ALM-AHME's picture
End of training
b08b42c
{
"best_metric": 0.9856115107913669,
"best_model_checkpoint": "beit-large-patch16-224-finetuned-eurosat-50/checkpoint-1464",
"epoch": 12.0,
"global_step": 1464,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 2.996965098634295e-07,
"loss": 2.0438,
"step": 5
},
{
"epoch": 0.08,
"learning_rate": 3.1866464339908956e-07,
"loss": 2.0206,
"step": 10
},
{
"epoch": 0.12,
"learning_rate": 3.376327769347496e-07,
"loss": 2.0757,
"step": 15
},
{
"epoch": 0.16,
"learning_rate": 3.566009104704098e-07,
"loss": 2.0002,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 3.7556904400606983e-07,
"loss": 2.0448,
"step": 25
},
{
"epoch": 0.25,
"learning_rate": 3.945371775417299e-07,
"loss": 2.0736,
"step": 30
},
{
"epoch": 0.29,
"learning_rate": 4.1350531107739005e-07,
"loss": 2.0419,
"step": 35
},
{
"epoch": 0.33,
"learning_rate": 4.324734446130501e-07,
"loss": 1.8874,
"step": 40
},
{
"epoch": 0.37,
"learning_rate": 4.5144157814871016e-07,
"loss": 1.9695,
"step": 45
},
{
"epoch": 0.41,
"learning_rate": 4.7040971168437033e-07,
"loss": 1.9367,
"step": 50
},
{
"epoch": 0.45,
"learning_rate": 4.893778452200304e-07,
"loss": 1.9436,
"step": 55
},
{
"epoch": 0.49,
"learning_rate": 5.083459787556905e-07,
"loss": 1.9032,
"step": 60
},
{
"epoch": 0.53,
"learning_rate": 5.273141122913505e-07,
"loss": 1.9033,
"step": 65
},
{
"epoch": 0.57,
"learning_rate": 5.462822458270107e-07,
"loss": 1.8211,
"step": 70
},
{
"epoch": 0.61,
"learning_rate": 5.652503793626707e-07,
"loss": 1.8974,
"step": 75
},
{
"epoch": 0.66,
"learning_rate": 5.842185128983308e-07,
"loss": 1.8411,
"step": 80
},
{
"epoch": 0.7,
"learning_rate": 6.031866464339909e-07,
"loss": 1.8139,
"step": 85
},
{
"epoch": 0.74,
"learning_rate": 6.22154779969651e-07,
"loss": 1.716,
"step": 90
},
{
"epoch": 0.78,
"learning_rate": 6.411229135053112e-07,
"loss": 1.8489,
"step": 95
},
{
"epoch": 0.82,
"learning_rate": 6.600910470409712e-07,
"loss": 1.7175,
"step": 100
},
{
"epoch": 0.86,
"learning_rate": 6.790591805766313e-07,
"loss": 1.7794,
"step": 105
},
{
"epoch": 0.9,
"learning_rate": 6.980273141122915e-07,
"loss": 1.6816,
"step": 110
},
{
"epoch": 0.94,
"learning_rate": 7.169954476479515e-07,
"loss": 1.7085,
"step": 115
},
{
"epoch": 0.98,
"learning_rate": 7.359635811836116e-07,
"loss": 1.7148,
"step": 120
},
{
"epoch": 1.0,
"eval_accuracy": 0.39157245632065774,
"eval_loss": 1.6402440071105957,
"eval_runtime": 532.0523,
"eval_samples_per_second": 1.829,
"eval_steps_per_second": 0.115,
"step": 122
},
{
"epoch": 1.02,
"learning_rate": 7.549317147192717e-07,
"loss": 1.7102,
"step": 125
},
{
"epoch": 1.07,
"learning_rate": 7.738998482549317e-07,
"loss": 1.7429,
"step": 130
},
{
"epoch": 1.11,
"learning_rate": 7.928679817905918e-07,
"loss": 1.6153,
"step": 135
},
{
"epoch": 1.15,
"learning_rate": 8.118361153262519e-07,
"loss": 1.6106,
"step": 140
},
{
"epoch": 1.19,
"learning_rate": 8.308042488619121e-07,
"loss": 1.616,
"step": 145
},
{
"epoch": 1.23,
"learning_rate": 8.497723823975721e-07,
"loss": 1.6453,
"step": 150
},
{
"epoch": 1.27,
"learning_rate": 8.687405159332323e-07,
"loss": 1.5465,
"step": 155
},
{
"epoch": 1.31,
"learning_rate": 8.877086494688924e-07,
"loss": 1.4909,
"step": 160
},
{
"epoch": 1.35,
"learning_rate": 9.066767830045524e-07,
"loss": 1.547,
"step": 165
},
{
"epoch": 1.39,
"learning_rate": 9.256449165402125e-07,
"loss": 1.4908,
"step": 170
},
{
"epoch": 1.43,
"learning_rate": 9.446130500758727e-07,
"loss": 1.4469,
"step": 175
},
{
"epoch": 1.48,
"learning_rate": 9.635811836115328e-07,
"loss": 1.4792,
