{
  "best_metric": 0.2977409064769745,
  "best_model_checkpoint": "finetuned-ConvNext-Indian-food/checkpoint-1300",
  "epoch": 10.0,
  "global_step": 3330,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 0.0001993993993993994,
      "loss": 2.9337,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001987987987987988,
      "loss": 2.7589,
      "step": 20
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001981981981981982,
      "loss": 2.4837,
      "step": 30
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001975975975975976,
      "loss": 2.2961,
      "step": 40
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019699699699699701,
      "loss": 2.1006,
      "step": 50
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001963963963963964,
      "loss": 1.8491,
      "step": 60
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019579579579579582,
      "loss": 1.637,
      "step": 70
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001951951951951952,
      "loss": 1.4124,
      "step": 80
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019459459459459462,
      "loss": 1.3069,
      "step": 90
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019399399399399402,
      "loss": 1.3145,
      "step": 100
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.8150903294367694,
      "eval_loss": 1.0460201501846313,
      "eval_runtime": 32.8174,
      "eval_samples_per_second": 28.674,
      "eval_steps_per_second": 3.596,
      "step": 100
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001933933933933934,
      "loss": 1.2783,
      "step": 110
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019279279279279282,
      "loss": 1.0547,
      "step": 120
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019225225225225225,
      "loss": 1.0143,
      "step": 130
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019165165165165165,
      "loss": 0.9249,
      "step": 140
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019105105105105105,
      "loss": 0.8564,
      "step": 150
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019045045045045048,
      "loss": 0.8674,
      "step": 160
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00018984984984984985,
      "loss": 0.9013,
      "step": 170
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00018924924924924925,
      "loss": 0.7794,
      "step": 180
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00018864864864864866,
      "loss": 0.7487,
      "step": 190
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00018804804804804806,
      "loss": 0.6694,
      "step": 200
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.8756641870350691,
      "eval_loss": 0.5439420938491821,
      "eval_runtime": 32.7659,
      "eval_samples_per_second": 28.719,
      "eval_steps_per_second": 3.601,
      "step": 200
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00018744744744744746,
      "loss": 0.7296,
      "step": 210
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018684684684684686,
      "loss": 0.6253,
      "step": 220
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018624624624624626,
      "loss": 0.5394,
      "step": 230
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00018564564564564566,
      "loss": 0.6077,
      "step": 240
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00018504504504504506,
      "loss": 0.6871,
      "step": 250
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018444444444444446,
      "loss": 0.6409,
      "step": 260
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00018384384384384386,
      "loss": 0.6477,
      "step": 270
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00018324324324324324,
      "loss": 0.7863,
      "step": 280
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018264264264264266,
      "loss": 0.563,
      "step": 290
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00018204204204204204,
      "loss": 0.5057,
      "step": 300
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.8831030818278427,
      "eval_loss": 0.439805269241333,
      "eval_runtime": 33.0097,
      "eval_samples_per_second": 28.507,
      "eval_steps_per_second": 3.575,
      "step": 300
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00018144144144144144,
      "loss": 0.4947,
      "step": 310
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00018084084084084084,
      "loss": 0.5115,
      "step": 320
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00018024024024024024,
      "loss": 0.4236,
      "step": 330
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00017963963963963967,
      "loss": 0.4525,
      "step": 340
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00017903903903903904,
      "loss": 0.5395,
      "step": 350
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00017843843843843844,
      "loss": 0.3434,
      "step": 360
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00017783783783783784,
      "loss": 0.2957,
      "step": 370
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00017723723723723724,
      "loss": 0.3677,
      "step": 380
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00017663663663663665,
      "loss": 0.4456,
      "step": 390
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001760960960960961,
      "loss": 0.4381,
      "step": 400
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.8820403825717322,
      "eval_loss": 0.42858749628067017,
      "eval_runtime": 32.2263,
      "eval_samples_per_second": 29.2,
      "eval_steps_per_second": 3.662,
      "step": 400
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001754954954954955,
      "loss": 0.5108,
      "step": 410
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001748948948948949,
      "loss": 0.501,
      "step": 420
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001742942942942943,
      "loss": 0.319,
      "step": 430
