ALM-AHME's picture
End of training
cd1eb92
{
"best_metric": 0.9901477832512315,
"best_model_checkpoint": "beit-large-patch16-224-finetuned-Lesion-Classification-HAM10000-AH-60-20-20-Shuffled-3rd/checkpoint-1368",
"epoch": 11.947598253275109,
"global_step": 1368,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 2.029220779220779e-08,
"loss": 2.3507,
"step": 5
},
{
"epoch": 0.09,
"learning_rate": 4.058441558441558e-08,
"loss": 2.3465,
"step": 10
},
{
"epoch": 0.13,
"learning_rate": 6.087662337662339e-08,
"loss": 2.2753,
"step": 15
},
{
"epoch": 0.17,
"learning_rate": 8.116883116883116e-08,
"loss": 2.3029,
"step": 20
},
{
"epoch": 0.22,
"learning_rate": 1.0146103896103897e-07,
"loss": 2.2669,
"step": 25
},
{
"epoch": 0.26,
"learning_rate": 1.2175324675324677e-07,
"loss": 2.2821,
"step": 30
},
{
"epoch": 0.31,
"learning_rate": 1.4204545454545455e-07,
"loss": 2.1245,
"step": 35
},
{
"epoch": 0.35,
"learning_rate": 1.6233766233766232e-07,
"loss": 2.3123,
"step": 40
},
{
"epoch": 0.39,
"learning_rate": 1.8262987012987013e-07,
"loss": 2.177,
"step": 45
},
{
"epoch": 0.44,
"learning_rate": 2.0292207792207793e-07,
"loss": 2.1838,
"step": 50
},
{
"epoch": 0.48,
"learning_rate": 2.2321428571428574e-07,
"loss": 2.2212,
"step": 55
},
{
"epoch": 0.52,
"learning_rate": 2.4350649350649354e-07,
"loss": 2.1027,
"step": 60
},
{
"epoch": 0.57,
"learning_rate": 2.637987012987013e-07,
"loss": 2.1215,
"step": 65
},
{
"epoch": 0.61,
"learning_rate": 2.840909090909091e-07,
"loss": 2.1549,
"step": 70
},
{
"epoch": 0.66,
"learning_rate": 3.043831168831169e-07,
"loss": 2.0832,
"step": 75
},
{
"epoch": 0.7,
"learning_rate": 3.2467532467532465e-07,
"loss": 1.9727,
"step": 80
},
{
"epoch": 0.74,
"learning_rate": 3.4496753246753253e-07,
"loss": 2.0903,
"step": 85
},
{
"epoch": 0.79,
"learning_rate": 3.6525974025974026e-07,
"loss": 2.0775,
"step": 90
},
{
"epoch": 0.83,
"learning_rate": 3.855519480519481e-07,
"loss": 2.1454,
"step": 95
},
{
"epoch": 0.87,
"learning_rate": 4.0584415584415586e-07,
"loss": 2.0931,
"step": 100
},
{
"epoch": 0.92,
"learning_rate": 4.2613636363636364e-07,
"loss": 2.0124,
"step": 105
},
{
"epoch": 0.96,
"learning_rate": 4.4642857142857147e-07,
"loss": 1.9835,
"step": 110
},
{
"epoch": 1.0,
"eval_accuracy": 0.2315270935960591,
"eval_loss": 1.9295728206634521,
"eval_runtime": 389.2292,
"eval_samples_per_second": 3.129,
"eval_steps_per_second": 0.198,
"step": 114
},
{
"epoch": 1.0,
"learning_rate": 4.6672077922077925e-07,
"loss": 2.0623,
"step": 115
},
{
"epoch": 1.05,
"learning_rate": 4.870129870129871e-07,
"loss": 1.9206,
"step": 120
},
{
"epoch": 1.09,
"learning_rate": 5.073051948051948e-07,
"loss": 1.9793,
"step": 125
},
{
"epoch": 1.14,
"learning_rate": 5.275974025974026e-07,
"loss": 1.8406,
"step": 130
},
{
"epoch": 1.18,
"learning_rate": 5.478896103896105e-07,
"loss": 1.8882,
"step": 135
},
{
"epoch": 1.22,
"learning_rate": 5.681818181818182e-07,
"loss": 1.958,
"step": 140
},
{
"epoch": 1.27,
"learning_rate": 5.88474025974026e-07,
"loss": 1.8282,
"step": 145
},
{
"epoch": 1.31,
"learning_rate": 6.087662337662339e-07,
"loss": 1.8004,
"step": 150
},
{
"epoch": 1.35,
"learning_rate": 6.290584415584417e-07,
"loss": 1.8836,
"step": 155
},
{
"epoch": 1.4,
"learning_rate": 6.493506493506493e-07,
"loss": 1.8654,
"step": 160
},
{
"epoch": 1.44,
"learning_rate": 6.696428571428571e-07,
"loss": 1.7986,
"step": 165
},
{
"epoch": 1.48,