"step": 180
},
{
"epoch": 1.52,
"learning_rate": 9.82549317147193e-07,
"loss": 1.4072,
"step": 185
},
{
"epoch": 1.56,
"learning_rate": 1.0015174506828528e-06,
"loss": 1.4486,
"step": 190
},
{
"epoch": 1.6,
"learning_rate": 1.020485584218513e-06,
"loss": 1.3872,
"step": 195
},
{
"epoch": 1.64,
"learning_rate": 1.039453717754173e-06,
"loss": 1.356,
"step": 200
},
{
"epoch": 1.68,
"learning_rate": 1.0584218512898331e-06,
"loss": 1.3684,
"step": 205
},
{
"epoch": 1.72,
"learning_rate": 1.0773899848254933e-06,
"loss": 1.3258,
"step": 210
},
{
"epoch": 1.76,
"learning_rate": 1.0963581183611534e-06,
"loss": 1.2382,
"step": 215
},
{
"epoch": 1.8,
"learning_rate": 1.1153262518968135e-06,
"loss": 1.2238,
"step": 220
},
{
"epoch": 1.84,
"learning_rate": 1.1342943854324736e-06,
"loss": 1.2492,
"step": 225
},
{
"epoch": 1.89,
"learning_rate": 1.1532625189681337e-06,
"loss": 1.1968,
"step": 230
},
{
"epoch": 1.93,
"learning_rate": 1.1722306525037936e-06,
"loss": 1.2322,
"step": 235
},
{
"epoch": 1.97,
"learning_rate": 1.191198786039454e-06,
"loss": 1.1543,
"step": 240
},
{
"epoch": 2.0,
"eval_accuracy": 0.6207605344295992,
"eval_loss": 1.0717859268188477,
"eval_runtime": 11.9932,
"eval_samples_per_second": 81.129,
"eval_steps_per_second": 5.086,
"step": 244
},
{
"epoch": 2.01,
"learning_rate": 1.210166919575114e-06,
"loss": 1.2166,
"step": 245
},
{
"epoch": 2.05,
"learning_rate": 1.229135053110774e-06,
"loss": 1.2192,
"step": 250
},
{
"epoch": 2.09,
"learning_rate": 1.248103186646434e-06,
"loss": 1.1991,
"step": 255
},
{
"epoch": 2.13,
"learning_rate": 1.2670713201820941e-06,
"loss": 1.0528,
"step": 260
},
{
"epoch": 2.17,
"learning_rate": 1.2860394537177542e-06,
"loss": 1.1812,
"step": 265
},
{
"epoch": 2.21,
"learning_rate": 1.3050075872534144e-06,
"loss": 1.1829,
"step": 270
},
{
"epoch": 2.25,
"learning_rate": 1.3239757207890745e-06,
"loss": 1.073,
"step": 275
},
{
"epoch": 2.3,
"learning_rate": 1.3429438543247344e-06,
"loss": 1.0199,
"step": 280
},
{
"epoch": 2.34,
"learning_rate": 1.3619119878603945e-06,
"loss": 1.1072,
"step": 285
},
{
"epoch": 2.38,
"learning_rate": 1.3808801213960548e-06,
"loss": 1.0258,
"step": 290
},
{
"epoch": 2.42,
"learning_rate": 1.399848254931715e-06,
"loss": 0.9957,
"step": 295
},
{
"epoch": 2.46,
"learning_rate": 1.418816388467375e-06,
"loss": 0.9576,
"step": 300
},
{
"epoch": 2.5,
"learning_rate": 1.4377845220030351e-06,
"loss": 1.06,
"step": 305
},
{
"epoch": 2.54,
"learning_rate": 1.4567526555386952e-06,
"loss": 1.0144,
"step": 310
},
{
"epoch": 2.58,
"learning_rate": 1.4757207890743551e-06,
"loss": 0.9347,
"step": 315
},
{
"epoch": 2.62,
"learning_rate": 1.4946889226100152e-06,
"loss": 0.959,
"step": 320
},
{
"epoch": 2.66,
"learning_rate": 1.5136570561456754e-06,
"loss": 0.9115,
"step": 325
},
{
"epoch": 2.7,
"learning_rate": 1.5326251896813355e-06,
"loss": 1.0356,
"step": 330
},
{
"epoch": 2.75,
"learning_rate": 1.5515933232169956e-06,
"loss": 0.8808,
"step": 335
},
{
"epoch": 2.79,
"learning_rate": 1.5705614567526557e-06,
"loss": 0.9193,
"step": 340
},
{
"epoch": 2.83,
"learning_rate": 1.5895295902883156e-06,
"loss": 0.9572,
"step": 345
},
{
"epoch": 2.87,
"learning_rate": 1.6084977238239757e-06,
"loss": 0.7693,
"step": 350
},
{
"epoch": 2.91,
"learning_rate": 1.627465857359636e-06,
"loss": 0.9359,
"step": 355
},
{
"epoch": 2.95,
"learning_rate": 1.6464339908952961e-06,
"loss": 0.8272,
"step": 360
},
{
"epoch": 2.99,
"learning_rate": 1.6654021244309562e-06,
"loss": 0.8948,
"step": 365
},
{
"epoch": 3.0,
"eval_accuracy": 0.7564234326824255,
"eval_loss": 0.722839891910553,
