ALM-AHME's picture
End of training
bec82ee
{
"best_metric": 0.9893267651888341,
"best_model_checkpoint": "convnextv2-large-1k-224-finetuned-Lesion-Classification-HAM10000-AH-60-20-20-Shuffled-3rd/checkpoint-1710",
"epoch": 14.934497816593886,
"eval_steps": 500,
"global_step": 1710,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 1.6244314489928526e-07,
"loss": 1.9525,
"step": 5
},
{
"epoch": 0.09,
"learning_rate": 3.248862897985705e-07,
"loss": 1.9512,
"step": 10
},
{
"epoch": 0.13,
"learning_rate": 4.873294346978557e-07,
"loss": 1.9512,
"step": 15
},
{
"epoch": 0.17,
"learning_rate": 6.49772579597141e-07,
"loss": 1.9484,
"step": 20
},
{
"epoch": 0.22,
"learning_rate": 8.122157244964262e-07,
"loss": 1.9582,
"step": 25
},
{
"epoch": 0.26,
"learning_rate": 9.746588693957115e-07,
"loss": 1.9481,
"step": 30
},
{
"epoch": 0.31,
"learning_rate": 1.1371020142949968e-06,
"loss": 1.9428,
"step": 35
},
{
"epoch": 0.35,
"learning_rate": 1.299545159194282e-06,
"loss": 1.9426,
"step": 40
},
{
"epoch": 0.39,
"learning_rate": 1.4619883040935671e-06,
"loss": 1.949,
"step": 45
},
{
"epoch": 0.44,
"learning_rate": 1.6244314489928524e-06,
"loss": 1.9452,
"step": 50
},
{
"epoch": 0.48,
"learning_rate": 1.7868745938921377e-06,
"loss": 1.9423,
"step": 55
},
{
"epoch": 0.52,
"learning_rate": 1.949317738791423e-06,
"loss": 1.9445,
"step": 60
},
{
"epoch": 0.57,
"learning_rate": 2.1117608836907083e-06,
"loss": 1.9363,
"step": 65
},
{
"epoch": 0.61,
"learning_rate": 2.2742040285899936e-06,
"loss": 1.9339,
"step": 70
},
{
"epoch": 0.66,
"learning_rate": 2.436647173489279e-06,
"loss": 1.9354,
"step": 75
},
{
"epoch": 0.7,
"learning_rate": 2.599090318388564e-06,
"loss": 1.9331,
"step": 80
},
{
"epoch": 0.74,
"learning_rate": 2.7615334632878494e-06,
"loss": 1.9369,
"step": 85
},
{
"epoch": 0.79,
"learning_rate": 2.9239766081871343e-06,
"loss": 1.9274,
"step": 90
},
{
"epoch": 0.83,
"learning_rate": 3.0864197530864196e-06,
"loss": 1.9253,
"step": 95
},
{
"epoch": 0.87,
"learning_rate": 3.248862897985705e-06,
"loss": 1.915,
"step": 100
},
{
"epoch": 0.92,
"learning_rate": 3.41130604288499e-06,
"loss": 1.927,
"step": 105
},
{
"epoch": 0.96,
"learning_rate": 3.5737491877842754e-06,
"loss": 1.9123,
"step": 110
},
{
"epoch": 1.0,
"eval_accuracy": 0.22413793103448276,
"eval_loss": 1.914461374282837,
"eval_runtime": 377.1313,
"eval_samples_per_second": 3.23,
"eval_steps_per_second": 0.204,
"step": 114
},
{
"epoch": 1.0,
"learning_rate": 3.7361923326835607e-06,
"loss": 1.9076,
"step": 115
},
{
"epoch": 1.05,
"learning_rate": 3.898635477582846e-06,
"loss": 1.9081,
"step": 120
},
{
"epoch": 1.09,
"learning_rate": 4.061078622482131e-06,
"loss": 1.8984,
"step": 125
},
{
"epoch": 1.14,
"learning_rate": 4.2235217673814166e-06,
"loss": 1.8989,
"step": 130
},
{
"epoch": 1.18,
"learning_rate": 4.3859649122807014e-06,
"loss": 1.8838,
"step": 135
},
{
"epoch": 1.22,
"learning_rate": 4.548408057179987e-06,
"loss": 1.879,
"step": 140
},
{
"epoch": 1.27,
"learning_rate": 4.710851202079272e-06,
"loss": 1.885,
"step": 145
},
{
"epoch": 1.31,
"learning_rate": 4.873294346978558e-06,
"loss": 1.8768,
"step": 150
},
{
"epoch": 1.35,
"learning_rate": 5.0357374918778426e-06,
"loss": 1.8656,
"step": 155
},
{
"epoch": 1.4,
"learning_rate": 5.198180636777128e-06,
"loss": 1.859,
"step": 160
},
{
"epoch": 1.44,
"learning_rate": 5.360623781676413e-06,
"loss": 1.8556,
"step": 165
},
{
"epoch": 1.48,
"learning_rate": 5.523066926575699e-06,
"loss": 1.8514,
"step": 170
},
{
"epoch": 1.53,
"learning_rate": 5.685510071474984e-06,
"loss": 1.842,
"step": 175
},
{
"epoch": 1.57,
"learning_rate": 5.8479532163742686e-06,
"loss": 1.8187,
"step": 180
},
{
"epoch": 1.62,
"learning_rate": 6.010396361273554e-06,
"loss": 1.8254,
"step": 185
},
{
"epoch": 1.66,
"learning_rate": 6.172839506172839e-06,
"loss": 1.8086,
"step": 190
},
{
"epoch": 1.7,
"learning_rate": 6.335282651072125e-06,
"loss": 1.7918,
"step": 195
},
{
"epoch": 1.75,
"learning_rate": 6.49772579597141e-06,
"loss": 1.7671,
"step": 200
},
{
"epoch": 1.79,
"learning_rate": 6.660168940870695e-06,
"loss": 1.7675,
"step": 205
},
{
"epoch": 1.83,
"learning_rate": 6.82261208576998e-06,
"loss": 1.7677,
"step": 210
},
{
"epoch": 1.88,
"learning_rate": 6.985055230669266e-06,
"loss": 1.7123,