"learning_rate": 6.899350649350651e-07,
"loss": 1.7501,
"step": 170
},
{
"epoch": 1.53,
"learning_rate": 7.102272727272729e-07,
"loss": 1.8718,
"step": 175
},
{
"epoch": 1.57,
"learning_rate": 7.305194805194805e-07,
"loss": 1.8201,
"step": 180
},
{
"epoch": 1.62,
"learning_rate": 7.508116883116883e-07,
"loss": 1.6845,
"step": 185
},
{
"epoch": 1.66,
"learning_rate": 7.711038961038962e-07,
"loss": 1.7282,
"step": 190
},
{
"epoch": 1.7,
"learning_rate": 7.913961038961039e-07,
"loss": 1.7062,
"step": 195
},
{
"epoch": 1.75,
"learning_rate": 8.116883116883117e-07,
"loss": 1.6732,
"step": 200
},
{
"epoch": 1.79,
"learning_rate": 8.319805194805196e-07,
"loss": 1.6762,
"step": 205
},
{
"epoch": 1.83,
"learning_rate": 8.522727272727273e-07,
"loss": 1.613,
"step": 210
},
{
"epoch": 1.88,
"learning_rate": 8.725649350649351e-07,
"loss": 1.5492,
"step": 215
},
{
"epoch": 1.92,
"learning_rate": 8.928571428571429e-07,
"loss": 1.5562,
"step": 220
},
{
"epoch": 1.97,
"learning_rate": 9.131493506493508e-07,
"loss": 1.6045,
"step": 225
},
{
"epoch": 2.0,
"eval_accuracy": 0.5172413793103449,
"eval_loss": 1.4333993196487427,
"eval_runtime": 47.4589,
"eval_samples_per_second": 25.664,
"eval_steps_per_second": 1.622,
"step": 229
},
{
"epoch": 2.01,
"learning_rate": 9.334415584415585e-07,
"loss": 1.4204,
"step": 230
},
{
"epoch": 2.05,
"learning_rate": 9.537337662337662e-07,
"loss": 1.4782,
"step": 235
},
{
"epoch": 2.1,
"learning_rate": 9.740259740259742e-07,
"loss": 1.4652,
"step": 240
},
{
"epoch": 2.14,
"learning_rate": 9.943181818181819e-07,
"loss": 1.4294,
"step": 245
},
{
"epoch": 2.18,
"learning_rate": 1.0146103896103896e-06,
"loss": 1.3151,
"step": 250
},
{
"epoch": 2.23,
"learning_rate": 1.0349025974025975e-06,
"loss": 1.3861,
"step": 255
},
{
"epoch": 2.27,
"learning_rate": 1.0551948051948053e-06,
"loss": 1.4704,
"step": 260
},
{
"epoch": 2.31,
"learning_rate": 1.075487012987013e-06,
"loss": 1.373,
"step": 265
},
{
"epoch": 2.36,
"learning_rate": 1.095779220779221e-06,
"loss": 1.3331,
"step": 270
},
{
"epoch": 2.4,
"learning_rate": 1.1160714285714287e-06,
"loss": 1.3973,
"step": 275
},
{
"epoch": 2.45,
"learning_rate": 1.1363636363636364e-06,
"loss": 1.2745,
"step": 280
},
{
"epoch": 2.49,
"learning_rate": 1.1566558441558443e-06,
"loss": 1.2652,
"step": 285
},
{
"epoch": 2.53,
"learning_rate": 1.176948051948052e-06,
"loss": 1.1772,
"step": 290
},
{
"epoch": 2.58,
"learning_rate": 1.1972402597402598e-06,
"loss": 1.2118,
"step": 295
},
{
"epoch": 2.62,
"learning_rate": 1.2175324675324677e-06,
"loss": 1.1511,
"step": 300
},
{
"epoch": 2.66,
"learning_rate": 1.2378246753246754e-06,
"loss": 1.1987,
"step": 305
},
{
"epoch": 2.71,
"learning_rate": 1.2581168831168834e-06,
"loss": 1.1258,
"step": 310
},
{
"epoch": 2.75,
"learning_rate": 1.278409090909091e-06,
"loss": 1.1209,
"step": 315
},
{
"epoch": 2.79,
"learning_rate": 1.2987012987012986e-06,
"loss": 1.1496,
"step": 320
},
{
"epoch": 2.84,
"learning_rate": 1.3189935064935067e-06,
"loss": 1.0379,
"step": 325
},
{
"epoch": 2.88,
"learning_rate": 1.3392857142857143e-06,
"loss": 1.1059,
"step": 330
},
{
"epoch": 2.93,
"learning_rate": 1.359577922077922e-06,
"loss": 1.1168,
"step": 335
},
{
"epoch": 2.97,
"learning_rate": 1.3798701298701301e-06,
"loss": 1.0525,
"step": 340
},
{
"epoch": 3.0,
"eval_accuracy": 0.6962233169129721,
"eval_loss": 0.9297933578491211,
"eval_runtime": 47.245,