"eval_runtime": 12.1701,
"eval_samples_per_second": 79.95,
"eval_steps_per_second": 5.012,
"step": 366
},
{
"epoch": 3.03,
"learning_rate": 1.6843702579666163e-06,
"loss": 0.9244,
"step": 370
},
{
"epoch": 3.07,
"learning_rate": 1.7033383915022764e-06,
"loss": 0.8253,
"step": 375
},
{
"epoch": 3.11,
"learning_rate": 1.7223065250379363e-06,
"loss": 0.7977,
"step": 380
},
{
"epoch": 3.16,
"learning_rate": 1.7412746585735965e-06,
"loss": 0.8684,
"step": 385
},
{
"epoch": 3.2,
"learning_rate": 1.7602427921092566e-06,
"loss": 0.7826,
"step": 390
},
{
"epoch": 3.24,
"learning_rate": 1.7792109256449167e-06,
"loss": 0.7662,
"step": 395
},
{
"epoch": 3.28,
"learning_rate": 1.7981790591805768e-06,
"loss": 0.7828,
"step": 400
},
{
"epoch": 3.32,
"learning_rate": 1.8171471927162369e-06,
"loss": 0.7999,
"step": 405
},
{
"epoch": 3.36,
"learning_rate": 1.8361153262518968e-06,
"loss": 0.7535,
"step": 410
},
{
"epoch": 3.4,
"learning_rate": 1.855083459787557e-06,
"loss": 0.6339,
"step": 415
},
{
"epoch": 3.44,
"learning_rate": 1.874051593323217e-06,
"loss": 0.8628,
"step": 420
},
{
"epoch": 3.48,
"learning_rate": 1.8930197268588773e-06,
"loss": 0.8602,
"step": 425
},
{
"epoch": 3.52,
"learning_rate": 1.9119878603945374e-06,
"loss": 0.6121,
"step": 430
},
{
"epoch": 3.57,
"learning_rate": 1.9309559939301973e-06,
"loss": 0.7386,
"step": 435
},
{
"epoch": 3.61,
"learning_rate": 1.9499241274658577e-06,
"loss": 0.7043,
"step": 440
},
{
"epoch": 3.65,
"learning_rate": 1.9688922610015176e-06,
"loss": 0.7084,
"step": 445
},
{
"epoch": 3.69,
"learning_rate": 1.987860394537178e-06,
"loss": 0.722,
"step": 450
},
{
"epoch": 3.73,
"learning_rate": 2.0068285280728378e-06,
"loss": 0.7191,
"step": 455
},
{
"epoch": 3.77,
"learning_rate": 2.0257966616084977e-06,
"loss": 0.6511,
"step": 460
},
{
"epoch": 3.81,
"learning_rate": 2.044764795144158e-06,
"loss": 0.7341,
"step": 465
},
{
"epoch": 3.85,
"learning_rate": 2.063732928679818e-06,
"loss": 0.7016,
"step": 470
},
{
"epoch": 3.89,
"learning_rate": 2.0827010622154782e-06,
"loss": 0.7606,
"step": 475
},
{
"epoch": 3.93,
"learning_rate": 2.101669195751138e-06,
"loss": 0.6985,
"step": 480
},
{
"epoch": 3.98,
"learning_rate": 2.120637329286798e-06,
"loss": 0.6348,
"step": 485
},
{
"epoch": 4.0,
"eval_accuracy": 0.816032887975334,
"eval_loss": 0.5326985120773315,
"eval_runtime": 12.039,
"eval_samples_per_second": 80.82,
"eval_steps_per_second": 5.067,
"step": 488
},
{
"epoch": 4.02,
"learning_rate": 2.1396054628224588e-06,
"loss": 0.7858,
"step": 490
},
{
"epoch": 4.06,
"learning_rate": 2.1585735963581187e-06,
"loss": 0.6641,
"step": 495
},
{
"epoch": 4.1,
"learning_rate": 2.1775417298937786e-06,
"loss": 0.6902,
"step": 500
},
{
"epoch": 4.14,
"learning_rate": 2.196509863429439e-06,
"loss": 0.5704,
"step": 505
},
{
"epoch": 4.18,
"learning_rate": 2.2154779969650988e-06,
"loss": 0.5771,
"step": 510
},
{
"epoch": 4.22,
"learning_rate": 2.234446130500759e-06,
"loss": 0.5898,
"step": 515
},
{
"epoch": 4.26,
"learning_rate": 2.253414264036419e-06,
"loss": 0.6716,
"step": 520
},
{
"epoch": 4.3,
"learning_rate": 2.272382397572079e-06,
"loss": 0.6628,
"step": 525
},
{
"epoch": 4.34,
"learning_rate": 2.291350531107739e-06,
"loss": 0.5503,
"step": 530
},
{
"epoch": 4.39,
"learning_rate": 2.310318664643399e-06,
"loss": 0.5619,
"step": 535
},
{
"epoch": 4.43,
"learning_rate": 2.3292867981790594e-06,
"loss": 0.6089,
"step": 540
},
{
"epoch": 4.47,
"learning_rate": 2.3482549317147193e-06,
"loss": 0.5298,
"step": 545
},
{
"epoch": 4.51,
"learning_rate": 2.3672230652503792e-06,
"loss": 0.7065,