"step": 215
},
{
"epoch": 1.92,
"learning_rate": 7.147498375568551e-06,
"loss": 1.7342,
"step": 220
},
{
"epoch": 1.97,
"learning_rate": 7.3099415204678366e-06,
"loss": 1.7059,
"step": 225
},
{
"epoch": 2.0,
"eval_accuracy": 0.5599343185550082,
"eval_loss": 1.6894936561584473,
"eval_runtime": 14.2791,
"eval_samples_per_second": 85.299,
"eval_steps_per_second": 5.392,
"step": 229
},
{
"epoch": 2.01,
"learning_rate": 7.4723846653671214e-06,
"loss": 1.671,
"step": 230
},
{
"epoch": 2.05,
"learning_rate": 7.634827810266407e-06,
"loss": 1.6611,
"step": 235
},
{
"epoch": 2.1,
"learning_rate": 7.797270955165692e-06,
"loss": 1.6648,
"step": 240
},
{
"epoch": 2.14,
"learning_rate": 7.959714100064977e-06,
"loss": 1.6345,
"step": 245
},
{
"epoch": 2.18,
"learning_rate": 8.122157244964262e-06,
"loss": 1.6003,
"step": 250
},
{
"epoch": 2.23,
"learning_rate": 8.284600389863548e-06,
"loss": 1.6394,
"step": 255
},
{
"epoch": 2.27,
"learning_rate": 8.447043534762833e-06,
"loss": 1.6096,
"step": 260
},
{
"epoch": 2.31,
"learning_rate": 8.609486679662118e-06,
"loss": 1.5623,
"step": 265
},
{
"epoch": 2.36,
"learning_rate": 8.771929824561403e-06,
"loss": 1.5776,
"step": 270
},
{
"epoch": 2.4,
"learning_rate": 8.93437296946069e-06,
"loss": 1.5345,
"step": 275
},
{
"epoch": 2.45,
"learning_rate": 9.096816114359974e-06,
"loss": 1.5747,
"step": 280
},
{
"epoch": 2.49,
"learning_rate": 9.259259259259259e-06,
"loss": 1.5132,
"step": 285
},
{
"epoch": 2.53,
"learning_rate": 9.421702404158544e-06,
"loss": 1.4826,
"step": 290
},
{
"epoch": 2.58,
"learning_rate": 9.584145549057829e-06,
"loss": 1.5075,
"step": 295
},
{
"epoch": 2.62,
"learning_rate": 9.746588693957115e-06,
"loss": 1.4527,
"step": 300
},
{
"epoch": 2.66,
"learning_rate": 9.9090318388564e-06,
"loss": 1.4805,
"step": 305
},
{
"epoch": 2.71,
"learning_rate": 1.0071474983755685e-05,
"loss": 1.4585,
"step": 310
},
{
"epoch": 2.75,
"learning_rate": 1.023391812865497e-05,
"loss": 1.4536,
"step": 315
},
{
"epoch": 2.79,
"learning_rate": 1.0396361273554257e-05,
"loss": 1.3963,
"step": 320
},
{
"epoch": 2.84,
"learning_rate": 1.0558804418453541e-05,
"loss": 1.405,
"step": 325
},
{
"epoch": 2.88,
"learning_rate": 1.0721247563352826e-05,
"loss": 1.4295,
"step": 330
},
{
"epoch": 2.93,
"learning_rate": 1.0883690708252111e-05,
"loss": 1.3754,
"step": 335
},
{
"epoch": 2.97,
"learning_rate": 1.1046133853151398e-05,
"loss": 1.3919,
"step": 340
},
{
"epoch": 3.0,
"eval_accuracy": 0.6633825944170771,
"eval_loss": 1.319370985031128,
"eval_runtime": 13.9747,
"eval_samples_per_second": 87.158,
"eval_steps_per_second": 5.51,
"step": 343
},
{
"epoch": 3.01,
"learning_rate": 1.1208576998050683e-05,
"loss": 1.3057,
"step": 345
},
{
"epoch": 3.06,
"learning_rate": 1.1371020142949967e-05,
"loss": 1.2981,
"step": 350
},
{
"epoch": 3.1,
"learning_rate": 1.1533463287849252e-05,
"loss": 1.3214,
"step": 355
},
{
"epoch": 3.14,
"learning_rate": 1.1695906432748537e-05,
"loss": 1.3028,
"step": 360
},
{
"epoch": 3.19,
"learning_rate": 1.1858349577647824e-05,
"loss": 1.2512,
"step": 365
},
{
"epoch": 3.23,
"learning_rate": 1.2020792722547109e-05,
"loss": 1.2558,
"step": 370
},
{
"epoch": 3.28,
"learning_rate": 1.2183235867446393e-05,
"loss": 1.2195,
"step": 375
},
{
"epoch": 3.32,
"learning_rate": 1.2345679012345678e-05,
"loss": 1.2963,
"step": 380
},
{
"epoch": 3.36,
"learning_rate": 1.2508122157244965e-05,
"loss": 1.2549,
"step": 385
},
{
"epoch": 3.41,
"learning_rate": 1.267056530214425e-05,
"loss": 1.2753,
"step": 390
},
{
"epoch": 3.45,
"learning_rate": 1.2833008447043535e-05,
"loss": 1.2366,
"step": 395
},
{
"epoch": 3.49,
"learning_rate": 1.299545159194282e-05,
"loss": 1.1684,
"step": 400
},
{
"epoch": 3.54,
"learning_rate": 1.3157894736842106e-05,
"loss": 1.1474,
"step": 405
},
{
"epoch": 3.58,
"learning_rate": 1.332033788174139e-05,
"loss": 1.2366,
"step": 410
},
{
"epoch": 3.62,
"learning_rate": 1.3482781026640676e-05,
"loss": 1.1156,
"step": 415
},
{
"epoch": 3.67,
"learning_rate": 1.364522417153996e-05,
"loss": 1.0825,
"step": 420
},
{
"epoch": 3.71,
"learning_rate": 1.3807667316439245e-05,
"loss": 1.1678,
"step": 425
},
{
"epoch": 3.76,
"learning_rate": 1.3970110461338532e-05,
"loss": 1.1521,
"step": 430
},
{
"epoch": 3.8,
"learning_rate": 1.4132553606237817e-05,