"eval_samples_per_second": 25.781,
"eval_steps_per_second": 1.63,
"step": 343
},
{
"epoch": 3.01,
"learning_rate": 1.4001623376623376e-06,
"loss": 1.0494,
"step": 345
},
{
"epoch": 3.06,
"learning_rate": 1.4204545454545458e-06,
"loss": 1.0545,
"step": 350
},
{
"epoch": 3.1,
"learning_rate": 1.4407467532467533e-06,
"loss": 0.9587,
"step": 355
},
{
"epoch": 3.14,
"learning_rate": 1.461038961038961e-06,
"loss": 0.9282,
"step": 360
},
{
"epoch": 3.19,
"learning_rate": 1.481331168831169e-06,
"loss": 0.9952,
"step": 365
},
{
"epoch": 3.23,
"learning_rate": 1.5016233766233767e-06,
"loss": 1.0251,
"step": 370
},
{
"epoch": 3.28,
"learning_rate": 1.5219155844155844e-06,
"loss": 0.9288,
"step": 375
},
{
"epoch": 3.32,
"learning_rate": 1.5422077922077924e-06,
"loss": 0.9591,
"step": 380
},
{
"epoch": 3.36,
"learning_rate": 1.5625e-06,
"loss": 0.9768,
"step": 385
},
{
"epoch": 3.41,
"learning_rate": 1.5827922077922078e-06,
"loss": 0.94,
"step": 390
},
{
"epoch": 3.45,
"learning_rate": 1.6030844155844157e-06,
"loss": 0.9394,
"step": 395
},
{
"epoch": 3.49,
"learning_rate": 1.6233766233766235e-06,
"loss": 0.9356,
"step": 400
},
{
"epoch": 3.54,
"learning_rate": 1.6436688311688312e-06,
"loss": 0.8767,
"step": 405
},
{
"epoch": 3.58,
"learning_rate": 1.6639610389610391e-06,
"loss": 0.757,
"step": 410
},
{
"epoch": 3.62,
"learning_rate": 1.6842532467532468e-06,
"loss": 0.8279,
"step": 415
},
{
"epoch": 3.67,
"learning_rate": 1.7045454545454546e-06,
"loss": 0.8506,
"step": 420
},
{
"epoch": 3.71,
"learning_rate": 1.7248376623376625e-06,
"loss": 0.8643,
"step": 425
},
{
"epoch": 3.76,
"learning_rate": 1.7451298701298702e-06,
"loss": 0.9211,
"step": 430
},
{
"epoch": 3.8,
"learning_rate": 1.7654220779220782e-06,
"loss": 0.8413,
"step": 435
},
{
"epoch": 3.84,
"learning_rate": 1.7857142857142859e-06,
"loss": 0.7988,
"step": 440
},
{
"epoch": 3.89,
"learning_rate": 1.8060064935064936e-06,
"loss": 0.7708,
"step": 445
},
{
"epoch": 3.93,
"learning_rate": 1.8262987012987016e-06,
"loss": 0.6857,
"step": 450
},
{
"epoch": 3.97,
"learning_rate": 1.8465909090909093e-06,
"loss": 0.795,
"step": 455
},
{
"epoch": 4.0,
"eval_accuracy": 0.770935960591133,
"eval_loss": 0.6579709649085999,
"eval_runtime": 47.7238,
"eval_samples_per_second": 25.522,
"eval_steps_per_second": 1.613,
"step": 458
},
{
"epoch": 4.02,
"learning_rate": 1.866883116883117e-06,
"loss": 0.6999,
"step": 460
},
{
"epoch": 4.06,
"learning_rate": 1.887175324675325e-06,
"loss": 0.7453,
"step": 465
},
{
"epoch": 4.1,
"learning_rate": 1.9074675324675324e-06,
"loss": 0.618,
"step": 470
},
{
"epoch": 4.15,
"learning_rate": 1.92775974025974e-06,
"loss": 0.7828,
"step": 475
},
{
"epoch": 4.19,
"learning_rate": 1.9480519480519483e-06,
"loss": 0.7425,
"step": 480
},
{
"epoch": 4.24,
"learning_rate": 1.968344155844156e-06,
"loss": 0.6702,
"step": 485
},
{
"epoch": 4.28,
"learning_rate": 1.9886363636363638e-06,
"loss": 0.8407,
"step": 490
},
{
"epoch": 4.32,
"learning_rate": 2.0089285714285715e-06,
"loss": 0.7639,
"step": 495
},
{
"epoch": 4.37,
"learning_rate": 2.0292207792207792e-06,
"loss": 0.7345,
"step": 500
},
{
"epoch": 4.41,
"learning_rate": 2.049512987012987e-06,
"loss": 0.8418,
"step": 505
},
{
"epoch": 4.45,
"learning_rate": 2.069805194805195e-06,
"loss": 0.6487,
"step": 510
},
{
"epoch": 4.5,
"learning_rate": 2.090097402597403e-06,
"loss": 0.6811,
"step": 515
},
{
"epoch": 4.54,
"learning_rate": 2.1103896103896105e-06,