"step": 550
},
{
"epoch": 4.55,
"learning_rate": 2.38619119878604e-06,
"loss": 0.5326,
"step": 555
},
{
"epoch": 4.59,
"learning_rate": 2.4051593323217e-06,
"loss": 0.6341,
"step": 560
},
{
"epoch": 4.63,
"learning_rate": 2.4241274658573598e-06,
"loss": 0.5714,
"step": 565
},
{
"epoch": 4.67,
"learning_rate": 2.44309559939302e-06,
"loss": 0.6201,
"step": 570
},
{
"epoch": 4.71,
"learning_rate": 2.46206373292868e-06,
"loss": 0.6174,
"step": 575
},
{
"epoch": 4.75,
"learning_rate": 2.4810318664643403e-06,
"loss": 0.4883,
"step": 580
},
{
"epoch": 4.8,
"learning_rate": 2.5e-06,
"loss": 0.5766,
"step": 585
},
{
"epoch": 4.84,
"learning_rate": 2.51896813353566e-06,
"loss": 0.6456,
"step": 590
},
{
"epoch": 4.88,
"learning_rate": 2.5379362670713204e-06,
"loss": 0.4569,
"step": 595
},
{
"epoch": 4.92,
"learning_rate": 2.5569044006069803e-06,
"loss": 0.5008,
"step": 600
},
{
"epoch": 4.96,
"learning_rate": 2.5758725341426406e-06,
"loss": 0.5066,
"step": 605
},
{
"epoch": 5.0,
"learning_rate": 2.5948406676783005e-06,
"loss": 0.647,
"step": 610
},
{
"epoch": 5.0,
"eval_accuracy": 0.8550873586844809,
"eval_loss": 0.4080745279788971,
"eval_runtime": 12.0595,
"eval_samples_per_second": 80.684,
"eval_steps_per_second": 5.058,
"step": 610
},
{
"epoch": 5.04,
"learning_rate": 2.6138088012139604e-06,
"loss": 0.5447,
"step": 615
},
{
"epoch": 5.08,
"learning_rate": 2.6327769347496208e-06,
"loss": 0.4686,
"step": 620
},
{
"epoch": 5.12,
"learning_rate": 2.6517450682852807e-06,
"loss": 0.5135,
"step": 625
},
{
"epoch": 5.16,
"learning_rate": 2.670713201820941e-06,
"loss": 0.5673,
"step": 630
},
{
"epoch": 5.2,
"learning_rate": 2.689681335356601e-06,
"loss": 0.449,
"step": 635
},
{
"epoch": 5.25,
"learning_rate": 2.708649468892261e-06,
"loss": 0.4626,
"step": 640
},
{
"epoch": 5.29,
"learning_rate": 2.727617602427921e-06,
"loss": 0.4035,
"step": 645
},
{
"epoch": 5.33,
"learning_rate": 2.746585735963581e-06,
"loss": 0.5546,
"step": 650
},
{
"epoch": 5.37,
"learning_rate": 2.7655538694992417e-06,
"loss": 0.568,
"step": 655
},
{
"epoch": 5.41,
"learning_rate": 2.7845220030349016e-06,
"loss": 0.4888,
"step": 660
},
{
"epoch": 5.45,
"learning_rate": 2.803490136570562e-06,
"loss": 0.3702,
"step": 665
},
{
"epoch": 5.49,
"learning_rate": 2.822458270106222e-06,
"loss": 0.3705,
"step": 670
},
{
"epoch": 5.53,
"learning_rate": 2.841426403641882e-06,
"loss": 0.4295,
"step": 675
},
{
"epoch": 5.57,
"learning_rate": 2.860394537177542e-06,
"loss": 0.4891,
"step": 680
},
{
"epoch": 5.61,
"learning_rate": 2.8793626707132024e-06,
"loss": 0.4497,
"step": 685
},
{
"epoch": 5.66,
"learning_rate": 2.8983308042488623e-06,
"loss": 0.4685,
"step": 690
},
{
"epoch": 5.7,
"learning_rate": 2.917298937784522e-06,
"loss": 0.5381,
"step": 695
},
{
"epoch": 5.74,
"learning_rate": 2.9362670713201825e-06,
"loss": 0.4436,
"step": 700
},
{
"epoch": 5.78,
"learning_rate": 2.9552352048558424e-06,
"loss": 0.3855,
"step": 705
},
{
"epoch": 5.82,
"learning_rate": 2.9742033383915027e-06,
"loss": 0.4104,
"step": 710
},
{
"epoch": 5.86,
"learning_rate": 2.9931714719271626e-06,
"loss": 0.3901,
"step": 715
},
{
"epoch": 5.9,
"learning_rate": 3.0121396054628225e-06,
"loss": 0.3812,
"step": 720
},
{
"epoch": 5.94,
"learning_rate": 3.031107738998483e-06,
"loss": 0.5535,
"step": 725
},
{
"epoch": 5.98,
"learning_rate": 3.0500758725341427e-06,
"loss": 0.3244,
"step": 730
},
{
"epoch": 6.0,
"eval_accuracy": 0.9095580678314491,
"eval_loss": 0.2964554727077484,
"eval_runtime": 12.0232,
"eval_samples_per_second": 80.927,
"eval_steps_per_second": 5.074,
