ALM-AHME's picture
End of training
8c4b59a
{
"best_metric": 0.9971724787935909,
"best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-BreastCancer-BreakHis-AH-Shuff-3/checkpoint-2189",
"epoch": 12.0,
"eval_steps": 500,
"global_step": 2388,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 2.0938023450586265e-07,
"loss": 0.7386,
"step": 5
},
{
"epoch": 0.05,
"learning_rate": 4.187604690117253e-07,
"loss": 0.7211,
"step": 10
},
{
"epoch": 0.08,
"learning_rate": 6.28140703517588e-07,
"loss": 0.7177,
"step": 15
},
{
"epoch": 0.1,
"learning_rate": 8.375209380234506e-07,
"loss": 0.7531,
"step": 20
},
{
"epoch": 0.13,
"learning_rate": 1.0469011725293133e-06,
"loss": 0.6788,
"step": 25
},
{
"epoch": 0.15,
"learning_rate": 1.256281407035176e-06,
"loss": 0.7,
"step": 30
},
{
"epoch": 0.18,
"learning_rate": 1.4656616415410386e-06,
"loss": 0.6658,
"step": 35
},
{
"epoch": 0.2,
"learning_rate": 1.6750418760469012e-06,
"loss": 0.6642,
"step": 40
},
{
"epoch": 0.23,
"learning_rate": 1.8844221105527638e-06,
"loss": 0.641,
"step": 45
},
{
"epoch": 0.25,
"learning_rate": 2.0938023450586266e-06,
"loss": 0.6131,
"step": 50
},
{
"epoch": 0.28,
"learning_rate": 2.3031825795644893e-06,
"loss": 0.5781,
"step": 55
},
{
"epoch": 0.3,
"learning_rate": 2.512562814070352e-06,
"loss": 0.5835,
"step": 60
},
{
"epoch": 0.33,
"learning_rate": 2.7219430485762145e-06,
"loss": 0.5773,
"step": 65
},
{
"epoch": 0.35,
"learning_rate": 2.931323283082077e-06,
"loss": 0.55,
"step": 70
},
{
"epoch": 0.38,
"learning_rate": 3.1407035175879398e-06,
"loss": 0.4751,
"step": 75
},
{
"epoch": 0.4,
"learning_rate": 3.3500837520938024e-06,
"loss": 0.4837,
"step": 80
},
{
"epoch": 0.43,
"learning_rate": 3.559463986599665e-06,
"loss": 0.4874,
"step": 85
},
{
"epoch": 0.45,
"learning_rate": 3.7688442211055276e-06,
"loss": 0.4528,
"step": 90
},
{
"epoch": 0.48,
"learning_rate": 3.978224455611391e-06,
"loss": 0.4497,
"step": 95
},
{
"epoch": 0.5,
"learning_rate": 4.187604690117253e-06,
"loss": 0.407,
"step": 100
},
{
"epoch": 0.53,
"learning_rate": 4.396984924623116e-06,
"loss": 0.3557,
"step": 105
},
{
"epoch": 0.55,
"learning_rate": 4.6063651591289785e-06,
"loss": 0.3435,
"step": 110
},
{
"epoch": 0.58,
"learning_rate": 4.815745393634841e-06,
"loss": 0.334,
"step": 115
},
{
"epoch": 0.6,
"learning_rate": 5.025125628140704e-06,
"loss": 0.2579,
"step": 120
},
{
"epoch": 0.63,
"learning_rate": 5.234505862646566e-06,
"loss": 0.298,
"step": 125
},
{
"epoch": 0.65,
"learning_rate": 5.443886097152429e-06,
"loss": 0.2904,
"step": 130
},
{
"epoch": 0.68,
"learning_rate": 5.653266331658292e-06,
"loss": 0.2653,
"step": 135
},
{
"epoch": 0.7,
"learning_rate": 5.862646566164154e-06,
"loss": 0.2814,
"step": 140
},
{
"epoch": 0.73,
"learning_rate": 6.072026800670017e-06,
"loss": 0.2375,
"step": 145
},
{
"epoch": 0.75,
"learning_rate": 6.2814070351758795e-06,
"loss": 0.2557,
"step": 150
},
{
"epoch": 0.78,
"learning_rate": 6.490787269681742e-06,
"loss": 0.301,
"step": 155
},
{
"epoch": 0.8,
"learning_rate": 6.700167504187605e-06,
"loss": 0.2171,
"step": 160
},
{
"epoch": 0.83,
"learning_rate": 6.909547738693467e-06,
"loss": 0.1899,
"step": 165
},
{
"epoch": 0.85,
"learning_rate": 7.11892797319933e-06,
"loss": 0.2101,
"step": 170
},
{
"epoch": 0.88,
"learning_rate": 7.328308207705193e-06,
"loss": 0.2239,
"step": 175
},
{
"epoch": 0.9,
"learning_rate": 7.537688442211055e-06,
"loss": 0.2252,
"step": 180
},
{
"epoch": 0.93,
"learning_rate": 7.747068676716917e-06,
"loss": 0.2012,
"step": 185
},
{
"epoch": 0.95,
"learning_rate": 7.956448911222781e-06,
"loss": 0.1873,
"step": 190
},
{
"epoch": 0.98,
"learning_rate": 8.165829145728644e-06,
"loss": 0.2598,
"step": 195
},
{
"epoch": 1.0,
"eval_accuracy": 0.941564561734213,
"eval_loss": 0.1445263922214508,
"eval_runtime": 535.3563,
"eval_samples_per_second": 3.964,
"eval_steps_per_second": 0.497,
"step": 199
},
{
"epoch": 1.01,
"learning_rate": 8.375209380234507e-06,
"loss": 0.2636,
"step": 200
},
{
"epoch": 1.03,
"learning_rate": 8.58458961474037e-06,
"loss": 0.193,
"step": 205
},
{
"epoch": 1.06,
"learning_rate": 8.793969849246232e-06,
"loss": 0.2517,
"step": 210
},
{
"epoch": 1.08,
"learning_rate": 9.003350083752094e-06,
"loss": 0.1317,
"step": 215
},
{
"epoch": 1.11,
"learning_rate": 9.212730318257957e-06,
"loss": 0.2586,
"step": 220
},
{
"epoch": 1.13,
"learning_rate": 9.42211055276382e-06,
"loss": 0.199,
"step": 225
},
{
"epoch": 1.16,
"learning_rate": 9.631490787269682e-06,
"loss": 0.1048,
"step": 230
},
{
"epoch": 1.18,
"learning_rate": 9.840871021775545e-06,
"loss": 0.1902,
"step": 235
},
{
"epoch": 1.21,
"learning_rate": 1.0050251256281408e-05,
"loss": 0.2831,
"step": 240
},
{
"epoch": 1.23,
"learning_rate": 1.025963149078727e-05,
"loss": 0.279,
"step": 245
},
{
"epoch": 1.26,
"learning_rate": 1.0469011725293133e-05,
"loss": 0.3919,
"step": 250
},
{
"epoch": 1.28,
"learning_rate": 1.0678391959798995e-05,
"loss": 0.2474,
"step": 255
},
{
"epoch": 1.31,
"learning_rate": 1.0887772194304858e-05,
"loss": 0.1412,
"step": 260
},
{
"epoch": 1.33,
"learning_rate": 1.109715242881072e-05,
"loss": 0.2199,
"step": 265
},
{
"epoch": 1.36,
"learning_rate": 1.1306532663316583e-05,
"loss": 0.1251,
"step": 270
},
{
"epoch": 1.38,
"learning_rate": 1.1515912897822446e-05,
"loss": 0.2052,
"step": 275
},
{
"epoch": 1.41,
"learning_rate": 1.1725293132328309e-05,
"loss": 0.1302,
"step": 280
},
{
"epoch": 1.43,
"learning_rate": 1.1934673366834171e-05,
"loss": 0.1979,
"step": 285
},
{
"epoch": 1.46,
"learning_rate": 1.2144053601340034e-05,
"loss": 0.1516,
"step": 290
},
{
"epoch": 1.48,
"learning_rate": 1.2353433835845896e-05,
"loss": 0.2435,
"step": 295
},
{
"epoch": 1.51,
"learning_rate": 1.2562814070351759e-05,
