jayanta's picture
End of training
6ed2cc8
raw
history blame
38.5 kB
{
"best_metric": 1.0,
"best_model_checkpoint": "vit-base-patch16-224-in21k-face-recognition/checkpoint-1488",
"epoch": 8.0,
"global_step": 2976,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 4.026845637583892e-06,
"loss": 2.214,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 8.053691275167785e-06,
"loss": 2.1532,
"step": 20
},
{
"epoch": 0.08,
"learning_rate": 1.2080536912751678e-05,
"loss": 2.0195,
"step": 30
},
{
"epoch": 0.11,
"learning_rate": 1.610738255033557e-05,
"loss": 1.7918,
"step": 40
},
{
"epoch": 0.13,
"learning_rate": 2.0134228187919465e-05,
"loss": 1.4794,
"step": 50
},
{
"epoch": 0.16,
"learning_rate": 2.4161073825503356e-05,
"loss": 1.1516,
"step": 60
},
{
"epoch": 0.19,
"learning_rate": 2.8187919463087248e-05,
"loss": 0.8457,
"step": 70
},
{
"epoch": 0.22,
"learning_rate": 3.221476510067114e-05,
"loss": 0.6145,
"step": 80
},
{
"epoch": 0.24,
"learning_rate": 3.6241610738255034e-05,
"loss": 0.446,
"step": 90
},
{
"epoch": 0.27,
"learning_rate": 4.026845637583893e-05,
"loss": 0.3573,
"step": 100
},
{
"epoch": 0.3,
"learning_rate": 4.4295302013422824e-05,
"loss": 0.289,
"step": 110
},
{
"epoch": 0.32,
"learning_rate": 4.832214765100671e-05,
"loss": 0.2559,
"step": 120
},
{
"epoch": 0.35,
"learning_rate": 5.23489932885906e-05,
"loss": 0.2186,
"step": 130
},
{
"epoch": 0.38,
"learning_rate": 5.6375838926174495e-05,
"loss": 0.1936,
"step": 140
},
{
"epoch": 0.4,
"learning_rate": 6.04026845637584e-05,
"loss": 0.181,
"step": 150
},
{
"epoch": 0.43,
"learning_rate": 6.442953020134228e-05,
"loss": 0.1621,
"step": 160
},
{
"epoch": 0.46,
"learning_rate": 6.845637583892617e-05,
"loss": 0.1518,
"step": 170
},
{
"epoch": 0.48,
"learning_rate": 7.248322147651007e-05,
"loss": 0.1361,
"step": 180
},
{
"epoch": 0.51,
"learning_rate": 7.651006711409396e-05,
"loss": 0.1268,
"step": 190
},
{
"epoch": 0.54,
"learning_rate": 8.053691275167786e-05,
"loss": 0.1159,
"step": 200
},
{
"epoch": 0.56,
"learning_rate": 8.456375838926175e-05,
"loss": 0.1088,
"step": 210
},
{
"epoch": 0.59,
"learning_rate": 8.859060402684565e-05,
"loss": 0.1015,
"step": 220
},
{
"epoch": 0.62,
"learning_rate": 9.261744966442954e-05,
"loss": 0.0947,
"step": 230
},
{
"epoch": 0.65,
"learning_rate": 9.664429530201342e-05,
"loss": 0.087,
"step": 240
},
{
"epoch": 0.67,
"learning_rate": 0.00010067114093959731,
"loss": 0.0823,
"step": 250
},
{
"epoch": 0.7,
"learning_rate": 0.0001046979865771812,
"loss": 0.0772,
"step": 260
},
{
"epoch": 0.73,
"learning_rate": 0.0001087248322147651,
"loss": 0.0694,
"step": 270
},
{
"epoch": 0.75,
"learning_rate": 0.00011275167785234899,
"loss": 0.0642,
"step": 280
},
{
"epoch": 0.78,
"learning_rate": 0.00011677852348993289,
"loss": 0.0592,
"step": 290
},
{
"epoch": 0.81,
"learning_rate": 0.00011991038088125467,
"loss": 0.0556,
"step": 300
},
{
"epoch": 0.83,
"learning_rate": 0.000119462285287528,
"loss": 0.0516,
"step": 310
},
{
"epoch": 0.86,
"learning_rate": 0.00011901418969380135,
"loss": 0.0679,
"step": 320
},
{
"epoch": 0.89,
"learning_rate": 0.00011856609410007469,
"loss": 0.0582,
"step": 330
},
{
"epoch": 0.91,
"learning_rate": 0.00011811799850634802,
"loss": 0.0487,
"step": 340
},
{
"epoch": 0.94,
"learning_rate": 0.00011766990291262137,
"loss": 0.0462,
"step": 350
},
{
"epoch": 0.97,
"learning_rate": 0.0001172218073188947,
"loss": 0.0411,
"step": 360
},
{
"epoch": 0.99,
"learning_rate": 0.00011677371172516803,
"loss": 0.0368,
"step": 370
},
{
"epoch": 1.0,
"eval_accuracy": 0.999957997311828,