"loss": 0.6427,
"step": 520
},
{
"epoch": 4.59,
"learning_rate": 2.1306818181818183e-06,
"loss": 0.6892,
"step": 525
},
{
"epoch": 4.63,
"learning_rate": 2.150974025974026e-06,
"loss": 0.7313,
"step": 530
},
{
"epoch": 4.67,
"learning_rate": 2.171266233766234e-06,
"loss": 0.7844,
"step": 535
},
{
"epoch": 4.72,
"learning_rate": 2.191558441558442e-06,
"loss": 0.6754,
"step": 540
},
{
"epoch": 4.76,
"learning_rate": 2.2118506493506496e-06,
"loss": 0.7047,
"step": 545
},
{
"epoch": 4.8,
"learning_rate": 2.2321428571428573e-06,
"loss": 0.6764,
"step": 550
},
{
"epoch": 4.85,
"learning_rate": 2.252435064935065e-06,
"loss": 0.5848,
"step": 555
},
{
"epoch": 4.89,
"learning_rate": 2.2727272727272728e-06,
"loss": 0.6101,
"step": 560
},
{
"epoch": 4.93,
"learning_rate": 2.293019480519481e-06,
"loss": 0.5865,
"step": 565
},
{
"epoch": 4.98,
"learning_rate": 2.3133116883116886e-06,
"loss": 0.5739,
"step": 570
},
{
"epoch": 5.0,
"eval_accuracy": 0.8366174055829229,
"eval_loss": 0.4716862440109253,
"eval_runtime": 47.4118,
"eval_samples_per_second": 25.69,
"eval_steps_per_second": 1.624,
"step": 572
},
{
"epoch": 5.02,
"learning_rate": 2.3336038961038964e-06,
"loss": 0.5091,
"step": 575
},
{
"epoch": 5.07,
"learning_rate": 2.353896103896104e-06,
"loss": 0.5958,
"step": 580
},
{
"epoch": 5.11,
"learning_rate": 2.374188311688312e-06,
"loss": 0.5332,
"step": 585
},
{
"epoch": 5.15,
"learning_rate": 2.3944805194805195e-06,
"loss": 0.5888,
"step": 590
},
{
"epoch": 5.2,
"learning_rate": 2.4147727272727277e-06,
"loss": 0.6929,
"step": 595
},
{
"epoch": 5.24,
"learning_rate": 2.4350649350649354e-06,
"loss": 0.6091,
"step": 600
},
{
"epoch": 5.28,
"learning_rate": 2.455357142857143e-06,
"loss": 0.5664,
"step": 605
},
{
"epoch": 5.33,
"learning_rate": 2.475649350649351e-06,
"loss": 0.5503,
"step": 610
},
{
"epoch": 5.37,
"learning_rate": 2.4959415584415586e-06,
"loss": 0.583,
"step": 615
},
{
"epoch": 5.41,
"learning_rate": 2.5162337662337667e-06,
"loss": 0.5986,
"step": 620
},
{
"epoch": 5.46,
"learning_rate": 2.536525974025974e-06,
"loss": 0.6175,
"step": 625
},
{
"epoch": 5.5,
"learning_rate": 2.556818181818182e-06,
"loss": 0.5965,
"step": 630
},
{
"epoch": 5.55,
"learning_rate": 2.57711038961039e-06,
"loss": 0.4783,
"step": 635
},
{
"epoch": 5.59,
"learning_rate": 2.597402597402597e-06,
"loss": 0.5228,
"step": 640
},
{
"epoch": 5.63,
"learning_rate": 2.6176948051948053e-06,
"loss": 0.6572,
"step": 645
},
{
"epoch": 5.68,
"learning_rate": 2.6379870129870135e-06,
"loss": 0.5293,
"step": 650
},
{
"epoch": 5.72,
"learning_rate": 2.658279220779221e-06,
"loss": 0.4884,
"step": 655
},
{
"epoch": 5.76,
"learning_rate": 2.6785714285714285e-06,
"loss": 0.5902,
"step": 660
},
{
"epoch": 5.81,
"learning_rate": 2.6988636363636367e-06,
"loss": 0.4701,
"step": 665
},
{
"epoch": 5.85,
"learning_rate": 2.719155844155844e-06,
"loss": 0.4623,
"step": 670
},
{
"epoch": 5.9,
"learning_rate": 2.739448051948052e-06,
"loss": 0.4476,
"step": 675
},
{
"epoch": 5.94,
"learning_rate": 2.7597402597402603e-06,
"loss": 0.4749,
"step": 680
},
{
"epoch": 5.98,
"learning_rate": 2.780032467532468e-06,
"loss": 0.5821,
"step": 685
},
{
"epoch": 6.0,
"eval_accuracy": 0.8850574712643678,
"eval_loss": 0.3510984480381012,
"eval_runtime": 47.4539,
"eval_samples_per_second": 25.667,
"eval_steps_per_second": 1.623,
"step": 687
},
{
"epoch": 6.03,
"learning_rate": 2.8003246753246753e-06,
