": 3040
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.00017399323701198895,
      "loss": 0.1726,
      "step": 3050
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.00017091915155241318,
      "loss": 0.1552,
      "step": 3060
    },
    {
      "epoch": 4.25,
      "learning_rate": 0.00016784506609283738,
      "loss": 0.1761,
      "step": 3070
    },
    {
      "epoch": 4.26,
      "learning_rate": 0.0001647709806332616,
      "loss": 0.1269,
      "step": 3080
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.00016169689517368582,
      "loss": 0.131,
      "step": 3090
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.00015862280971411005,
      "loss": 0.1513,
      "step": 3100
    },
    {
      "epoch": 4.3,
      "learning_rate": 0.00015554872425453428,
      "loss": 0.154,
      "step": 3110
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00015247463879495852,
      "loss": 0.1658,
      "step": 3120
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.00014940055333538272,
      "loss": 0.1544,
      "step": 3130
    },
    {
      "epoch": 4.34,
      "learning_rate": 0.00014632646787580695,
      "loss": 0.1497,
      "step": 3140
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.00014325238241623119,
      "loss": 0.1534,
      "step": 3150
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.00014017829695665542,
      "loss": 0.1518,
      "step": 3160
    },
    {
      "epoch": 4.38,
      "learning_rate": 0.00013710421149707962,
      "loss": 0.1346,
      "step": 3170
    },
    {
      "epoch": 4.4,
      "learning_rate": 0.00013403012603750383,
      "loss": 0.1416,
      "step": 3180
    },
    {
      "epoch": 4.41,
      "learning_rate": 0.00013095604057792806,
      "loss": 0.1483,
      "step": 3190
    },
    {
      "epoch": 4.43,
      "learning_rate": 0.0001278819551183523,
      "loss": 0.1473,
      "step": 3200
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.0001248078696587765,
      "loss": 0.1481,
      "step": 3210
    },
    {
      "epoch": 4.45,
      "learning_rate": 0.00012173378419920074,
      "loss": 0.1452,
      "step": 3220
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.00011865969873962497,
      "loss": 0.1305,
      "step": 3230
    },
    {
      "epoch": 4.48,
      "learning_rate": 0.00011558561328004918,
      "loss": 0.1579,
      "step": 3240
    },
    {
      "epoch": 4.5,
      "learning_rate": 0.00011251152782047341,
      "loss": 0.1436,
      "step": 3250
    },
    {
      "epoch": 4.51,
      "learning_rate": 0.00010943744236089763,
      "loss": 0.1808,
      "step": 3260
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00010636335690132186,
      "loss": 0.1617,
      "step": 3270
    },
    {
      "epoch": 4.54,
      "learning_rate": 0.0001032892714417461,
      "loss": 0.1579,
      "step": 3280
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.0001002151859821703,
      "loss": 0.1535,
      "step": 3290
    },
    {
      "epoch": 4.56,
      "learning_rate": 9.714110052259453e-05,
      "loss": 0.1448,
      "step": 3300
    },
    {
      "epoch": 4.58,
      "learning_rate": 9.406701506301875e-05,
      "loss": 0.1381,
      "step": 3310
    },
    {
      "epoch": 4.59,
      "learning_rate": 9.099292960344298e-05,
      "loss": 0.1419,
      "step": 3320
    },
    {
      "epoch": 4.61,
      "learning_rate": 8.791884414386721e-05,
      "loss": 0.1439,
      "step": 3330
    },
    {
      "epoch": 4.62,
      "learning_rate": 8.484475868429142e-05,
      "loss": 0.1354,
      "step": 3340
    },
    {
      "epoch": 4.63,
      "learning_rate": 8.177067322471565e-05,
      "loss": 0.1363,
      "step": 3350
    },
    {
      "epoch": 4.65,
