{
  "best_metric": 0.7152466367713004,
  "best_model_checkpoint": "/content/drive/MyDrive/vit-base-valence/checkpoint-850",
  "epoch": 8.0,
  "eval_steps": 50,
  "global_step": 1008,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 1.9801587301587305e-05,
      "loss": 1.1812,
      "step": 10
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9603174603174602e-05,
      "loss": 1.138,
      "step": 20
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.9404761904761906e-05,
      "loss": 1.0099,
      "step": 30
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.920634920634921e-05,
      "loss": 1.0198,
      "step": 40
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.900793650793651e-05,
      "loss": 0.9472,
      "step": 50
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.6031390134529148,
      "eval_loss": 0.8905579447746277,
      "eval_runtime": 12.7388,
      "eval_samples_per_second": 35.011,
      "eval_steps_per_second": 4.396,
      "step": 50
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.880952380952381e-05,
      "loss": 0.907,
      "step": 60
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.8611111111111114e-05,
      "loss": 0.87,
      "step": 70
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8412698412698415e-05,
      "loss": 0.8141,
      "step": 80
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.8214285714285715e-05,
      "loss": 0.8191,
      "step": 90
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.801587301587302e-05,
      "loss": 0.8528,
      "step": 100
    },
    {
      "epoch": 0.79,
      "eval_accuracy": 0.6614349775784754,
      "eval_loss": 0.7742370367050171,
      "eval_runtime": 12.7432,
      "eval_samples_per_second": 34.999,
      "eval_steps_per_second": 4.394,
      "step": 100
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.781746031746032e-05,
      "loss": 0.7775,
      "step": 110
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.761904761904762e-05,
      "loss": 0.7763,
      "step": 120
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.7420634920634923e-05,
      "loss": 0.654,
      "step": 130
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.7222222222222224e-05,
      "loss": 0.5683,
      "step": 140
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.7023809523809524e-05,
      "loss": 0.544,
      "step": 150
    },
    {
      "epoch": 1.19,
      "eval_accuracy": 0.6659192825112108,
      "eval_loss": 0.7118873000144958,
      "eval_runtime": 12.762,
      "eval_samples_per_second": 34.947,
      "eval_steps_per_second": 4.388,
      "step": 150
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.6825396825396828e-05,
      "loss": 0.5242,
      "step": 160
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.6626984126984128e-05,
      "loss": 0.569,
      "step": 170
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.642857142857143e-05,
      "loss": 0.4828,
      "step": 180
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.6230158730158732e-05,
      "loss": 0.4871,
      "step": 190
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.6051587301587302e-05,
      "loss": 0.5249,
      "step": 200
    },
    {
      "epoch": 1.59,
      "eval_accuracy": 0.6905829596412556,
      "eval_loss": 0.6951313018798828,
      "eval_runtime": 12.8506,
      "eval_samples_per_second": 34.707,
      "eval_steps_per_second": 4.358,
      "step": 200
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.5853174603174603e-05,
      "loss": 0.5511,
      "step": 210
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.5654761904761906e-05,
      "loss": 0.4835,
      "step": 220
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.5456349206349207e-05,
      "loss": 0.4923,
      "step": 230
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.5257936507936509e-05,
      "loss": 0.5137,
      "step": 240
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.5059523809523811e-05,
      "loss": 0.4744,
      "step": 250
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.7062780269058296,
      "eval_loss": 0.6830359697341919,
      "eval_runtime": 12.7409,
      "eval_samples_per_second": 35.005,
      "eval_steps_per_second": 4.395,
      "step": 250
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.4861111111111113e-05,
      "loss": 0.3409,
      "step": 260
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.4662698412698413e-05,