"loss": 0.1363,
"step": 300
},
{
"epoch": 1.53,
"learning_rate": 1.2772194304857622e-05,
"loss": 0.1761,
"step": 305
},
{
"epoch": 1.56,
"learning_rate": 1.2981574539363484e-05,
"loss": 0.1312,
"step": 310
},
{
"epoch": 1.58,
"learning_rate": 1.3190954773869347e-05,
"loss": 0.1731,
"step": 315
},
{
"epoch": 1.61,
"learning_rate": 1.340033500837521e-05,
"loss": 0.1808,
"step": 320
},
{
"epoch": 1.63,
"learning_rate": 1.3609715242881072e-05,
"loss": 0.2518,
"step": 325
},
{
"epoch": 1.66,
"learning_rate": 1.3819095477386935e-05,
"loss": 0.1414,
"step": 330
},
{
"epoch": 1.68,
"learning_rate": 1.4028475711892797e-05,
"loss": 0.1738,
"step": 335
},
{
"epoch": 1.71,
"learning_rate": 1.423785594639866e-05,
"loss": 0.2038,
"step": 340
},
{
"epoch": 1.73,
"learning_rate": 1.4447236180904523e-05,
"loss": 0.1958,
"step": 345
},
{
"epoch": 1.76,
"learning_rate": 1.4656616415410385e-05,
"loss": 0.1442,
"step": 350
},
{
"epoch": 1.78,
"learning_rate": 1.4865996649916248e-05,
"loss": 0.1938,
"step": 355
},
{
"epoch": 1.81,
"learning_rate": 1.507537688442211e-05,
"loss": 0.1803,
"step": 360
},
{
"epoch": 1.83,
"learning_rate": 1.528475711892797e-05,
"loss": 0.1918,
"step": 365
},
{
"epoch": 1.86,
"learning_rate": 1.5494137353433834e-05,
"loss": 0.2579,
"step": 370
},
{
"epoch": 1.88,
"learning_rate": 1.57035175879397e-05,
"loss": 0.2417,
"step": 375
},
{
"epoch": 1.91,
"learning_rate": 1.5912897822445563e-05,
"loss": 0.116,
"step": 380
},
{
"epoch": 1.93,
"learning_rate": 1.6122278056951425e-05,
"loss": 0.3688,
"step": 385
},
{
"epoch": 1.96,
"learning_rate": 1.6331658291457288e-05,
"loss": 0.1942,
"step": 390
},
{
"epoch": 1.98,
"learning_rate": 1.654103852596315e-05,
"loss": 0.194,
"step": 395
},
{
"epoch": 2.0,
"eval_accuracy": 0.9524033930254477,
"eval_loss": 0.11212717741727829,
"eval_runtime": 91.4664,
"eval_samples_per_second": 23.2,
"eval_steps_per_second": 2.908,
"step": 398
},
{
"epoch": 2.01,
"learning_rate": 1.6750418760469013e-05,
"loss": 0.1765,
"step": 400
},
{
"epoch": 2.04,
"learning_rate": 1.6959798994974876e-05,
"loss": 0.1709,
"step": 405
},
{
"epoch": 2.06,
"learning_rate": 1.716917922948074e-05,
"loss": 0.1229,
"step": 410
},
{
"epoch": 2.09,
"learning_rate": 1.73785594639866e-05,
"loss": 0.0914,
"step": 415
},
{
"epoch": 2.11,
"learning_rate": 1.7587939698492464e-05,
"loss": 0.1755,
"step": 420
},
{
"epoch": 2.14,
"learning_rate": 1.7797319932998326e-05,
"loss": 0.2204,
"step": 425
},
{
"epoch": 2.16,
"learning_rate": 1.800670016750419e-05,
"loss": 0.1749,
"step": 430
},
{
"epoch": 2.19,
"learning_rate": 1.821608040201005e-05,
"loss": 0.0697,
"step": 435
},
{
"epoch": 2.21,
"learning_rate": 1.8425460636515914e-05,
"loss": 0.1144,
"step": 440
},
{
"epoch": 2.24,
"learning_rate": 1.8634840871021777e-05,
"loss": 0.1236,
"step": 445
},
{
"epoch": 2.26,
"learning_rate": 1.884422110552764e-05,
"loss": 0.2386,
"step": 450
},
{
"epoch": 2.29,
"learning_rate": 1.9053601340033502e-05,
"loss": 0.156,
"step": 455
},
{
"epoch": 2.31,
"learning_rate": 1.9262981574539365e-05,
"loss": 0.2102,
"step": 460
},
{
"epoch": 2.34,
"learning_rate": 1.9472361809045227e-05,
"loss": 0.1128,
"step": 465
},
{
"epoch": 2.36,
"learning_rate": 1.968174204355109e-05,
"loss": 0.112,
"step": 470
},
{
"epoch": 2.39,
"learning_rate": 1.9891122278056953e-05,
"loss": 0.1304,
"step": 475
},
{
"epoch": 2.41,
"learning_rate": 2.0100502512562815e-05,
"loss": 0.1724,
"step": 480
},
{
"epoch": 2.44,
"learning_rate": 2.0309882747068678e-05,
"loss": 0.3854,
"step": 485
},
{
"epoch": 2.46,
"learning_rate": 2.051926298157454e-05,
"loss": 0.2293,
"step": 490
},
{
"epoch": 2.49,
"learning_rate": 2.0728643216080403e-05,
"loss": 0.1072,
"step": 495
},
{
"epoch": 2.51,
"learning_rate": 2.0938023450586266e-05,
"loss": 0.1487,
"step": 500
},
{
"epoch": 2.54,
"learning_rate": 2.1147403685092128e-05,
"loss": 0.1017,
"step": 505
},
{
"epoch": 2.56,
"learning_rate": 2.135678391959799e-05,
"loss": 0.1227,
"step": 510
},
{
"epoch": 2.59,
"learning_rate": 2.1566164154103854e-05,
"loss": 0.1068,
"step": 515
},
{
"epoch": 2.61,
"learning_rate": 2.1775544388609716e-05,
"loss": 0.0725,
"step": 520
},
{
"epoch": 2.64,
"learning_rate": 2.198492462311558e-05,
"loss": 0.1454,
"step": 525
},
{
"epoch": 2.66,
"learning_rate": 2.219430485762144e-05,
"loss": 0.1174,
"step": 530
},
{
"epoch": 2.69,
"learning_rate": 2.2403685092127304e-05,
"loss": 0.1054,
"step": 535
},
{
"epoch": 2.71,
"learning_rate": 2.2613065326633167e-05,
"loss": 0.1123,
"step": 540
},
{
"epoch": 2.74,
"learning_rate": 2.282244556113903e-05,
"loss": 0.1119,
"step": 545
},
{
"epoch": 2.76,
"learning_rate": 2.3031825795644892e-05,
"loss": 0.0954,
"step": 550
},
{
"epoch": 2.79,
"learning_rate": 2.3241206030150754e-05,
"loss": 0.1004,
"step": 555
},
{
"epoch": 2.81,
"learning_rate": 2.3450586264656617e-05,
"loss": 0.1512,
"step": 560
},
{
"epoch": 2.84,
"learning_rate": 2.365996649916248e-05,
"loss": 0.1917,
"step": 565
},
{
"epoch": 2.86,
"learning_rate": 2.3869346733668342e-05,
"loss": 0.0982,
"step": 570
},
{
"epoch": 2.89,
"learning_rate": 2.4078726968174205e-05,
"loss": 0.0821,
"step": 575
},
{
"epoch": 2.91,
"learning_rate": 2.4288107202680068e-05,
"loss": 0.119,
"step": 580
},
{
"epoch": 2.94,
"learning_rate": 2.449748743718593e-05,
"loss": 0.142,
"step": 585
},
{
"epoch": 2.96,
"learning_rate": 2.4706867671691793e-05,
"loss": 0.0796,
"step": 590
},
{
"epoch": 2.99,
"learning_rate": 2.4916247906197655e-05,
"loss": 0.0967,
"step": 595
},
{
"epoch": 3.0,
"eval_accuracy": 0.9825636192271442,
"eval_loss": 0.050433963537216187,
"eval_runtime": 91.4591,
"eval_samples_per_second": 23.202,
"eval_steps_per_second": 2.908,
"step": 597
},
{
"epoch": 3.02,