"eval_loss": 0.034596964716911316,
"eval_runtime": 326.4848,
"eval_samples_per_second": 72.922,
"eval_steps_per_second": 1.139,
"step": 372
},
{
"epoch": 1.02,
"learning_rate": 0.00011632561613144138,
"loss": 0.0335,
"step": 380
},
{
"epoch": 1.05,
"learning_rate": 0.00011587752053771471,
"loss": 0.0316,
"step": 390
},
{
"epoch": 1.08,
"learning_rate": 0.00011542942494398806,
"loss": 0.0303,
"step": 400
},
{
"epoch": 1.1,
"learning_rate": 0.00011498132935026138,
"loss": 0.0285,
"step": 410
},
{
"epoch": 1.13,
"learning_rate": 0.00011453323375653473,
"loss": 0.0273,
"step": 420
},
{
"epoch": 1.16,
"learning_rate": 0.00011408513816280807,
"loss": 0.0258,
"step": 430
},
{
"epoch": 1.18,
"learning_rate": 0.00011363704256908141,
"loss": 0.0248,
"step": 440
},
{
"epoch": 1.21,
"learning_rate": 0.00011318894697535474,
"loss": 0.0257,
"step": 450
},
{
"epoch": 1.24,
"learning_rate": 0.00011274085138162807,
"loss": 0.0251,
"step": 460
},
{
"epoch": 1.26,
"learning_rate": 0.00011229275578790142,
"loss": 0.0225,
"step": 470
},
{
"epoch": 1.29,
"learning_rate": 0.00011184466019417477,
"loss": 0.0212,
"step": 480
},
{
"epoch": 1.32,
"learning_rate": 0.00011139656460044809,
"loss": 0.0214,
"step": 490
},
{
"epoch": 1.34,
"learning_rate": 0.00011094846900672144,
"loss": 0.0197,
"step": 500
},
{
"epoch": 1.37,
"learning_rate": 0.00011050037341299477,
"loss": 0.0203,
"step": 510
},
{
"epoch": 1.4,
"learning_rate": 0.00011005227781926812,
"loss": 0.0183,
"step": 520
},
{
"epoch": 1.42,
"learning_rate": 0.00010960418222554145,
"loss": 0.0202,
"step": 530
},
{
"epoch": 1.45,
"learning_rate": 0.00010915608663181478,
"loss": 0.0258,
"step": 540
},
{
"epoch": 1.48,
"learning_rate": 0.00010870799103808813,
"loss": 0.0198,
"step": 550
},
{
"epoch": 1.51,
"learning_rate": 0.00010825989544436148,
"loss": 0.0194,
"step": 560
},
{
"epoch": 1.53,
"learning_rate": 0.00010781179985063481,
"loss": 0.0171,
"step": 570
},
{
"epoch": 1.56,
"learning_rate": 0.00010736370425690814,
"loss": 0.0152,
"step": 580
},
{
"epoch": 1.59,
"learning_rate": 0.00010691560866318148,
"loss": 0.0162,
"step": 590
},
{
"epoch": 1.61,
"learning_rate": 0.00010646751306945482,
"loss": 0.0143,
"step": 600
},
{
"epoch": 1.64,
"learning_rate": 0.00010601941747572817,
"loss": 0.014,
"step": 610
},
{
"epoch": 1.67,
"learning_rate": 0.00010557132188200149,
"loss": 0.0132,
"step": 620
},
{
"epoch": 1.69,
"learning_rate": 0.00010512322628827484,
"loss": 0.0127,
"step": 630
},
{
"epoch": 1.72,
"learning_rate": 0.00010467513069454817,
"loss": 0.0123,
"step": 640
},
{
"epoch": 1.75,
"learning_rate": 0.00010422703510082152,
"loss": 0.0124,
"step": 650
},
{
"epoch": 1.77,
"learning_rate": 0.00010377893950709485,
"loss": 0.0118,
"step": 660
},
{
"epoch": 1.8,
"learning_rate": 0.00010333084391336818,
"loss": 0.0128,
"step": 670
},
{
"epoch": 1.83,
"learning_rate": 0.00010288274831964153,
"loss": 0.0139,
"step": 680
},
{
"epoch": 1.85,
"learning_rate": 0.00010243465272591486,
"loss": 0.0158,
"step": 690
},
{
"epoch": 1.88,
"learning_rate": 0.0001019865571321882,
"loss": 0.0112,
"step": 700
},
{
"epoch": 1.91,
"learning_rate": 0.00010153846153846155,
"loss": 0.0104,
"step": 710
},
{
"epoch": 1.94,
"learning_rate": 0.00010109036594473488,
"loss": 0.0099,
"step": 720
},
{
"epoch": 1.96,
"learning_rate": 0.00010064227035100823,
"loss": 0.0095,
"step": 730
},
{
"epoch": 1.99,
"learning_rate": 0.00010019417475728155,
"loss": 0.0094,
"step": 740
},
{
"epoch": 2.0,
"eval_accuracy": 0.999957997311828,
"eval_loss": 0.009242160245776176,
"eval_runtime": 315.5591,
"eval_samples_per_second": 75.447,
