ALM-AHME's picture
End of training
ec58cc3
{
"best_metric": 0.9893267651888341,
"best_model_checkpoint": "beit-large-patch16-224-finetuned-Lesion-Classification-HAM10000-AH-60-20-20-Shuffled/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": 2.108,
"step": 5
},
{
"epoch": 0.09,
"learning_rate": 4.058441558441558e-08,
"loss": 2.2072,
"step": 10
},
{
"epoch": 0.13,
"learning_rate": 6.087662337662339e-08,
"loss": 2.2095,
"step": 15
},
{
"epoch": 0.17,
"learning_rate": 8.116883116883116e-08,
"loss": 2.285,
"step": 20
},
{
"epoch": 0.22,
"learning_rate": 1.0146103896103897e-07,
"loss": 2.1893,
"step": 25
},
{
"epoch": 0.26,
"learning_rate": 1.2175324675324677e-07,
"loss": 2.2163,
"step": 30
},
{
"epoch": 0.31,
"learning_rate": 1.4204545454545455e-07,
"loss": 2.2329,
"step": 35
},
{
"epoch": 0.35,
"learning_rate": 1.6233766233766232e-07,
"loss": 2.1437,
"step": 40
},
{
"epoch": 0.39,
"learning_rate": 1.8262987012987013e-07,
"loss": 2.2994,
"step": 45
},
{
"epoch": 0.44,
"learning_rate": 2.0292207792207793e-07,
"loss": 2.1974,
"step": 50
},
{
"epoch": 0.48,
"learning_rate": 2.2321428571428574e-07,
"loss": 2.1891,
"step": 55
},
{
"epoch": 0.52,
"learning_rate": 2.4350649350649354e-07,
"loss": 2.3163,
"step": 60
},
{
"epoch": 0.57,
"learning_rate": 2.637987012987013e-07,
"loss": 2.2002,
"step": 65
},
{
"epoch": 0.61,
"learning_rate": 2.840909090909091e-07,
"loss": 2.0097,
"step": 70
},
{
"epoch": 0.66,
"learning_rate": 3.043831168831169e-07,
"loss": 2.0603,
"step": 75
},
{
"epoch": 0.7,
"learning_rate": 3.2467532467532465e-07,
"loss": 1.9911,
"step": 80
},
{
"epoch": 0.74,
"learning_rate": 3.4496753246753253e-07,
"loss": 2.0949,
"step": 85
},
{
"epoch": 0.79,
"learning_rate": 3.6525974025974026e-07,
"loss": 2.0385,
"step": 90
},
{
"epoch": 0.83,
"learning_rate": 3.855519480519481e-07,
"loss": 2.0068,
"step": 95
},
{
"epoch": 0.87,
"learning_rate": 4.0584415584415586e-07,
"loss": 2.1027,
"step": 100
},
{
"epoch": 0.92,
"learning_rate": 4.2613636363636364e-07,
"loss": 1.9149,
"step": 105
},
{
"epoch": 0.96,
"learning_rate": 4.4642857142857147e-07,
"loss": 2.1055,
"step": 110
},
{
"epoch": 1.0,
"eval_accuracy": 0.16009852216748768,
"eval_loss": 2.0091114044189453,
"eval_runtime": 991.3741,
"eval_samples_per_second": 1.229,
"eval_steps_per_second": 0.078,
"step": 114
},
{
"epoch": 1.0,
"learning_rate": 4.6672077922077925e-07,
"loss": 1.89,
"step": 115
},
{
"epoch": 1.05,
"learning_rate": 4.870129870129871e-07,
"loss": 2.0264,
"step": 120
},
{
"epoch": 1.09,
"learning_rate": 5.073051948051948e-07,
"loss": 2.0185,
"step": 125
},
{
"epoch": 1.14,
"learning_rate": 5.275974025974026e-07,
"loss": 1.9248,
"step": 130
},
{
"epoch": 1.18,
"learning_rate": 5.478896103896105e-07,
"loss": 1.8667,
"step": 135
},
{
"epoch": 1.22,
"learning_rate": 5.681818181818182e-07,
"loss": 1.8826,
"step": 140
},
{
"epoch": 1.27,
"learning_rate": 5.88474025974026e-07,
"loss": 1.9451,
"step": 145
},
{
"epoch": 1.31,
"learning_rate": 6.087662337662339e-07,
"loss": 1.8617,
"step": 150
},
{
"epoch": 1.35,
"learning_rate": 6.290584415584417e-07,
"loss": 1.8339,
"step": 155
},
{
"epoch": 1.4,
"learning_rate": 6.493506493506493e-07,
"loss": 1.8251,
"step": 160
},
{
"epoch": 1.44,
"learning_rate": 6.696428571428571e-07,
"loss": 1.8671,
"step": 165
},
{
"epoch": 1.48,
"learning_rate": 6.899350649350651e-07,
"loss": 1.8487,
"step": 170
},
{
"epoch": 1.53,