      "loss": 0.3297,
      "step": 270
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.4464285714285715e-05,
      "loss": 0.3496,
      "step": 280
    },
    {
      "epoch": 2.3,
      "learning_rate": 1.4265873015873018e-05,
      "loss": 0.3384,
      "step": 290
    },
    {
      "epoch": 2.38,
      "learning_rate": 1.406746031746032e-05,
      "loss": 0.2939,
      "step": 300
    },
    {
      "epoch": 2.38,
      "eval_accuracy": 0.7130044843049327,
      "eval_loss": 0.6794872283935547,
      "eval_runtime": 12.7967,
      "eval_samples_per_second": 34.853,
      "eval_steps_per_second": 4.376,
      "step": 300
    },
    {
      "epoch": 2.46,
      "learning_rate": 1.386904761904762e-05,
      "loss": 0.3219,
      "step": 310
    },
    {
      "epoch": 2.54,
      "learning_rate": 1.3670634920634922e-05,
      "loss": 0.2759,
      "step": 320
    },
    {
      "epoch": 2.62,
      "learning_rate": 1.3472222222222224e-05,
      "loss": 0.252,
      "step": 330
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.3273809523809524e-05,
      "loss": 0.2574,
      "step": 340
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.3075396825396826e-05,
      "loss": 0.2802,
      "step": 350
    },
    {
      "epoch": 2.78,
      "eval_accuracy": 0.7062780269058296,
      "eval_loss": 0.6800991296768188,
      "eval_runtime": 12.7516,
      "eval_samples_per_second": 34.976,
      "eval_steps_per_second": 4.392,
      "step": 350
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.2876984126984129e-05,
      "loss": 0.3155,
      "step": 360
    },
    {
      "epoch": 2.94,
      "learning_rate": 1.2678571428571429e-05,
      "loss": 0.3315,
      "step": 370
    },
    {
      "epoch": 3.02,
      "learning_rate": 1.2480158730158731e-05,
      "loss": 0.2872,
      "step": 380
    },
    {
      "epoch": 3.1,
      "learning_rate": 1.2281746031746033e-05,
      "loss": 0.1899,
      "step": 390
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.2083333333333333e-05,
      "loss": 0.1755,
      "step": 400
    },
    {
      "epoch": 3.17,
      "eval_accuracy": 0.7040358744394619,
      "eval_loss": 0.6927831172943115,
      "eval_runtime": 12.8352,
      "eval_samples_per_second": 34.748,
      "eval_steps_per_second": 4.363,
      "step": 400
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.1884920634920635e-05,
      "loss": 0.1763,
      "step": 410
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.1686507936507938e-05,
      "loss": 0.1725,
      "step": 420
    },
    {
      "epoch": 3.41,
      "learning_rate": 1.1488095238095238e-05,
      "loss": 0.1742,
      "step": 430
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.128968253968254e-05,
      "loss": 0.162,
      "step": 440
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.1091269841269842e-05,
      "loss": 0.1921,
      "step": 450
    },
    {
      "epoch": 3.57,
      "eval_accuracy": 0.7040358744394619,
      "eval_loss": 0.7218114733695984,
      "eval_runtime": 12.8436,
      "eval_samples_per_second": 34.725,
      "eval_steps_per_second": 4.36,
      "step": 450
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.0892857142857142e-05,
      "loss": 0.1636,
      "step": 460
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.0694444444444444e-05,
      "loss": 0.1979,
      "step": 470
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.0496031746031747e-05,
      "loss": 0.1994,
      "step": 480
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.0297619047619047e-05,
      "loss": 0.1562,
      "step": 490
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.0099206349206349e-05,
      "loss": 0.1682,
      "step": 500
    },
    {
      "epoch": 3.97,
      "eval_accuracy": 0.7062780269058296,
      "eval_loss": 0.7192036509513855,
      "eval_runtime": 12.7187,
      "eval_samples_per_second": 35.066,
      "eval_steps_per_second": 4.403,
      "step": 500
    },
    {
      "epoch": 4.05,
      "learning_rate": 9.900793650793653e-06,
      "loss": 0.1473,
      "step": 510
    },
    {
      "epoch": 4.13,
      "learning_rate": 9.702380952380953e-06,
      "loss": 0.0982,
      "step": 520
    },
    {
      "epoch": 4.21,
      "learning_rate": 9.503968253968255e-06,
      "loss": 0.0967,
      "step": 530
    },
    {
      "epoch": 4.29,
      "learning_rate": 9.305555555555557e-06,