"eval_steps_per_second": 1.179,
"step": 744
},
{
"epoch": 2.02,
"learning_rate": 9.974607916355489e-05,
"loss": 0.009,
"step": 750
},
{
"epoch": 2.04,
"learning_rate": 9.929798356982824e-05,
"loss": 0.0089,
"step": 760
},
{
"epoch": 2.07,
"learning_rate": 9.884988797610157e-05,
"loss": 0.0086,
"step": 770
},
{
"epoch": 2.1,
"learning_rate": 9.84017923823749e-05,
"loss": 0.0086,
"step": 780
},
{
"epoch": 2.12,
"learning_rate": 9.795369678864824e-05,
"loss": 0.0082,
"step": 790
},
{
"epoch": 2.15,
"learning_rate": 9.750560119492159e-05,
"loss": 0.0081,
"step": 800
},
{
"epoch": 2.18,
"learning_rate": 9.705750560119493e-05,
"loss": 0.0081,
"step": 810
},
{
"epoch": 2.2,
"learning_rate": 9.660941000746825e-05,
"loss": 0.0078,
"step": 820
},
{
"epoch": 2.23,
"learning_rate": 9.61613144137416e-05,
"loss": 0.0077,
"step": 830
},
{
"epoch": 2.26,
"learning_rate": 9.571321882001495e-05,
"loss": 0.0074,
"step": 840
},
{
"epoch": 2.28,
"learning_rate": 9.526512322628828e-05,
"loss": 0.0073,
"step": 850
},
{
"epoch": 2.31,
"learning_rate": 9.481702763256161e-05,
"loss": 0.0096,
"step": 860
},
{
"epoch": 2.34,
"learning_rate": 9.436893203883495e-05,
"loss": 0.0078,
"step": 870
},
{
"epoch": 2.37,
"learning_rate": 9.39208364451083e-05,
"loss": 0.0073,
"step": 880
},
{
"epoch": 2.39,
"learning_rate": 9.347274085138164e-05,
"loss": 0.0069,
"step": 890
},
{
"epoch": 2.42,
"learning_rate": 9.302464525765496e-05,
"loss": 0.0067,
"step": 900
},
{
"epoch": 2.45,
"learning_rate": 9.257654966392831e-05,
"loss": 0.0066,
"step": 910
},
{
"epoch": 2.47,
"learning_rate": 9.212845407020164e-05,
"loss": 0.0064,
"step": 920
},
{
"epoch": 2.5,
"learning_rate": 9.168035847647499e-05,
"loss": 0.0063,
"step": 930
},
{
"epoch": 2.53,
"learning_rate": 9.123226288274832e-05,
"loss": 0.0061,
"step": 940
},
{
"epoch": 2.55,
"learning_rate": 9.078416728902166e-05,
"loss": 0.006,
"step": 950
},
{
"epoch": 2.58,
"learning_rate": 9.0336071695295e-05,
"loss": 0.0059,
"step": 960
},
{
"epoch": 2.61,
"learning_rate": 8.988797610156834e-05,
"loss": 0.0058,
"step": 970
},
{
"epoch": 2.63,
"learning_rate": 8.943988050784167e-05,
"loss": 0.0057,
"step": 980
},
{
"epoch": 2.66,
"learning_rate": 8.899178491411502e-05,
"loss": 0.0055,
"step": 990
},
{
"epoch": 2.69,
"learning_rate": 8.854368932038835e-05,
"loss": 0.0056,
"step": 1000
},
{
"epoch": 2.72,
"learning_rate": 8.80955937266617e-05,
"loss": 0.0055,
"step": 1010
},
{
"epoch": 2.74,
"learning_rate": 8.764749813293502e-05,
"loss": 0.0053,
"step": 1020
},
{
"epoch": 2.77,
"learning_rate": 8.719940253920836e-05,
"loss": 0.0052,
"step": 1030
},
{
"epoch": 2.8,
"learning_rate": 8.675130694548171e-05,
"loss": 0.0052,
"step": 1040
},
{
"epoch": 2.82,
"learning_rate": 8.630321135175504e-05,
"loss": 0.0051,
"step": 1050
},
{
"epoch": 2.85,
"learning_rate": 8.585511575802838e-05,
"loss": 0.0051,
"step": 1060
},
{
"epoch": 2.88,
"learning_rate": 8.540702016430171e-05,
"loss": 0.005,
"step": 1070
},
{
"epoch": 2.9,
"learning_rate": 8.495892457057506e-05,
"loss": 0.0052,
"step": 1080
},
{
"epoch": 2.93,
"learning_rate": 8.45108289768484e-05,
"loss": 0.0049,
"step": 1090
},
{
"epoch": 2.96,
"learning_rate": 8.406273338312172e-05,
"loss": 0.0048,
"step": 1100
},
{
"epoch": 2.98,
"learning_rate": 8.361463778939507e-05,
"loss": 0.0046,
"step": 1110
},
{
"epoch": 3.0,
"eval_accuracy": 0.999957997311828,
"eval_loss": 0.004723448771983385,
"eval_runtime": 316.511,
"eval_samples_per_second": 75.22,
"eval_steps_per_second": 1.175,
"step": 1116
},
{
"epoch": 3.01,
"learning_rate": 8.316654219566842e-05,