"learning_rate": 7.102272727272729e-07,
"loss": 1.7534,
"step": 175
},
{
"epoch": 1.57,
"learning_rate": 7.305194805194805e-07,
"loss": 1.7212,
"step": 180
},
{
"epoch": 1.62,
"learning_rate": 7.508116883116883e-07,
"loss": 1.8024,
"step": 185
},
{
"epoch": 1.66,
"learning_rate": 7.711038961038962e-07,
"loss": 1.7943,
"step": 190
},
{
"epoch": 1.7,
"learning_rate": 7.913961038961039e-07,
"loss": 1.6546,
"step": 195
},
{
"epoch": 1.75,
"learning_rate": 8.116883116883117e-07,
"loss": 1.678,
"step": 200
},
{
"epoch": 1.79,
"learning_rate": 8.319805194805196e-07,
"loss": 1.7724,
"step": 205
},
{
"epoch": 1.83,
"learning_rate": 8.522727272727273e-07,
"loss": 1.7413,
"step": 210
},
{
"epoch": 1.88,
"learning_rate": 8.725649350649351e-07,
"loss": 1.6184,
"step": 215
},
{
"epoch": 1.92,
"learning_rate": 8.928571428571429e-07,
"loss": 1.6969,
"step": 220
},
{
"epoch": 1.97,
"learning_rate": 9.131493506493508e-07,
"loss": 1.6582,
"step": 225
},
{
"epoch": 2.0,
"eval_accuracy": 0.4187192118226601,
"eval_loss": 1.5953072309494019,
"eval_runtime": 46.7943,
"eval_samples_per_second": 26.029,
"eval_steps_per_second": 1.645,
"step": 229
},
{
"epoch": 2.01,
"learning_rate": 9.334415584415585e-07,
"loss": 1.6087,
"step": 230
},
{
"epoch": 2.05,
"learning_rate": 9.537337662337662e-07,
"loss": 1.528,
"step": 235
},
{
"epoch": 2.1,
"learning_rate": 9.740259740259742e-07,
"loss": 1.4899,
"step": 240
},
{
"epoch": 2.14,
"learning_rate": 9.943181818181819e-07,
"loss": 1.5472,
"step": 245
},
{
"epoch": 2.18,
"learning_rate": 1.0146103896103896e-06,
"loss": 1.5591,
"step": 250
},
{
"epoch": 2.23,
"learning_rate": 1.0349025974025975e-06,
"loss": 1.4602,
"step": 255
},
{
"epoch": 2.27,
"learning_rate": 1.0551948051948053e-06,
"loss": 1.4764,
"step": 260
},
{
"epoch": 2.31,
"learning_rate": 1.075487012987013e-06,
"loss": 1.4756,
"step": 265
},
{
"epoch": 2.36,
"learning_rate": 1.095779220779221e-06,
"loss": 1.4322,
"step": 270
},
{
"epoch": 2.4,
"learning_rate": 1.1160714285714287e-06,
"loss": 1.4242,
"step": 275
},
{
"epoch": 2.45,
"learning_rate": 1.1363636363636364e-06,
"loss": 1.367,
"step": 280
},
{
"epoch": 2.49,
"learning_rate": 1.1566558441558443e-06,
"loss": 1.3979,
"step": 285
},
{
"epoch": 2.53,
"learning_rate": 1.176948051948052e-06,
"loss": 1.3273,
"step": 290
},
{
"epoch": 2.58,
"learning_rate": 1.1972402597402598e-06,
"loss": 1.2942,
"step": 295
},
{
"epoch": 2.62,
"learning_rate": 1.2175324675324677e-06,
"loss": 1.2446,
"step": 300
},
{
"epoch": 2.66,
"learning_rate": 1.2378246753246754e-06,
"loss": 1.2468,
"step": 305
},
{
"epoch": 2.71,
"learning_rate": 1.2581168831168834e-06,
"loss": 1.2983,
"step": 310
},
{
"epoch": 2.75,
"learning_rate": 1.278409090909091e-06,
"loss": 1.2012,
"step": 315
},
{
"epoch": 2.79,
"learning_rate": 1.2987012987012986e-06,
"loss": 1.1967,
"step": 320
},
{
"epoch": 2.84,
"learning_rate": 1.3189935064935067e-06,
"loss": 1.1737,
"step": 325
},
{
"epoch": 2.88,
"learning_rate": 1.3392857142857143e-06,
"loss": 1.3117,
"step": 330
},
{
"epoch": 2.93,
"learning_rate": 1.359577922077922e-06,
"loss": 1.1569,
"step": 335
},
{
"epoch": 2.97,
"learning_rate": 1.3798701298701301e-06,
"loss": 1.2399,
"step": 340
},
{
"epoch": 3.0,
"eval_accuracy": 0.5977011494252874,
"eval_loss": 1.1053130626678467,
"eval_runtime": 46.8918,
"eval_samples_per_second": 25.975,
"eval_steps_per_second": 1.642,
"step": 343
},
{