"loss": 0.4662,
"step": 690
},
{
"epoch": 6.07,
"learning_rate": 2.8206168831168834e-06,
"loss": 0.3655,
"step": 695
},
{
"epoch": 6.11,
"learning_rate": 2.8409090909090916e-06,
"loss": 0.5519,
"step": 700
},
{
"epoch": 6.16,
"learning_rate": 2.861201298701299e-06,
"loss": 0.5053,
"step": 705
},
{
"epoch": 6.2,
"learning_rate": 2.8814935064935066e-06,
"loss": 0.5677,
"step": 710
},
{
"epoch": 6.24,
"learning_rate": 2.9017857142857148e-06,
"loss": 0.4138,
"step": 715
},
{
"epoch": 6.29,
"learning_rate": 2.922077922077922e-06,
"loss": 0.4516,
"step": 720
},
{
"epoch": 6.33,
"learning_rate": 2.94237012987013e-06,
"loss": 0.4424,
"step": 725
},
{
"epoch": 6.38,
"learning_rate": 2.962662337662338e-06,
"loss": 0.4724,
"step": 730
},
{
"epoch": 6.42,
"learning_rate": 2.9829545454545457e-06,
"loss": 0.5281,
"step": 735
},
{
"epoch": 6.46,
"learning_rate": 3.0032467532467534e-06,
"loss": 0.3981,
"step": 740
},
{
"epoch": 6.51,
"learning_rate": 3.0235389610389615e-06,
"loss": 0.4369,
"step": 745
},
{
"epoch": 6.55,
"learning_rate": 3.043831168831169e-06,
"loss": 0.4717,
"step": 750
},
{
"epoch": 6.59,
"learning_rate": 3.064123376623377e-06,
"loss": 0.3732,
"step": 755
},
{
"epoch": 6.64,
"learning_rate": 3.0844155844155847e-06,
"loss": 0.5155,
"step": 760
},
{
"epoch": 6.68,
"learning_rate": 3.1047077922077924e-06,
"loss": 0.4014,
"step": 765
},
{
"epoch": 6.72,
"learning_rate": 3.125e-06,
"loss": 0.4267,
"step": 770
},
{
"epoch": 6.77,
"learning_rate": 3.1452922077922083e-06,
"loss": 0.4185,
"step": 775
},
{
"epoch": 6.81,
"learning_rate": 3.1655844155844156e-06,
"loss": 0.4573,
"step": 780
},
{
"epoch": 6.86,
"learning_rate": 3.1858766233766237e-06,
"loss": 0.4297,
"step": 785
},
{
"epoch": 6.9,
"learning_rate": 3.2061688311688315e-06,
"loss": 0.4326,
"step": 790
},
{
"epoch": 6.94,
"learning_rate": 3.226461038961039e-06,
"loss": 0.3424,
"step": 795
},
{
"epoch": 6.99,
"learning_rate": 3.246753246753247e-06,
"loss": 0.4566,
"step": 800
},
{
"epoch": 7.0,
"eval_accuracy": 0.9203612479474549,
"eval_loss": 0.2704623341560364,
"eval_runtime": 47.6081,
"eval_samples_per_second": 25.584,
"eval_steps_per_second": 1.617,
"step": 801
},
{
"epoch": 7.03,
"learning_rate": 3.267045454545455e-06,
"loss": 0.5424,
"step": 805
},
{
"epoch": 7.07,
"learning_rate": 3.2873376623376624e-06,
"loss": 0.3633,
"step": 810
},
{
"epoch": 7.12,
"learning_rate": 3.3076298701298705e-06,
"loss": 0.3712,
"step": 815
},
{
"epoch": 7.16,
"learning_rate": 3.3279220779220782e-06,
"loss": 0.3545,
"step": 820
},
{
"epoch": 7.21,
"learning_rate": 3.3482142857142855e-06,
"loss": 0.3979,
"step": 825
},
{
"epoch": 7.25,
"learning_rate": 3.3685064935064937e-06,
"loss": 0.3465,
"step": 830
},
{
"epoch": 7.29,
"learning_rate": 3.388798701298702e-06,
"loss": 0.355,
"step": 835
},
{
"epoch": 7.34,
"learning_rate": 3.409090909090909e-06,
"loss": 0.3266,
"step": 840
},
{
"epoch": 7.38,
"learning_rate": 3.4293831168831173e-06,
"loss": 0.3264,
"step": 845
},
{
"epoch": 7.42,
"learning_rate": 3.449675324675325e-06,
"loss": 0.4101,
"step": 850
},
{
"epoch": 7.47,
"learning_rate": 3.4699675324675323e-06,
"loss": 0.3877,
"step": 855
},
{
"epoch": 7.51,
"learning_rate": 3.4902597402597405e-06,
"loss": 0.3479,
"step": 860
},
{
"epoch": 7.55,
"learning_rate": 3.5105519480519486e-06,
"loss": 0.3163,
"step": 865
},
{
"epoch": 7.6,
"learning_rate": 3.5308441558441563e-06,