"learning_rate": 2.5125628140703518e-05,
"loss": 0.1725,
"step": 600
},
{
"epoch": 3.04,
"learning_rate": 2.533500837520938e-05,
"loss": 0.0554,
"step": 605
},
{
"epoch": 3.07,
"learning_rate": 2.5544388609715243e-05,
"loss": 0.4003,
"step": 610
},
{
"epoch": 3.09,
"learning_rate": 2.5753768844221106e-05,
"loss": 0.2184,
"step": 615
},
{
"epoch": 3.12,
"learning_rate": 2.596314907872697e-05,
"loss": 0.1262,
"step": 620
},
{
"epoch": 3.14,
"learning_rate": 2.617252931323283e-05,
"loss": 0.0931,
"step": 625
},
{
"epoch": 3.17,
"learning_rate": 2.6381909547738694e-05,
"loss": 0.0755,
"step": 630
},
{
"epoch": 3.19,
"learning_rate": 2.6591289782244556e-05,
"loss": 0.0888,
"step": 635
},
{
"epoch": 3.22,
"learning_rate": 2.680067001675042e-05,
"loss": 0.1824,
"step": 640
},
{
"epoch": 3.24,
"learning_rate": 2.7010050251256282e-05,
"loss": 0.1061,
"step": 645
},
{
"epoch": 3.27,
"learning_rate": 2.7219430485762144e-05,
"loss": 0.1057,
"step": 650
},
{
"epoch": 3.29,
"learning_rate": 2.7428810720268007e-05,
"loss": 0.2637,
"step": 655
},
{
"epoch": 3.32,
"learning_rate": 2.763819095477387e-05,
"loss": 0.1097,
"step": 660
},
{
"epoch": 3.34,
"learning_rate": 2.7847571189279732e-05,
"loss": 0.1488,
"step": 665
},
{
"epoch": 3.37,
"learning_rate": 2.8056951423785595e-05,
"loss": 0.0566,
"step": 670
},
{
"epoch": 3.39,
"learning_rate": 2.8266331658291457e-05,
"loss": 0.1574,
"step": 675
},
{
"epoch": 3.42,
"learning_rate": 2.847571189279732e-05,
"loss": 0.1386,
"step": 680
},
{
"epoch": 3.44,
"learning_rate": 2.8685092127303183e-05,
"loss": 0.1321,
"step": 685
},
{
"epoch": 3.47,
"learning_rate": 2.8894472361809045e-05,
"loss": 0.2852,
"step": 690
},
{
"epoch": 3.49,
"learning_rate": 2.9103852596314908e-05,
"loss": 0.1582,
"step": 695
},
{
"epoch": 3.52,
"learning_rate": 2.931323283082077e-05,
"loss": 0.2385,
"step": 700
},
{
"epoch": 3.54,
"learning_rate": 2.9522613065326633e-05,
"loss": 0.2513,
"step": 705
},
{
"epoch": 3.57,
"learning_rate": 2.9731993299832496e-05,
"loss": 0.0796,
"step": 710
},
{
"epoch": 3.59,
"learning_rate": 2.994137353433836e-05,
"loss": 0.0922,
"step": 715
},
{
"epoch": 3.62,
"learning_rate": 3.015075376884422e-05,
"loss": 0.1955,
"step": 720
},
{
"epoch": 3.64,
"learning_rate": 3.0360134003350084e-05,
"loss": 0.1581,
"step": 725
},
{
"epoch": 3.67,
"learning_rate": 3.056951423785594e-05,
"loss": 0.0307,
"step": 730
},
{
"epoch": 3.69,
"learning_rate": 3.077889447236181e-05,
"loss": 0.1641,
"step": 735
},
{
"epoch": 3.72,
"learning_rate": 3.098827470686767e-05,
"loss": 0.3599,
"step": 740
},
{
"epoch": 3.74,
"learning_rate": 3.1197654941373534e-05,
"loss": 0.112,
"step": 745
},
{
"epoch": 3.77,
"learning_rate": 3.14070351758794e-05,
"loss": 0.1384,
"step": 750
},
{
"epoch": 3.79,
"learning_rate": 3.1616415410385266e-05,
"loss": 0.2174,
"step": 755
},
{
"epoch": 3.82,
"learning_rate": 3.1825795644891125e-05,
"loss": 0.1579,
"step": 760
},
{
"epoch": 3.84,
"learning_rate": 3.203517587939699e-05,
"loss": 0.1173,
"step": 765
},
{
"epoch": 3.87,
"learning_rate": 3.224455611390285e-05,
"loss": 0.0456,
"step": 770
},
{
"epoch": 3.89,
"learning_rate": 3.245393634840872e-05,
"loss": 0.1281,
"step": 775
},
{
"epoch": 3.92,
"learning_rate": 3.2663316582914576e-05,
"loss": 0.1152,
"step": 780
},
{
"epoch": 3.94,
"learning_rate": 3.287269681742044e-05,
"loss": 0.1576,
"step": 785
},
{
"epoch": 3.97,
"learning_rate": 3.30820770519263e-05,
"loss": 0.1334,
"step": 790
},
{
"epoch": 3.99,
"learning_rate": 3.329145728643217e-05,
"loss": 0.0809,
"step": 795
},
{
"epoch": 4.0,
"eval_accuracy": 0.9448633364750235,
"eval_loss": 0.1603740155696869,
"eval_runtime": 91.2767,
"eval_samples_per_second": 23.248,
"eval_steps_per_second": 2.914,
"step": 796
},
{
"epoch": 4.02,
"learning_rate": 3.3500837520938026e-05,
"loss": 0.1905,
"step": 800
},
{
"epoch": 4.05,
"learning_rate": 3.371021775544389e-05,
"loss": 0.1124,
"step": 805
},
{
"epoch": 4.07,
"learning_rate": 3.391959798994975e-05,
"loss": 0.1095,
"step": 810
},
{
"epoch": 4.1,
"learning_rate": 3.412897822445562e-05,
"loss": 0.1477,
"step": 815
},
{
"epoch": 4.12,
"learning_rate": 3.433835845896148e-05,
"loss": 0.0943,
"step": 820
},
{
"epoch": 4.15,
"learning_rate": 3.454773869346734e-05,
"loss": 0.1347,
"step": 825
},
{
"epoch": 4.17,
"learning_rate": 3.47571189279732e-05,
"loss": 0.0956,
"step": 830
},
{
"epoch": 4.2,
"learning_rate": 3.496649916247907e-05,
"loss": 0.2531,
"step": 835
},
{
"epoch": 4.22,
"learning_rate": 3.517587939698493e-05,
"loss": 0.2285,
"step": 840
},
{
"epoch": 4.25,
"learning_rate": 3.538525963149079e-05,
"loss": 0.1616,
"step": 845
},
{
"epoch": 4.27,
"learning_rate": 3.559463986599665e-05,
"loss": 0.0924,
"step": 850
},
{
"epoch": 4.3,
"learning_rate": 3.580402010050252e-05,
"loss": 0.167,
"step": 855
},
{
"epoch": 4.32,
"learning_rate": 3.601340033500838e-05,
"loss": 0.0795,
"step": 860
},
{
"epoch": 4.35,
"learning_rate": 3.6222780569514244e-05,
"loss": 0.059,
"step": 865
},
{
"epoch": 4.37,
"learning_rate": 3.64321608040201e-05,
"loss": 0.069,
"step": 870
},
{
"epoch": 4.4,
"learning_rate": 3.664154103852597e-05,
"loss": 0.083,
"step": 875
},
{
"epoch": 4.42,
"learning_rate": 3.685092127303183e-05,
"loss": 0.1194,
"step": 880
},
{
"epoch": 4.45,
"learning_rate": 3.7060301507537694e-05,
"loss": 0.1297,
"step": 885
},
{
"epoch": 4.47,
"learning_rate": 3.7269681742043554e-05,
"loss": 0.2179,
"step": 890
},
{
"epoch": 4.5,
"learning_rate": 3.747906197654942e-05,
"loss": 0.2596,
"step": 895
},
{
"epoch": 4.52,
"learning_rate": 3.768844221105528e-05,
"loss": 0.1906,
"step": 900
},
{
"epoch": 4.55,
"learning_rate": 3.7897822445561145e-05,
"loss": 0.0549,
"step": 905
},
{
"epoch": 4.57,