"loss": 0.0046,
"step": 1120
},
{
"epoch": 3.04,
"learning_rate": 8.271844660194175e-05,
"loss": 0.0046,
"step": 1130
},
{
"epoch": 3.06,
"learning_rate": 8.22703510082151e-05,
"loss": 0.0045,
"step": 1140
},
{
"epoch": 3.09,
"learning_rate": 8.182225541448842e-05,
"loss": 0.0044,
"step": 1150
},
{
"epoch": 3.12,
"learning_rate": 8.137415982076177e-05,
"loss": 0.0044,
"step": 1160
},
{
"epoch": 3.15,
"learning_rate": 8.092606422703511e-05,
"loss": 0.0043,
"step": 1170
},
{
"epoch": 3.17,
"learning_rate": 8.047796863330845e-05,
"loss": 0.0042,
"step": 1180
},
{
"epoch": 3.2,
"learning_rate": 8.002987303958178e-05,
"loss": 0.0041,
"step": 1190
},
{
"epoch": 3.23,
"learning_rate": 7.958177744585511e-05,
"loss": 0.0041,
"step": 1200
},
{
"epoch": 3.25,
"learning_rate": 7.913368185212846e-05,
"loss": 0.004,
"step": 1210
},
{
"epoch": 3.28,
"learning_rate": 7.86855862584018e-05,
"loss": 0.004,
"step": 1220
},
{
"epoch": 3.31,
"learning_rate": 7.823749066467513e-05,
"loss": 0.004,
"step": 1230
},
{
"epoch": 3.33,
"learning_rate": 7.778939507094847e-05,
"loss": 0.0039,
"step": 1240
},
{
"epoch": 3.36,
"learning_rate": 7.73412994772218e-05,
"loss": 0.0039,
"step": 1250
},
{
"epoch": 3.39,
"learning_rate": 7.689320388349515e-05,
"loss": 0.0038,
"step": 1260
},
{
"epoch": 3.41,
"learning_rate": 7.644510828976849e-05,
"loss": 0.0037,
"step": 1270
},
{
"epoch": 3.44,
"learning_rate": 7.599701269604182e-05,
"loss": 0.0036,
"step": 1280
},
{
"epoch": 3.47,
"learning_rate": 7.554891710231517e-05,
"loss": 0.0036,
"step": 1290
},
{
"epoch": 3.49,
"learning_rate": 7.51008215085885e-05,
"loss": 0.0036,
"step": 1300
},
{
"epoch": 3.52,
"learning_rate": 7.465272591486183e-05,
"loss": 0.0036,
"step": 1310
},
{
"epoch": 3.55,
"learning_rate": 7.420463032113518e-05,
"loss": 0.0035,
"step": 1320
},
{
"epoch": 3.58,
"learning_rate": 7.375653472740851e-05,
"loss": 0.0034,
"step": 1330
},
{
"epoch": 3.6,
"learning_rate": 7.330843913368186e-05,
"loss": 0.0034,
"step": 1340
},
{
"epoch": 3.63,
"learning_rate": 7.286034353995518e-05,
"loss": 0.0036,
"step": 1350
},
{
"epoch": 3.66,
"learning_rate": 7.241224794622853e-05,
"loss": 0.0035,
"step": 1360
},
{
"epoch": 3.68,
"learning_rate": 7.196415235250188e-05,
"loss": 0.0034,
"step": 1370
},
{
"epoch": 3.71,
"learning_rate": 7.151605675877521e-05,
"loss": 0.0033,
"step": 1380
},
{
"epoch": 3.74,
"learning_rate": 7.106796116504854e-05,
"loss": 0.0033,
"step": 1390
},
{
"epoch": 3.76,
"learning_rate": 7.061986557132189e-05,
"loss": 0.0032,
"step": 1400
},
{
"epoch": 3.79,
"learning_rate": 7.017176997759522e-05,
"loss": 0.0032,
"step": 1410
},
{
"epoch": 3.82,
"learning_rate": 6.972367438386857e-05,
"loss": 0.0031,
"step": 1420
},
{
"epoch": 3.84,
"learning_rate": 6.927557879014189e-05,
"loss": 0.0031,
"step": 1430
},
{
"epoch": 3.87,
"learning_rate": 6.882748319641524e-05,
"loss": 0.0031,
"step": 1440
},
{
"epoch": 3.9,
"learning_rate": 6.837938760268858e-05,
"loss": 0.003,
"step": 1450
},
{
"epoch": 3.92,
"learning_rate": 6.793129200896192e-05,
"loss": 0.0031,
"step": 1460
},
{
"epoch": 3.95,
"learning_rate": 6.748319641523525e-05,
"loss": 0.003,
"step": 1470
},
{
"epoch": 3.98,
"learning_rate": 6.703510082150858e-05,
"loss": 0.0029,
"step": 1480
},
{
"epoch": 4.0,
"eval_accuracy": 1.0,
"eval_loss": 0.002898953389376402,
"eval_runtime": 315.5752,
"eval_samples_per_second": 75.443,
"eval_steps_per_second": 1.179,
"step": 1488
},
{
"epoch": 4.01,
"learning_rate": 6.658700522778193e-05,
"loss": 0.003,
"step": 1490
},
{
"epoch": 4.03,
"learning_rate": 6.613890963405528e-05,