      "loss": 0.1019,
      "step": 540
    },
    {
      "epoch": 4.37,
      "learning_rate": 9.107142857142858e-06,
      "loss": 0.1369,
      "step": 550
    },
    {
      "epoch": 4.37,
      "eval_accuracy": 0.7040358744394619,
      "eval_loss": 0.7594305276870728,
      "eval_runtime": 12.7574,
      "eval_samples_per_second": 34.96,
      "eval_steps_per_second": 4.39,
      "step": 550
    },
    {
      "epoch": 4.44,
      "learning_rate": 8.90873015873016e-06,
      "loss": 0.0901,
      "step": 560
    },
    {
      "epoch": 4.52,
      "learning_rate": 8.710317460317462e-06,
      "loss": 0.121,
      "step": 570
    },
    {
      "epoch": 4.6,
      "learning_rate": 8.511904761904762e-06,
      "loss": 0.0922,
      "step": 580
    },
    {
      "epoch": 4.68,
      "learning_rate": 8.313492063492064e-06,
      "loss": 0.112,
      "step": 590
    },
    {
      "epoch": 4.76,
      "learning_rate": 8.115079365079366e-06,
      "loss": 0.1177,
      "step": 600
    },
    {
      "epoch": 4.76,
      "eval_accuracy": 0.7062780269058296,
      "eval_loss": 0.7682589292526245,
      "eval_runtime": 12.7777,
      "eval_samples_per_second": 34.905,
      "eval_steps_per_second": 4.383,
      "step": 600
    },
    {
      "epoch": 4.84,
      "learning_rate": 7.916666666666667e-06,
      "loss": 0.1057,
      "step": 610
    },
    {
      "epoch": 4.92,
      "learning_rate": 7.718253968253969e-06,
      "loss": 0.1122,
      "step": 620
    },
    {
      "epoch": 5.0,
      "learning_rate": 7.519841269841271e-06,
      "loss": 0.1105,
      "step": 630
    },
    {
      "epoch": 5.08,
      "learning_rate": 7.321428571428572e-06,
      "loss": 0.0735,
      "step": 640
    },
    {
      "epoch": 5.16,
      "learning_rate": 7.123015873015874e-06,
      "loss": 0.0504,
      "step": 650
    },
    {
      "epoch": 5.16,
      "eval_accuracy": 0.7040358744394619,
      "eval_loss": 0.7936971783638,
      "eval_runtime": 12.9729,
      "eval_samples_per_second": 34.379,
      "eval_steps_per_second": 4.317,
      "step": 650
    },
    {
      "epoch": 5.24,
      "learning_rate": 6.924603174603175e-06,
      "loss": 0.0705,
      "step": 660
    },
    {
      "epoch": 5.32,
      "learning_rate": 6.726190476190477e-06,
      "loss": 0.0563,
      "step": 670
    },
    {
      "epoch": 5.4,
      "learning_rate": 6.5277777777777784e-06,
      "loss": 0.0683,
      "step": 680
    },
    {
      "epoch": 5.48,
      "learning_rate": 6.32936507936508e-06,
      "loss": 0.0698,
      "step": 690
    },
    {
      "epoch": 5.56,
      "learning_rate": 6.130952380952382e-06,
      "loss": 0.0797,
      "step": 700
    },
    {
      "epoch": 5.56,
      "eval_accuracy": 0.7062780269058296,
      "eval_loss": 0.8110105395317078,
      "eval_runtime": 12.8549,
      "eval_samples_per_second": 34.695,
      "eval_steps_per_second": 4.356,
      "step": 700
    },
    {
      "epoch": 5.63,
      "learning_rate": 5.932539682539683e-06,
      "loss": 0.0628,
      "step": 710
    },
    {
      "epoch": 5.71,
      "learning_rate": 5.734126984126984e-06,
      "loss": 0.0794,
      "step": 720
    },
    {
      "epoch": 5.79,
      "learning_rate": 5.535714285714286e-06,
      "loss": 0.0809,
      "step": 730
    },
    {
      "epoch": 5.87,
      "learning_rate": 5.337301587301587e-06,
      "loss": 0.1132,
      "step": 740
    },
    {
      "epoch": 5.95,
      "learning_rate": 5.138888888888889e-06,
      "loss": 0.0767,
      "step": 750
    },
    {
      "epoch": 5.95,
      "eval_accuracy": 0.7107623318385651,
      "eval_loss": 0.8268046975135803,
      "eval_runtime": 12.7986,
      "eval_samples_per_second": 34.848,
      "eval_steps_per_second": 4.375,
      "step": 750
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.940476190476191e-06,
      "loss": 0.0694,
      "step": 760
    },
    {
      "epoch": 6.11,
      "learning_rate": 4.742063492063492e-06,
      "loss": 0.0573,
      "step": 770
    },
    {
      "epoch": 6.19,
      "learning_rate": 4.543650793650794e-06,
      "loss": 0.0548,
      "step": 780
    },
    {
      "epoch": 6.27,
      "learning_rate": 4.345238095238096e-06,
      "loss": 0.0539,
      "step": 790
    },
    {
      "epoch": 6.35,
      "learning_rate": 4.146825396825397e-06,
      "loss": 0.0597,
      "step": 800
    },
    {
      "epoch": 6.35,
      "eval_accuracy": 0.7062780269058296,