"loss": 0.3154,
"step": 870
},
{
"epoch": 7.64,
"learning_rate": 3.5511363636363636e-06,
"loss": 0.3221,
"step": 875
},
{
"epoch": 7.69,
"learning_rate": 3.5714285714285718e-06,
"loss": 0.3248,
"step": 880
},
{
"epoch": 7.73,
"learning_rate": 3.59172077922078e-06,
"loss": 0.3938,
"step": 885
},
{
"epoch": 7.77,
"learning_rate": 3.6120129870129872e-06,
"loss": 0.3077,
"step": 890
},
{
"epoch": 7.82,
"learning_rate": 3.6323051948051954e-06,
"loss": 0.3841,
"step": 895
},
{
"epoch": 7.86,
"learning_rate": 3.652597402597403e-06,
"loss": 0.3472,
"step": 900
},
{
"epoch": 7.9,
"learning_rate": 3.6728896103896104e-06,
"loss": 0.2981,
"step": 905
},
{
"epoch": 7.95,
"learning_rate": 3.6931818181818186e-06,
"loss": 0.2397,
"step": 910
},
{
"epoch": 7.99,
"learning_rate": 3.7134740259740267e-06,
"loss": 0.2751,
"step": 915
},
{
"epoch": 8.0,
"eval_accuracy": 0.9384236453201971,
"eval_loss": 0.21142302453517914,
"eval_runtime": 47.4205,
"eval_samples_per_second": 25.685,
"eval_steps_per_second": 1.624,
"step": 916
},
{
"epoch": 8.03,
"learning_rate": 3.733766233766234e-06,
"loss": 0.3534,
"step": 920
},
{
"epoch": 8.08,
"learning_rate": 3.7540584415584417e-06,
"loss": 0.2201,
"step": 925
},
{
"epoch": 8.12,
"learning_rate": 3.77435064935065e-06,
"loss": 0.284,
"step": 930
},
{
"epoch": 8.17,
"learning_rate": 3.794642857142857e-06,
"loss": 0.2844,
"step": 935
},
{
"epoch": 8.21,
"learning_rate": 3.814935064935065e-06,
"loss": 0.2889,
"step": 940
},
{
"epoch": 8.25,
"learning_rate": 3.8352272727272735e-06,
"loss": 0.3146,
"step": 945
},
{
"epoch": 8.3,
"learning_rate": 3.85551948051948e-06,
"loss": 0.3071,
"step": 950
},
{
"epoch": 8.34,
"learning_rate": 3.875811688311689e-06,
"loss": 0.3546,
"step": 955
},
{
"epoch": 8.38,
"learning_rate": 3.896103896103897e-06,
"loss": 0.3454,
"step": 960
},
{
"epoch": 8.43,
"learning_rate": 3.916396103896104e-06,
"loss": 0.3196,
"step": 965
},
{
"epoch": 8.47,
"learning_rate": 3.936688311688312e-06,
"loss": 0.276,
"step": 970
},
{
"epoch": 8.52,
"learning_rate": 3.95698051948052e-06,
"loss": 0.2985,
"step": 975
},
{
"epoch": 8.56,
"learning_rate": 3.9772727272727275e-06,
"loss": 0.332,
"step": 980
},
{
"epoch": 8.6,
"learning_rate": 3.997564935064935e-06,
"loss": 0.2749,
"step": 985
},
{
"epoch": 8.65,
"learning_rate": 4.017857142857143e-06,
"loss": 0.2819,
"step": 990
},
{
"epoch": 8.69,
"learning_rate": 4.038149350649351e-06,
"loss": 0.3132,
"step": 995
},
{
"epoch": 8.73,
"learning_rate": 4.0584415584415584e-06,
"loss": 0.3288,
"step": 1000
},
{
"epoch": 8.78,
"learning_rate": 4.078733766233767e-06,
"loss": 0.2985,
"step": 1005
},
{
"epoch": 8.82,
"learning_rate": 4.099025974025974e-06,
"loss": 0.3344,
"step": 1010
},
{
"epoch": 8.86,
"learning_rate": 4.1193181818181825e-06,
"loss": 0.2911,
"step": 1015
},
{
"epoch": 8.91,
"learning_rate": 4.13961038961039e-06,
"loss": 0.2336,
"step": 1020
},
{
"epoch": 8.95,
"learning_rate": 4.159902597402598e-06,
"loss": 0.3129,
"step": 1025
},
{
"epoch": 9.0,
"learning_rate": 4.180194805194806e-06,
"loss": 0.2352,
"step": 1030
},
{
"epoch": 9.0,
"eval_accuracy": 0.9688013136288999,
"eval_loss": 0.13030777871608734,
"eval_runtime": 47.3249,
"eval_samples_per_second": 25.737,
"eval_steps_per_second": 1.627,
"step": 1030
},
{
"epoch": 9.04,
"learning_rate": 4.200487012987013e-06,
"loss": 0.3073,
"step": 1035
},
{
"epoch": 9.08,
"learning_rate": 4.220779220779221e-06,