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001736936936936937,
      "loss": 0.3744,
      "step": 440
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001730930930930931,
      "loss": 0.4664,
      "step": 450
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001724924924924925,
      "loss": 0.5735,
      "step": 460
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001718918918918919,
      "loss": 0.2927,
      "step": 470
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00017129129129129128,
      "loss": 0.5256,
      "step": 480
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001706906906906907,
      "loss": 0.4009,
      "step": 490
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00017009009009009008,
      "loss": 0.4376,
      "step": 500
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.9043570669500531,
      "eval_loss": 0.34000876545906067,
      "eval_runtime": 32.3032,
      "eval_samples_per_second": 29.13,
      "eval_steps_per_second": 3.653,
      "step": 500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001694894894894895,
      "loss": 0.4056,
      "step": 510
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00016888888888888889,
      "loss": 0.4832,
      "step": 520
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0001682882882882883,
      "loss": 0.488,
      "step": 530
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001676876876876877,
      "loss": 0.3927,
      "step": 540
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001670870870870871,
      "loss": 0.3167,
      "step": 550
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00016648648648648652,
      "loss": 0.3411,
      "step": 560
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0001658858858858859,
      "loss": 0.3771,
      "step": 570
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001652852852852853,
      "loss": 0.3965,
      "step": 580
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001646846846846847,
      "loss": 0.5483,
      "step": 590
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001640840840840841,
      "loss": 0.2499,
      "step": 600
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.9064824654622742,
      "eval_loss": 0.3312312662601471,
      "eval_runtime": 32.7174,
      "eval_samples_per_second": 28.761,
      "eval_steps_per_second": 3.607,
      "step": 600
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001634834834834835,
      "loss": 0.3226,
      "step": 610
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001628828828828829,
      "loss": 0.2392,
      "step": 620
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001622822822822823,
      "loss": 0.5277,
      "step": 630
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001616816816816817,
      "loss": 0.3779,
      "step": 640
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001610810810810811,
      "loss": 0.362,
      "step": 650
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00016048048048048047,
      "loss": 0.3427,
      "step": 660
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001598798798798799,
      "loss": 0.437,
      "step": 670
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.00015927927927927927,
      "loss": 0.4169,
      "step": 680
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0001586786786786787,
      "loss": 0.311,
      "step": 690
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.00015807807807807807,
      "loss": 0.2802,
      "step": 700
    },
    {
      "epoch": 2.1,
      "eval_accuracy": 0.9032943676939427,
      "eval_loss": 0.33384454250335693,
      "eval_runtime": 33.4958,
      "eval_samples_per_second": 28.093,
      "eval_steps_per_second": 3.523,
      "step": 700
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.00015747747747747747,
      "loss": 0.1493,
      "step": 710
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00015687687687687688,
      "loss": 0.272,
      "step": 720
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00015627627627627628,
      "loss": 0.2145,
      "step": 730
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0001556756756756757,
      "loss": 0.1844,
      "step": 740
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.00015507507507507508,
      "loss": 0.2618,
      "step": 750
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.00015447447447447448,
      "loss": 0.3342,
      "step": 760
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.00015387387387387388,
      "loss": 0.3784,
      "step": 770
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00015327327327327328,
      "loss": 0.306,
      "step": 780
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.00015267267267267268,
      "loss": 0.3756,
      "step": 790
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.00015207207207207208,
      "loss": 0.3014,
      "step": 800
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 0.8947927736450585,
      "eval_loss": 0.35719022154808044,
      "eval_runtime": 32.3776,
      "eval_samples_per_second": 29.063,
      "eval_steps_per_second": 3.644,
      "step": 800
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.00015147147147147148,
      "loss": 0.2036,
      "step": 810
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.00015087087087087088,
      "loss": 0.239,
      "step": 820
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.00015027027027027028,
      "loss": 0.2958,
      "step": 830
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.00014966966966966969,
      "loss": 0.3529,
      "step": 840
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0001490690690690691,
      "loss": 0.2531,
      "step": 850
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.00014846846846846846,
      "loss": 0.2603,
      "step": 860
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0001478678678678679,
      "loss": 0.2643,
      "step": 870