"learning_rate": 3.8107202680067004e-05,
"loss": 0.1066,
"step": 910
},
{
"epoch": 4.6,
"learning_rate": 3.831658291457287e-05,
"loss": 0.1355,
"step": 915
},
{
"epoch": 4.62,
"learning_rate": 3.852596314907873e-05,
"loss": 0.1011,
"step": 920
},
{
"epoch": 4.65,
"learning_rate": 3.8735343383584595e-05,
"loss": 0.1629,
"step": 925
},
{
"epoch": 4.67,
"learning_rate": 3.8944723618090455e-05,
"loss": 0.1952,
"step": 930
},
{
"epoch": 4.7,
"learning_rate": 3.915410385259632e-05,
"loss": 0.1166,
"step": 935
},
{
"epoch": 4.72,
"learning_rate": 3.936348408710218e-05,
"loss": 0.1182,
"step": 940
},
{
"epoch": 4.75,
"learning_rate": 3.9572864321608046e-05,
"loss": 0.1669,
"step": 945
},
{
"epoch": 4.77,
"learning_rate": 3.9782244556113905e-05,
"loss": 0.1174,
"step": 950
},
{
"epoch": 4.8,
"learning_rate": 3.999162479061977e-05,
"loss": 0.0891,
"step": 955
},
{
"epoch": 4.82,
"learning_rate": 4.020100502512563e-05,
"loss": 0.2477,
"step": 960
},
{
"epoch": 4.85,
"learning_rate": 4.0410385259631496e-05,
"loss": 0.2259,
"step": 965
},
{
"epoch": 4.87,
"learning_rate": 4.0619765494137356e-05,
"loss": 0.0424,
"step": 970
},
{
"epoch": 4.9,
"learning_rate": 4.082914572864322e-05,
"loss": 0.0697,
"step": 975
},
{
"epoch": 4.92,
"learning_rate": 4.103852596314908e-05,
"loss": 0.1165,
"step": 980
},
{
"epoch": 4.95,
"learning_rate": 4.124790619765495e-05,
"loss": 0.1085,
"step": 985
},
{
"epoch": 4.97,
"learning_rate": 4.1457286432160806e-05,
"loss": 0.0367,
"step": 990
},
{
"epoch": 5.0,
"learning_rate": 4.166666666666667e-05,
"loss": 0.1531,
"step": 995
},
{
"epoch": 5.0,
"eval_accuracy": 0.9806786050895382,
"eval_loss": 0.06729210168123245,
"eval_runtime": 91.2343,
"eval_samples_per_second": 23.259,
"eval_steps_per_second": 2.916,
"step": 995
},
{
"epoch": 5.03,
"learning_rate": 4.187604690117253e-05,
"loss": 0.0984,
"step": 1000
},
{
"epoch": 5.05,
"learning_rate": 4.20854271356784e-05,
"loss": 0.1212,
"step": 1005
},
{
"epoch": 5.08,
"learning_rate": 4.2294807370184257e-05,
"loss": 0.1852,
"step": 1010
},
{
"epoch": 5.1,
"learning_rate": 4.250418760469012e-05,
"loss": 0.1014,
"step": 1015
},
{
"epoch": 5.13,
"learning_rate": 4.271356783919598e-05,
"loss": 0.1069,
"step": 1020
},
{
"epoch": 5.15,
"learning_rate": 4.292294807370185e-05,
"loss": 0.05,
"step": 1025
},
{
"epoch": 5.18,
"learning_rate": 4.313232830820771e-05,
"loss": 0.1147,
"step": 1030
},
{
"epoch": 5.2,
"learning_rate": 4.334170854271357e-05,
"loss": 0.1277,
"step": 1035
},
{
"epoch": 5.23,
"learning_rate": 4.355108877721943e-05,
"loss": 0.1375,
"step": 1040
},
{
"epoch": 5.25,
"learning_rate": 4.37604690117253e-05,
"loss": 0.0598,
"step": 1045
},
{
"epoch": 5.28,
"learning_rate": 4.396984924623116e-05,
"loss": 0.0845,
"step": 1050
},
{
"epoch": 5.3,
"learning_rate": 4.4179229480737024e-05,
"loss": 0.0603,
"step": 1055
},
{
"epoch": 5.33,
"learning_rate": 4.438860971524288e-05,
"loss": 0.1557,
"step": 1060
},
{
"epoch": 5.35,
"learning_rate": 4.459798994974875e-05,
"loss": 0.0632,
"step": 1065
},
{
"epoch": 5.38,
"learning_rate": 4.480737018425461e-05,
"loss": 0.1375,
"step": 1070
},
{
"epoch": 5.4,
"learning_rate": 4.5016750418760474e-05,
"loss": 0.1563,
"step": 1075
},
{
"epoch": 5.43,
"learning_rate": 4.522613065326633e-05,
"loss": 0.0756,
"step": 1080
},
{
"epoch": 5.45,
"learning_rate": 4.54355108877722e-05,
"loss": 0.1598,
"step": 1085
},
{
"epoch": 5.48,
"learning_rate": 4.564489112227806e-05,
"loss": 0.1176,
"step": 1090
},
{
"epoch": 5.5,
"learning_rate": 4.5854271356783925e-05,
"loss": 0.0973,
"step": 1095
},
{
"epoch": 5.53,
"learning_rate": 4.6063651591289784e-05,
"loss": 0.1257,
"step": 1100
},
{
"epoch": 5.55,
"learning_rate": 4.627303182579565e-05,
"loss": 0.0995,
"step": 1105
},
{
"epoch": 5.58,
"learning_rate": 4.648241206030151e-05,
"loss": 0.0587,
"step": 1110
},
{
"epoch": 5.6,
"learning_rate": 4.6691792294807375e-05,
"loss": 0.053,
"step": 1115
},
{
"epoch": 5.63,
"learning_rate": 4.6901172529313234e-05,
"loss": 0.0598,
"step": 1120
},
{
"epoch": 5.65,
"learning_rate": 4.71105527638191e-05,
"loss": 0.0378,
"step": 1125
},
{
"epoch": 5.68,
"learning_rate": 4.731993299832496e-05,
"loss": 0.1141,
"step": 1130
},
{
"epoch": 5.7,
"learning_rate": 4.7529313232830825e-05,
"loss": 0.1103,
"step": 1135
},
{
"epoch": 5.73,
"learning_rate": 4.7738693467336685e-05,
"loss": 0.0986,
"step": 1140
},
{
"epoch": 5.75,
"learning_rate": 4.794807370184255e-05,
"loss": 0.0762,
"step": 1145
},
{
"epoch": 5.78,
"learning_rate": 4.815745393634841e-05,
"loss": 0.0736,
"step": 1150
},
{
"epoch": 5.8,
"learning_rate": 4.8366834170854276e-05,
"loss": 0.1732,
"step": 1155
},
{
"epoch": 5.83,
"learning_rate": 4.8576214405360135e-05,
"loss": 0.1787,
"step": 1160
},
{
"epoch": 5.85,
"learning_rate": 4.8785594639866e-05,
"loss": 0.2364,
"step": 1165
},
{
"epoch": 5.88,
"learning_rate": 4.899497487437186e-05,
"loss": 0.0701,
"step": 1170
},
{
"epoch": 5.9,
"learning_rate": 4.9204355108877726e-05,
"loss": 0.065,
"step": 1175
},
{
"epoch": 5.93,
"learning_rate": 4.9413735343383586e-05,
"loss": 0.2027,
"step": 1180
},
{
"epoch": 5.95,
"learning_rate": 4.962311557788945e-05,
"loss": 0.1818,
"step": 1185
},
{
"epoch": 5.98,
"learning_rate": 4.983249581239531e-05,
"loss": 0.0941,
"step": 1190
},
{
"epoch": 6.0,
"eval_accuracy": 0.9679547596606974,
"eval_loss": 0.08656678348779678,
"eval_runtime": 91.4801,
"eval_samples_per_second": 23.196,
"eval_steps_per_second": 2.908,
"step": 1194
},
{
"epoch": 6.01,
"learning_rate": 4.995812395309883e-05,
"loss": 0.0423,
"step": 1195
},
{
"epoch": 6.03,
"learning_rate": 4.974874371859297e-05,
"loss": 0.1607,
"step": 1200
},
{
"epoch": 6.06,
"learning_rate": 4.95393634840871e-05,
"loss": 0.2889,