"loss": 0.2497,
"step": 1040
},
{
"epoch": 9.13,
"learning_rate": 4.241071428571429e-06,
"loss": 0.2582,
"step": 1045
},
{
"epoch": 9.17,
"learning_rate": 4.2613636363636365e-06,
"loss": 0.2629,
"step": 1050
},
{
"epoch": 9.21,
"learning_rate": 4.281655844155844e-06,
"loss": 0.3715,
"step": 1055
},
{
"epoch": 9.26,
"learning_rate": 4.301948051948052e-06,
"loss": 0.2179,
"step": 1060
},
{
"epoch": 9.3,
"learning_rate": 4.3222402597402605e-06,
"loss": 0.2298,
"step": 1065
},
{
"epoch": 9.34,
"learning_rate": 4.342532467532468e-06,
"loss": 0.3508,
"step": 1070
},
{
"epoch": 9.39,
"learning_rate": 4.362824675324675e-06,
"loss": 0.2501,
"step": 1075
},
{
"epoch": 9.43,
"learning_rate": 4.383116883116884e-06,
"loss": 0.2063,
"step": 1080
},
{
"epoch": 9.48,
"learning_rate": 4.4034090909090914e-06,
"loss": 0.1413,
"step": 1085
},
{
"epoch": 9.52,
"learning_rate": 4.423701298701299e-06,
"loss": 0.2101,
"step": 1090
},
{
"epoch": 9.56,
"learning_rate": 4.443993506493507e-06,
"loss": 0.2291,
"step": 1095
},
{
"epoch": 9.61,
"learning_rate": 4.464285714285715e-06,
"loss": 0.2546,
"step": 1100
},
{
"epoch": 9.65,
"learning_rate": 4.484577922077922e-06,
"loss": 0.287,
"step": 1105
},
{
"epoch": 9.69,
"learning_rate": 4.50487012987013e-06,
"loss": 0.2189,
"step": 1110
},
{
"epoch": 9.74,
"learning_rate": 4.525162337662339e-06,
"loss": 0.211,
"step": 1115
},
{
"epoch": 9.78,
"learning_rate": 4.5454545454545455e-06,
"loss": 0.226,
"step": 1120
},
{
"epoch": 9.83,
"learning_rate": 4.565746753246753e-06,
"loss": 0.3594,
"step": 1125
},
{
"epoch": 9.87,
"learning_rate": 4.586038961038962e-06,
"loss": 0.2186,
"step": 1130
},
{
"epoch": 9.91,
"learning_rate": 4.606331168831169e-06,
"loss": 0.2522,
"step": 1135
},
{
"epoch": 9.96,
"learning_rate": 4.626623376623377e-06,
"loss": 0.1542,
"step": 1140
},
{
"epoch": 10.0,
"learning_rate": 4.646915584415585e-06,
"loss": 0.1831,
"step": 1145
},
{
"epoch": 10.0,
"eval_accuracy": 0.9688013136288999,
"eval_loss": 0.11936096101999283,
"eval_runtime": 47.5458,
"eval_samples_per_second": 25.617,
"eval_steps_per_second": 1.619,
"step": 1145
},
{
"epoch": 10.04,
"learning_rate": 4.667207792207793e-06,
"loss": 0.1742,
"step": 1150
},
{
"epoch": 10.09,
"learning_rate": 4.6875000000000004e-06,
"loss": 0.2575,
"step": 1155
},
{
"epoch": 10.13,
"learning_rate": 4.707792207792208e-06,
"loss": 0.1794,
"step": 1160
},
{
"epoch": 10.17,
"learning_rate": 4.728084415584416e-06,
"loss": 0.2451,
"step": 1165
},
{
"epoch": 10.22,
"learning_rate": 4.748376623376624e-06,
"loss": 0.2675,
"step": 1170
},
{
"epoch": 10.26,
"learning_rate": 4.768668831168831e-06,
"loss": 0.2399,
"step": 1175
},
{
"epoch": 10.31,
"learning_rate": 4.788961038961039e-06,
"loss": 0.2345,
"step": 1180
},
{
"epoch": 10.35,
"learning_rate": 4.809253246753247e-06,
"loss": 0.2109,
"step": 1185
},
{
"epoch": 10.39,
"learning_rate": 4.829545454545455e-06,
"loss": 0.2612,
"step": 1190
},
{
"epoch": 10.44,
"learning_rate": 4.849837662337662e-06,
"loss": 0.1786,
"step": 1195
},
{
"epoch": 10.48,
"learning_rate": 4.870129870129871e-06,
"loss": 0.2758,
"step": 1200
},
{
"epoch": 10.52,
"learning_rate": 4.8904220779220785e-06,
"loss": 0.2478,
"step": 1205
},
{
"epoch": 10.57,
"learning_rate": 4.910714285714286e-06,
"loss": 0.2829,
"step": 1210
},
{
"epoch": 10.61,
"learning_rate": 4.931006493506494e-06,
"loss": 0.2255,
"step": 1215
},
{
"epoch": 10.66,
