eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
6d2c1bd
raw
history blame
22.6 kB
{
"best_metric": 0.21034620702266693,
"best_model_checkpoint": "output/eminem/checkpoint-912",
"epoch": 2.0,
"global_step": 912,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.00010098945265056683,
"loss": 0.4239,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 9.890598330141942e-05,
"loss": 0.484,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 9.678717853140213e-05,
"loss": 0.4876,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 9.46355087720941e-05,
"loss": 0.4738,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 9.245348277418726e-05,
"loss": 0.4579,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 9.024364468238985e-05,
"loss": 0.4577,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 8.800857106906198e-05,
"loss": 0.4851,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 8.575086793006042e-05,
"loss": 0.4626,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 8.34731676462491e-05,
"loss": 0.4782,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 8.117812591427207e-05,
"loss": 0.5103,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 7.886841865013191e-05,
"loss": 0.5388,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 7.654673886917425e-05,
"loss": 0.4255,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 7.421579354617336e-05,
"loss": 0.4845,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 7.187830045910335e-05,
"loss": 0.4882,
"step": 70
},
{
"epoch": 0.16,
"learning_rate": 6.953698502034407e-05,
"loss": 0.4911,
"step": 75
},
{
"epoch": 0.17,
"learning_rate": 6.719457709896656e-05,
"loss": 0.445,
"step": 80
},
{
"epoch": 0.18,
"learning_rate": 6.485380783782185e-05,
"loss": 0.4824,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 6.251740646916227e-05,
"loss": 0.478,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 6.0188097132472896e-05,
"loss": 0.5299,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 5.786859569824095e-05,
"loss": 0.5087,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 5.556160660138586e-05,
"loss": 0.4854,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 5.326981968799179e-05,
"loss": 0.5017,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 5.099590707908848e-05,
"loss": 0.4689,
"step": 115
},
{
"epoch": 0.26,
"learning_rate": 4.874252005506042e-05,
"loss": 0.4921,
"step": 120
},
{
"epoch": 0.27,
"learning_rate": 4.651228596437459e-05,
"loss": 0.4873,
"step": 125
},
{
"epoch": 0.28,
"learning_rate": 4.4307805160220805e-05,
"loss": 0.4682,
"step": 130
},
{
"epoch": 0.29,
"learning_rate": 4.2131647968601464e-05,
"loss": 0.5228,
"step": 135
},
{
"epoch": 0.3,
"learning_rate": 3.998635169145922e-05,
"loss": 0.4468,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 3.7874417648290806e-05,
"loss": 0.452,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 3.57983082597153e-05,
"loss": 0.4363,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 3.376044417641229e-05,
"loss": 0.4868,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 3.176320145674611e-05,
"loss": 0.4906,
"step": 160
},
{
"epoch": 0.36,
"learning_rate": 2.98089087963816e-05,
"loss": 0.5023,
"step": 165
},
{
"epoch": 0.37,
"learning_rate": 2.7899844813147157e-05,
"loss": 0.4712,
"step": 170
},
{
"epoch": 0.38,
"learning_rate": 2.603823539024549e-05,
"loss": 0.4571,
"step": 175
},
{
"epoch": 0.39,
"learning_rate": 2.422625108099115e-05,
"loss": 0.4262,
"step": 180
},
{
"epoch": 0.4,
"learning_rate": 2.2466004578035872e-05,
"loss": 0.455,
"step": 185
},
{
"epoch": 0.41,
"learning_rate": 2.0759548250059194e-05,
"loss": 0.4506,
"step": 190
},
{
"epoch": 0.42,
"learning_rate": 1.910887174880859e-05,
"loss": 0.4219,
"step": 195