"epoch": 3.01,
"learning_rate": 1.4001623376623376e-06,
"loss": 1.0869,
"step": 345
},
{
"epoch": 3.06,
"learning_rate": 1.4204545454545458e-06,
"loss": 1.1094,
"step": 350
},
{
"epoch": 3.1,
"learning_rate": 1.4407467532467533e-06,
"loss": 1.1013,
"step": 355
},
{
"epoch": 3.14,
"learning_rate": 1.461038961038961e-06,
"loss": 1.1792,
"step": 360
},
{
"epoch": 3.19,
"learning_rate": 1.481331168831169e-06,
"loss": 1.0155,
"step": 365
},
{
"epoch": 3.23,
"learning_rate": 1.5016233766233767e-06,
"loss": 1.0912,
"step": 370
},
{
"epoch": 3.28,
"learning_rate": 1.5219155844155844e-06,
"loss": 1.0208,
"step": 375
},
{
"epoch": 3.32,
"learning_rate": 1.5422077922077924e-06,
"loss": 1.0427,
"step": 380
},
{
"epoch": 3.36,
"learning_rate": 1.5625e-06,
"loss": 1.0104,
"step": 385
},
{
"epoch": 3.41,
"learning_rate": 1.5827922077922078e-06,
"loss": 0.9838,
"step": 390
},
{
"epoch": 3.45,
"learning_rate": 1.6030844155844157e-06,
"loss": 0.9334,
"step": 395
},
{
"epoch": 3.49,
"learning_rate": 1.6233766233766235e-06,
"loss": 1.0505,
"step": 400
},
{
"epoch": 3.54,
"learning_rate": 1.6436688311688312e-06,
"loss": 0.8974,
"step": 405
},
{
"epoch": 3.58,
"learning_rate": 1.6639610389610391e-06,
"loss": 0.9257,
"step": 410
},
{
"epoch": 3.62,
"learning_rate": 1.6842532467532468e-06,
"loss": 0.88,
"step": 415
},
{
"epoch": 3.67,
"learning_rate": 1.7045454545454546e-06,
"loss": 0.9421,
"step": 420
},
{
"epoch": 3.71,
"learning_rate": 1.7248376623376625e-06,
"loss": 0.8861,
"step": 425
},
{
"epoch": 3.76,
"learning_rate": 1.7451298701298702e-06,
"loss": 0.7577,
"step": 430
},
{
"epoch": 3.8,
"learning_rate": 1.7654220779220782e-06,
"loss": 0.8573,
"step": 435
},
{
"epoch": 3.84,
"learning_rate": 1.7857142857142859e-06,
"loss": 0.8664,
"step": 440
},
{
"epoch": 3.89,
"learning_rate": 1.8060064935064936e-06,
"loss": 0.859,
"step": 445
},
{
"epoch": 3.93,
"learning_rate": 1.8262987012987016e-06,
"loss": 0.8762,
"step": 450
},
{
"epoch": 3.97,
"learning_rate": 1.8465909090909093e-06,
"loss": 0.8417,
"step": 455
},
{
"epoch": 4.0,
"eval_accuracy": 0.7241379310344828,
"eval_loss": 0.7601869702339172,
"eval_runtime": 46.7277,
"eval_samples_per_second": 26.066,
"eval_steps_per_second": 1.648,
"step": 458
},
{
"epoch": 4.02,
"learning_rate": 1.866883116883117e-06,
"loss": 0.8784,
"step": 460
},
{
"epoch": 4.06,
"learning_rate": 1.887175324675325e-06,
"loss": 0.9804,
"step": 465
},
{
"epoch": 4.1,
"learning_rate": 1.9074675324675324e-06,
"loss": 0.7837,
"step": 470
},
{
"epoch": 4.15,
"learning_rate": 1.92775974025974e-06,
"loss": 0.7528,
"step": 475
},
{
"epoch": 4.19,
"learning_rate": 1.9480519480519483e-06,
"loss": 0.6939,
"step": 480
},
{
"epoch": 4.24,
"learning_rate": 1.968344155844156e-06,
"loss": 0.8287,
"step": 485
},
{
"epoch": 4.28,
"learning_rate": 1.9886363636363638e-06,
"loss": 0.8655,
"step": 490
},
{
"epoch": 4.32,
"learning_rate": 2.0089285714285715e-06,
"loss": 0.7398,
"step": 495
},
{
"epoch": 4.37,
"learning_rate": 2.0292207792207792e-06,
"loss": 0.6669,
"step": 500
},
{
"epoch": 4.41,
"learning_rate": 2.049512987012987e-06,
"loss": 0.7623,
"step": 505
},
{
"epoch": 4.45,
"learning_rate": 2.069805194805195e-06,
"loss": 0.7007,
"step": 510
},
{
"epoch": 4.5,
"learning_rate": 2.090097402597403e-06,
"loss": 0.7341,
"step": 515
},
{
"epoch": 4.54,
"learning_rate": 2.1103896103896105e-06,
"loss": 0.6959,
"step": 520
},
{
"epoch": 4.59,