      "eval_loss": 0.8549516797065735,
      "eval_runtime": 12.8211,
      "eval_samples_per_second": 34.786,
      "eval_steps_per_second": 4.368,
      "step": 800
    },
    {
      "epoch": 6.43,
      "learning_rate": 3.9484126984126985e-06,
      "loss": 0.0572,
      "step": 810
    },
    {
      "epoch": 6.51,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.0579,
      "step": 820
    },
    {
      "epoch": 6.59,
      "learning_rate": 3.551587301587302e-06,
      "loss": 0.0727,
      "step": 830
    },
    {
      "epoch": 6.67,
      "learning_rate": 3.3531746031746034e-06,
      "loss": 0.0522,
      "step": 840
    },
    {
      "epoch": 6.75,
      "learning_rate": 3.154761904761905e-06,
      "loss": 0.0669,
      "step": 850
    },
    {
      "epoch": 6.75,
      "eval_accuracy": 0.7152466367713004,
      "eval_loss": 0.8530937433242798,
      "eval_runtime": 12.7041,
      "eval_samples_per_second": 35.107,
      "eval_steps_per_second": 4.408,
      "step": 850
    },
    {
      "epoch": 6.83,
      "learning_rate": 2.9563492063492066e-06,
      "loss": 0.0586,
      "step": 860
    },
    {
      "epoch": 6.9,
      "learning_rate": 2.7579365079365083e-06,
      "loss": 0.0445,
      "step": 870
    },
    {
      "epoch": 6.98,
      "learning_rate": 2.5595238095238095e-06,
      "loss": 0.0552,
      "step": 880
    },
    {
      "epoch": 7.06,
      "learning_rate": 2.361111111111111e-06,
      "loss": 0.0302,
      "step": 890
    },
    {
      "epoch": 7.14,
      "learning_rate": 2.1626984126984128e-06,
      "loss": 0.0483,
      "step": 900
    },
    {
      "epoch": 7.14,
      "eval_accuracy": 0.7130044843049327,
      "eval_loss": 0.8547905683517456,
      "eval_runtime": 12.8017,
      "eval_samples_per_second": 34.839,
      "eval_steps_per_second": 4.374,
      "step": 900
    },
    {
      "epoch": 7.22,
      "learning_rate": 1.9642857142857144e-06,
      "loss": 0.0585,
      "step": 910
    },
    {
      "epoch": 7.3,
      "learning_rate": 1.765873015873016e-06,
      "loss": 0.0335,
      "step": 920
    },
    {
      "epoch": 7.38,
      "learning_rate": 1.5674603174603175e-06,
      "loss": 0.037,
      "step": 930
    },
    {
      "epoch": 7.46,
      "learning_rate": 1.3690476190476193e-06,
      "loss": 0.0396,
      "step": 940
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.1706349206349208e-06,
      "loss": 0.0587,
      "step": 950
    },
    {
      "epoch": 7.54,
      "eval_accuracy": 0.7152466367713004,
      "eval_loss": 0.8677387237548828,
      "eval_runtime": 12.7829,
      "eval_samples_per_second": 34.89,
      "eval_steps_per_second": 4.381,
      "step": 950
    },
    {
      "epoch": 7.62,
      "learning_rate": 9.722222222222224e-07,
      "loss": 0.0596,
      "step": 960
    },
    {
      "epoch": 7.7,
      "learning_rate": 7.738095238095239e-07,
      "loss": 0.0418,
      "step": 970
    },
    {
      "epoch": 7.78,
      "learning_rate": 5.753968253968255e-07,
      "loss": 0.0619,
      "step": 980
    },
    {
      "epoch": 7.86,
      "learning_rate": 3.76984126984127e-07,
      "loss": 0.0508,
      "step": 990
    },
    {
      "epoch": 7.94,
      "learning_rate": 1.7857142857142858e-07,
      "loss": 0.038,
      "step": 1000
    },
    {
      "epoch": 7.94,
      "eval_accuracy": 0.7130044843049327,
      "eval_loss": 0.8715421557426453,
      "eval_runtime": 12.7857,
      "eval_samples_per_second": 34.883,
      "eval_steps_per_second": 4.38,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "step": 1008,
      "total_flos": 7.325059066442809e+18,
      "train_loss": 0.2753036499082569,
      "train_runtime": 1319.5023,
      "train_samples_per_second": 24.294,
      "train_steps_per_second": 0.764
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7152466367713004,
      "eval_loss": 0.8530937433242798,
      "eval_runtime": 12.7761,
      "eval_samples_per_second": 34.909,
      "eval_steps_per_second": 4.383,
      "step": 1008
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9852757674070377,
      "eval_loss": 0.04634059593081474,
      "eval_runtime": 115.7634,
      "eval_samples_per_second": 34.614,
      "eval_steps_per_second": 4.328,
      "step": 1008
    }
  ],
  "logging_steps": 10,
  "max_steps": 1008,
  "num_train_epochs": 8,
  "save_steps": 50,
  "total_flos": 7.325059066442809e+18,
  "trial_name": null,
  "trial_params": null
}