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.00014726726726726726,
      "loss": 0.3614,
      "step": 880
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.00014666666666666666,
      "loss": 0.1766,
      "step": 890
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.00014606606606606606,
      "loss": 0.2508,
      "step": 900
    },
    {
      "epoch": 2.7,
      "eval_accuracy": 0.9022316684378321,
      "eval_loss": 0.34318017959594727,
      "eval_runtime": 32.6394,
      "eval_samples_per_second": 28.83,
      "eval_steps_per_second": 3.615,
      "step": 900
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.00014546546546546546,
      "loss": 0.3306,
      "step": 910
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0001448648648648649,
      "loss": 0.2941,
      "step": 920
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.00014426426426426427,
      "loss": 0.4128,
      "step": 930
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.00014366366366366367,
      "loss": 0.3812,
      "step": 940
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.00014306306306306307,
      "loss": 0.2912,
      "step": 950
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.00014246246246246247,
      "loss": 0.3563,
      "step": 960
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.00014186186186186187,
      "loss": 0.3287,
      "step": 970
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.00014126126126126127,
      "loss": 0.4866,
      "step": 980
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.00014066066066066067,
      "loss": 0.2232,
      "step": 990
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00014006006006006007,
      "loss": 0.2012,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9086078639744952,
      "eval_loss": 0.30601173639297485,
      "eval_runtime": 33.0716,
      "eval_samples_per_second": 28.453,
      "eval_steps_per_second": 3.568,
      "step": 1000
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00013945945945945947,
      "loss": 0.1947,
      "step": 1010
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.00013885885885885887,
      "loss": 0.3934,
      "step": 1020
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.00013825825825825827,
      "loss": 0.1187,
      "step": 1030
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.00013765765765765765,
      "loss": 0.236,
      "step": 1040
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.00013705705705705708,
      "loss": 0.2484,
      "step": 1050
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.00013645645645645645,
      "loss": 0.1168,
      "step": 1060
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.00013585585585585585,
      "loss": 0.1394,
      "step": 1070
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.00013525525525525525,
      "loss": 0.246,
      "step": 1080
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.00013465465465465465,
      "loss": 0.3722,
      "step": 1090
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.00013405405405405408,
      "loss": 0.2634,
      "step": 1100
    },
    {
      "epoch": 3.3,
      "eval_accuracy": 0.9086078639744952,
      "eval_loss": 0.34511974453926086,
      "eval_runtime": 32.7524,
      "eval_samples_per_second": 28.731,
      "eval_steps_per_second": 3.603,
      "step": 1100
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.00013345345345345345,
      "loss": 0.2849,
      "step": 1110
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.00013285285285285285,
      "loss": 0.165,
      "step": 1120
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.00013225225225225226,
      "loss": 0.2819,
      "step": 1130
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.00013165165165165166,
      "loss": 0.2486,
      "step": 1140
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.00013105105105105106,
      "loss": 0.1947,
      "step": 1150
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.00013045045045045046,
      "loss": 0.1998,
      "step": 1160
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.00012984984984984986,
      "loss": 0.1896,
      "step": 1170
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.00012924924924924926,
      "loss": 0.1579,
      "step": 1180
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.00012864864864864866,
      "loss": 0.1477,
      "step": 1190
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.00012804804804804806,
      "loss": 0.2483,
      "step": 1200
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.9043570669500531,
      "eval_loss": 0.35497206449508667,
      "eval_runtime": 32.9774,
      "eval_samples_per_second": 28.535,
      "eval_steps_per_second": 3.578,
      "step": 1200
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.00012744744744744746,
      "loss": 0.2252,
      "step": 1210
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.00012684684684684684,
      "loss": 0.1764,
      "step": 1220
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.00012624624624624626,
      "loss": 0.2994,
      "step": 1230
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.00012564564564564564,
      "loss": 0.2404,
      "step": 1240
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.00012504504504504507,
      "loss": 0.1576,
      "step": 1250
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.00012444444444444444,
      "loss": 0.1595,
      "step": 1260
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.00012384384384384384,
      "loss": 0.2793,
      "step": 1270
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.00012324324324324327,
      "loss": 0.2766,
      "step": 1280
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.00012264264264264264,
      "loss": 0.1626,
      "step": 1290
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.00012204204204204204,
      "loss": 0.2273,
      "step": 1300