"loss": 1.0757,
"step": 435
},
{
"epoch": 3.84,
"learning_rate": 1.4294996751137102e-05,
"loss": 1.1882,
"step": 440
},
{
"epoch": 3.89,
"learning_rate": 1.4457439896036387e-05,
"loss": 1.0494,
"step": 445
},
{
"epoch": 3.93,
"learning_rate": 1.4619883040935673e-05,
"loss": 1.0706,
"step": 450
},
{
"epoch": 3.97,
"learning_rate": 1.4782326185834958e-05,
"loss": 1.0811,
"step": 455
},
{
"epoch": 4.0,
"eval_accuracy": 0.7151067323481116,
"eval_loss": 1.030263066291809,
"eval_runtime": 14.185,
"eval_samples_per_second": 85.865,
"eval_steps_per_second": 5.428,
"step": 458
},
{
"epoch": 4.02,
"learning_rate": 1.4944769330734243e-05,
"loss": 1.087,
"step": 460
},
{
"epoch": 4.06,
"learning_rate": 1.5107212475633528e-05,
"loss": 1.048,
"step": 465
},
{
"epoch": 4.1,
"learning_rate": 1.5269655620532814e-05,
"loss": 0.9958,
"step": 470
},
{
"epoch": 4.15,
"learning_rate": 1.54320987654321e-05,
"loss": 0.9875,
"step": 475
},
{
"epoch": 4.19,
"learning_rate": 1.5594541910331384e-05,
"loss": 0.974,
"step": 480
},
{
"epoch": 4.24,
"learning_rate": 1.575698505523067e-05,
"loss": 1.0003,
"step": 485
},
{
"epoch": 4.28,
"learning_rate": 1.5919428200129954e-05,
"loss": 1.1908,
"step": 490
},
{
"epoch": 4.32,
"learning_rate": 1.608187134502924e-05,
"loss": 0.9534,
"step": 495
},
{
"epoch": 4.37,
"learning_rate": 1.6244314489928523e-05,
"loss": 0.9967,
"step": 500
},
{
"epoch": 4.41,
"learning_rate": 1.640675763482781e-05,
"loss": 0.9369,
"step": 505
},
{
"epoch": 4.45,
"learning_rate": 1.6569200779727097e-05,
"loss": 0.923,
"step": 510
},
{
"epoch": 4.5,
"learning_rate": 1.673164392462638e-05,
"loss": 0.9534,
"step": 515
},
{
"epoch": 4.54,
"learning_rate": 1.6894087069525666e-05,
"loss": 0.8914,
"step": 520
},
{
"epoch": 4.59,
"learning_rate": 1.705653021442495e-05,
"loss": 0.9275,
"step": 525
},
{
"epoch": 4.63,
"learning_rate": 1.7218973359324236e-05,
"loss": 0.9442,
"step": 530
},
{
"epoch": 4.67,
"learning_rate": 1.738141650422352e-05,
"loss": 0.9798,
"step": 535
},
{
"epoch": 4.72,
"learning_rate": 1.7543859649122806e-05,
"loss": 0.9844,
"step": 540
},
{
"epoch": 4.76,
"learning_rate": 1.770630279402209e-05,
"loss": 0.9517,
"step": 545
},
{
"epoch": 4.8,
"learning_rate": 1.786874593892138e-05,
"loss": 0.8746,
"step": 550
},
{
"epoch": 4.85,
"learning_rate": 1.8031189083820664e-05,
"loss": 0.9737,
"step": 555
},
{
"epoch": 4.89,
"learning_rate": 1.819363222871995e-05,
"loss": 1.0315,
"step": 560
},
{
"epoch": 4.93,
"learning_rate": 1.8356075373619233e-05,
"loss": 0.9309,
"step": 565
},
{
"epoch": 4.98,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.9301,
"step": 570
},
{
"epoch": 5.0,
"eval_accuracy": 0.7561576354679803,
"eval_loss": 0.8376412987709045,
"eval_runtime": 14.173,
"eval_samples_per_second": 85.938,
"eval_steps_per_second": 5.433,
"step": 572
},
{
"epoch": 5.02,
"learning_rate": 1.8680961663417803e-05,
"loss": 0.9062,
"step": 575
},
{
"epoch": 5.07,
"learning_rate": 1.8843404808317088e-05,
"loss": 0.9207,
"step": 580
},
{
"epoch": 5.11,
"learning_rate": 1.9005847953216373e-05,
"loss": 0.9726,
"step": 585
},
{
"epoch": 5.15,
"learning_rate": 1.9168291098115658e-05,
"loss": 1.001,
"step": 590
},
{
"epoch": 5.2,
"learning_rate": 1.9330734243014946e-05,
"loss": 0.9216,
"step": 595
},
{
"epoch": 5.24,
"learning_rate": 1.949317738791423e-05,
"loss": 0.8762,
"step": 600
},
{
"epoch": 5.28,
"learning_rate": 1.9655620532813516e-05,
"loss": 0.9166,
"step": 605
},
{
"epoch": 5.33,
"learning_rate": 1.98180636777128e-05,
"loss": 0.8569,
"step": 610
},
{
"epoch": 5.37,
"learning_rate": 1.9980506822612085e-05,
"loss": 0.7795,
"step": 615
},
{
"epoch": 5.41,
"learning_rate": 2.014294996751137e-05,
"loss": 0.826,
"step": 620
},
{
"epoch": 5.46,
"learning_rate": 2.0305393112410655e-05,
"loss": 0.8161,
"step": 625
},
{
"epoch": 5.5,
"learning_rate": 2.046783625730994e-05,
"loss": 0.7988,
"step": 630
},
{
"epoch": 5.55,
"learning_rate": 2.0630279402209228e-05,
"loss": 0.8475,
"step": 635
},
{
"epoch": 5.59,
"learning_rate": 2.0792722547108513e-05,
"loss": 0.7824,
"step": 640
},
{
"epoch": 5.63,
"learning_rate": 2.0955165692007798e-05,
"loss": 0.7193,
"step": 645
},
{
"epoch": 5.68,
"learning_rate": 2.1117608836907083e-05,
"loss": 0.828,
"step": 650
},
{
"epoch": 5.72,