"loss": 0.0029,
"step": 1500
},
{
"epoch": 4.06,
"learning_rate": 6.56908140403286e-05,
"loss": 0.0029,
"step": 1510
},
{
"epoch": 4.09,
"learning_rate": 6.524271844660194e-05,
"loss": 0.0028,
"step": 1520
},
{
"epoch": 4.11,
"learning_rate": 6.479462285287528e-05,
"loss": 0.0028,
"step": 1530
},
{
"epoch": 4.14,
"learning_rate": 6.434652725914862e-05,
"loss": 0.0028,
"step": 1540
},
{
"epoch": 4.17,
"learning_rate": 6.389843166542196e-05,
"loss": 0.0028,
"step": 1550
},
{
"epoch": 4.19,
"learning_rate": 6.345033607169529e-05,
"loss": 0.0027,
"step": 1560
},
{
"epoch": 4.22,
"learning_rate": 6.300224047796864e-05,
"loss": 0.0027,
"step": 1570
},
{
"epoch": 4.25,
"learning_rate": 6.255414488424197e-05,
"loss": 0.0027,
"step": 1580
},
{
"epoch": 4.27,
"learning_rate": 6.21060492905153e-05,
"loss": 0.0027,
"step": 1590
},
{
"epoch": 4.3,
"learning_rate": 6.165795369678865e-05,
"loss": 0.0026,
"step": 1600
},
{
"epoch": 4.33,
"learning_rate": 6.120985810306199e-05,
"loss": 0.0026,
"step": 1610
},
{
"epoch": 4.35,
"learning_rate": 6.076176250933533e-05,
"loss": 0.0026,
"step": 1620
},
{
"epoch": 4.38,
"learning_rate": 6.031366691560866e-05,
"loss": 0.0026,
"step": 1630
},
{
"epoch": 4.41,
"learning_rate": 5.9865571321882e-05,
"loss": 0.0025,
"step": 1640
},
{
"epoch": 4.44,
"learning_rate": 5.941747572815534e-05,
"loss": 0.0025,
"step": 1650
},
{
"epoch": 4.46,
"learning_rate": 5.896938013442868e-05,
"loss": 0.0025,
"step": 1660
},
{
"epoch": 4.49,
"learning_rate": 5.852128454070202e-05,
"loss": 0.0025,
"step": 1670
},
{
"epoch": 4.52,
"learning_rate": 5.807318894697535e-05,
"loss": 0.0024,
"step": 1680
},
{
"epoch": 4.54,
"learning_rate": 5.762509335324869e-05,
"loss": 0.0024,
"step": 1690
},
{
"epoch": 4.57,
"learning_rate": 5.717699775952203e-05,
"loss": 0.0032,
"step": 1700
},
{
"epoch": 4.6,
"learning_rate": 5.672890216579537e-05,
"loss": 0.003,
"step": 1710
},
{
"epoch": 4.62,
"learning_rate": 5.6280806572068713e-05,
"loss": 0.0028,
"step": 1720
},
{
"epoch": 4.65,
"learning_rate": 5.583271097834205e-05,
"loss": 0.0055,
"step": 1730
},
{
"epoch": 4.68,
"learning_rate": 5.538461538461539e-05,
"loss": 0.0036,
"step": 1740
},
{
"epoch": 4.7,
"learning_rate": 5.493651979088873e-05,
"loss": 0.0029,
"step": 1750
},
{
"epoch": 4.73,
"learning_rate": 5.448842419716207e-05,
"loss": 0.0059,
"step": 1760
},
{
"epoch": 4.76,
"learning_rate": 5.40403286034354e-05,
"loss": 0.0042,
"step": 1770
},
{
"epoch": 4.78,
"learning_rate": 5.359223300970874e-05,
"loss": 0.0031,
"step": 1780
},
{
"epoch": 4.81,
"learning_rate": 5.3144137415982074e-05,
"loss": 0.0026,
"step": 1790
},
{
"epoch": 4.84,
"learning_rate": 5.269604182225542e-05,
"loss": 0.0025,
"step": 1800
},
{
"epoch": 4.87,
"learning_rate": 5.2247946228528755e-05,
"loss": 0.0025,
"step": 1810
},
{
"epoch": 4.89,
"learning_rate": 5.1799850634802095e-05,
"loss": 0.0023,
"step": 1820
},
{
"epoch": 4.92,
"learning_rate": 5.135175504107543e-05,
"loss": 0.0024,
"step": 1830
},
{
"epoch": 4.95,
"learning_rate": 5.090365944734877e-05,
"loss": 0.0023,
"step": 1840
},
{
"epoch": 4.97,
"learning_rate": 5.045556385362211e-05,
"loss": 0.0022,
"step": 1850
},
{
"epoch": 5.0,
"learning_rate": 5.000746825989545e-05,
"loss": 0.0022,
"step": 1860
},
{
"epoch": 5.0,
"eval_accuracy": 0.9999159946236559,
"eval_loss": 0.0023123060818761587,
"eval_runtime": 314.9106,
"eval_samples_per_second": 75.602,
"eval_steps_per_second": 1.181,
"step": 1860
},
{
"epoch": 5.03,
"learning_rate": 4.955937266616878e-05,
"loss": 0.0023,
"step": 1870
},
{
"epoch": 5.05,
"learning_rate": 4.911127707244212e-05,