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.9107332624867163,
      "eval_loss": 0.2977409064769745,
      "eval_runtime": 33.1212,
      "eval_samples_per_second": 28.411,
      "eval_steps_per_second": 3.563,
      "step": 1300
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.00012144144144144144,
      "loss": 0.3854,
      "step": 1310
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.00012084084084084084,
      "loss": 0.326,
      "step": 1320
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.00012024024024024026,
      "loss": 0.1902,
      "step": 1330
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.00011963963963963965,
      "loss": 0.2819,
      "step": 1340
    },
    {
      "epoch": 4.05,
      "learning_rate": 0.00011903903903903903,
      "loss": 0.0952,
      "step": 1350
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.00011843843843843845,
      "loss": 0.2189,
      "step": 1360
    },
    {
      "epoch": 4.11,
      "learning_rate": 0.00011783783783783784,
      "loss": 0.1538,
      "step": 1370
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.00011723723723723725,
      "loss": 0.292,
      "step": 1380
    },
    {
      "epoch": 4.17,
      "learning_rate": 0.00011663663663663664,
      "loss": 0.1944,
      "step": 1390
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.00011603603603603604,
      "loss": 0.1214,
      "step": 1400
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 0.9160467587672688,
      "eval_loss": 0.3265420198440552,
      "eval_runtime": 32.8927,
      "eval_samples_per_second": 28.608,
      "eval_steps_per_second": 3.587,
      "step": 1400
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.00011543543543543545,
      "loss": 0.1619,
      "step": 1410
    },
    {
      "epoch": 4.26,
      "learning_rate": 0.00011483483483483484,
      "loss": 0.1994,
      "step": 1420
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.00011423423423423425,
      "loss": 0.1492,
      "step": 1430
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00011363363363363364,
      "loss": 0.185,
      "step": 1440
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.00011303303303303303,
      "loss": 0.096,
      "step": 1450
    },
    {
      "epoch": 4.38,
      "learning_rate": 0.00011243243243243244,
      "loss": 0.2023,
      "step": 1460
    },
    {
      "epoch": 4.41,
      "learning_rate": 0.00011183183183183183,
      "loss": 0.172,
      "step": 1470
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.00011123123123123123,
      "loss": 0.2173,
      "step": 1480
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.00011063063063063063,
      "loss": 0.1906,
      "step": 1490
    },
    {
      "epoch": 4.5,
      "learning_rate": 0.00011003003003003003,
      "loss": 0.2048,
      "step": 1500
    },
    {
      "epoch": 4.5,
      "eval_accuracy": 0.9213602550478215,
      "eval_loss": 0.31257954239845276,
      "eval_runtime": 33.0709,
      "eval_samples_per_second": 28.454,
      "eval_steps_per_second": 3.568,
      "step": 1500
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.00010942942942942945,
      "loss": 0.1993,
      "step": 1510
    },
    {
      "epoch": 4.56,
      "learning_rate": 0.00010882882882882883,
      "loss": 0.2563,
      "step": 1520
    },
    {
      "epoch": 4.59,
      "learning_rate": 0.00010822822822822822,
      "loss": 0.1404,
      "step": 1530
    },
    {
      "epoch": 4.62,
      "learning_rate": 0.00010762762762762764,
      "loss": 0.3455,
      "step": 1540
    },
    {
      "epoch": 4.65,
      "learning_rate": 0.00010702702702702702,
      "loss": 0.1722,
      "step": 1550
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.00010642642642642644,
      "loss": 0.2154,
      "step": 1560
    },
    {
      "epoch": 4.71,
      "learning_rate": 0.00010582582582582582,
      "loss": 0.1499,
      "step": 1570
    },
    {
      "epoch": 4.74,
      "learning_rate": 0.00010522522522522523,
      "loss": 0.106,
      "step": 1580
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.00010462462462462464,
      "loss": 0.0827,
      "step": 1590
    },
    {
      "epoch": 4.8,
      "learning_rate": 0.00010402402402402403,
      "loss": 0.0997,
      "step": 1600
    },
    {
      "epoch": 4.8,
      "eval_accuracy": 0.9160467587672688,
      "eval_loss": 0.3163708448410034,
      "eval_runtime": 32.9295,
      "eval_samples_per_second": 28.576,
      "eval_steps_per_second": 3.583,
      "step": 1600
    },
    {
      "epoch": 4.83,
      "learning_rate": 0.00010342342342342344,
      "loss": 0.2504,
      "step": 1610
    },
    {
      "epoch": 4.86,
      "learning_rate": 0.00010282282282282283,
      "loss": 0.156,
      "step": 1620
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.00010222222222222222,
      "loss": 0.1835,
      "step": 1630
    },
    {
      "epoch": 4.92,
      "learning_rate": 0.00010162162162162163,
      "loss": 0.2362,
      "step": 1640
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.00010102102102102102,
      "loss": 0.1598,
      "step": 1650
    },
    {
      "epoch": 4.98,
      "learning_rate": 0.00010042042042042043,
      "loss": 0.1116,
      "step": 1660
    },
    {
      "epoch": 5.02,
      "learning_rate": 9.981981981981983e-05,
      "loss": 0.0431,
      "step": 1670
    },
    {
      "epoch": 5.05,
      "learning_rate": 9.921921921921923e-05,
      "loss": 0.2061,
      "step": 1680
    },
    {
      "epoch": 5.08,
      "learning_rate": 9.861861861861862e-05,
      "loss": 0.117,
      "step": 1690
    },
    {
      "epoch": 5.11,
      "learning_rate": 9.801801801801802e-05,
      "loss": 0.1145,
      "step": 1700
    },
    {
      "epoch": 5.11,
      "eval_accuracy": 0.9139213602550478,
      "eval_loss": 0.30551373958587646,
      "eval_runtime": 32.9051,
      "eval_samples_per_second": 28.597,
      "eval_steps_per_second": 3.586,
      "step": 1700
    },
    {