"step": 1205
},
{
"epoch": 6.08,
"learning_rate": 4.932998324958124e-05,
"loss": 0.1447,
"step": 1210
},
{
"epoch": 6.11,
"learning_rate": 4.912060301507538e-05,
"loss": 0.0918,
"step": 1215
},
{
"epoch": 6.13,
"learning_rate": 4.891122278056952e-05,
"loss": 0.1959,
"step": 1220
},
{
"epoch": 6.16,
"learning_rate": 4.870184254606365e-05,
"loss": 0.0893,
"step": 1225
},
{
"epoch": 6.18,
"learning_rate": 4.849246231155779e-05,
"loss": 0.0551,
"step": 1230
},
{
"epoch": 6.21,
"learning_rate": 4.828308207705193e-05,
"loss": 0.0628,
"step": 1235
},
{
"epoch": 6.23,
"learning_rate": 4.807370184254607e-05,
"loss": 0.0512,
"step": 1240
},
{
"epoch": 6.26,
"learning_rate": 4.78643216080402e-05,
"loss": 0.2542,
"step": 1245
},
{
"epoch": 6.28,
"learning_rate": 4.765494137353434e-05,
"loss": 0.0462,
"step": 1250
},
{
"epoch": 6.31,
"learning_rate": 4.7445561139028476e-05,
"loss": 0.1519,
"step": 1255
},
{
"epoch": 6.33,
"learning_rate": 4.723618090452262e-05,
"loss": 0.0917,
"step": 1260
},
{
"epoch": 6.36,
"learning_rate": 4.702680067001675e-05,
"loss": 0.2339,
"step": 1265
},
{
"epoch": 6.38,
"learning_rate": 4.681742043551089e-05,
"loss": 0.1873,
"step": 1270
},
{
"epoch": 6.41,
"learning_rate": 4.6608040201005026e-05,
"loss": 0.0984,
"step": 1275
},
{
"epoch": 6.43,
"learning_rate": 4.639865996649917e-05,
"loss": 0.1295,
"step": 1280
},
{
"epoch": 6.46,
"learning_rate": 4.61892797319933e-05,
"loss": 0.0943,
"step": 1285
},
{
"epoch": 6.48,
"learning_rate": 4.597989949748744e-05,
"loss": 0.1148,
"step": 1290
},
{
"epoch": 6.51,
"learning_rate": 4.5770519262981575e-05,
"loss": 0.0837,
"step": 1295
},
{
"epoch": 6.53,
"learning_rate": 4.5561139028475716e-05,
"loss": 0.1483,
"step": 1300
},
{
"epoch": 6.56,
"learning_rate": 4.535175879396985e-05,
"loss": 0.0798,
"step": 1305
},
{
"epoch": 6.58,
"learning_rate": 4.514237855946399e-05,
"loss": 0.0404,
"step": 1310
},
{
"epoch": 6.61,
"learning_rate": 4.4932998324958125e-05,
"loss": 0.0347,
"step": 1315
},
{
"epoch": 6.63,
"learning_rate": 4.4723618090452266e-05,
"loss": 0.0504,
"step": 1320
},
{
"epoch": 6.66,
"learning_rate": 4.45142378559464e-05,
"loss": 0.1347,
"step": 1325
},
{
"epoch": 6.68,
"learning_rate": 4.430485762144054e-05,
"loss": 0.0369,
"step": 1330
},
{
"epoch": 6.71,
"learning_rate": 4.4095477386934674e-05,
"loss": 0.1117,
"step": 1335
},
{
"epoch": 6.73,
"learning_rate": 4.3886097152428815e-05,
"loss": 0.175,
"step": 1340
},
{
"epoch": 6.76,
"learning_rate": 4.367671691792295e-05,
"loss": 0.0575,
"step": 1345
},
{
"epoch": 6.78,
"learning_rate": 4.346733668341709e-05,
"loss": 0.0205,
"step": 1350
},
{
"epoch": 6.81,
"learning_rate": 4.3257956448911224e-05,
"loss": 0.0523,
"step": 1355
},
{
"epoch": 6.83,
"learning_rate": 4.3048576214405365e-05,
"loss": 0.0719,
"step": 1360
},
{
"epoch": 6.86,
"learning_rate": 4.28391959798995e-05,
"loss": 0.0567,
"step": 1365
},
{
"epoch": 6.88,
"learning_rate": 4.262981574539364e-05,
"loss": 0.0985,
"step": 1370
},
{
"epoch": 6.91,
"learning_rate": 4.2420435510887773e-05,
"loss": 0.2237,
"step": 1375
},
{
"epoch": 6.93,
"learning_rate": 4.2211055276381914e-05,
"loss": 0.0886,
"step": 1380
},
{
"epoch": 6.96,
"learning_rate": 4.200167504187605e-05,
"loss": 0.0679,
"step": 1385
},
{
"epoch": 6.98,
"learning_rate": 4.179229480737019e-05,
"loss": 0.1157,
"step": 1390
},
{
"epoch": 7.0,
"eval_accuracy": 0.9844486333647502,
"eval_loss": 0.05253003165125847,
"eval_runtime": 91.534,
"eval_samples_per_second": 23.183,
"eval_steps_per_second": 2.906,
"step": 1393
},
{
"epoch": 7.01,
"learning_rate": 4.158291457286432e-05,
"loss": 0.141,
"step": 1395
},
{
"epoch": 7.04,
"learning_rate": 4.1373534338358464e-05,
"loss": 0.0745,
"step": 1400
},
{
"epoch": 7.06,
"learning_rate": 4.11641541038526e-05,
"loss": 0.0342,
"step": 1405
},
{
"epoch": 7.09,
"learning_rate": 4.095477386934674e-05,
"loss": 0.056,
"step": 1410
},
{
"epoch": 7.11,
"learning_rate": 4.074539363484087e-05,
"loss": 0.16,
"step": 1415
},
{
"epoch": 7.14,
"learning_rate": 4.053601340033501e-05,
"loss": 0.109,
"step": 1420
},
{
"epoch": 7.16,
"learning_rate": 4.032663316582915e-05,
"loss": 0.152,
"step": 1425
},
{
"epoch": 7.19,
"learning_rate": 4.011725293132329e-05,
"loss": 0.0268,
"step": 1430
},
{
"epoch": 7.21,
"learning_rate": 3.990787269681742e-05,
"loss": 0.0469,
"step": 1435
},
{
"epoch": 7.24,
"learning_rate": 3.969849246231156e-05,
"loss": 0.0486,
"step": 1440
},
{
"epoch": 7.26,
"learning_rate": 3.94891122278057e-05,
"loss": 0.0799,
"step": 1445
},
{
"epoch": 7.29,
"learning_rate": 3.927973199329984e-05,
"loss": 0.039,
"step": 1450
},
{
"epoch": 7.31,
"learning_rate": 3.907035175879397e-05,
"loss": 0.1076,
"step": 1455
},
{
"epoch": 7.34,
"learning_rate": 3.886097152428811e-05,
"loss": 0.0738,
"step": 1460
},
{
"epoch": 7.36,
"learning_rate": 3.8651591289782246e-05,
"loss": 0.1493,
"step": 1465
},
{
"epoch": 7.39,
"learning_rate": 3.844221105527639e-05,
"loss": 0.1027,
"step": 1470
},
{
"epoch": 7.41,
"learning_rate": 3.823283082077052e-05,
"loss": 0.0343,
"step": 1475
},
{
"epoch": 7.44,
"learning_rate": 3.802345058626466e-05,
"loss": 0.1157,
"step": 1480
},
{
"epoch": 7.46,
"learning_rate": 3.7814070351758796e-05,
"loss": 0.1568,
"step": 1485
},
{
"epoch": 7.49,
"learning_rate": 3.7604690117252936e-05,
"loss": 0.0784,
"step": 1490
},
{
"epoch": 7.51,
"learning_rate": 3.739530988274707e-05,
"loss": 0.0659,
"step": 1495
},
{
"epoch": 7.54,
"learning_rate": 3.7185929648241204e-05,
"loss": 0.1141,
"step": 1500
},
{
"epoch": 7.56,
"learning_rate": 3.6976549413735345e-05,
"loss": 0.1225,
"step": 1505
},
{
"epoch": 7.59,
"learning_rate": 3.676716917922948e-05,
"loss": 0.112,
"step": 1510
},
{
"epoch": 7.61,