"learning_rate": 2.1280051981806368e-05,
"loss": 0.7638,
"step": 655
},
{
"epoch": 5.76,
"learning_rate": 2.1442495126705653e-05,
"loss": 0.8164,
"step": 660
},
{
"epoch": 5.81,
"learning_rate": 2.1604938271604937e-05,
"loss": 0.6875,
"step": 665
},
{
"epoch": 5.85,
"learning_rate": 2.1767381416504222e-05,
"loss": 0.6429,
"step": 670
},
{
"epoch": 5.9,
"learning_rate": 2.1929824561403507e-05,
"loss": 0.6859,
"step": 675
},
{
"epoch": 5.94,
"learning_rate": 2.2092267706302795e-05,
"loss": 0.7012,
"step": 680
},
{
"epoch": 5.98,
"learning_rate": 2.225471085120208e-05,
"loss": 0.7336,
"step": 685
},
{
"epoch": 6.0,
"eval_accuracy": 0.8374384236453202,
"eval_loss": 0.6449394226074219,
"eval_runtime": 14.2954,
"eval_samples_per_second": 85.202,
"eval_steps_per_second": 5.386,
"step": 687
},
{
"epoch": 6.03,
"learning_rate": 2.2417153996101365e-05,
"loss": 0.7113,
"step": 690
},
{
"epoch": 6.07,
"learning_rate": 2.257959714100065e-05,
"loss": 0.6328,
"step": 695
},
{
"epoch": 6.11,
"learning_rate": 2.2742040285899935e-05,
"loss": 0.7496,
"step": 700
},
{
"epoch": 6.16,
"learning_rate": 2.290448343079922e-05,
"loss": 0.8733,
"step": 705
},
{
"epoch": 6.2,
"learning_rate": 2.3066926575698505e-05,
"loss": 0.7935,
"step": 710
},
{
"epoch": 6.24,
"learning_rate": 2.322936972059779e-05,
"loss": 0.8732,
"step": 715
},
{
"epoch": 6.29,
"learning_rate": 2.3391812865497074e-05,
"loss": 0.7589,
"step": 720
},
{
"epoch": 6.33,
"learning_rate": 2.3554256010396363e-05,
"loss": 0.6315,
"step": 725
},
{
"epoch": 6.38,
"learning_rate": 2.3716699155295647e-05,
"loss": 0.7392,
"step": 730
},
{
"epoch": 6.42,
"learning_rate": 2.3879142300194932e-05,
"loss": 0.672,
"step": 735
},
{
"epoch": 6.46,
"learning_rate": 2.4041585445094217e-05,
"loss": 0.6742,
"step": 740
},
{
"epoch": 6.51,
"learning_rate": 2.4204028589993502e-05,
"loss": 0.5929,
"step": 745
},
{
"epoch": 6.55,
"learning_rate": 2.4366471734892787e-05,
"loss": 0.6387,
"step": 750
},
{
"epoch": 6.59,
"learning_rate": 2.4528914879792072e-05,
"loss": 0.6444,
"step": 755
},
{
"epoch": 6.64,
"learning_rate": 2.4691358024691357e-05,
"loss": 0.6301,
"step": 760
},
{
"epoch": 6.68,
"learning_rate": 2.485380116959064e-05,
"loss": 0.66,
"step": 765
},
{
"epoch": 6.72,
"learning_rate": 2.501624431448993e-05,
"loss": 0.6949,
"step": 770
},
{
"epoch": 6.77,
"learning_rate": 2.5178687459389218e-05,
"loss": 0.6736,
"step": 775
},
{
"epoch": 6.81,
"learning_rate": 2.53411306042885e-05,
"loss": 0.8662,
"step": 780
},
{
"epoch": 6.86,
"learning_rate": 2.5503573749187788e-05,
"loss": 0.5951,
"step": 785
},
{
"epoch": 6.9,
"learning_rate": 2.566601689408707e-05,
"loss": 0.671,
"step": 790
},
{
"epoch": 6.94,
"learning_rate": 2.5828460038986357e-05,
"loss": 0.5656,
"step": 795
},
{
"epoch": 6.99,
"learning_rate": 2.599090318388564e-05,
"loss": 0.5805,
"step": 800
},
{
"epoch": 7.0,
"eval_accuracy": 0.8735632183908046,
"eval_loss": 0.5090786814689636,
"eval_runtime": 14.2885,
"eval_samples_per_second": 85.243,
"eval_steps_per_second": 5.389,
"step": 801
},
{
"epoch": 7.03,
"learning_rate": 2.6153346328784927e-05,
"loss": 0.5979,
"step": 805
},
{
"epoch": 7.07,
"learning_rate": 2.6315789473684212e-05,
"loss": 0.6142,
"step": 810
},
{
"epoch": 7.12,
"learning_rate": 2.6478232618583497e-05,
"loss": 0.5754,
"step": 815
},
{
"epoch": 7.16,
"learning_rate": 2.664067576348278e-05,
"loss": 0.6194,
"step": 820
},
{
"epoch": 7.21,
"learning_rate": 2.680311890838207e-05,
"loss": 0.7505,
"step": 825
},
{
"epoch": 7.25,
"learning_rate": 2.696556205328135e-05,
"loss": 0.6599,
"step": 830
},
{
"epoch": 7.29,
"learning_rate": 2.712800519818064e-05,
"loss": 0.6005,
"step": 835
},
{
"epoch": 7.34,
"learning_rate": 2.729044834307992e-05,
"loss": 0.5826,
"step": 840
},
{
"epoch": 7.38,
"learning_rate": 2.745289148797921e-05,
"loss": 0.4662,
"step": 845
},
{
"epoch": 7.42,
"learning_rate": 2.761533463287849e-05,
"loss": 0.4749,
"step": 850
},
{
"epoch": 7.47,
"learning_rate": 2.777777777777778e-05,
"loss": 0.5808,
"step": 855
},
{
"epoch": 7.51,
"learning_rate": 2.7940220922677064e-05,
"loss": 0.5508,
"step": 860
},
{
"epoch": 7.55,
"learning_rate": 2.8102664067576352e-05,
"loss": 0.562,
"step": 865
},
{
"epoch": 7.6,
"learning_rate": 2.8265107212475634e-05,
"loss": 0.4951,