      "epoch": 5.14,
      "learning_rate": 9.741741741741742e-05,
      "loss": 0.1235,
      "step": 1710
    },
    {
      "epoch": 5.17,
      "learning_rate": 9.681681681681682e-05,
      "loss": 0.0922,
      "step": 1720
    },
    {
      "epoch": 5.2,
      "learning_rate": 9.621621621621622e-05,
      "loss": 0.2855,
      "step": 1730
    },
    {
      "epoch": 5.23,
      "learning_rate": 9.561561561561561e-05,
      "loss": 0.1353,
      "step": 1740
    },
    {
      "epoch": 5.26,
      "learning_rate": 9.501501501501501e-05,
      "loss": 0.098,
      "step": 1750
    },
    {
      "epoch": 5.29,
      "learning_rate": 9.441441441441443e-05,
      "loss": 0.1862,
      "step": 1760
    },
    {
      "epoch": 5.32,
      "learning_rate": 9.381381381381383e-05,
      "loss": 0.2086,
      "step": 1770
    },
    {
      "epoch": 5.35,
      "learning_rate": 9.321321321321322e-05,
      "loss": 0.1683,
      "step": 1780
    },
    {
      "epoch": 5.38,
      "learning_rate": 9.261261261261262e-05,
      "loss": 0.1101,
      "step": 1790
    },
    {
      "epoch": 5.41,
      "learning_rate": 9.201201201201202e-05,
      "loss": 0.1578,
      "step": 1800
    },
    {
      "epoch": 5.41,
      "eval_accuracy": 0.9171094580233794,
      "eval_loss": 0.3195084035396576,
      "eval_runtime": 33.6527,
      "eval_samples_per_second": 27.962,
      "eval_steps_per_second": 3.506,
      "step": 1800
    },
    {
      "epoch": 5.44,
      "learning_rate": 9.141141141141142e-05,
      "loss": 0.2522,
      "step": 1810
    },
    {
      "epoch": 5.47,
      "learning_rate": 9.081081081081082e-05,
      "loss": 0.0995,
      "step": 1820
    },
    {
      "epoch": 5.5,
      "learning_rate": 9.02102102102102e-05,
      "loss": 0.1168,
      "step": 1830
    },
    {
      "epoch": 5.53,
      "learning_rate": 8.96096096096096e-05,
      "loss": 0.17,
      "step": 1840
    },
    {
      "epoch": 5.56,
      "learning_rate": 8.900900900900902e-05,
      "loss": 0.0545,
      "step": 1850
    },
    {
      "epoch": 5.59,
      "learning_rate": 8.840840840840842e-05,
      "loss": 0.1607,
      "step": 1860
    },
    {
      "epoch": 5.62,
      "learning_rate": 8.780780780780781e-05,
      "loss": 0.0549,
      "step": 1870
    },
    {
      "epoch": 5.65,
      "learning_rate": 8.720720720720721e-05,
      "loss": 0.1503,
      "step": 1880
    },
    {
      "epoch": 5.68,
      "learning_rate": 8.660660660660661e-05,
      "loss": 0.2044,
      "step": 1890
    },
    {
      "epoch": 5.71,
      "learning_rate": 8.600600600600601e-05,
      "loss": 0.0615,
      "step": 1900
    },
    {
      "epoch": 5.71,
      "eval_accuracy": 0.9107332624867163,
      "eval_loss": 0.3401225805282593,
      "eval_runtime": 33.1832,
      "eval_samples_per_second": 28.358,
      "eval_steps_per_second": 3.556,
      "step": 1900
    },
    {
      "epoch": 5.74,
      "learning_rate": 8.540540540540541e-05,
      "loss": 0.1611,
      "step": 1910
    },
    {
      "epoch": 5.77,
      "learning_rate": 8.48048048048048e-05,
      "loss": 0.1693,
      "step": 1920
    },
    {
      "epoch": 5.8,
      "learning_rate": 8.42042042042042e-05,
      "loss": 0.079,
      "step": 1930
    },
    {
      "epoch": 5.83,
      "learning_rate": 8.360360360360362e-05,
      "loss": 0.2182,
      "step": 1940
    },
    {
      "epoch": 5.86,
      "learning_rate": 8.300300300300302e-05,
      "loss": 0.2156,
      "step": 1950
    },
    {
      "epoch": 5.89,
      "learning_rate": 8.24024024024024e-05,
      "loss": 0.1422,
      "step": 1960
    },
    {
      "epoch": 5.92,
      "learning_rate": 8.18018018018018e-05,
      "loss": 0.1465,
      "step": 1970
    },
    {
      "epoch": 5.95,
      "learning_rate": 8.12012012012012e-05,
      "loss": 0.2347,
      "step": 1980
    },
    {
      "epoch": 5.98,
      "learning_rate": 8.06006006006006e-05,
      "loss": 0.1994,
      "step": 1990
    },
    {
      "epoch": 6.01,
      "learning_rate": 8e-05,
      "loss": 0.1537,
      "step": 2000
    },
    {
      "epoch": 6.01,
      "eval_accuracy": 0.9096705632306057,
      "eval_loss": 0.3427741527557373,
      "eval_runtime": 33.1728,
      "eval_samples_per_second": 28.367,
      "eval_steps_per_second": 3.557,
      "step": 2000
    },
    {
      "epoch": 6.04,
      "learning_rate": 7.93993993993994e-05,
      "loss": 0.1196,
      "step": 2010
    },
    {
      "epoch": 6.07,
      "learning_rate": 7.87987987987988e-05,
      "loss": 0.1979,
      "step": 2020
    },
    {
      "epoch": 6.1,
      "learning_rate": 7.819819819819821e-05,
      "loss": 0.123,
      "step": 2030
    },
    {
      "epoch": 6.13,
      "learning_rate": 7.759759759759761e-05,
      "loss": 0.1802,
      "step": 2040
    },
    {
      "epoch": 6.16,
      "learning_rate": 7.699699699699701e-05,
      "loss": 0.1126,
      "step": 2050
    },
    {
      "epoch": 6.19,
      "learning_rate": 7.63963963963964e-05,
      "loss": 0.1223,
      "step": 2060
    },
    {
      "epoch": 6.22,
      "learning_rate": 7.57957957957958e-05,
      "loss": 0.1536,
      "step": 2070
    },
    {
      "epoch": 6.25,
      "learning_rate": 7.51951951951952e-05,
      "loss": 0.0856,
      "step": 2080
    },
    {
      "epoch": 6.28,
      "learning_rate": 7.45945945945946e-05,
      "loss": 0.0928,
      "step": 2090
    },
    {
      "epoch": 6.31,
      "learning_rate": 7.399399399399399e-05,
      "loss": 0.1278,
      "step": 2100
    },
    {
      "epoch": 6.31,
      "eval_accuracy": 0.9192348565356004,
      "eval_loss": 0.3057916462421417,
      "eval_runtime": 32.9085,
      "eval_samples_per_second": 28.594,
      "eval_steps_per_second": 3.586,
      "step": 2100
    },
    {
      "epoch": 6.34,
      "learning_rate": 7.339339339339339e-05,
      "loss": 0.1626,
      "step": 2110
    },
    {
      "epoch": 6.37,
      "learning_rate": 7.27927927927928e-05,
      "loss": 0.1518,