"step": 732
},
{
"epoch": 6.02,
"learning_rate": 3.069044006069803e-06,
"loss": 0.4418,
"step": 735
},
{
"epoch": 6.07,
"learning_rate": 3.088012139605463e-06,
"loss": 0.4396,
"step": 740
},
{
"epoch": 6.11,
"learning_rate": 3.106980273141123e-06,
"loss": 0.3343,
"step": 745
},
{
"epoch": 6.15,
"learning_rate": 3.125948406676783e-06,
"loss": 0.2852,
"step": 750
},
{
"epoch": 6.19,
"learning_rate": 3.144916540212443e-06,
"loss": 0.478,
"step": 755
},
{
"epoch": 6.23,
"learning_rate": 3.1638846737481034e-06,
"loss": 0.4008,
"step": 760
},
{
"epoch": 6.27,
"learning_rate": 3.1828528072837633e-06,
"loss": 0.3887,
"step": 765
},
{
"epoch": 6.31,
"learning_rate": 3.201820940819423e-06,
"loss": 0.3685,
"step": 770
},
{
"epoch": 6.35,
"learning_rate": 3.2207890743550835e-06,
"loss": 0.4041,
"step": 775
},
{
"epoch": 6.39,
"learning_rate": 3.2397572078907434e-06,
"loss": 0.4459,
"step": 780
},
{
"epoch": 6.43,
"learning_rate": 3.258725341426404e-06,
"loss": 0.4183,
"step": 785
},
{
"epoch": 6.48,
"learning_rate": 3.277693474962064e-06,
"loss": 0.3246,
"step": 790
},
{
"epoch": 6.52,
"learning_rate": 3.2966616084977244e-06,
"loss": 0.4094,
"step": 795
},
{
"epoch": 6.56,
"learning_rate": 3.3156297420333843e-06,
"loss": 0.3565,
"step": 800
},
{
"epoch": 6.6,
"learning_rate": 3.3345978755690446e-06,
"loss": 0.3411,
"step": 805
},
{
"epoch": 6.64,
"learning_rate": 3.3535660091047045e-06,
"loss": 0.4017,
"step": 810
},
{
"epoch": 6.68,
"learning_rate": 3.3725341426403644e-06,
"loss": 0.343,
"step": 815
},
{
"epoch": 6.72,
"learning_rate": 3.3915022761760247e-06,
"loss": 0.3491,
"step": 820
},
{
"epoch": 6.76,
"learning_rate": 3.4104704097116846e-06,
"loss": 0.4143,
"step": 825
},
{
"epoch": 6.8,
"learning_rate": 3.429438543247345e-06,
"loss": 0.357,
"step": 830
},
{
"epoch": 6.84,
"learning_rate": 3.448406676783005e-06,
"loss": 0.4457,
"step": 835
},
{
"epoch": 6.89,
"learning_rate": 3.467374810318665e-06,
"loss": 0.3565,
"step": 840
},
{
"epoch": 6.93,
"learning_rate": 3.486342943854325e-06,
"loss": 0.3373,
"step": 845
},
{
"epoch": 6.97,
"learning_rate": 3.505311077389985e-06,
"loss": 0.305,
"step": 850
},
{
"epoch": 7.0,
"eval_accuracy": 0.934224049331963,
"eval_loss": 0.2515142858028412,
"eval_runtime": 11.9514,
"eval_samples_per_second": 81.413,
"eval_steps_per_second": 5.104,
"step": 854
},
{
"epoch": 7.01,
"learning_rate": 3.5242792109256453e-06,
"loss": 0.3753,
"step": 855
},
{
"epoch": 7.05,
"learning_rate": 3.543247344461305e-06,
"loss": 0.3247,
"step": 860
},
{
"epoch": 7.09,
"learning_rate": 3.5622154779969655e-06,
"loss": 0.3262,
"step": 865
},
{
"epoch": 7.13,
"learning_rate": 3.5811836115326254e-06,
"loss": 0.3195,
"step": 870
},
{
"epoch": 7.17,
"learning_rate": 3.6001517450682853e-06,
"loss": 0.2624,
"step": 875
},
{
"epoch": 7.21,
"learning_rate": 3.6191198786039456e-06,
"loss": 0.2721,
"step": 880
},
{
"epoch": 7.25,
"learning_rate": 3.6380880121396055e-06,
"loss": 0.2785,
"step": 885
},
{
"epoch": 7.3,
"learning_rate": 3.657056145675266e-06,
"loss": 0.4016,
"step": 890
},
{
"epoch": 7.34,
"learning_rate": 3.6760242792109257e-06,
"loss": 0.2788,
"step": 895
},
{
"epoch": 7.38,
"learning_rate": 3.6949924127465856e-06,
"loss": 0.3418,
"step": 900
},
{
"epoch": 7.42,
"learning_rate": 3.713960546282246e-06,
"loss": 0.3171,
"step": 905
},
{
"epoch": 7.46,
"learning_rate": 3.732928679817906e-06,
"loss": 0.3527,
"step": 910
},
{
"epoch": 7.5,
"learning_rate": 3.7518968133535666e-06,
"loss": 0.3782,
"step": 915
},
{
"epoch": 7.54,
"learning_rate": 3.7708649468892265e-06,
"loss": 0.3037,
"step": 920