"learning_rate": 4.951298701298702e-06,
"loss": 0.207,
"step": 1220
},
{
"epoch": 10.7,
"learning_rate": 4.9715909090909094e-06,
"loss": 0.2014,
"step": 1225
},
{
"epoch": 10.74,
"learning_rate": 4.991883116883117e-06,
"loss": 0.2433,
"step": 1230
},
{
"epoch": 10.79,
"learning_rate": 4.889705882352942e-06,
"loss": 0.2324,
"step": 1235
},
{
"epoch": 10.83,
"learning_rate": 4.705882352941177e-06,
"loss": 0.2457,
"step": 1240
},
{
"epoch": 10.87,
"learning_rate": 4.522058823529412e-06,
"loss": 0.2169,
"step": 1245
},
{
"epoch": 10.92,
"learning_rate": 4.3382352941176475e-06,
"loss": 0.2155,
"step": 1250
},
{
"epoch": 10.96,
"learning_rate": 4.154411764705883e-06,
"loss": 0.1515,
"step": 1255
},
{
"epoch": 11.0,
"eval_accuracy": 0.986863711001642,
"eval_loss": 0.06730249524116516,
"eval_runtime": 47.4532,
"eval_samples_per_second": 25.667,
"eval_steps_per_second": 1.623,
"step": 1259
},
{
"epoch": 11.0,
"learning_rate": 3.970588235294118e-06,
"loss": 0.1638,
"step": 1260
},
{
"epoch": 11.05,
"learning_rate": 3.786764705882353e-06,
"loss": 0.1321,
"step": 1265
},
{
"epoch": 11.09,
"learning_rate": 3.6029411764705883e-06,
"loss": 0.1586,
"step": 1270
},
{
"epoch": 11.14,
"learning_rate": 3.419117647058824e-06,
"loss": 0.1369,
"step": 1275
},
{
"epoch": 11.18,
"learning_rate": 3.2352941176470594e-06,
"loss": 0.1217,
"step": 1280
},
{
"epoch": 11.22,
"learning_rate": 3.0514705882352947e-06,
"loss": 0.1539,
"step": 1285
},
{
"epoch": 11.27,
"learning_rate": 2.8676470588235296e-06,
"loss": 0.1937,
"step": 1290
},
{
"epoch": 11.31,
"learning_rate": 2.683823529411765e-06,
"loss": 0.1711,
"step": 1295
},
{
"epoch": 11.35,
"learning_rate": 2.5e-06,
"loss": 0.1968,
"step": 1300
},
{
"epoch": 11.4,
"learning_rate": 2.3161764705882355e-06,
"loss": 0.1667,
"step": 1305
},
{
"epoch": 11.44,
"learning_rate": 2.132352941176471e-06,
"loss": 0.1743,
"step": 1310
},
{
"epoch": 11.48,
"learning_rate": 1.948529411764706e-06,
"loss": 0.1271,
"step": 1315
},
{
"epoch": 11.53,
"learning_rate": 1.7647058823529414e-06,
"loss": 0.1085,
"step": 1320
},
{
"epoch": 11.57,
"learning_rate": 1.5808823529411765e-06,
"loss": 0.2339,
"step": 1325
},
{
"epoch": 11.62,
"learning_rate": 1.3970588235294119e-06,
"loss": 0.1406,
"step": 1330
},
{
"epoch": 11.66,
"learning_rate": 1.2132352941176472e-06,
"loss": 0.088,
"step": 1335
},
{
"epoch": 11.7,
"learning_rate": 1.0294117647058825e-06,
"loss": 0.1128,
"step": 1340
},
{
"epoch": 11.75,
"learning_rate": 8.455882352941178e-07,
"loss": 0.131,
"step": 1345
},
{
"epoch": 11.79,
"learning_rate": 6.61764705882353e-07,
"loss": 0.1513,
"step": 1350
},
{
"epoch": 11.83,
"learning_rate": 4.779411764705882e-07,
"loss": 0.1848,
"step": 1355
},
{
"epoch": 11.88,
"learning_rate": 2.9411764705882356e-07,
"loss": 0.1777,
"step": 1360
},
{
"epoch": 11.92,
"learning_rate": 1.1029411764705884e-07,
"loss": 0.204,
"step": 1365
},
{
"epoch": 11.95,
"eval_accuracy": 0.9901477832512315,
"eval_loss": 0.048802148550748825,
"eval_runtime": 47.5685,
"eval_samples_per_second": 25.605,
"eval_steps_per_second": 1.619,
"step": 1368
},
{
"epoch": 11.95,
"step": 1368,
"total_flos": 1.195686793821323e+19,
"train_loss": 0.7552440928088294,
"train_runtime": 8820.1099,
"train_samples_per_second": 4.967,
"train_steps_per_second": 0.155
}
],
"max_steps": 1368,
"num_train_epochs": 12,
"total_flos": 1.195686793821323e+19,
"trial_name": null,
"trial_params": null
}