"loss": 0.0021,
"step": 1880
},
{
"epoch": 5.08,
"learning_rate": 4.866318147871546e-05,
"loss": 0.0044,
"step": 1890
},
{
"epoch": 5.11,
"learning_rate": 4.82150858849888e-05,
"loss": 0.0021,
"step": 1900
},
{
"epoch": 5.13,
"learning_rate": 4.7766990291262136e-05,
"loss": 0.0022,
"step": 1910
},
{
"epoch": 5.16,
"learning_rate": 4.7318894697535476e-05,
"loss": 0.0021,
"step": 1920
},
{
"epoch": 5.19,
"learning_rate": 4.687079910380881e-05,
"loss": 0.0021,
"step": 1930
},
{
"epoch": 5.22,
"learning_rate": 4.642270351008216e-05,
"loss": 0.002,
"step": 1940
},
{
"epoch": 5.24,
"learning_rate": 4.597460791635549e-05,
"loss": 0.0023,
"step": 1950
},
{
"epoch": 5.27,
"learning_rate": 4.552651232262883e-05,
"loss": 0.002,
"step": 1960
},
{
"epoch": 5.3,
"learning_rate": 4.5078416728902164e-05,
"loss": 0.002,
"step": 1970
},
{
"epoch": 5.32,
"learning_rate": 4.4630321135175504e-05,
"loss": 0.0019,
"step": 1980
},
{
"epoch": 5.35,
"learning_rate": 4.4182225541448844e-05,
"loss": 0.0019,
"step": 1990
},
{
"epoch": 5.38,
"learning_rate": 4.3734129947722184e-05,
"loss": 0.0019,
"step": 2000
},
{
"epoch": 5.4,
"learning_rate": 4.328603435399552e-05,
"loss": 0.0019,
"step": 2010
},
{
"epoch": 5.43,
"learning_rate": 4.283793876026886e-05,
"loss": 0.0019,
"step": 2020
},
{
"epoch": 5.46,
"learning_rate": 4.23898431665422e-05,
"loss": 0.0019,
"step": 2030
},
{
"epoch": 5.48,
"learning_rate": 4.194174757281554e-05,
"loss": 0.0019,
"step": 2040
},
{
"epoch": 5.51,
"learning_rate": 4.149365197908887e-05,
"loss": 0.0019,
"step": 2050
},
{
"epoch": 5.54,
"learning_rate": 4.104555638536221e-05,
"loss": 0.0019,
"step": 2060
},
{
"epoch": 5.56,
"learning_rate": 4.0597460791635545e-05,
"loss": 0.0019,
"step": 2070
},
{
"epoch": 5.59,
"learning_rate": 4.014936519790889e-05,
"loss": 0.0019,
"step": 2080
},
{
"epoch": 5.62,
"learning_rate": 3.9701269604182226e-05,
"loss": 0.0018,
"step": 2090
},
{
"epoch": 5.65,
"learning_rate": 3.9253174010455566e-05,
"loss": 0.0018,
"step": 2100
},
{
"epoch": 5.67,
"learning_rate": 3.88050784167289e-05,
"loss": 0.0018,
"step": 2110
},
{
"epoch": 5.7,
"learning_rate": 3.835698282300224e-05,
"loss": 0.0018,
"step": 2120
},
{
"epoch": 5.73,
"learning_rate": 3.790888722927558e-05,
"loss": 0.0018,
"step": 2130
},
{
"epoch": 5.75,
"learning_rate": 3.746079163554892e-05,
"loss": 0.0018,
"step": 2140
},
{
"epoch": 5.78,
"learning_rate": 3.701269604182225e-05,
"loss": 0.0018,
"step": 2150
},
{
"epoch": 5.81,
"learning_rate": 3.656460044809559e-05,
"loss": 0.0017,
"step": 2160
},
{
"epoch": 5.83,
"learning_rate": 3.6116504854368933e-05,
"loss": 0.0018,
"step": 2170
},
{
"epoch": 5.86,
"learning_rate": 3.5668409260642274e-05,
"loss": 0.0017,
"step": 2180
},
{
"epoch": 5.89,
"learning_rate": 3.522031366691561e-05,
"loss": 0.0017,
"step": 2190
},
{
"epoch": 5.91,
"learning_rate": 3.477221807318895e-05,
"loss": 0.0017,
"step": 2200
},
{
"epoch": 5.94,
"learning_rate": 3.432412247946228e-05,
"loss": 0.0017,
"step": 2210
},
{
"epoch": 5.97,
"learning_rate": 3.387602688573563e-05,
"loss": 0.0017,
"step": 2220
},
{
"epoch": 5.99,
"learning_rate": 3.342793129200896e-05,
"loss": 0.0017,
"step": 2230
},
{
"epoch": 6.0,
"eval_accuracy": 1.0,
"eval_loss": 0.0016880695475265384,
"eval_runtime": 314.9291,
"eval_samples_per_second": 75.598,
"eval_steps_per_second": 1.181,
"step": 2232
},
{
"epoch": 6.02,
"learning_rate": 3.29798356982823e-05,
"loss": 0.0017,
"step": 2240
},
{
"epoch": 6.05,
"learning_rate": 3.2531740104555635e-05,
"loss": 0.0017,
"step": 2250
},
{
"epoch": 6.08,