},
{
"epoch": 7.58,
"learning_rate": 3.789833080424887e-06,
"loss": 0.262,
"step": 925
},
{
"epoch": 7.62,
"learning_rate": 3.8088012139605467e-06,
"loss": 0.2344,
"step": 930
},
{
"epoch": 7.66,
"learning_rate": 3.827769347496207e-06,
"loss": 0.2259,
"step": 935
},
{
"epoch": 7.7,
"learning_rate": 3.846737481031867e-06,
"loss": 0.3134,
"step": 940
},
{
"epoch": 7.75,
"learning_rate": 3.865705614567527e-06,
"loss": 0.3532,
"step": 945
},
{
"epoch": 7.79,
"learning_rate": 3.884673748103187e-06,
"loss": 0.2763,
"step": 950
},
{
"epoch": 7.83,
"learning_rate": 3.9036418816388475e-06,
"loss": 0.2797,
"step": 955
},
{
"epoch": 7.87,
"learning_rate": 3.922610015174507e-06,
"loss": 0.3303,
"step": 960
},
{
"epoch": 7.91,
"learning_rate": 3.941578148710167e-06,
"loss": 0.345,
"step": 965
},
{
"epoch": 7.95,
"learning_rate": 3.960546282245827e-06,
"loss": 0.275,
"step": 970
},
{
"epoch": 7.99,
"learning_rate": 3.979514415781487e-06,
"loss": 0.3522,
"step": 975
},
{
"epoch": 8.0,
"eval_accuracy": 0.9568345323741008,
"eval_loss": 0.16673944890499115,
"eval_runtime": 12.0466,
"eval_samples_per_second": 80.77,
"eval_steps_per_second": 5.064,
"step": 976
},
{
"epoch": 8.03,
"learning_rate": 3.998482549317148e-06,
"loss": 0.3001,
"step": 980
},
{
"epoch": 8.07,
"learning_rate": 4.017450682852808e-06,
"loss": 0.2947,
"step": 985
},
{
"epoch": 8.11,
"learning_rate": 4.036418816388468e-06,
"loss": 0.2497,
"step": 990
},
{
"epoch": 8.16,
"learning_rate": 4.0553869499241275e-06,
"loss": 0.304,
"step": 995
},
{
"epoch": 8.2,
"learning_rate": 4.074355083459787e-06,
"loss": 0.2297,
"step": 1000
},
{
"epoch": 8.24,
"learning_rate": 4.093323216995448e-06,
"loss": 0.2439,
"step": 1005
},
{
"epoch": 8.28,
"learning_rate": 4.112291350531108e-06,
"loss": 0.2543,
"step": 1010
},
{
"epoch": 8.32,
"learning_rate": 4.131259484066768e-06,
"loss": 0.3068,
"step": 1015
},
{
"epoch": 8.36,
"learning_rate": 4.150227617602428e-06,
"loss": 0.3072,
"step": 1020
},
{
"epoch": 8.4,
"learning_rate": 4.169195751138089e-06,
"loss": 0.2947,
"step": 1025
},
{
"epoch": 8.44,
"learning_rate": 4.1881638846737485e-06,
"loss": 0.2332,
"step": 1030
},
{
"epoch": 8.48,
"learning_rate": 4.207132018209408e-06,
"loss": 0.2252,
"step": 1035
},
{
"epoch": 8.52,
"learning_rate": 4.226100151745068e-06,
"loss": 0.2784,
"step": 1040
},
{
"epoch": 8.57,
"learning_rate": 4.245068285280728e-06,
"loss": 0.2483,
"step": 1045
},
{
"epoch": 8.61,
"learning_rate": 4.264036418816389e-06,
"loss": 0.2148,
"step": 1050
},
{
"epoch": 8.65,
"learning_rate": 4.283004552352049e-06,
"loss": 0.2397,
"step": 1055
},
{
"epoch": 8.69,
"learning_rate": 4.3019726858877096e-06,
"loss": 0.2262,
"step": 1060
},
{
"epoch": 8.73,
"learning_rate": 4.3209408194233695e-06,
"loss": 0.2538,
"step": 1065
},
{
"epoch": 8.77,
"learning_rate": 4.339908952959029e-06,
"loss": 0.2504,
"step": 1070
},
{
"epoch": 8.81,
"learning_rate": 4.358877086494689e-06,
"loss": 0.2153,
"step": 1075
},
{
"epoch": 8.85,
"learning_rate": 4.377845220030349e-06,
"loss": 0.2901,
"step": 1080
},
{
"epoch": 8.89,
"learning_rate": 4.39681335356601e-06,
"loss": 0.1936,
"step": 1085
},
{
"epoch": 8.93,
"learning_rate": 4.41578148710167e-06,
"loss": 0.3549,
"step": 1090
},
{
"epoch": 8.98,
"learning_rate": 4.43474962063733e-06,
"loss": 0.1782,
"step": 1095
},
{
"epoch": 9.0,
"eval_accuracy": 0.9568345323741008,
"eval_loss": 0.1493809074163437,
"eval_runtime": 11.8998,
"eval_samples_per_second": 81.766,
"eval_steps_per_second": 5.126,
"step": 1098
},
{
"epoch": 9.02,
"learning_rate": 4.45371775417299e-06,
"loss": 0.3434,