"learning_rate": 2.1306818181818183e-06,
"loss": 0.7649,
"step": 525
},
{
"epoch": 4.63,
"learning_rate": 2.150974025974026e-06,
"loss": 0.7243,
"step": 530
},
{
"epoch": 4.67,
"learning_rate": 2.171266233766234e-06,
"loss": 0.6517,
"step": 535
},
{
"epoch": 4.72,
"learning_rate": 2.191558441558442e-06,
"loss": 0.735,
"step": 540
},
{
"epoch": 4.76,
"learning_rate": 2.2118506493506496e-06,
"loss": 0.6925,
"step": 545
},
{
"epoch": 4.8,
"learning_rate": 2.2321428571428573e-06,
"loss": 0.6783,
"step": 550
},
{
"epoch": 4.85,
"learning_rate": 2.252435064935065e-06,
"loss": 0.5866,
"step": 555
},
{
"epoch": 4.89,
"learning_rate": 2.2727272727272728e-06,
"loss": 0.5415,
"step": 560
},
{
"epoch": 4.93,
"learning_rate": 2.293019480519481e-06,
"loss": 0.5739,
"step": 565
},
{
"epoch": 4.98,
"learning_rate": 2.3133116883116886e-06,
"loss": 0.5517,
"step": 570
},
{
"epoch": 5.0,
"eval_accuracy": 0.8013136288998358,
"eval_loss": 0.5651180744171143,
"eval_runtime": 46.8697,
"eval_samples_per_second": 25.987,
"eval_steps_per_second": 1.643,
"step": 572
},
{
"epoch": 5.02,
"learning_rate": 2.3336038961038964e-06,
"loss": 0.7433,
"step": 575
},
{
"epoch": 5.07,
"learning_rate": 2.353896103896104e-06,
"loss": 0.5786,
"step": 580
},
{
"epoch": 5.11,
"learning_rate": 2.374188311688312e-06,
"loss": 0.6458,
"step": 585
},
{
"epoch": 5.15,
"learning_rate": 2.3944805194805195e-06,
"loss": 0.6321,
"step": 590
},
{
"epoch": 5.2,
"learning_rate": 2.4147727272727277e-06,
"loss": 0.5119,
"step": 595
},
{
"epoch": 5.24,
"learning_rate": 2.4350649350649354e-06,
"loss": 0.627,
"step": 600
},
{
"epoch": 5.28,
"learning_rate": 2.455357142857143e-06,
"loss": 0.6677,
"step": 605
},
{
"epoch": 5.33,
"learning_rate": 2.475649350649351e-06,
"loss": 0.4946,
"step": 610
},
{
"epoch": 5.37,
"learning_rate": 2.4959415584415586e-06,
"loss": 0.5979,
"step": 615
},
{
"epoch": 5.41,
"learning_rate": 2.5162337662337667e-06,
"loss": 0.5592,
"step": 620
},
{
"epoch": 5.46,
"learning_rate": 2.536525974025974e-06,
"loss": 0.5518,
"step": 625
},
{
"epoch": 5.5,
"learning_rate": 2.556818181818182e-06,
"loss": 0.6418,
"step": 630
},
{
"epoch": 5.55,
"learning_rate": 2.57711038961039e-06,
"loss": 0.7065,
"step": 635
},
{
"epoch": 5.59,
"learning_rate": 2.597402597402597e-06,
"loss": 0.6154,
"step": 640
},
{
"epoch": 5.63,
"learning_rate": 2.6176948051948053e-06,
"loss": 0.6088,
"step": 645
},
{
"epoch": 5.68,
"learning_rate": 2.6379870129870135e-06,
"loss": 0.5844,
"step": 650
},
{
"epoch": 5.72,
"learning_rate": 2.658279220779221e-06,
"loss": 0.5261,
"step": 655
},
{
"epoch": 5.76,
"learning_rate": 2.6785714285714285e-06,
"loss": 0.5826,
"step": 660
},
{
"epoch": 5.81,
"learning_rate": 2.6988636363636367e-06,
"loss": 0.4568,
"step": 665
},
{
"epoch": 5.85,
"learning_rate": 2.719155844155844e-06,
"loss": 0.4887,
"step": 670
},
{
"epoch": 5.9,
"learning_rate": 2.739448051948052e-06,
"loss": 0.4853,
"step": 675
},
{
"epoch": 5.94,
"learning_rate": 2.7597402597402603e-06,
"loss": 0.5716,
"step": 680
},
{
"epoch": 5.98,
"learning_rate": 2.780032467532468e-06,
"loss": 0.5777,
"step": 685
},
{
"epoch": 6.0,
"eval_accuracy": 0.8768472906403941,
"eval_loss": 0.39802247285842896,
"eval_runtime": 46.8639,
"eval_samples_per_second": 25.99,
"eval_steps_per_second": 1.643,
"step": 687
},
{
"epoch": 6.03,
"learning_rate": 2.8003246753246753e-06,
"loss": 0.5484,
"step": 690