"learning_rate": 3.2083644510828975e-05,
"loss": 0.0016,
"step": 2260
},
{
"epoch": 6.1,
"learning_rate": 3.1635548917102315e-05,
"loss": 0.0017,
"step": 2270
},
{
"epoch": 6.13,
"learning_rate": 3.1187453323375655e-05,
"loss": 0.0016,
"step": 2280
},
{
"epoch": 6.16,
"learning_rate": 3.0739357729648995e-05,
"loss": 0.0016,
"step": 2290
},
{
"epoch": 6.18,
"learning_rate": 3.0291262135922332e-05,
"loss": 0.0016,
"step": 2300
},
{
"epoch": 6.21,
"learning_rate": 2.984316654219567e-05,
"loss": 0.0016,
"step": 2310
},
{
"epoch": 6.24,
"learning_rate": 2.9395070948469006e-05,
"loss": 0.0016,
"step": 2320
},
{
"epoch": 6.26,
"learning_rate": 2.8946975354742346e-05,
"loss": 0.0016,
"step": 2330
},
{
"epoch": 6.29,
"learning_rate": 2.8498879761015683e-05,
"loss": 0.0016,
"step": 2340
},
{
"epoch": 6.32,
"learning_rate": 2.8050784167289023e-05,
"loss": 0.0016,
"step": 2350
},
{
"epoch": 6.34,
"learning_rate": 2.760268857356236e-05,
"loss": 0.0016,
"step": 2360
},
{
"epoch": 6.37,
"learning_rate": 2.71545929798357e-05,
"loss": 0.0016,
"step": 2370
},
{
"epoch": 6.4,
"learning_rate": 2.6706497386109037e-05,
"loss": 0.0016,
"step": 2380
},
{
"epoch": 6.42,
"learning_rate": 2.6258401792382373e-05,
"loss": 0.0016,
"step": 2390
},
{
"epoch": 6.45,
"learning_rate": 2.5810306198655713e-05,
"loss": 0.0016,
"step": 2400
},
{
"epoch": 6.48,
"learning_rate": 2.536221060492905e-05,
"loss": 0.0016,
"step": 2410
},
{
"epoch": 6.51,
"learning_rate": 2.491411501120239e-05,
"loss": 0.0016,
"step": 2420
},
{
"epoch": 6.53,
"learning_rate": 2.4466019417475727e-05,
"loss": 0.0015,
"step": 2430
},
{
"epoch": 6.56,
"learning_rate": 2.4017923823749067e-05,
"loss": 0.0015,
"step": 2440
},
{
"epoch": 6.59,
"learning_rate": 2.3569828230022404e-05,
"loss": 0.0015,
"step": 2450
},
{
"epoch": 6.61,
"learning_rate": 2.312173263629574e-05,
"loss": 0.0015,
"step": 2460
},
{
"epoch": 6.64,
"learning_rate": 2.267363704256908e-05,
"loss": 0.0015,
"step": 2470
},
{
"epoch": 6.67,
"learning_rate": 2.2225541448842418e-05,
"loss": 0.0015,
"step": 2480
},
{
"epoch": 6.69,
"learning_rate": 2.1777445855115758e-05,
"loss": 0.0015,
"step": 2490
},
{
"epoch": 6.72,
"learning_rate": 2.1329350261389095e-05,
"loss": 0.0015,
"step": 2500
},
{
"epoch": 6.75,
"learning_rate": 2.0881254667662435e-05,
"loss": 0.0015,
"step": 2510
},
{
"epoch": 6.77,
"learning_rate": 2.0433159073935772e-05,
"loss": 0.0015,
"step": 2520
},
{
"epoch": 6.8,
"learning_rate": 1.998506348020911e-05,
"loss": 0.0015,
"step": 2530
},
{
"epoch": 6.83,
"learning_rate": 1.953696788648245e-05,
"loss": 0.0015,
"step": 2540
},
{
"epoch": 6.85,
"learning_rate": 1.9088872292755786e-05,
"loss": 0.0015,
"step": 2550
},
{
"epoch": 6.88,
"learning_rate": 1.8640776699029126e-05,
"loss": 0.0015,
"step": 2560
},
{
"epoch": 6.91,
"learning_rate": 1.8192681105302466e-05,
"loss": 0.0015,
"step": 2570
},
{
"epoch": 6.94,
"learning_rate": 1.7744585511575806e-05,
"loss": 0.0015,
"step": 2580
},
{
"epoch": 6.96,
"learning_rate": 1.7296489917849143e-05,
"loss": 0.0015,
"step": 2590
},
{
"epoch": 6.99,
"learning_rate": 1.684839432412248e-05,
"loss": 0.0015,
"step": 2600
},
{
"epoch": 7.0,
"eval_accuracy": 1.0,
"eval_loss": 0.001458011451177299,
"eval_runtime": 317.9295,
"eval_samples_per_second": 74.885,
"eval_steps_per_second": 1.17,
"step": 2604
},
{
"epoch": 7.02,
"learning_rate": 1.640029873039582e-05,
"loss": 0.0015,
"step": 2610
},
{
"epoch": 7.04,
"learning_rate": 1.5952203136669157e-05,
"loss": 0.0015,
"step": 2620
},
{
"epoch": 7.07,
"learning_rate": 1.5504107542942497e-05,
"loss": 0.0014,
"step": 2630
},