"step": 870
},
{
"epoch": 7.64,
"learning_rate": 2.8427550357374922e-05,
"loss": 0.8779,
"step": 875
},
{
"epoch": 7.69,
"learning_rate": 2.8589993502274203e-05,
"loss": 0.6447,
"step": 880
},
{
"epoch": 7.73,
"learning_rate": 2.875243664717349e-05,
"loss": 0.6258,
"step": 885
},
{
"epoch": 7.77,
"learning_rate": 2.8914879792072773e-05,
"loss": 0.6469,
"step": 890
},
{
"epoch": 7.82,
"learning_rate": 2.907732293697206e-05,
"loss": 0.5322,
"step": 895
},
{
"epoch": 7.86,
"learning_rate": 2.9239766081871346e-05,
"loss": 0.5236,
"step": 900
},
{
"epoch": 7.9,
"learning_rate": 2.9402209226770635e-05,
"loss": 0.6151,
"step": 905
},
{
"epoch": 7.95,
"learning_rate": 2.9564652371669916e-05,
"loss": 0.5125,
"step": 910
},
{
"epoch": 7.99,
"learning_rate": 2.9727095516569204e-05,
"loss": 0.5982,
"step": 915
},
{
"epoch": 8.0,
"eval_accuracy": 0.8825944170771757,
"eval_loss": 0.4677470922470093,
"eval_runtime": 14.3782,
"eval_samples_per_second": 84.711,
"eval_steps_per_second": 5.355,
"step": 916
},
{
"epoch": 8.03,
"learning_rate": 2.9889538661468486e-05,
"loss": 0.5948,
"step": 920
},
{
"epoch": 8.08,
"learning_rate": 3.0051981806367774e-05,
"loss": 0.6046,
"step": 925
},
{
"epoch": 8.12,
"learning_rate": 3.0214424951267055e-05,
"loss": 0.7253,
"step": 930
},
{
"epoch": 8.17,
"learning_rate": 3.0376868096166344e-05,
"loss": 0.5269,
"step": 935
},
{
"epoch": 8.21,
"learning_rate": 3.053931124106563e-05,
"loss": 0.665,
"step": 940
},
{
"epoch": 8.25,
"learning_rate": 3.0701754385964913e-05,
"loss": 0.4819,
"step": 945
},
{
"epoch": 8.3,
"learning_rate": 3.08641975308642e-05,
"loss": 0.558,
"step": 950
},
{
"epoch": 8.34,
"learning_rate": 3.102664067576348e-05,
"loss": 0.5623,
"step": 955
},
{
"epoch": 8.38,
"learning_rate": 3.118908382066277e-05,
"loss": 0.6116,
"step": 960
},
{
"epoch": 8.43,
"learning_rate": 3.135152696556205e-05,
"loss": 0.5544,
"step": 965
},
{
"epoch": 8.47,
"learning_rate": 3.151397011046134e-05,
"loss": 0.4559,
"step": 970
},
{
"epoch": 8.52,
"learning_rate": 3.167641325536063e-05,
"loss": 0.5572,
"step": 975
},
{
"epoch": 8.56,
"learning_rate": 3.183885640025991e-05,
"loss": 0.5206,
"step": 980
},
{
"epoch": 8.6,
"learning_rate": 3.20012995451592e-05,
"loss": 0.5465,
"step": 985
},
{
"epoch": 8.65,
"learning_rate": 3.216374269005848e-05,
"loss": 0.5293,
"step": 990
},
{
"epoch": 8.69,
"learning_rate": 3.232618583495777e-05,
"loss": 0.4627,
"step": 995
},
{
"epoch": 8.73,
"learning_rate": 3.248862897985705e-05,
"loss": 0.6437,
"step": 1000
},
{
"epoch": 8.78,
"learning_rate": 3.265107212475634e-05,
"loss": 0.4954,
"step": 1005
},
{
"epoch": 8.82,
"learning_rate": 3.281351526965562e-05,
"loss": 0.5089,
"step": 1010
},
{
"epoch": 8.86,
"learning_rate": 3.297595841455491e-05,
"loss": 0.5259,
"step": 1015
},
{
"epoch": 8.91,
"learning_rate": 3.313840155945419e-05,
"loss": 0.4522,
"step": 1020
},
{
"epoch": 8.95,
"learning_rate": 3.330084470435348e-05,
"loss": 0.5106,
"step": 1025
},
{
"epoch": 9.0,
"learning_rate": 3.346328784925276e-05,
"loss": 0.4939,
"step": 1030
},
{
"epoch": 9.0,
"eval_accuracy": 0.8842364532019704,
"eval_loss": 0.4029363989830017,
"eval_runtime": 14.153,
"eval_samples_per_second": 86.059,
"eval_steps_per_second": 5.441,
"step": 1030
},
{
"epoch": 9.04,
"learning_rate": 3.362573099415205e-05,
"loss": 0.4433,
"step": 1035
},
{
"epoch": 9.08,
"learning_rate": 3.378817413905133e-05,
"loss": 0.3981,
"step": 1040
},
{
"epoch": 9.13,
"learning_rate": 3.395061728395062e-05,
"loss": 0.3883,
"step": 1045
},
{
"epoch": 9.17,
"learning_rate": 3.41130604288499e-05,
"loss": 0.4423,
"step": 1050
},
{
"epoch": 9.21,
"learning_rate": 3.427550357374919e-05,
"loss": 0.4946,
"step": 1055
},
{
"epoch": 9.26,
"learning_rate": 3.443794671864847e-05,
"loss": 0.3786,
"step": 1060
},
{
"epoch": 9.3,
"learning_rate": 3.4600389863547764e-05,
"loss": 0.4355,
"step": 1065
},
{
"epoch": 9.34,
"learning_rate": 3.476283300844704e-05,
"loss": 0.4098,
"step": 1070
},
{
"epoch": 9.39,
"learning_rate": 3.492527615334633e-05,
"loss": 0.465,
"step": 1075
},
{
"epoch": 9.43,
"learning_rate": 3.508771929824561e-05,
"loss": 0.4482,
"step": 1080
},
{
"epoch": 9.48,
"learning_rate": 3.52501624431449e-05,
"loss": 0.4087,
"step": 1085
},
{
"epoch": 9.52,
"learning_rate": 3.541260558804418e-05,