},
{
"epoch": 6.07,
"learning_rate": 2.8206168831168834e-06,
"loss": 0.4529,
"step": 695
},
{
"epoch": 6.11,
"learning_rate": 2.8409090909090916e-06,
"loss": 0.4813,
"step": 700
},
{
"epoch": 6.16,
"learning_rate": 2.861201298701299e-06,
"loss": 0.5704,
"step": 705
},
{
"epoch": 6.2,
"learning_rate": 2.8814935064935066e-06,
"loss": 0.5216,
"step": 710
},
{
"epoch": 6.24,
"learning_rate": 2.9017857142857148e-06,
"loss": 0.4718,
"step": 715
},
{
"epoch": 6.29,
"learning_rate": 2.922077922077922e-06,
"loss": 0.4797,
"step": 720
},
{
"epoch": 6.33,
"learning_rate": 2.94237012987013e-06,
"loss": 0.4987,
"step": 725
},
{
"epoch": 6.38,
"learning_rate": 2.962662337662338e-06,
"loss": 0.5066,
"step": 730
},
{
"epoch": 6.42,
"learning_rate": 2.9829545454545457e-06,
"loss": 0.4662,
"step": 735
},
{
"epoch": 6.46,
"learning_rate": 3.0032467532467534e-06,
"loss": 0.4463,
"step": 740
},
{
"epoch": 6.51,
"learning_rate": 3.0235389610389615e-06,
"loss": 0.3972,
"step": 745
},
{
"epoch": 6.55,
"learning_rate": 3.043831168831169e-06,
"loss": 0.4549,
"step": 750
},
{
"epoch": 6.59,
"learning_rate": 3.064123376623377e-06,
"loss": 0.4817,
"step": 755
},
{
"epoch": 6.64,
"learning_rate": 3.0844155844155847e-06,
"loss": 0.4626,
"step": 760
},
{
"epoch": 6.68,
"learning_rate": 3.1047077922077924e-06,
"loss": 0.3278,
"step": 765
},
{
"epoch": 6.72,
"learning_rate": 3.125e-06,
"loss": 0.4929,
"step": 770
},
{
"epoch": 6.77,
"learning_rate": 3.1452922077922083e-06,
"loss": 0.5325,
"step": 775
},
{
"epoch": 6.81,
"learning_rate": 3.1655844155844156e-06,
"loss": 0.3779,
"step": 780
},
{
"epoch": 6.86,
"learning_rate": 3.1858766233766237e-06,
"loss": 0.3715,
"step": 785
},
{
"epoch": 6.9,
"learning_rate": 3.2061688311688315e-06,
"loss": 0.3513,
"step": 790
},
{
"epoch": 6.94,
"learning_rate": 3.226461038961039e-06,
"loss": 0.3366,
"step": 795
},
{
"epoch": 6.99,
"learning_rate": 3.246753246753247e-06,
"loss": 0.408,
"step": 800
},
{
"epoch": 7.0,
"eval_accuracy": 0.9154351395730707,
"eval_loss": 0.2912403345108032,
"eval_runtime": 46.6881,
"eval_samples_per_second": 26.088,
"eval_steps_per_second": 1.649,
"step": 801
},
{
"epoch": 7.03,
"learning_rate": 3.267045454545455e-06,
"loss": 0.2852,
"step": 805
},
{
"epoch": 7.07,
"learning_rate": 3.2873376623376624e-06,
"loss": 0.4248,
"step": 810
},
{
"epoch": 7.12,
"learning_rate": 3.3076298701298705e-06,
"loss": 0.3769,
"step": 815
},
{
"epoch": 7.16,
"learning_rate": 3.3279220779220782e-06,
"loss": 0.4492,
"step": 820
},
{
"epoch": 7.21,
"learning_rate": 3.3482142857142855e-06,
"loss": 0.4322,
"step": 825
},
{
"epoch": 7.25,
"learning_rate": 3.3685064935064937e-06,
"loss": 0.4586,
"step": 830
},
{
"epoch": 7.29,
"learning_rate": 3.388798701298702e-06,
"loss": 0.3588,
"step": 835
},
{
"epoch": 7.34,
"learning_rate": 3.409090909090909e-06,
"loss": 0.3841,
"step": 840
},
{
"epoch": 7.38,
"learning_rate": 3.4293831168831173e-06,
"loss": 0.3817,
"step": 845
},
{
"epoch": 7.42,
"learning_rate": 3.449675324675325e-06,
"loss": 0.326,
"step": 850
},
{
"epoch": 7.47,
"learning_rate": 3.4699675324675323e-06,
"loss": 0.3644,
"step": 855
},
{
"epoch": 7.51,
"learning_rate": 3.4902597402597405e-06,
"loss": 0.3328,
"step": 860
},
{
"epoch": 7.55,
"learning_rate": 3.5105519480519486e-06,
"loss": 0.4362,
"step": 865
},
{
"epoch": 7.6,
"learning_rate": 3.5308441558441563e-06,
"loss": 0.3732,
"step": 870
},
{