"step": 1100
},
{
"epoch": 9.06,
"learning_rate": 4.4726858877086495e-06,
"loss": 0.2269,
"step": 1105
},
{
"epoch": 9.1,
"learning_rate": 4.49165402124431e-06,
"loss": 0.2592,
"step": 1110
},
{
"epoch": 9.14,
"learning_rate": 4.51062215477997e-06,
"loss": 0.3053,
"step": 1115
},
{
"epoch": 9.18,
"learning_rate": 4.52959028831563e-06,
"loss": 0.2418,
"step": 1120
},
{
"epoch": 9.22,
"learning_rate": 4.54855842185129e-06,
"loss": 0.2827,
"step": 1125
},
{
"epoch": 9.26,
"learning_rate": 4.56752655538695e-06,
"loss": 0.1982,
"step": 1130
},
{
"epoch": 9.3,
"learning_rate": 4.5864946889226106e-06,
"loss": 0.2264,
"step": 1135
},
{
"epoch": 9.34,
"learning_rate": 4.6054628224582705e-06,
"loss": 0.1667,
"step": 1140
},
{
"epoch": 9.39,
"learning_rate": 4.62443095599393e-06,
"loss": 0.1866,
"step": 1145
},
{
"epoch": 9.43,
"learning_rate": 4.64339908952959e-06,
"loss": 0.1695,
"step": 1150
},
{
"epoch": 9.47,
"learning_rate": 4.66236722306525e-06,
"loss": 0.1994,
"step": 1155
},
{
"epoch": 9.51,
"learning_rate": 4.681335356600911e-06,
"loss": 0.2105,
"step": 1160
},
{
"epoch": 9.55,
"learning_rate": 4.700303490136571e-06,
"loss": 0.2704,
"step": 1165
},
{
"epoch": 9.59,
"learning_rate": 4.719271623672231e-06,
"loss": 0.2072,
"step": 1170
},
{
"epoch": 9.63,
"learning_rate": 4.738239757207891e-06,
"loss": 0.201,
"step": 1175
},
{
"epoch": 9.67,
"learning_rate": 4.757207890743551e-06,
"loss": 0.2818,
"step": 1180
},
{
"epoch": 9.71,
"learning_rate": 4.776176024279211e-06,
"loss": 0.2726,
"step": 1185
},
{
"epoch": 9.75,
"learning_rate": 4.795144157814872e-06,
"loss": 0.2102,
"step": 1190
},
{
"epoch": 9.8,
"learning_rate": 4.814112291350532e-06,
"loss": 0.2046,
"step": 1195
},
{
"epoch": 9.84,
"learning_rate": 4.833080424886192e-06,
"loss": 0.2335,
"step": 1200
},
{
"epoch": 9.88,
"learning_rate": 4.852048558421852e-06,
"loss": 0.1967,
"step": 1205
},
{
"epoch": 9.92,
"learning_rate": 4.871016691957512e-06,
"loss": 0.1611,
"step": 1210
},
{
"epoch": 9.96,
"learning_rate": 4.889984825493172e-06,
"loss": 0.1757,
"step": 1215
},
{
"epoch": 10.0,
"learning_rate": 4.908952959028832e-06,
"loss": 0.1849,
"step": 1220
},
{
"epoch": 10.0,
"eval_accuracy": 0.9712230215827338,
"eval_loss": 0.09722532331943512,
"eval_runtime": 11.9324,
"eval_samples_per_second": 81.543,
"eval_steps_per_second": 5.112,
"step": 1220
},
{
"epoch": 10.04,
"learning_rate": 4.927921092564492e-06,
"loss": 0.1319,
"step": 1225
},
{
"epoch": 10.08,
"learning_rate": 4.946889226100152e-06,
"loss": 0.2066,
"step": 1230
},
{
"epoch": 10.12,
"learning_rate": 4.965857359635812e-06,
"loss": 0.1481,
"step": 1235
},
{
"epoch": 10.16,
"learning_rate": 4.984825493171473e-06,
"loss": 0.1144,
"step": 1240
},
{
"epoch": 10.2,
"learning_rate": 4.965753424657534e-06,
"loss": 0.2347,
"step": 1245
},
{
"epoch": 10.25,
"learning_rate": 4.7945205479452054e-06,
"loss": 0.1637,
"step": 1250
},
{
"epoch": 10.29,
"learning_rate": 4.6232876712328774e-06,
"loss": 0.2435,
"step": 1255
},
{
"epoch": 10.33,
"learning_rate": 4.4520547945205486e-06,
"loss": 0.1673,
"step": 1260
},
{
"epoch": 10.37,
"learning_rate": 4.28082191780822e-06,
"loss": 0.1852,
"step": 1265
},
{
"epoch": 10.41,
"learning_rate": 4.109589041095891e-06,
"loss": 0.2216,
"step": 1270
},
{
"epoch": 10.45,
"learning_rate": 3.938356164383562e-06,
"loss": 0.1129,
"step": 1275
},
{
"epoch": 10.49,
"learning_rate": 3.767123287671233e-06,
"loss": 0.1451,
"step": 1280
},
{
"epoch": 10.53,
"learning_rate": 3.5958904109589043e-06,
"loss": 0.1597,
"step": 1285
},
{
"epoch": 10.57,
"learning_rate": 3.4246575342465754e-06,