"learning_rate": 3.655778894472362e-05,
"loss": 0.1435,
"step": 1515
},
{
"epoch": 7.64,
"learning_rate": 3.6348408710217754e-05,
"loss": 0.0802,
"step": 1520
},
{
"epoch": 7.66,
"learning_rate": 3.6139028475711895e-05,
"loss": 0.0237,
"step": 1525
},
{
"epoch": 7.69,
"learning_rate": 3.592964824120603e-05,
"loss": 0.1263,
"step": 1530
},
{
"epoch": 7.71,
"learning_rate": 3.572026800670017e-05,
"loss": 0.0624,
"step": 1535
},
{
"epoch": 7.74,
"learning_rate": 3.5510887772194303e-05,
"loss": 0.0963,
"step": 1540
},
{
"epoch": 7.76,
"learning_rate": 3.5301507537688444e-05,
"loss": 0.0227,
"step": 1545
},
{
"epoch": 7.79,
"learning_rate": 3.509212730318258e-05,
"loss": 0.0827,
"step": 1550
},
{
"epoch": 7.81,
"learning_rate": 3.488274706867672e-05,
"loss": 0.1123,
"step": 1555
},
{
"epoch": 7.84,
"learning_rate": 3.467336683417085e-05,
"loss": 0.0659,
"step": 1560
},
{
"epoch": 7.86,
"learning_rate": 3.4463986599664994e-05,
"loss": 0.0376,
"step": 1565
},
{
"epoch": 7.89,
"learning_rate": 3.425460636515913e-05,
"loss": 0.0089,
"step": 1570
},
{
"epoch": 7.91,
"learning_rate": 3.404522613065327e-05,
"loss": 0.0912,
"step": 1575
},
{
"epoch": 7.94,
"learning_rate": 3.38358458961474e-05,
"loss": 0.073,
"step": 1580
},
{
"epoch": 7.96,
"learning_rate": 3.362646566164154e-05,
"loss": 0.0478,
"step": 1585
},
{
"epoch": 7.99,
"learning_rate": 3.341708542713568e-05,
"loss": 0.0684,
"step": 1590
},
{
"epoch": 8.0,
"eval_accuracy": 0.9759660697455231,
"eval_loss": 0.1003836989402771,
"eval_runtime": 91.5881,
"eval_samples_per_second": 23.169,
"eval_steps_per_second": 2.904,
"step": 1592
},
{
"epoch": 8.02,
"learning_rate": 3.320770519262982e-05,
"loss": 0.0423,
"step": 1595
},
{
"epoch": 8.04,
"learning_rate": 3.299832495812395e-05,
"loss": 0.0516,
"step": 1600
},
{
"epoch": 8.07,
"learning_rate": 3.278894472361809e-05,
"loss": 0.0599,
"step": 1605
},
{
"epoch": 8.09,
"learning_rate": 3.257956448911223e-05,
"loss": 0.1022,
"step": 1610
},
{
"epoch": 8.12,
"learning_rate": 3.237018425460637e-05,
"loss": 0.0228,
"step": 1615
},
{
"epoch": 8.14,
"learning_rate": 3.21608040201005e-05,
"loss": 0.0516,
"step": 1620
},
{
"epoch": 8.17,
"learning_rate": 3.195142378559464e-05,
"loss": 0.0463,
"step": 1625
},
{
"epoch": 8.19,
"learning_rate": 3.1742043551088776e-05,
"loss": 0.0959,
"step": 1630
},
{
"epoch": 8.22,
"learning_rate": 3.153266331658292e-05,
"loss": 0.0109,
"step": 1635
},
{
"epoch": 8.24,
"learning_rate": 3.132328308207705e-05,
"loss": 0.059,
"step": 1640
},
{
"epoch": 8.27,
"learning_rate": 3.111390284757119e-05,
"loss": 0.0971,
"step": 1645
},
{
"epoch": 8.29,
"learning_rate": 3.0904522613065326e-05,
"loss": 0.0721,
"step": 1650
},
{
"epoch": 8.32,
"learning_rate": 3.0695142378559467e-05,
"loss": 0.1205,
"step": 1655
},
{
"epoch": 8.34,
"learning_rate": 3.0485762144053604e-05,
"loss": 0.0985,
"step": 1660
},
{
"epoch": 8.37,
"learning_rate": 3.027638190954774e-05,
"loss": 0.0457,
"step": 1665
},
{
"epoch": 8.39,
"learning_rate": 3.006700167504188e-05,
"loss": 0.1018,
"step": 1670
},
{
"epoch": 8.42,
"learning_rate": 2.9857621440536016e-05,
"loss": 0.1062,
"step": 1675
},
{
"epoch": 8.44,
"learning_rate": 2.9648241206030153e-05,
"loss": 0.0627,
"step": 1680
},
{
"epoch": 8.47,
"learning_rate": 2.943886097152429e-05,
"loss": 0.0603,
"step": 1685
},
{
"epoch": 8.49,
"learning_rate": 2.9229480737018428e-05,
"loss": 0.0543,
"step": 1690
},
{
"epoch": 8.52,
"learning_rate": 2.9020100502512566e-05,
"loss": 0.0975,
"step": 1695
},
{
"epoch": 8.54,
"learning_rate": 2.8810720268006703e-05,
"loss": 0.0124,
"step": 1700
},
{
"epoch": 8.57,
"learning_rate": 2.860134003350084e-05,
"loss": 0.0914,
"step": 1705
},
{
"epoch": 8.59,
"learning_rate": 2.8391959798994978e-05,
"loss": 0.0461,
"step": 1710
},
{
"epoch": 8.62,
"learning_rate": 2.8182579564489115e-05,
"loss": 0.0174,
"step": 1715
},
{
"epoch": 8.64,
"learning_rate": 2.7973199329983252e-05,
"loss": 0.0043,
"step": 1720
},
{
"epoch": 8.67,
"learning_rate": 2.776381909547739e-05,
"loss": 0.048,
"step": 1725
},
{
"epoch": 8.69,
"learning_rate": 2.7554438860971527e-05,
"loss": 0.0277,
"step": 1730
},
{
"epoch": 8.72,
"learning_rate": 2.7345058626465665e-05,
"loss": 0.125,
"step": 1735
},
{
"epoch": 8.74,
"learning_rate": 2.7135678391959802e-05,
"loss": 0.0723,
"step": 1740
},
{
"epoch": 8.77,
"learning_rate": 2.692629815745394e-05,
"loss": 0.1748,
"step": 1745
},
{
"epoch": 8.79,
"learning_rate": 2.6716917922948077e-05,
"loss": 0.0646,
"step": 1750
},
{
"epoch": 8.82,
"learning_rate": 2.6507537688442214e-05,
"loss": 0.0415,
"step": 1755
},
{
"epoch": 8.84,
"learning_rate": 2.629815745393635e-05,
"loss": 0.0726,
"step": 1760
},
{
"epoch": 8.87,
"learning_rate": 2.608877721943049e-05,
"loss": 0.0047,
"step": 1765
},
{
"epoch": 8.89,
"learning_rate": 2.5879396984924626e-05,
"loss": 0.0561,
"step": 1770
},
{
"epoch": 8.92,
"learning_rate": 2.5670016750418764e-05,
"loss": 0.0529,
"step": 1775
},
{
"epoch": 8.94,
"learning_rate": 2.54606365159129e-05,
"loss": 0.0009,
"step": 1780
},
{
"epoch": 8.97,
"learning_rate": 2.525125628140704e-05,
"loss": 0.0378,
"step": 1785
},
{
"epoch": 8.99,
"learning_rate": 2.5041876046901176e-05,
"loss": 0.0141,
"step": 1790
},
{
"epoch": 9.0,
"eval_accuracy": 0.9872761545711592,
"eval_loss": 0.05207793414592743,
"eval_runtime": 91.6702,
"eval_samples_per_second": 23.148,
"eval_steps_per_second": 2.902,
"step": 1791
},
{
"epoch": 9.02,
"learning_rate": 2.4832495812395313e-05,
"loss": 0.0666,
"step": 1795
},
{
"epoch": 9.05,
"learning_rate": 2.462311557788945e-05,
"loss": 0.0038,
"step": 1800
},
{
"epoch": 9.07,
"learning_rate": 2.4413735343383588e-05,
"loss": 0.1081,
"step": 1805
},
{
"epoch": 9.1,
"learning_rate": 2.4204355108877725e-05,