"epoch": 7.64,
"learning_rate": 3.5511363636363636e-06,
"loss": 0.3732,
"step": 875
},
{
"epoch": 7.69,
"learning_rate": 3.5714285714285718e-06,
"loss": 0.3757,
"step": 880
},
{
"epoch": 7.73,
"learning_rate": 3.59172077922078e-06,
"loss": 0.3383,
"step": 885
},
{
"epoch": 7.77,
"learning_rate": 3.6120129870129872e-06,
"loss": 0.3281,
"step": 890
},
{
"epoch": 7.82,
"learning_rate": 3.6323051948051954e-06,
"loss": 0.2599,
"step": 895
},
{
"epoch": 7.86,
"learning_rate": 3.652597402597403e-06,
"loss": 0.3384,
"step": 900
},
{
"epoch": 7.9,
"learning_rate": 3.6728896103896104e-06,
"loss": 0.3587,
"step": 905
},
{
"epoch": 7.95,
"learning_rate": 3.6931818181818186e-06,
"loss": 0.3419,
"step": 910
},
{
"epoch": 7.99,
"learning_rate": 3.7134740259740267e-06,
"loss": 0.2395,
"step": 915
},
{
"epoch": 8.0,
"eval_accuracy": 0.9417077175697866,
"eval_loss": 0.21853221952915192,
"eval_runtime": 46.8025,
"eval_samples_per_second": 26.024,
"eval_steps_per_second": 1.645,
"step": 916
},
{
"epoch": 8.03,
"learning_rate": 3.733766233766234e-06,
"loss": 0.3855,
"step": 920
},
{
"epoch": 8.08,
"learning_rate": 3.7540584415584417e-06,
"loss": 0.2681,
"step": 925
},
{
"epoch": 8.12,
"learning_rate": 3.77435064935065e-06,
"loss": 0.2996,
"step": 930
},
{
"epoch": 8.17,
"learning_rate": 3.794642857142857e-06,
"loss": 0.2894,
"step": 935
},
{
"epoch": 8.21,
"learning_rate": 3.814935064935065e-06,
"loss": 0.2479,
"step": 940
},
{
"epoch": 8.25,
"learning_rate": 3.8352272727272735e-06,
"loss": 0.3879,
"step": 945
},
{
"epoch": 8.3,
"learning_rate": 3.85551948051948e-06,
"loss": 0.218,
"step": 950
},
{
"epoch": 8.34,
"learning_rate": 3.875811688311689e-06,
"loss": 0.283,
"step": 955
},
{
"epoch": 8.38,
"learning_rate": 3.896103896103897e-06,
"loss": 0.3067,
"step": 960
},
{
"epoch": 8.43,
"learning_rate": 3.916396103896104e-06,
"loss": 0.2708,
"step": 965
},
{
"epoch": 8.47,
"learning_rate": 3.936688311688312e-06,
"loss": 0.2923,
"step": 970
},
{
"epoch": 8.52,
"learning_rate": 3.95698051948052e-06,
"loss": 0.352,
"step": 975
},
{
"epoch": 8.56,
"learning_rate": 3.9772727272727275e-06,
"loss": 0.3366,
"step": 980
},
{
"epoch": 8.6,
"learning_rate": 3.997564935064935e-06,
"loss": 0.3538,
"step": 985
},
{
"epoch": 8.65,
"learning_rate": 4.017857142857143e-06,
"loss": 0.3784,
"step": 990
},
{
"epoch": 8.69,
"learning_rate": 4.038149350649351e-06,
"loss": 0.2374,
"step": 995
},
{
"epoch": 8.73,
"learning_rate": 4.0584415584415584e-06,
"loss": 0.2373,
"step": 1000
},
{
"epoch": 8.78,
"learning_rate": 4.078733766233767e-06,
"loss": 0.2347,
"step": 1005
},
{
"epoch": 8.82,
"learning_rate": 4.099025974025974e-06,
"loss": 0.2406,
"step": 1010
},
{
"epoch": 8.86,
"learning_rate": 4.1193181818181825e-06,
"loss": 0.2949,
"step": 1015
},
{
"epoch": 8.91,
"learning_rate": 4.13961038961039e-06,
"loss": 0.2925,
"step": 1020
},
{
"epoch": 8.95,
"learning_rate": 4.159902597402598e-06,
"loss": 0.343,
"step": 1025
},
{
"epoch": 9.0,
"learning_rate": 4.180194805194806e-06,
"loss": 0.3613,
"step": 1030
},
{
"epoch": 9.0,
"eval_accuracy": 0.9474548440065681,
"eval_loss": 0.17530561983585358,
"eval_runtime": 46.7366,
"eval_samples_per_second": 26.061,
"eval_steps_per_second": 1.648,
"step": 1030
},
{
"epoch": 9.04,
"learning_rate": 4.200487012987013e-06,
"loss": 0.3092,
"step": 1035
},
{
"epoch": 9.08,
"learning_rate": 4.220779220779221e-06,
"loss": 0.2786,