"loss": 0.3359,
"step": 1090
},
{
"epoch": 9.56,
"learning_rate": 3.557504873294347e-05,
"loss": 0.3615,
"step": 1095
},
{
"epoch": 9.61,
"learning_rate": 3.573749187784276e-05,
"loss": 0.4745,
"step": 1100
},
{
"epoch": 9.65,
"learning_rate": 3.589993502274204e-05,
"loss": 0.5523,
"step": 1105
},
{
"epoch": 9.69,
"learning_rate": 3.606237816764133e-05,
"loss": 0.4844,
"step": 1110
},
{
"epoch": 9.74,
"learning_rate": 3.622482131254061e-05,
"loss": 0.4739,
"step": 1115
},
{
"epoch": 9.78,
"learning_rate": 3.63872644574399e-05,
"loss": 0.516,
"step": 1120
},
{
"epoch": 9.83,
"learning_rate": 3.654970760233918e-05,
"loss": 0.4514,
"step": 1125
},
{
"epoch": 9.87,
"learning_rate": 3.671215074723847e-05,
"loss": 0.4085,
"step": 1130
},
{
"epoch": 9.91,
"learning_rate": 3.687459389213775e-05,
"loss": 0.3573,
"step": 1135
},
{
"epoch": 9.96,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.3216,
"step": 1140
},
{
"epoch": 10.0,
"learning_rate": 3.719948018193632e-05,
"loss": 0.3296,
"step": 1145
},
{
"epoch": 10.0,
"eval_accuracy": 0.9203612479474549,
"eval_loss": 0.30508941411972046,
"eval_runtime": 14.4142,
"eval_samples_per_second": 84.5,
"eval_steps_per_second": 5.342,
"step": 1145
},
{
"epoch": 10.04,
"learning_rate": 3.7361923326835606e-05,
"loss": 0.4359,
"step": 1150
},
{
"epoch": 10.09,
"learning_rate": 3.75243664717349e-05,
"loss": 0.3197,
"step": 1155
},
{
"epoch": 10.13,
"learning_rate": 3.7686809616634176e-05,
"loss": 0.2599,
"step": 1160
},
{
"epoch": 10.17,
"learning_rate": 3.784925276153347e-05,
"loss": 0.4766,
"step": 1165
},
{
"epoch": 10.22,
"learning_rate": 3.8011695906432746e-05,
"loss": 0.3636,
"step": 1170
},
{
"epoch": 10.26,
"learning_rate": 3.817413905133204e-05,
"loss": 0.3543,
"step": 1175
},
{
"epoch": 10.31,
"learning_rate": 3.8336582196231315e-05,
"loss": 0.4652,
"step": 1180
},
{
"epoch": 10.35,
"learning_rate": 3.849902534113061e-05,
"loss": 0.4386,
"step": 1185
},
{
"epoch": 10.39,
"learning_rate": 3.866146848602989e-05,
"loss": 0.5715,
"step": 1190
},
{
"epoch": 10.44,
"learning_rate": 3.882391163092918e-05,
"loss": 0.3944,
"step": 1195
},
{
"epoch": 10.48,
"learning_rate": 3.898635477582846e-05,
"loss": 0.4055,
"step": 1200
},
{
"epoch": 10.52,
"learning_rate": 3.9148797920727747e-05,
"loss": 0.473,
"step": 1205
},
{
"epoch": 10.57,
"learning_rate": 3.931124106562703e-05,
"loss": 0.4065,
"step": 1210
},
{
"epoch": 10.61,
"learning_rate": 3.9473684210526316e-05,
"loss": 0.3315,
"step": 1215
},
{
"epoch": 10.66,
"learning_rate": 3.96361273554256e-05,
"loss": 0.3246,
"step": 1220
},
{
"epoch": 10.7,
"learning_rate": 3.9798570500324886e-05,
"loss": 0.5089,
"step": 1225
},
{
"epoch": 10.74,
"learning_rate": 3.996101364522417e-05,
"loss": 0.3709,
"step": 1230
},
{
"epoch": 10.79,
"learning_rate": 4.012345679012346e-05,
"loss": 0.4847,
"step": 1235
},
{
"epoch": 10.83,
"learning_rate": 4.028589993502274e-05,
"loss": 0.3866,
"step": 1240
},
{
"epoch": 10.87,
"learning_rate": 4.044834307992203e-05,
"loss": 0.2914,
"step": 1245
},
{
"epoch": 10.92,
"learning_rate": 4.061078622482131e-05,
"loss": 0.3748,
"step": 1250
},
{
"epoch": 10.96,
"learning_rate": 4.07732293697206e-05,
"loss": 0.3559,
"step": 1255
},
{
"epoch": 11.0,
"eval_accuracy": 0.8366174055829229,
"eval_loss": 0.5266081094741821,
"eval_runtime": 14.2921,
"eval_samples_per_second": 85.222,
"eval_steps_per_second": 5.388,
"step": 1259
},
{
"epoch": 11.0,
"learning_rate": 4.093567251461988e-05,
"loss": 0.5677,
"step": 1260
},
{
"epoch": 11.05,
"learning_rate": 4.109811565951917e-05,
"loss": 0.6239,
"step": 1265
},
{
"epoch": 11.09,
"learning_rate": 4.1260558804418457e-05,
"loss": 0.5781,
"step": 1270
},
{
"epoch": 11.14,
"learning_rate": 4.142300194931774e-05,
"loss": 0.341,
"step": 1275
},
{
"epoch": 11.18,
"learning_rate": 4.1585445094217026e-05,
"loss": 0.31,
"step": 1280
},
{
"epoch": 11.22,
"learning_rate": 4.174788823911631e-05,
"loss": 0.3522,
"step": 1285
},
{
"epoch": 11.27,
"learning_rate": 4.1910331384015596e-05,
"loss": 0.3488,
"step": 1290
},
{
"epoch": 11.31,
"learning_rate": 4.207277452891488e-05,
"loss": 0.2973,
"step": 1295
},
{
"epoch": 11.35,
"learning_rate": 4.2235217673814166e-05,
"loss": 0.2906,
"step": 1300
},
{
"epoch": 11.4,
"learning_rate": 4.239766081871345e-05,