      "learning_rate": 7.869658776513987e-05,
      "loss": 0.1395,
      "step": 3360
    },
    {
      "epoch": 4.66,
      "learning_rate": 7.56225023055641e-05,
      "loss": 0.146,
      "step": 3370
    },
    {
      "epoch": 4.67,
      "learning_rate": 7.254841684598831e-05,
      "loss": 0.1472,
      "step": 3380
    },
    {
      "epoch": 4.69,
      "learning_rate": 6.947433138641254e-05,
      "loss": 0.1328,
      "step": 3390
    },
    {
      "epoch": 4.7,
      "learning_rate": 6.640024592683677e-05,
      "loss": 0.143,
      "step": 3400
    },
    {
      "epoch": 4.72,
      "learning_rate": 6.332616046726099e-05,
      "loss": 0.1339,
      "step": 3410
    },
    {
      "epoch": 4.73,
      "learning_rate": 6.0252075007685215e-05,
      "loss": 0.1298,
      "step": 3420
    },
    {
      "epoch": 4.74,
      "learning_rate": 5.717798954810944e-05,
      "loss": 0.121,
      "step": 3430
    },
    {
      "epoch": 4.76,
      "learning_rate": 5.410390408853366e-05,
      "loss": 0.1523,
      "step": 3440
    },
    {
      "epoch": 4.77,
      "learning_rate": 5.1029818628957884e-05,
      "loss": 0.147,
      "step": 3450
    },
    {
      "epoch": 4.79,
      "learning_rate": 4.795573316938211e-05,
      "loss": 0.1253,
      "step": 3460
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.4881647709806335e-05,
      "loss": 0.1386,
      "step": 3470
    },
    {
      "epoch": 4.81,
      "learning_rate": 4.180756225023056e-05,
      "loss": 0.1316,
      "step": 3480
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.873347679065478e-05,
      "loss": 0.1338,
      "step": 3490
    },
    {
      "epoch": 4.84,
      "learning_rate": 3.5659391331079004e-05,
      "loss": 0.1354,
      "step": 3500
    },
    {
      "epoch": 4.85,
      "learning_rate": 3.258530587150323e-05,
      "loss": 0.1512,
      "step": 3510
    },
    {
      "epoch": 4.87,
      "learning_rate": 2.951122041192745e-05,
      "loss": 0.1288,
      "step": 3520
    },
    {
      "epoch": 4.88,
      "learning_rate": 2.6437134952351673e-05,
      "loss": 0.1418,
      "step": 3530
    },
    {
      "epoch": 4.9,
      "learning_rate": 2.3363049492775902e-05,
      "loss": 0.122,
      "step": 3540
    },
    {
      "epoch": 4.91,
      "learning_rate": 2.0288964033200124e-05,
      "loss": 0.1276,
      "step": 3550
    },
    {
      "epoch": 4.92,
      "learning_rate": 1.7214878573624346e-05,
      "loss": 0.1306,
      "step": 3560
    },
    {
      "epoch": 4.94,
      "learning_rate": 1.4140793114048571e-05,
      "loss": 0.1367,
      "step": 3570
    },
    {
      "epoch": 4.95,
      "learning_rate": 1.1066707654472795e-05,
      "loss": 0.1482,
      "step": 3580
    },
    {
      "epoch": 4.97,
      "learning_rate": 7.992622194897018e-06,
      "loss": 0.1459,
      "step": 3590
    },
    {
      "epoch": 4.98,
      "learning_rate": 4.918536735321242e-06,
      "loss": 0.1087,
      "step": 3600
    },
    {
      "epoch": 4.99,
      "learning_rate": 1.8444512757454657e-06,
      "loss": 0.1525,
      "step": 3610
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8999603017070266,
      "eval_loss": 0.2823958098888397,
      "eval_runtime": 482.9204,
      "eval_samples_per_second": 26.081,
      "eval_steps_per_second": 0.816,
      "step": 3615
    },
    {
      "epoch": 5.0,
      "step": 3615,
      "total_flos": 5.8101857302104e+19,
      "train_loss": 0.20637444648505246,
      "train_runtime": 19742.095,
      "train_samples_per_second": 23.438,
      "train_steps_per_second": 0.183
    }
  ],
  "logging_steps": 10,
  "max_steps": 3615,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5.8101857302104e+19,
  "trial_name": null,
  "trial_params": null
}