"step": 1040
},
{
"epoch": 9.13,
"learning_rate": 4.241071428571429e-06,
"loss": 0.2372,
"step": 1045
},
{
"epoch": 9.17,
"learning_rate": 4.2613636363636365e-06,
"loss": 0.2254,
"step": 1050
},
{
"epoch": 9.21,
"learning_rate": 4.281655844155844e-06,
"loss": 0.3296,
"step": 1055
},
{
"epoch": 9.26,
"learning_rate": 4.301948051948052e-06,
"loss": 0.2538,
"step": 1060
},
{
"epoch": 9.3,
"learning_rate": 4.3222402597402605e-06,
"loss": 0.2594,
"step": 1065
},
{
"epoch": 9.34,
"learning_rate": 4.342532467532468e-06,
"loss": 0.2818,
"step": 1070
},
{
"epoch": 9.39,
"learning_rate": 4.362824675324675e-06,
"loss": 0.225,
"step": 1075
},
{
"epoch": 9.43,
"learning_rate": 4.383116883116884e-06,
"loss": 0.1928,
"step": 1080
},
{
"epoch": 9.48,
"learning_rate": 4.4034090909090914e-06,
"loss": 0.1809,
"step": 1085
},
{
"epoch": 9.52,
"learning_rate": 4.423701298701299e-06,
"loss": 0.289,
"step": 1090
},
{
"epoch": 9.56,
"learning_rate": 4.443993506493507e-06,
"loss": 0.2408,
"step": 1095
},
{
"epoch": 9.61,
"learning_rate": 4.464285714285715e-06,
"loss": 0.2477,
"step": 1100
},
{
"epoch": 9.65,
"learning_rate": 4.484577922077922e-06,
"loss": 0.2281,
"step": 1105
},
{
"epoch": 9.69,
"learning_rate": 4.50487012987013e-06,
"loss": 0.2231,
"step": 1110
},
{
"epoch": 9.74,
"learning_rate": 4.525162337662339e-06,
"loss": 0.1817,
"step": 1115
},
{
"epoch": 9.78,
"learning_rate": 4.5454545454545455e-06,
"loss": 0.2242,
"step": 1120
},
{
"epoch": 9.83,
"learning_rate": 4.565746753246753e-06,
"loss": 0.314,
"step": 1125
},
{
"epoch": 9.87,
"learning_rate": 4.586038961038962e-06,
"loss": 0.1974,
"step": 1130
},
{
"epoch": 9.91,
"learning_rate": 4.606331168831169e-06,
"loss": 0.2517,
"step": 1135
},
{
"epoch": 9.96,
"learning_rate": 4.626623376623377e-06,
"loss": 0.1563,
"step": 1140
},
{
"epoch": 10.0,
"learning_rate": 4.646915584415585e-06,
"loss": 0.2408,
"step": 1145
},
{
"epoch": 10.0,
"eval_accuracy": 0.9614121510673235,
"eval_loss": 0.13528668880462646,
"eval_runtime": 46.7513,
"eval_samples_per_second": 26.053,
"eval_steps_per_second": 1.647,
"step": 1145
},
{
"epoch": 10.04,
"learning_rate": 4.667207792207793e-06,
"loss": 0.2764,
"step": 1150
},
{
"epoch": 10.09,
"learning_rate": 4.6875000000000004e-06,
"loss": 0.2571,
"step": 1155
},
{
"epoch": 10.13,
"learning_rate": 4.707792207792208e-06,
"loss": 0.153,
"step": 1160
},
{
"epoch": 10.17,
"learning_rate": 4.728084415584416e-06,
"loss": 0.2705,
"step": 1165
},
{
"epoch": 10.22,
"learning_rate": 4.748376623376624e-06,
"loss": 0.2665,
"step": 1170
},
{
"epoch": 10.26,
"learning_rate": 4.768668831168831e-06,
"loss": 0.2902,
"step": 1175
},
{
"epoch": 10.31,
"learning_rate": 4.788961038961039e-06,
"loss": 0.2662,
"step": 1180
},
{
"epoch": 10.35,
"learning_rate": 4.809253246753247e-06,
"loss": 0.2087,
"step": 1185
},
{
"epoch": 10.39,
"learning_rate": 4.829545454545455e-06,
"loss": 0.2877,
"step": 1190
},
{
"epoch": 10.44,
"learning_rate": 4.849837662337662e-06,
"loss": 0.1659,
"step": 1195
},
{
"epoch": 10.48,
"learning_rate": 4.870129870129871e-06,
"loss": 0.1714,
"step": 1200
},
{
"epoch": 10.52,
"learning_rate": 4.8904220779220785e-06,
"loss": 0.2592,
"step": 1205
},
{
"epoch": 10.57,
"learning_rate": 4.910714285714286e-06,
"loss": 0.2289,
"step": 1210
},
{
"epoch": 10.61,
"learning_rate": 4.931006493506494e-06,
"loss": 0.2291,
"step": 1215
},
{
"epoch": 10.66,