"loss": 0.3714,
"step": 1305
},
{
"epoch": 11.44,
"learning_rate": 4.2560103963612735e-05,
"loss": 0.3396,
"step": 1310
},
{
"epoch": 11.48,
"learning_rate": 4.272254710851202e-05,
"loss": 0.3847,
"step": 1315
},
{
"epoch": 11.53,
"learning_rate": 4.2884990253411305e-05,
"loss": 0.2346,
"step": 1320
},
{
"epoch": 11.57,
"learning_rate": 4.30474333983106e-05,
"loss": 0.4439,
"step": 1325
},
{
"epoch": 11.62,
"learning_rate": 4.3209876543209875e-05,
"loss": 0.3444,
"step": 1330
},
{
"epoch": 11.66,
"learning_rate": 4.3372319688109166e-05,
"loss": 0.3281,
"step": 1335
},
{
"epoch": 11.7,
"learning_rate": 4.3534762833008445e-05,
"loss": 0.263,
"step": 1340
},
{
"epoch": 11.75,
"learning_rate": 4.3697205977907736e-05,
"loss": 0.4037,
"step": 1345
},
{
"epoch": 11.79,
"learning_rate": 4.3859649122807014e-05,
"loss": 0.2882,
"step": 1350
},
{
"epoch": 11.83,
"learning_rate": 4.4022092267706306e-05,
"loss": 0.2378,
"step": 1355
},
{
"epoch": 11.88,
"learning_rate": 4.418453541260559e-05,
"loss": 0.3295,
"step": 1360
},
{
"epoch": 11.92,
"learning_rate": 4.4346978557504876e-05,
"loss": 0.2535,
"step": 1365
},
{
"epoch": 11.97,
"learning_rate": 4.450942170240416e-05,
"loss": 0.3089,
"step": 1370
},
{
"epoch": 12.0,
"eval_accuracy": 0.8883415435139573,
"eval_loss": 0.3568573296070099,
"eval_runtime": 14.1852,
"eval_samples_per_second": 85.864,
"eval_steps_per_second": 5.428,
"step": 1374
},
{
"epoch": 12.01,
"learning_rate": 4.4671864847303445e-05,
"loss": 0.4128,
"step": 1375
},
{
"epoch": 12.05,
"learning_rate": 4.483430799220273e-05,
"loss": 0.5395,
"step": 1380
},
{
"epoch": 12.1,
"learning_rate": 4.4996751137102015e-05,
"loss": 0.4097,
"step": 1385
},
{
"epoch": 12.14,
"learning_rate": 4.51591942820013e-05,
"loss": 0.3277,
"step": 1390
},
{
"epoch": 12.18,
"learning_rate": 4.5321637426900585e-05,
"loss": 0.3105,
"step": 1395
},
{
"epoch": 12.23,
"learning_rate": 4.548408057179987e-05,
"loss": 0.2685,
"step": 1400
},
{
"epoch": 12.27,
"learning_rate": 4.5646523716699155e-05,
"loss": 0.3484,
"step": 1405
},
{
"epoch": 12.31,
"learning_rate": 4.580896686159844e-05,
"loss": 0.3437,
"step": 1410
},
{
"epoch": 12.36,
"learning_rate": 4.597141000649773e-05,
"loss": 0.3515,
"step": 1415
},
{
"epoch": 12.4,
"learning_rate": 4.613385315139701e-05,
"loss": 0.3601,
"step": 1420
},
{
"epoch": 12.45,
"learning_rate": 4.62962962962963e-05,
"loss": 0.3287,
"step": 1425
},
{
"epoch": 12.49,
"learning_rate": 4.645873944119558e-05,
"loss": 0.3401,
"step": 1430
},
{
"epoch": 12.53,
"learning_rate": 4.662118258609487e-05,
"loss": 0.2752,
"step": 1435
},
{
"epoch": 12.58,
"learning_rate": 4.678362573099415e-05,
"loss": 0.4052,
"step": 1440
},
{
"epoch": 12.62,
"learning_rate": 4.694606887589344e-05,
"loss": 0.3117,
"step": 1445
},
{
"epoch": 12.66,
"learning_rate": 4.7108512020792725e-05,
"loss": 0.2903,
"step": 1450
},
{
"epoch": 12.71,
"learning_rate": 4.727095516569201e-05,
"loss": 0.2411,
"step": 1455
},
{
"epoch": 12.75,
"learning_rate": 4.7433398310591295e-05,
"loss": 0.3215,
"step": 1460
},
{
"epoch": 12.79,
"learning_rate": 4.759584145549058e-05,
"loss": 0.3449,
"step": 1465
},
{
"epoch": 12.84,
"learning_rate": 4.7758284600389865e-05,
"loss": 0.3509,
"step": 1470
},
{
"epoch": 12.88,
"learning_rate": 4.792072774528915e-05,
"loss": 0.3198,
"step": 1475
},
{
"epoch": 12.93,
"learning_rate": 4.8083170890188434e-05,
"loss": 0.3255,
"step": 1480
},
{
"epoch": 12.97,
"learning_rate": 4.824561403508772e-05,
"loss": 0.2272,
"step": 1485
},
{
"epoch": 13.0,
"eval_accuracy": 0.9679802955665024,
"eval_loss": 0.14868870377540588,
"eval_runtime": 14.3655,
"eval_samples_per_second": 84.787,
"eval_steps_per_second": 5.36,
"step": 1488
},
{
"epoch": 13.01,
"learning_rate": 4.8408057179987004e-05,
"loss": 0.2218,
"step": 1490
},
{
"epoch": 13.06,
"learning_rate": 4.857050032488629e-05,
"loss": 0.4013,
"step": 1495
},
{
"epoch": 13.1,
"learning_rate": 4.8732943469785574e-05,
"loss": 0.3278,
"step": 1500
},
{
"epoch": 13.14,
"learning_rate": 4.8895386614684865e-05,
"loss": 0.31,
"step": 1505
},
{
"epoch": 13.19,
"learning_rate": 4.9057829759584143e-05,
"loss": 0.2165,
"step": 1510
},
{
"epoch": 13.23,
"learning_rate": 4.9220272904483435e-05,
"loss": 0.1984,
"step": 1515
},
{
"epoch": 13.28,