"loss": 0.0273,
"step": 1810
},
{
"epoch": 9.12,
"learning_rate": 2.3994974874371863e-05,
"loss": 0.0519,
"step": 1815
},
{
"epoch": 9.15,
"learning_rate": 2.3785594639866e-05,
"loss": 0.0233,
"step": 1820
},
{
"epoch": 9.17,
"learning_rate": 2.3576214405360137e-05,
"loss": 0.0128,
"step": 1825
},
{
"epoch": 9.2,
"learning_rate": 2.3366834170854275e-05,
"loss": 0.0657,
"step": 1830
},
{
"epoch": 9.22,
"learning_rate": 2.3157453936348412e-05,
"loss": 0.016,
"step": 1835
},
{
"epoch": 9.25,
"learning_rate": 2.294807370184255e-05,
"loss": 0.0528,
"step": 1840
},
{
"epoch": 9.27,
"learning_rate": 2.2738693467336687e-05,
"loss": 0.0561,
"step": 1845
},
{
"epoch": 9.3,
"learning_rate": 2.2529313232830824e-05,
"loss": 0.0826,
"step": 1850
},
{
"epoch": 9.32,
"learning_rate": 2.231993299832496e-05,
"loss": 0.0392,
"step": 1855
},
{
"epoch": 9.35,
"learning_rate": 2.21105527638191e-05,
"loss": 0.0815,
"step": 1860
},
{
"epoch": 9.37,
"learning_rate": 2.1901172529313236e-05,
"loss": 0.0392,
"step": 1865
},
{
"epoch": 9.4,
"learning_rate": 2.169179229480737e-05,
"loss": 0.0209,
"step": 1870
},
{
"epoch": 9.42,
"learning_rate": 2.1482412060301508e-05,
"loss": 0.1453,
"step": 1875
},
{
"epoch": 9.45,
"learning_rate": 2.1273031825795645e-05,
"loss": 0.0127,
"step": 1880
},
{
"epoch": 9.47,
"learning_rate": 2.1063651591289783e-05,
"loss": 0.0012,
"step": 1885
},
{
"epoch": 9.5,
"learning_rate": 2.085427135678392e-05,
"loss": 0.0525,
"step": 1890
},
{
"epoch": 9.52,
"learning_rate": 2.0644891122278057e-05,
"loss": 0.0213,
"step": 1895
},
{
"epoch": 9.55,
"learning_rate": 2.0435510887772195e-05,
"loss": 0.0595,
"step": 1900
},
{
"epoch": 9.57,
"learning_rate": 2.0226130653266332e-05,
"loss": 0.0434,
"step": 1905
},
{
"epoch": 9.6,
"learning_rate": 2.001675041876047e-05,
"loss": 0.0081,
"step": 1910
},
{
"epoch": 9.62,
"learning_rate": 1.9807370184254607e-05,
"loss": 0.0385,
"step": 1915
},
{
"epoch": 9.65,
"learning_rate": 1.9597989949748744e-05,
"loss": 0.054,
"step": 1920
},
{
"epoch": 9.67,
"learning_rate": 1.938860971524288e-05,
"loss": 0.0406,
"step": 1925
},
{
"epoch": 9.7,
"learning_rate": 1.917922948073702e-05,
"loss": 0.0373,
"step": 1930
},
{
"epoch": 9.72,
"learning_rate": 1.8969849246231156e-05,
"loss": 0.0314,
"step": 1935
},
{
"epoch": 9.75,
"learning_rate": 1.8760469011725294e-05,
"loss": 0.0017,
"step": 1940
},
{
"epoch": 9.77,
"learning_rate": 1.855108877721943e-05,
"loss": 0.0315,
"step": 1945
},
{
"epoch": 9.8,
"learning_rate": 1.834170854271357e-05,
"loss": 0.002,
"step": 1950
},
{
"epoch": 9.82,
"learning_rate": 1.8132328308207706e-05,
"loss": 0.076,
"step": 1955
},
{
"epoch": 9.85,
"learning_rate": 1.7922948073701843e-05,
"loss": 0.049,
"step": 1960
},
{
"epoch": 9.87,
"learning_rate": 1.771356783919598e-05,
"loss": 0.0174,
"step": 1965
},
{
"epoch": 9.9,
"learning_rate": 1.7504187604690118e-05,
"loss": 0.0064,
"step": 1970
},
{
"epoch": 9.92,
"learning_rate": 1.7294807370184255e-05,
"loss": 0.0382,
"step": 1975
},
{
"epoch": 9.95,
"learning_rate": 1.7085427135678393e-05,
"loss": 0.0646,
"step": 1980
},
{
"epoch": 9.97,
"learning_rate": 1.687604690117253e-05,
"loss": 0.003,
"step": 1985
},
{
"epoch": 10.0,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.038,
"step": 1990
},
{
"epoch": 10.0,
"eval_accuracy": 0.9934024505183789,
"eval_loss": 0.02926916442811489,
"eval_runtime": 91.6568,
"eval_samples_per_second": 23.152,
"eval_steps_per_second": 2.902,
"step": 1990
},
{
"epoch": 10.03,
"learning_rate": 1.6457286432160805e-05,
"loss": 0.0329,
"step": 1995
},
{
"epoch": 10.05,
"learning_rate": 1.6247906197654942e-05,
"loss": 0.0641,
"step": 2000
},
{
"epoch": 10.08,
"learning_rate": 1.603852596314908e-05,
"loss": 0.0806,
"step": 2005
},
{
"epoch": 10.1,
"learning_rate": 1.5829145728643217e-05,
"loss": 0.0285,
"step": 2010
},
{
"epoch": 10.13,
"learning_rate": 1.5619765494137354e-05,
"loss": 0.0191,
"step": 2015
},
{
"epoch": 10.15,
"learning_rate": 1.5410385259631492e-05,
"loss": 0.0372,
"step": 2020
},
{
"epoch": 10.18,
"learning_rate": 1.5201005025125627e-05,
"loss": 0.0013,
"step": 2025
},
{
"epoch": 10.2,
"learning_rate": 1.4991624790619765e-05,
"loss": 0.0009,
"step": 2030
},
{
"epoch": 10.23,
"learning_rate": 1.4782244556113902e-05,
"loss": 0.0111,
"step": 2035
},
{
"epoch": 10.25,
"learning_rate": 1.457286432160804e-05,
"loss": 0.0003,
"step": 2040
},
{
"epoch": 10.28,
"learning_rate": 1.4363484087102177e-05,
"loss": 0.0002,
"step": 2045
},
{
"epoch": 10.3,
"learning_rate": 1.4154103852596314e-05,
"loss": 0.005,
"step": 2050
},
{
"epoch": 10.33,
"learning_rate": 1.3944723618090452e-05,
"loss": 0.0094,
"step": 2055
},
{
"epoch": 10.35,
"learning_rate": 1.3735343383584589e-05,
"loss": 0.0001,
"step": 2060
},
{
"epoch": 10.38,
"learning_rate": 1.3525963149078726e-05,
"loss": 0.0084,
"step": 2065
},
{
"epoch": 10.4,
"learning_rate": 1.3316582914572864e-05,
"loss": 0.0192,
"step": 2070
},
{
"epoch": 10.43,
"learning_rate": 1.3107202680067001e-05,
"loss": 0.0324,
"step": 2075
},
{
"epoch": 10.45,
"learning_rate": 1.2897822445561139e-05,
"loss": 0.0463,
"step": 2080
},
{
"epoch": 10.48,
"learning_rate": 1.2688442211055276e-05,
"loss": 0.0644,
"step": 2085
},
{
"epoch": 10.5,
"learning_rate": 1.2479061976549413e-05,
"loss": 0.0702,
"step": 2090
},
{
"epoch": 10.53,
"learning_rate": 1.226968174204355e-05,
"loss": 0.0377,
"step": 2095
},
{
"epoch": 10.55,
"learning_rate": 1.2060301507537688e-05,
"loss": 0.0688,
"step": 2100
},
{
"epoch": 10.58,
"learning_rate": 1.1850921273031825e-05,
"loss": 0.0039,
"step": 2105
},
{
"epoch": 10.6,
"learning_rate": 1.1641541038525963e-05,
"loss": 0.0028,
"step": 2110
},
{
"epoch": 10.63,
"learning_rate": 1.14321608040201e-05,