"learning_rate": 4.951298701298702e-06,
"loss": 0.1795,
"step": 1220
},
{
"epoch": 10.7,
"learning_rate": 4.9715909090909094e-06,
"loss": 0.2915,
"step": 1225
},
{
"epoch": 10.74,
"learning_rate": 4.991883116883117e-06,
"loss": 0.1961,
"step": 1230
},
{
"epoch": 10.79,
"learning_rate": 4.889705882352942e-06,
"loss": 0.2525,
"step": 1235
},
{
"epoch": 10.83,
"learning_rate": 4.705882352941177e-06,
"loss": 0.1203,
"step": 1240
},
{
"epoch": 10.87,
"learning_rate": 4.522058823529412e-06,
"loss": 0.2102,
"step": 1245
},
{
"epoch": 10.92,
"learning_rate": 4.3382352941176475e-06,
"loss": 0.2728,
"step": 1250
},
{
"epoch": 10.96,
"learning_rate": 4.154411764705883e-06,
"loss": 0.2777,
"step": 1255
},
{
"epoch": 11.0,
"eval_accuracy": 0.9860426929392446,
"eval_loss": 0.0699055939912796,
"eval_runtime": 46.7733,
"eval_samples_per_second": 26.04,
"eval_steps_per_second": 1.646,
"step": 1259
},
{
"epoch": 11.0,
"learning_rate": 3.970588235294118e-06,
"loss": 0.1461,
"step": 1260
},
{
"epoch": 11.05,
"learning_rate": 3.786764705882353e-06,
"loss": 0.2089,
"step": 1265
},
{
"epoch": 11.09,
"learning_rate": 3.6029411764705883e-06,
"loss": 0.1504,
"step": 1270
},
{
"epoch": 11.14,
"learning_rate": 3.419117647058824e-06,
"loss": 0.1899,
"step": 1275
},
{
"epoch": 11.18,
"learning_rate": 3.2352941176470594e-06,
"loss": 0.1597,
"step": 1280
},
{
"epoch": 11.22,
"learning_rate": 3.0514705882352947e-06,
"loss": 0.1332,
"step": 1285
},
{
"epoch": 11.27,
"learning_rate": 2.8676470588235296e-06,
"loss": 0.1553,
"step": 1290
},
{
"epoch": 11.31,
"learning_rate": 2.683823529411765e-06,
"loss": 0.1826,
"step": 1295
},
{
"epoch": 11.35,
"learning_rate": 2.5e-06,
"loss": 0.2113,
"step": 1300
},
{
"epoch": 11.4,
"learning_rate": 2.3161764705882355e-06,
"loss": 0.1745,
"step": 1305
},
{
"epoch": 11.44,
"learning_rate": 2.132352941176471e-06,
"loss": 0.2032,
"step": 1310
},
{
"epoch": 11.48,
"learning_rate": 1.948529411764706e-06,
"loss": 0.1668,
"step": 1315
},
{
"epoch": 11.53,
"learning_rate": 1.7647058823529414e-06,
"loss": 0.0964,
"step": 1320
},
{
"epoch": 11.57,
"learning_rate": 1.5808823529411765e-06,
"loss": 0.0863,
"step": 1325
},
{
"epoch": 11.62,
"learning_rate": 1.3970588235294119e-06,
"loss": 0.1895,
"step": 1330
},
{
"epoch": 11.66,
"learning_rate": 1.2132352941176472e-06,
"loss": 0.0883,
"step": 1335
},
{
"epoch": 11.7,
"learning_rate": 1.0294117647058825e-06,
"loss": 0.1567,
"step": 1340
},
{
"epoch": 11.75,
"learning_rate": 8.455882352941178e-07,
"loss": 0.1907,
"step": 1345
},
{
"epoch": 11.79,
"learning_rate": 6.61764705882353e-07,
"loss": 0.0819,
"step": 1350
},
{
"epoch": 11.83,
"learning_rate": 4.779411764705882e-07,
"loss": 0.16,
"step": 1355
},
{
"epoch": 11.88,
"learning_rate": 2.9411764705882356e-07,
"loss": 0.1602,
"step": 1360
},
{
"epoch": 11.92,
"learning_rate": 1.1029411764705884e-07,
"loss": 0.1528,
"step": 1365
},
{
"epoch": 11.95,
"eval_accuracy": 0.9893267651888341,
"eval_loss": 0.048675429075956345,
"eval_runtime": 47.0491,
"eval_samples_per_second": 25.888,
"eval_steps_per_second": 1.637,
"step": 1368
},
{
"epoch": 11.95,
"step": 1368,
"total_flos": 1.195686793821323e+19,
"train_loss": 0.7763721961201283,
"train_runtime": 9836.7757,
"train_samples_per_second": 4.454,
"train_steps_per_second": 0.139
}
],
"max_steps": 1368,
"num_train_epochs": 12,
"total_flos": 1.195686793821323e+19,
"trial_name": null,
"trial_params": null
}