"learning_rate": 4.938271604938271e-05,
"loss": 0.2901,
"step": 1520
},
{
"epoch": 13.32,
"learning_rate": 4.9545159194282005e-05,
"loss": 0.3497,
"step": 1525
},
{
"epoch": 13.36,
"learning_rate": 4.970760233918128e-05,
"loss": 0.2879,
"step": 1530
},
{
"epoch": 13.41,
"learning_rate": 4.9870045484080575e-05,
"loss": 0.3517,
"step": 1535
},
{
"epoch": 13.45,
"learning_rate": 4.970760233918128e-05,
"loss": 0.2498,
"step": 1540
},
{
"epoch": 13.49,
"learning_rate": 4.824561403508772e-05,
"loss": 0.2148,
"step": 1545
},
{
"epoch": 13.54,
"learning_rate": 4.678362573099415e-05,
"loss": 0.2684,
"step": 1550
},
{
"epoch": 13.58,
"learning_rate": 4.5321637426900585e-05,
"loss": 0.2943,
"step": 1555
},
{
"epoch": 13.62,
"learning_rate": 4.3859649122807014e-05,
"loss": 0.2528,
"step": 1560
},
{
"epoch": 13.67,
"learning_rate": 4.239766081871345e-05,
"loss": 0.1341,
"step": 1565
},
{
"epoch": 13.71,
"learning_rate": 4.093567251461988e-05,
"loss": 0.2324,
"step": 1570
},
{
"epoch": 13.76,
"learning_rate": 3.9473684210526316e-05,
"loss": 0.3278,
"step": 1575
},
{
"epoch": 13.8,
"learning_rate": 3.8011695906432746e-05,
"loss": 0.3827,
"step": 1580
},
{
"epoch": 13.84,
"learning_rate": 3.654970760233918e-05,
"loss": 0.2841,
"step": 1585
},
{
"epoch": 13.89,
"learning_rate": 3.508771929824561e-05,
"loss": 0.2648,
"step": 1590
},
{
"epoch": 13.93,
"learning_rate": 3.362573099415205e-05,
"loss": 0.314,
"step": 1595
},
{
"epoch": 13.97,
"learning_rate": 3.216374269005848e-05,
"loss": 0.3355,
"step": 1600
},
{
"epoch": 14.0,
"eval_accuracy": 0.9269293924466339,
"eval_loss": 0.234413281083107,
"eval_runtime": 14.1453,
"eval_samples_per_second": 86.106,
"eval_steps_per_second": 5.444,
"step": 1603
},
{
"epoch": 14.02,
"learning_rate": 3.0701754385964913e-05,
"loss": 0.2411,
"step": 1605
},
{
"epoch": 14.06,
"learning_rate": 2.9239766081871346e-05,
"loss": 0.1699,
"step": 1610
},
{
"epoch": 14.1,
"learning_rate": 2.777777777777778e-05,
"loss": 0.239,
"step": 1615
},
{
"epoch": 14.15,
"learning_rate": 2.6315789473684212e-05,
"loss": 0.2076,
"step": 1620
},
{
"epoch": 14.19,
"learning_rate": 2.485380116959064e-05,
"loss": 0.2498,
"step": 1625
},
{
"epoch": 14.24,
"learning_rate": 2.3391812865497074e-05,
"loss": 0.2281,
"step": 1630
},
{
"epoch": 14.28,
"learning_rate": 2.1929824561403507e-05,
"loss": 0.1815,
"step": 1635
},
{
"epoch": 14.32,
"learning_rate": 2.046783625730994e-05,
"loss": 0.1548,
"step": 1640
},
{
"epoch": 14.37,
"learning_rate": 1.9005847953216373e-05,
"loss": 0.1867,
"step": 1645
},
{
"epoch": 14.41,
"learning_rate": 1.7543859649122806e-05,
"loss": 0.1237,
"step": 1650
},
{
"epoch": 14.45,
"learning_rate": 1.608187134502924e-05,
"loss": 0.1528,
"step": 1655
},
{
"epoch": 14.5,
"learning_rate": 1.4619883040935673e-05,
"loss": 0.1484,
"step": 1660
},
{
"epoch": 14.54,
"learning_rate": 1.3157894736842106e-05,
"loss": 0.208,
"step": 1665
},
{
"epoch": 14.59,
"learning_rate": 1.1695906432748537e-05,
"loss": 0.1522,
"step": 1670
},
{
"epoch": 14.63,
"learning_rate": 1.023391812865497e-05,
"loss": 0.1902,
"step": 1675
},
{
"epoch": 14.67,
"learning_rate": 8.771929824561403e-06,
"loss": 0.1111,
"step": 1680
},
{
"epoch": 14.72,
"learning_rate": 7.3099415204678366e-06,
"loss": 0.1581,
"step": 1685
},
{
"epoch": 14.76,
"learning_rate": 5.8479532163742686e-06,
"loss": 0.1491,
"step": 1690
},
{
"epoch": 14.8,
"learning_rate": 4.3859649122807014e-06,
"loss": 0.1382,
"step": 1695
},
{
"epoch": 14.85,
"learning_rate": 2.9239766081871343e-06,
"loss": 0.151,
"step": 1700
},
{
"epoch": 14.89,
"learning_rate": 1.4619883040935671e-06,
"loss": 0.1941,
"step": 1705
},
{
"epoch": 14.93,
"learning_rate": 0.0,
"loss": 0.1581,
"step": 1710
},
{
"epoch": 14.93,
"eval_accuracy": 0.9893267651888341,
"eval_loss": 0.06719201803207397,
"eval_runtime": 14.1195,
"eval_samples_per_second": 86.264,
"eval_steps_per_second": 5.453,
"step": 1710
},
{
"epoch": 14.93,
"step": 1710,
"total_flos": 9.675577376037974e+18,
"train_loss": 0.806596442168219,
"train_runtime": 3496.1772,
"train_samples_per_second": 15.664,
"train_steps_per_second": 0.489
}
],
"logging_steps": 5,
"max_steps": 1710,
"num_train_epochs": 15,
"save_steps": 500,
"total_flos": 9.675577376037974e+18,
"trial_name": null,
"trial_params": null
}