      "step": 2120
    },
    {
      "epoch": 6.4,
      "learning_rate": 7.21921921921922e-05,
      "loss": 0.1275,
      "step": 2130
    },
    {
      "epoch": 6.43,
      "learning_rate": 7.15915915915916e-05,
      "loss": 0.1691,
      "step": 2140
    },
    {
      "epoch": 6.46,
      "learning_rate": 7.099099099099099e-05,
      "loss": 0.0378,
      "step": 2150
    },
    {
      "epoch": 6.49,
      "learning_rate": 7.039039039039039e-05,
      "loss": 0.0721,
      "step": 2160
    },
    {
      "epoch": 6.52,
      "learning_rate": 6.97897897897898e-05,
      "loss": 0.116,
      "step": 2170
    },
    {
      "epoch": 6.55,
      "learning_rate": 6.91891891891892e-05,
      "loss": 0.0625,
      "step": 2180
    },
    {
      "epoch": 6.58,
      "learning_rate": 6.858858858858858e-05,
      "loss": 0.0699,
      "step": 2190
    },
    {
      "epoch": 6.61,
      "learning_rate": 6.798798798798798e-05,
      "loss": 0.1274,
      "step": 2200
    },
    {
      "epoch": 6.61,
      "eval_accuracy": 0.9192348565356004,
      "eval_loss": 0.3189365565776825,
      "eval_runtime": 32.9229,
      "eval_samples_per_second": 28.582,
      "eval_steps_per_second": 3.584,
      "step": 2200
    },
    {
      "epoch": 6.64,
      "learning_rate": 6.73873873873874e-05,
      "loss": 0.0926,
      "step": 2210
    },
    {
      "epoch": 6.67,
      "learning_rate": 6.67867867867868e-05,
      "loss": 0.0402,
      "step": 2220
    },
    {
      "epoch": 6.7,
      "learning_rate": 6.61861861861862e-05,
      "loss": 0.2149,
      "step": 2230
    },
    {
      "epoch": 6.73,
      "learning_rate": 6.558558558558559e-05,
      "loss": 0.2664,
      "step": 2240
    },
    {
      "epoch": 6.76,
      "learning_rate": 6.498498498498499e-05,
      "loss": 0.1215,
      "step": 2250
    },
    {
      "epoch": 6.79,
      "learning_rate": 6.438438438438439e-05,
      "loss": 0.0846,
      "step": 2260
    },
    {
      "epoch": 6.82,
      "learning_rate": 6.378378378378379e-05,
      "loss": 0.0453,
      "step": 2270
    },
    {
      "epoch": 6.85,
      "learning_rate": 6.318318318318318e-05,
      "loss": 0.122,
      "step": 2280
    },
    {
      "epoch": 6.88,
      "learning_rate": 6.258258258258258e-05,
      "loss": 0.2334,
      "step": 2290
    },
    {
      "epoch": 6.91,
      "learning_rate": 6.198198198198199e-05,
      "loss": 0.0877,
      "step": 2300
    },
    {
      "epoch": 6.91,
      "eval_accuracy": 0.9181721572794899,
      "eval_loss": 0.33698955178260803,
      "eval_runtime": 33.8365,
      "eval_samples_per_second": 27.81,
      "eval_steps_per_second": 3.487,
      "step": 2300
    },
    {
      "epoch": 6.94,
      "learning_rate": 6.138138138138139e-05,
      "loss": 0.1992,
      "step": 2310
    },
    {
      "epoch": 6.97,
      "learning_rate": 6.0780780780780786e-05,
      "loss": 0.1207,
      "step": 2320
    },
    {
      "epoch": 7.0,
      "learning_rate": 6.018018018018018e-05,
      "loss": 0.231,
      "step": 2330
    },
    {
      "epoch": 7.03,
      "learning_rate": 5.957957957957958e-05,
      "loss": 0.0674,
      "step": 2340
    },
    {
      "epoch": 7.06,
      "learning_rate": 5.897897897897898e-05,
      "loss": 0.0818,
      "step": 2350
    },
    {
      "epoch": 7.09,
      "learning_rate": 5.837837837837838e-05,
      "loss": 0.2289,
      "step": 2360
    },
    {
      "epoch": 7.12,
      "learning_rate": 5.7777777777777776e-05,
      "loss": 0.0863,
      "step": 2370
    },
    {
      "epoch": 7.15,
      "learning_rate": 5.717717717717718e-05,
      "loss": 0.1641,
      "step": 2380
    },
    {
      "epoch": 7.18,
      "learning_rate": 5.657657657657658e-05,
      "loss": 0.0895,
      "step": 2390
    },
    {
      "epoch": 7.21,
      "learning_rate": 5.597597597597598e-05,
      "loss": 0.1058,
      "step": 2400
    },
    {
      "epoch": 7.21,
      "eval_accuracy": 0.9192348565356004,
      "eval_loss": 0.3225291967391968,
      "eval_runtime": 33.1868,
      "eval_samples_per_second": 28.355,
      "eval_steps_per_second": 3.556,
      "step": 2400
    },
    {
      "epoch": 7.24,
      "learning_rate": 5.537537537537538e-05,
      "loss": 0.0884,
      "step": 2410
    },
    {
      "epoch": 7.27,
      "learning_rate": 5.4774774774774774e-05,
      "loss": 0.1579,
      "step": 2420
    },
    {
      "epoch": 7.3,
      "learning_rate": 5.4174174174174175e-05,
      "loss": 0.071,
      "step": 2430
    },
    {
      "epoch": 7.33,
      "learning_rate": 5.3573573573573576e-05,
      "loss": 0.0217,
      "step": 2440
    },
    {
      "epoch": 7.36,
      "learning_rate": 5.2972972972972976e-05,
      "loss": 0.0931,
      "step": 2450
    },
    {
      "epoch": 7.39,
      "learning_rate": 5.237237237237238e-05,
      "loss": 0.1673,
      "step": 2460
    },
    {
      "epoch": 7.42,
      "learning_rate": 5.177177177177177e-05,
      "loss": 0.1473,
      "step": 2470
    },
    {
      "epoch": 7.45,
      "learning_rate": 5.117117117117117e-05,
      "loss": 0.1578,
      "step": 2480
    },
    {
      "epoch": 7.48,
      "learning_rate": 5.063063063063064e-05,
      "loss": 0.1114,
      "step": 2490
    },
    {
      "epoch": 7.51,
      "learning_rate": 5.003003003003003e-05,
      "loss": 0.1742,
      "step": 2500
    },
    {
      "epoch": 7.51,
      "eval_accuracy": 0.9213602550478215,
      "eval_loss": 0.3341035842895508,
      "eval_runtime": 33.4101,
      "eval_samples_per_second": 28.165,
      "eval_steps_per_second": 3.532,
      "step": 2500
    },
    {
      "epoch": 7.54,
      "learning_rate": 4.942942942942943e-05,
      "loss": 0.12,
      "step": 2510
    },
    {
      "epoch": 7.57,
      "learning_rate": 4.882882882882883e-05,
      "loss": 0.1414,
      "step": 2520
    },
    {
      "epoch": 7.6,
      "learning_rate": 4.822822822822823e-05,
      "loss": 0.1844,
      "step": 2530
    },
    {
      "epoch": 7.63,