"loss": 0.0518,
"step": 2115
},
{
"epoch": 10.65,
"learning_rate": 1.1222780569514238e-05,
"loss": 0.0279,
"step": 2120
},
{
"epoch": 10.68,
"learning_rate": 1.1013400335008375e-05,
"loss": 0.0001,
"step": 2125
},
{
"epoch": 10.7,
"learning_rate": 1.0804020100502512e-05,
"loss": 0.0108,
"step": 2130
},
{
"epoch": 10.73,
"learning_rate": 1.059463986599665e-05,
"loss": 0.0513,
"step": 2135
},
{
"epoch": 10.75,
"learning_rate": 1.0385259631490787e-05,
"loss": 0.0208,
"step": 2140
},
{
"epoch": 10.78,
"learning_rate": 1.0175879396984924e-05,
"loss": 0.0141,
"step": 2145
},
{
"epoch": 10.8,
"learning_rate": 9.966499162479062e-06,
"loss": 0.0063,
"step": 2150
},
{
"epoch": 10.83,
"learning_rate": 9.7571189279732e-06,
"loss": 0.0723,
"step": 2155
},
{
"epoch": 10.85,
"learning_rate": 9.547738693467337e-06,
"loss": 0.0002,
"step": 2160
},
{
"epoch": 10.88,
"learning_rate": 9.338358458961474e-06,
"loss": 0.0477,
"step": 2165
},
{
"epoch": 10.9,
"learning_rate": 9.128978224455611e-06,
"loss": 0.0868,
"step": 2170
},
{
"epoch": 10.93,
"learning_rate": 8.919597989949749e-06,
"loss": 0.0424,
"step": 2175
},
{
"epoch": 10.95,
"learning_rate": 8.710217755443886e-06,
"loss": 0.0728,
"step": 2180
},
{
"epoch": 10.98,
"learning_rate": 8.500837520938023e-06,
"loss": 0.0044,
"step": 2185
},
{
"epoch": 11.0,
"eval_accuracy": 0.9971724787935909,
"eval_loss": 0.01605476438999176,
"eval_runtime": 91.6819,
"eval_samples_per_second": 23.145,
"eval_steps_per_second": 2.901,
"step": 2189
},
{
"epoch": 11.01,
"learning_rate": 8.291457286432161e-06,
"loss": 0.0324,
"step": 2190
},
{
"epoch": 11.03,
"learning_rate": 8.082077051926298e-06,
"loss": 0.0002,
"step": 2195
},
{
"epoch": 11.06,
"learning_rate": 7.872696817420436e-06,
"loss": 0.0001,
"step": 2200
},
{
"epoch": 11.08,
"learning_rate": 7.663316582914573e-06,
"loss": 0.0036,
"step": 2205
},
{
"epoch": 11.11,
"learning_rate": 7.45393634840871e-06,
"loss": 0.151,
"step": 2210
},
{
"epoch": 11.13,
"learning_rate": 7.244556113902848e-06,
"loss": 0.004,
"step": 2215
},
{
"epoch": 11.16,
"learning_rate": 7.035175879396985e-06,
"loss": 0.0008,
"step": 2220
},
{
"epoch": 11.18,
"learning_rate": 6.8257956448911225e-06,
"loss": 0.0643,
"step": 2225
},
{
"epoch": 11.21,
"learning_rate": 6.61641541038526e-06,
"loss": 0.0754,
"step": 2230
},
{
"epoch": 11.23,
"learning_rate": 6.407035175879397e-06,
"loss": 0.0048,
"step": 2235
},
{
"epoch": 11.26,
"learning_rate": 6.197654941373535e-06,
"loss": 0.0028,
"step": 2240
},
{
"epoch": 11.28,
"learning_rate": 5.988274706867672e-06,
"loss": 0.0378,
"step": 2245
},
{
"epoch": 11.31,
"learning_rate": 5.778894472361809e-06,
"loss": 0.0561,
"step": 2250
},
{
"epoch": 11.33,
"learning_rate": 5.569514237855947e-06,
"loss": 0.0005,
"step": 2255
},
{
"epoch": 11.36,
"learning_rate": 5.360134003350084e-06,
"loss": 0.0361,
"step": 2260
},
{
"epoch": 11.38,
"learning_rate": 5.1507537688442215e-06,
"loss": 0.0348,
"step": 2265
},
{
"epoch": 11.41,
"learning_rate": 4.941373534338359e-06,
"loss": 0.0509,
"step": 2270
},
{
"epoch": 11.43,
"learning_rate": 4.731993299832496e-06,
"loss": 0.0062,
"step": 2275
},
{
"epoch": 11.46,
"learning_rate": 4.522613065326634e-06,
"loss": 0.0228,
"step": 2280
},
{
"epoch": 11.48,
"learning_rate": 4.313232830820771e-06,
"loss": 0.0244,
"step": 2285
},
{
"epoch": 11.51,
"learning_rate": 4.103852596314908e-06,
"loss": 0.0003,
"step": 2290
},
{
"epoch": 11.53,
"learning_rate": 3.894472361809046e-06,
"loss": 0.0384,
"step": 2295
},
{
"epoch": 11.56,
"learning_rate": 3.685092127303183e-06,
"loss": 0.0251,
"step": 2300
},
{
"epoch": 11.58,
"learning_rate": 3.4757118927973206e-06,
"loss": 0.0548,
"step": 2305
},
{
"epoch": 11.61,
"learning_rate": 3.2663316582914575e-06,
"loss": 0.0022,
"step": 2310
},
{
"epoch": 11.63,
"learning_rate": 3.056951423785595e-06,
"loss": 0.0004,
"step": 2315
},
{
"epoch": 11.66,
"learning_rate": 2.8475711892797323e-06,
"loss": 0.0002,
"step": 2320
},
{
"epoch": 11.68,
"learning_rate": 2.6381909547738696e-06,
"loss": 0.0001,
"step": 2325
},
{
"epoch": 11.71,
"learning_rate": 2.428810720268007e-06,
"loss": 0.0004,
"step": 2330
},
{
"epoch": 11.73,
"learning_rate": 2.219430485762144e-06,
"loss": 0.0211,
"step": 2335
},
{
"epoch": 11.76,
"learning_rate": 2.0100502512562813e-06,
"loss": 0.0036,
"step": 2340
},
{
"epoch": 11.78,
"learning_rate": 1.800670016750419e-06,
"loss": 0.0122,
"step": 2345
},
{
"epoch": 11.81,
"learning_rate": 1.5912897822445563e-06,
"loss": 0.0265,
"step": 2350
},
{
"epoch": 11.83,
"learning_rate": 1.3819095477386937e-06,
"loss": 0.01,
"step": 2355
},
{
"epoch": 11.86,
"learning_rate": 1.1725293132328309e-06,
"loss": 0.0021,
"step": 2360
},
{
"epoch": 11.88,
"learning_rate": 9.631490787269682e-07,
"loss": 0.0,
"step": 2365
},
{
"epoch": 11.91,
"learning_rate": 7.537688442211055e-07,
"loss": 0.0001,
"step": 2370
},
{
"epoch": 11.93,
"learning_rate": 5.443886097152429e-07,
"loss": 0.0212,
"step": 2375
},
{
"epoch": 11.96,
"learning_rate": 3.3500837520938026e-07,
"loss": 0.0109,
"step": 2380
},
{
"epoch": 11.98,
"learning_rate": 1.2562814070351758e-07,
"loss": 0.0006,
"step": 2385
},
{
"epoch": 12.0,
"eval_accuracy": 0.9967012252591895,
"eval_loss": 0.019685085862874985,
"eval_runtime": 91.8593,
"eval_samples_per_second": 23.101,
"eval_steps_per_second": 2.896,
"step": 2388
},
{
"epoch": 12.0,
"step": 2388,
"total_flos": 1.7585593436847735e+19,
"train_loss": 0.12566840387472886,
"train_runtime": 12106.0173,
"train_samples_per_second": 6.308,
"train_steps_per_second": 0.197
}
],
"logging_steps": 5,
"max_steps": 2388,
"num_train_epochs": 12,
"save_steps": 500,
"total_flos": 1.7585593436847735e+19,
"trial_name": null,
"trial_params": null
}