{
"epoch": 7.1,
"learning_rate": 1.5056011949215834e-05,
"loss": 0.0014,
"step": 2640
},
{
"epoch": 7.12,
"learning_rate": 1.460791635548917e-05,
"loss": 0.0014,
"step": 2650
},
{
"epoch": 7.15,
"learning_rate": 1.4159820761762509e-05,
"loss": 0.0014,
"step": 2660
},
{
"epoch": 7.18,
"learning_rate": 1.3711725168035847e-05,
"loss": 0.0014,
"step": 2670
},
{
"epoch": 7.2,
"learning_rate": 1.3263629574309186e-05,
"loss": 0.0014,
"step": 2680
},
{
"epoch": 7.23,
"learning_rate": 1.2815533980582524e-05,
"loss": 0.0014,
"step": 2690
},
{
"epoch": 7.26,
"learning_rate": 1.2367438386855863e-05,
"loss": 0.0014,
"step": 2700
},
{
"epoch": 7.28,
"learning_rate": 1.19193427931292e-05,
"loss": 0.0014,
"step": 2710
},
{
"epoch": 7.31,
"learning_rate": 1.147124719940254e-05,
"loss": 0.0014,
"step": 2720
},
{
"epoch": 7.34,
"learning_rate": 1.1023151605675878e-05,
"loss": 0.0014,
"step": 2730
},
{
"epoch": 7.37,
"learning_rate": 1.0575056011949217e-05,
"loss": 0.0014,
"step": 2740
},
{
"epoch": 7.39,
"learning_rate": 1.0126960418222555e-05,
"loss": 0.0014,
"step": 2750
},
{
"epoch": 7.42,
"learning_rate": 9.678864824495894e-06,
"loss": 0.0014,
"step": 2760
},
{
"epoch": 7.45,
"learning_rate": 9.230769230769232e-06,
"loss": 0.0014,
"step": 2770
},
{
"epoch": 7.47,
"learning_rate": 8.782673637042569e-06,
"loss": 0.0014,
"step": 2780
},
{
"epoch": 7.5,
"learning_rate": 8.334578043315908e-06,
"loss": 0.0014,
"step": 2790
},
{
"epoch": 7.53,
"learning_rate": 7.886482449589246e-06,
"loss": 0.0014,
"step": 2800
},
{
"epoch": 7.55,
"learning_rate": 7.4383868558625845e-06,
"loss": 0.0014,
"step": 2810
},
{
"epoch": 7.58,
"learning_rate": 6.990291262135923e-06,
"loss": 0.0014,
"step": 2820
},
{
"epoch": 7.61,
"learning_rate": 6.542195668409261e-06,
"loss": 0.0014,
"step": 2830
},
{
"epoch": 7.63,
"learning_rate": 6.094100074682599e-06,
"loss": 0.0014,
"step": 2840
},
{
"epoch": 7.66,
"learning_rate": 5.646004480955938e-06,
"loss": 0.0014,
"step": 2850
},
{
"epoch": 7.69,
"learning_rate": 5.197908887229276e-06,
"loss": 0.0014,
"step": 2860
},
{
"epoch": 7.72,
"learning_rate": 4.749813293502614e-06,
"loss": 0.0014,
"step": 2870
},
{
"epoch": 7.74,
"learning_rate": 4.301717699775952e-06,
"loss": 0.0014,
"step": 2880
},
{
"epoch": 7.77,
"learning_rate": 3.853622106049291e-06,
"loss": 0.0014,
"step": 2890
},
{
"epoch": 7.8,
"learning_rate": 3.4055265123226292e-06,
"loss": 0.0014,
"step": 2900
},
{
"epoch": 7.82,
"learning_rate": 2.9574309185959673e-06,
"loss": 0.0014,
"step": 2910
},
{
"epoch": 7.85,
"learning_rate": 2.5093353248693058e-06,
"loss": 0.0014,
"step": 2920
},
{
"epoch": 7.88,
"learning_rate": 2.061239731142644e-06,
"loss": 0.0014,
"step": 2930
},
{
"epoch": 7.9,
"learning_rate": 1.613144137415982e-06,
"loss": 0.0014,
"step": 2940
},
{
"epoch": 7.93,
"learning_rate": 1.1650485436893204e-06,
"loss": 0.0014,
"step": 2950
},
{
"epoch": 7.96,
"learning_rate": 7.169529499626587e-07,
"loss": 0.0014,
"step": 2960
},
{
"epoch": 7.98,
"learning_rate": 2.68857356235997e-07,
"loss": 0.0014,
"step": 2970
},
{
"epoch": 8.0,
"eval_accuracy": 0.999957997311828,
"eval_loss": 0.0014714967692270875,
"eval_runtime": 317.4979,
"eval_samples_per_second": 74.986,
"eval_steps_per_second": 1.172,
"step": 2976
},
{
"epoch": 8.0,
"step": 2976,
"total_flos": 5.9038961296526475e+19,
"train_loss": 0.05853422665912207,
"train_runtime": 16155.4674,
"train_samples_per_second": 47.156,
"train_steps_per_second": 0.184
}
],
"max_steps": 2976,
"num_train_epochs": 8,
"total_flos": 5.9038961296526475e+19,
"trial_name": null,
"trial_params": null
}