      "learning_rate": 4.762762762762763e-05,
      "loss": 0.1866,
      "step": 2540
    },
    {
      "epoch": 7.66,
      "learning_rate": 4.7027027027027035e-05,
      "loss": 0.2651,
      "step": 2550
    },
    {
      "epoch": 7.69,
      "learning_rate": 4.642642642642643e-05,
      "loss": 0.0861,
      "step": 2560
    },
    {
      "epoch": 7.72,
      "learning_rate": 4.582582582582583e-05,
      "loss": 0.0685,
      "step": 2570
    },
    {
      "epoch": 7.75,
      "learning_rate": 4.5225225225225224e-05,
      "loss": 0.0992,
      "step": 2580
    },
    {
      "epoch": 7.78,
      "learning_rate": 4.4624624624624625e-05,
      "loss": 0.051,
      "step": 2590
    },
    {
      "epoch": 7.81,
      "learning_rate": 4.4024024024024026e-05,
      "loss": 0.0949,
      "step": 2600
    },
    {
      "epoch": 7.81,
      "eval_accuracy": 0.9256110520722636,
      "eval_loss": 0.3125704526901245,
      "eval_runtime": 33.5498,
      "eval_samples_per_second": 28.048,
      "eval_steps_per_second": 3.517,
      "step": 2600
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.342342342342343e-05,
      "loss": 0.0721,
      "step": 2610
    },
    {
      "epoch": 7.87,
      "learning_rate": 4.282282282282282e-05,
      "loss": 0.2377,
      "step": 2620
    },
    {
      "epoch": 7.9,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.1324,
      "step": 2630
    },
    {
      "epoch": 7.93,
      "learning_rate": 4.162162162162163e-05,
      "loss": 0.1696,
      "step": 2640
    },
    {
      "epoch": 7.96,
      "learning_rate": 4.102102102102102e-05,
      "loss": 0.1086,
      "step": 2650
    },
    {
      "epoch": 7.99,
      "learning_rate": 4.0420420420420424e-05,
      "loss": 0.2007,
      "step": 2660
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.981981981981982e-05,
      "loss": 0.1377,
      "step": 2670
    },
    {
      "epoch": 8.05,
      "learning_rate": 3.921921921921922e-05,
      "loss": 0.0684,
      "step": 2680
    },
    {
      "epoch": 8.08,
      "learning_rate": 3.861861861861862e-05,
      "loss": 0.0386,
      "step": 2690
    },
    {
      "epoch": 8.11,
      "learning_rate": 3.801801801801802e-05,
      "loss": 0.1732,
      "step": 2700
    },
    {
      "epoch": 8.11,
      "eval_accuracy": 0.9234856535600425,
      "eval_loss": 0.30780139565467834,
      "eval_runtime": 32.9779,
      "eval_samples_per_second": 28.534,
      "eval_steps_per_second": 3.578,
      "step": 2700
    },
    {
      "epoch": 8.14,
      "learning_rate": 3.741741741741742e-05,
      "loss": 0.0642,
      "step": 2710
    },
    {
      "epoch": 8.17,
      "learning_rate": 3.6816816816816816e-05,
      "loss": 0.16,
      "step": 2720
    },
    {
      "epoch": 8.2,
      "learning_rate": 3.621621621621622e-05,
      "loss": 0.0922,
      "step": 2730
    },
    {
      "epoch": 8.23,
      "learning_rate": 3.561561561561562e-05,
      "loss": 0.0865,
      "step": 2740
    },
    {
      "epoch": 8.26,
      "learning_rate": 3.501501501501502e-05,
      "loss": 0.1227,
      "step": 2750
    },
    {
      "epoch": 8.29,
      "learning_rate": 3.441441441441441e-05,
      "loss": 0.0769,
      "step": 2760
    },
    {
      "epoch": 8.32,
      "learning_rate": 3.381381381381381e-05,
      "loss": 0.1192,
      "step": 2770
    },
    {
      "epoch": 8.35,
      "learning_rate": 3.3213213213213214e-05,
      "loss": 0.1045,
      "step": 2780
    },
    {
      "epoch": 8.38,
      "learning_rate": 3.2612612612612615e-05,
      "loss": 0.1013,
      "step": 2790
    },
    {
      "epoch": 8.41,
      "learning_rate": 3.2012012012012015e-05,
      "loss": 0.0894,
      "step": 2800
    },
    {
      "epoch": 8.41,
      "eval_accuracy": 0.926673751328374,
      "eval_loss": 0.30979347229003906,
      "eval_runtime": 33.3735,
      "eval_samples_per_second": 28.196,
      "eval_steps_per_second": 3.536,
      "step": 2800
    },
    {
      "epoch": 8.44,
      "learning_rate": 3.141141141141141e-05,
      "loss": 0.0846,
      "step": 2810
    },
    {
      "epoch": 8.47,
      "learning_rate": 3.081081081081082e-05,
      "loss": 0.0786,
      "step": 2820
    },
    {
      "epoch": 8.5,
      "learning_rate": 3.021021021021021e-05,
      "loss": 0.1076,
      "step": 2830
    },
    {
      "epoch": 8.53,
      "learning_rate": 2.9609609609609612e-05,
      "loss": 0.0685,
      "step": 2840
    },
    {
      "epoch": 8.56,
      "learning_rate": 2.900900900900901e-05,
      "loss": 0.0529,
      "step": 2850
    },
    {
      "epoch": 8.59,
      "learning_rate": 2.840840840840841e-05,
      "loss": 0.1153,
      "step": 2860
    },
    {
      "epoch": 8.62,
      "learning_rate": 2.7807807807807808e-05,
      "loss": 0.0839,
      "step": 2870
    },
    {
      "epoch": 8.65,
      "learning_rate": 2.720720720720721e-05,
      "loss": 0.0996,
      "step": 2880
    },
    {
      "epoch": 8.68,
      "learning_rate": 2.660660660660661e-05,
      "loss": 0.1005,
      "step": 2890
    },
    {
      "epoch": 8.71,
      "learning_rate": 2.6006006006006007e-05,
      "loss": 0.1257,
      "step": 2900
    },
    {
      "epoch": 8.71,
      "eval_accuracy": 0.9319872476089267,
      "eval_loss": 0.30299508571624756,
      "eval_runtime": 33.1081,
      "eval_samples_per_second": 28.422,
      "eval_steps_per_second": 3.564,
      "step": 2900
    },
    {
      "epoch": 8.74,
      "learning_rate": 2.5405405405405408e-05,
      "loss": 0.16,
      "step": 2910
    },
    {
      "epoch": 8.77,
      "learning_rate": 2.480480480480481e-05,
      "loss": 0.1078,
      "step": 2920
    },
    {
      "epoch": 8.8,
      "learning_rate": 2.4204204204204206e-05,
      "loss": 0.0802,
      "step": 2930
    },
    {
      "epoch": 8.83,
      "learning_rate": 2.3603603603603607e-05,
      "loss": 0.0342,
      "step": 2940
    },
    {
      "epoch": 8.86,
      "learning_rate": 2.3003003003003004e-05,
      "loss": 0.2069,