"loss": 0.1701,
"step": 1290
},
{
"epoch": 10.61,
"learning_rate": 3.253424657534247e-06,
"loss": 0.1232,
"step": 1295
},
{
"epoch": 10.66,
"learning_rate": 3.082191780821918e-06,
"loss": 0.1724,
"step": 1300
},
{
"epoch": 10.7,
"learning_rate": 2.9109589041095893e-06,
"loss": 0.1786,
"step": 1305
},
{
"epoch": 10.74,
"learning_rate": 2.7397260273972604e-06,
"loss": 0.1959,
"step": 1310
},
{
"epoch": 10.78,
"learning_rate": 2.568493150684932e-06,
"loss": 0.2141,
"step": 1315
},
{
"epoch": 10.82,
"learning_rate": 2.3972602739726027e-06,
"loss": 0.2618,
"step": 1320
},
{
"epoch": 10.86,
"learning_rate": 2.2260273972602743e-06,
"loss": 0.233,
"step": 1325
},
{
"epoch": 10.9,
"learning_rate": 2.0547945205479454e-06,
"loss": 0.1443,
"step": 1330
},
{
"epoch": 10.94,
"learning_rate": 1.8835616438356166e-06,
"loss": 0.1439,
"step": 1335
},
{
"epoch": 10.98,
"learning_rate": 1.7123287671232877e-06,
"loss": 0.1814,
"step": 1340
},
{
"epoch": 11.0,
"eval_accuracy": 0.9845837615621789,
"eval_loss": 0.05592276155948639,
"eval_runtime": 11.975,
"eval_samples_per_second": 81.252,
"eval_steps_per_second": 5.094,
"step": 1342
},
{
"epoch": 11.02,
"learning_rate": 1.541095890410959e-06,
"loss": 0.1784,
"step": 1345
},
{
"epoch": 11.07,
"learning_rate": 1.3698630136986302e-06,
"loss": 0.1495,
"step": 1350
},
{
"epoch": 11.11,
"learning_rate": 1.1986301369863014e-06,
"loss": 0.1872,
"step": 1355
},
{
"epoch": 11.15,
"learning_rate": 1.0273972602739727e-06,
"loss": 0.0665,
"step": 1360
},
{
"epoch": 11.19,
"learning_rate": 8.561643835616439e-07,
"loss": 0.108,
"step": 1365
},
{
"epoch": 11.23,
"learning_rate": 6.849315068493151e-07,
"loss": 0.1019,
"step": 1370
},
{
"epoch": 11.27,
"learning_rate": 5.136986301369864e-07,
"loss": 0.1441,
"step": 1375
},
{
"epoch": 11.31,
"learning_rate": 3.4246575342465755e-07,
"loss": 0.1156,
"step": 1380
},
{
"epoch": 11.35,
"learning_rate": 1.7123287671232878e-07,
"loss": 0.1437,
"step": 1385
},
{
"epoch": 11.39,
"learning_rate": 0.0,
"loss": 0.1169,
"step": 1390
},
{
"epoch": 11.43,
"learning_rate": 0.0,
"loss": 0.1776,
"step": 1395
},
{
"epoch": 11.48,
"learning_rate": 0.0,
"loss": 0.0659,
"step": 1400
},
{
"epoch": 11.52,
"learning_rate": 0.0,
"loss": 0.0891,
"step": 1405
},
{
"epoch": 11.56,
"learning_rate": 0.0,
"loss": 0.1266,
"step": 1410
},
{
"epoch": 11.6,
"learning_rate": 0.0,
"loss": 0.1135,
"step": 1415
},
{
"epoch": 11.64,
"learning_rate": 0.0,
"loss": 0.1652,
"step": 1420
},
{
"epoch": 11.68,
"learning_rate": 0.0,
"loss": 0.2009,
"step": 1425
},
{
"epoch": 11.72,
"learning_rate": 0.0,
"loss": 0.1761,
"step": 1430
},
{
"epoch": 11.76,
"learning_rate": 0.0,
"loss": 0.182,
"step": 1435
},
{
"epoch": 11.8,
"learning_rate": 0.0,
"loss": 0.1208,
"step": 1440
},
{
"epoch": 11.84,
"learning_rate": 0.0,
"loss": 0.1124,
"step": 1445
},
{
"epoch": 11.89,
"learning_rate": 0.0,
"loss": 0.1923,
"step": 1450
},
{
"epoch": 11.93,
"learning_rate": 0.0,
"loss": 0.0787,
"step": 1455
},
{
"epoch": 11.97,
"learning_rate": 0.0,
"loss": 0.1682,
"step": 1460
},
{
"epoch": 12.0,
"eval_accuracy": 0.9856115107913669,
"eval_loss": 0.05682051554322243,
"eval_runtime": 12.083,
"eval_samples_per_second": 80.527,
"eval_steps_per_second": 5.048,
"step": 1464
},
{
"epoch": 12.0,
"step": 1464,
"total_flos": 1.2834594613117477e+19,
"train_loss": 0.6339893028180782,
"train_runtime": 3472.73,
"train_samples_per_second": 13.445,
"train_steps_per_second": 0.422
}
],
"max_steps": 1464,
"num_train_epochs": 12,
"total_flos": 1.2834594613117477e+19,
"trial_name": null,
"trial_params": null
}