      "step": 2950
    },
    {
      "epoch": 8.89,
      "learning_rate": 2.24024024024024e-05,
      "loss": 0.0605,
      "step": 2960
    },
    {
      "epoch": 8.92,
      "learning_rate": 2.1801801801801803e-05,
      "loss": 0.1067,
      "step": 2970
    },
    {
      "epoch": 8.95,
      "learning_rate": 2.12012012012012e-05,
      "loss": 0.0894,
      "step": 2980
    },
    {
      "epoch": 8.98,
      "learning_rate": 2.06006006006006e-05,
      "loss": 0.0995,
      "step": 2990
    },
    {
      "epoch": 9.01,
      "learning_rate": 2e-05,
      "loss": 0.1747,
      "step": 3000
    },
    {
      "epoch": 9.01,
      "eval_accuracy": 0.9256110520722636,
      "eval_loss": 0.3105961084365845,
      "eval_runtime": 33.3236,
      "eval_samples_per_second": 28.238,
      "eval_steps_per_second": 3.541,
      "step": 3000
    },
    {
      "epoch": 9.04,
      "learning_rate": 1.9399399399399402e-05,
      "loss": 0.0942,
      "step": 3010
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.87987987987988e-05,
      "loss": 0.0607,
      "step": 3020
    },
    {
      "epoch": 9.1,
      "learning_rate": 1.81981981981982e-05,
      "loss": 0.0353,
      "step": 3030
    },
    {
      "epoch": 9.13,
      "learning_rate": 1.7597597597597598e-05,
      "loss": 0.0738,
      "step": 3040
    },
    {
      "epoch": 9.16,
      "learning_rate": 1.6996996996996996e-05,
      "loss": 0.1516,
      "step": 3050
    },
    {
      "epoch": 9.19,
      "learning_rate": 1.6396396396396396e-05,
      "loss": 0.0942,
      "step": 3060
    },
    {
      "epoch": 9.22,
      "learning_rate": 1.5795795795795794e-05,
      "loss": 0.0873,
      "step": 3070
    },
    {
      "epoch": 9.25,
      "learning_rate": 1.5195195195195196e-05,
      "loss": 0.1092,
      "step": 3080
    },
    {
      "epoch": 9.28,
      "learning_rate": 1.4594594594594596e-05,
      "loss": 0.0645,
      "step": 3090
    },
    {
      "epoch": 9.31,
      "learning_rate": 1.3993993993993995e-05,
      "loss": 0.2119,
      "step": 3100
    },
    {
      "epoch": 9.31,
      "eval_accuracy": 0.9298618490967057,
      "eval_loss": 0.30373117327690125,
      "eval_runtime": 33.2692,
      "eval_samples_per_second": 28.284,
      "eval_steps_per_second": 3.547,
      "step": 3100
    },
    {
      "epoch": 9.34,
      "learning_rate": 1.3393393393393394e-05,
      "loss": 0.0756,
      "step": 3110
    },
    {
      "epoch": 9.37,
      "learning_rate": 1.2792792792792793e-05,
      "loss": 0.0866,
      "step": 3120
    },
    {
      "epoch": 9.4,
      "learning_rate": 1.2192192192192192e-05,
      "loss": 0.043,
      "step": 3130
    },
    {
      "epoch": 9.43,
      "learning_rate": 1.1591591591591593e-05,
      "loss": 0.0971,
      "step": 3140
    },
    {
      "epoch": 9.46,
      "learning_rate": 1.0990990990990992e-05,
      "loss": 0.0594,
      "step": 3150
    },
    {
      "epoch": 9.49,
      "learning_rate": 1.0390390390390391e-05,
      "loss": 0.0656,
      "step": 3160
    },
    {
      "epoch": 9.52,
      "learning_rate": 9.78978978978979e-06,
      "loss": 0.0711,
      "step": 3170
    },
    {
      "epoch": 9.55,
      "learning_rate": 9.18918918918919e-06,
      "loss": 0.1479,
      "step": 3180
    },
    {
      "epoch": 9.58,
      "learning_rate": 8.588588588588589e-06,
      "loss": 0.1114,
      "step": 3190
    },
    {
      "epoch": 9.61,
      "learning_rate": 7.987987987987988e-06,
      "loss": 0.1074,
      "step": 3200
    },
    {
      "epoch": 9.61,
      "eval_accuracy": 0.9277364505844846,
      "eval_loss": 0.30486080050468445,
      "eval_runtime": 33.3497,
      "eval_samples_per_second": 28.216,
      "eval_steps_per_second": 3.538,
      "step": 3200
    },
    {
      "epoch": 9.64,
      "learning_rate": 7.387387387387387e-06,
      "loss": 0.0991,
      "step": 3210
    },
    {
      "epoch": 9.67,
      "learning_rate": 6.786786786786788e-06,
      "loss": 0.0901,
      "step": 3220
    },
    {
      "epoch": 9.7,
      "learning_rate": 6.186186186186187e-06,
      "loss": 0.0995,
      "step": 3230
    },
    {
      "epoch": 9.73,
      "learning_rate": 5.585585585585586e-06,
      "loss": 0.0859,
      "step": 3240
    },
    {
      "epoch": 9.76,
      "learning_rate": 4.984984984984985e-06,
      "loss": 0.0628,
      "step": 3250
    },
    {
      "epoch": 9.79,
      "learning_rate": 4.384384384384384e-06,
      "loss": 0.0797,
      "step": 3260
    },
    {
      "epoch": 9.82,
      "learning_rate": 3.783783783783784e-06,
      "loss": 0.1148,
      "step": 3270
    },
    {
      "epoch": 9.85,
      "learning_rate": 3.183183183183183e-06,
      "loss": 0.0777,
      "step": 3280
    },
    {
      "epoch": 9.88,
      "learning_rate": 2.5825825825825827e-06,
      "loss": 0.0999,
      "step": 3290
    },
    {
      "epoch": 9.91,
      "learning_rate": 1.981981981981982e-06,
      "loss": 0.1275,
      "step": 3300
    },
    {
      "epoch": 9.91,
      "eval_accuracy": 0.9309245483528161,
      "eval_loss": 0.30457302927970886,
      "eval_runtime": 32.8441,
      "eval_samples_per_second": 28.65,
      "eval_steps_per_second": 3.593,
      "step": 3300
    },
    {
      "epoch": 9.94,
      "learning_rate": 1.3813813813813816e-06,
      "loss": 0.0675,
      "step": 3310
    },
    {
      "epoch": 9.97,
      "learning_rate": 7.807807807807808e-07,
      "loss": 0.059,
      "step": 3320
    },
    {
      "epoch": 10.0,
      "learning_rate": 1.801801801801802e-07,
      "loss": 0.0938,
      "step": 3330
    },
    {
      "epoch": 10.0,
      "step": 3330,
      "total_flos": 1.3394666639602483e+18,
      "train_loss": 0.2829396238287648,
      "train_runtime": 3129.5189,
      "train_samples_per_second": 17.025,
      "train_steps_per_second": 1.064
    }
  ],
  "max_steps": 3330,
  "num_train_epochs": 10,
  "total_flos": 1.3394666639602483e+18,
  "trial_name": null,
  "trial_params": null
}