},
{
"epoch": 0.43,
"learning_rate": 1.7515899689243134e-05,
"loss": 0.4164,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 1.5982489405528287e-05,
"loss": 0.4133,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 1.4510428785463036e-05,
"loss": 0.4375,
"step": 210
},
{
"epoch": 0.47,
"learning_rate": 1.3101434185879184e-05,
"loss": 0.4882,
"step": 215
},
{
"epoch": 0.48,
"learning_rate": 1.175714843145347e-05,
"loss": 0.445,
"step": 220
},
{
"epoch": 0.49,
"learning_rate": 1.0479138899242488e-05,
"loss": 0.46,
"step": 225
},
{
"epoch": 0.5,
"learning_rate": 9.268895691192436e-06,
"loss": 0.5074,
"step": 230
},
{
"epoch": 0.51,
"learning_rate": 8.127829896747804e-06,
"loss": 0.4527,
"step": 235
},
{
"epoch": 0.52,
"learning_rate": 7.057271947580458e-06,
"loss": 0.4939,
"step": 240
},
{
"epoch": 0.53,
"learning_rate": 6.058470066371539e-06,
"loss": 0.5338,
"step": 245
},
{
"epoch": 0.54,
"learning_rate": 5.132588811436518e-06,
"loss": 0.4298,
"step": 250
},
{
"epoch": 0.55,
"learning_rate": 4.280707718900271e-06,
"loss": 0.4572,
"step": 255
},
{
"epoch": 0.57,
"learning_rate": 3.5038200440114727e-06,
"loss": 0.4096,
"step": 260
},
{
"epoch": 0.58,
"learning_rate": 2.802831603045737e-06,
"loss": 0.5071,
"step": 265
},
{
"epoch": 0.59,
"learning_rate": 2.178559717169051e-06,
"loss": 0.444,
"step": 270
},
{
"epoch": 0.6,
"learning_rate": 1.6317322594756245e-06,
"loss": 0.4421,
"step": 275
},
{
"epoch": 0.61,
"learning_rate": 1.162986806318977e-06,
"loss": 0.4878,
"step": 280
},
{
"epoch": 0.62,
"learning_rate": 7.72869893930971e-07,
"loss": 0.4151,
"step": 285
},
{
"epoch": 0.63,
"learning_rate": 4.6183638117953145e-07,
"loss": 0.4066,
"step": 290
},
{
"epoch": 0.64,
"learning_rate": 2.3024891922762553e-07,
"loss": 0.4453,
"step": 295
},
{
"epoch": 0.65,
"learning_rate": 7.837752869653465e-08,
"loss": 0.5799,
"step": 300
},
{
"epoch": 0.66,
"learning_rate": 6.399284833793284e-09,
"loss": 0.4665,
"step": 305
},
{
"epoch": 0.67,
"learning_rate": 1.4398111052152874e-08,
"loss": 0.4856,
"step": 310
},
{
"epoch": 0.68,
"learning_rate": 1.0236468107794785e-07,
"loss": 0.4853,
"step": 315
},
{
"epoch": 0.7,
"learning_rate": 2.7019642982561925e-07,
"loss": 0.5332,
"step": 320
},
{
"epoch": 0.71,
"learning_rate": 5.176976729833616e-07,
"loss": 0.4575,
"step": 325
},
{
"epoch": 0.72,
"learning_rate": 8.445798351733662e-07,
"loss": 0.4864,
"step": 330
},
{
"epoch": 0.73,
"learning_rate": 1.2504617864165704e-06,
"loss": 0.442,
"step": 335
},
{
"epoch": 0.74,
"learning_rate": 1.7348702865138632e-06,
"loss": 0.4695,
"step": 340
},
{
"epoch": 0.75,
"learning_rate": 2.297240536822864e-06,
"loss": 0.5056,
"step": 345
},
{
"epoch": 0.76,
"learning_rate": 2.936916838787285e-06,
"loss": 0.4522,
"step": 350
},
{
"epoch": 0.77,
"learning_rate": 3.6531533584543354e-06,
"loss": 0.4949,
"step": 355
},
{
"epoch": 0.78,
"learning_rate": 4.445114996080512e-06,
"loss": 0.4966,
"step": 360
},
{
"epoch": 0.79,
"learning_rate": 5.3118783598219064e-06,
"loss": 0.493,
"step": 365
},
{
"epoch": 0.8,
"learning_rate": 6.252432842370185e-06,
"loss": 0.4831,
"step": 370
},
{
"epoch": 0.82,
"learning_rate": 7.265681799270346e-06,
"loss": 0.4715,
"step": 375
},
{
"epoch": 0.83,
"learning_rate": 8.350443827565338e-06,
"loss": 0.4506,
"step": 380
},
{
"epoch": 0.84,
"learning_rate": 9.505454143253312e-06,
"loss": 0.4343,
"step": 385
},
{
"epoch": 0.85,
"learning_rate": 1.072936605597144e-05,
"loss": 0.4487,
"step": 390
},
{
"epoch": 0.86,
"learning_rate": 1.2020752539181091e-05,
"loss": 0.5072,
"step": 395
},
{
"epoch": 0.87,
"learning_rate": 1.3378107894005304e-05,
"loss": 0.447,
"step": 400
},
{
"epoch": 0.88,
"learning_rate": 1.479984950482134e-05,
"loss": 0.4754,
"step": 405
},
{
"epoch": 0.89,
"learning_rate": 1.6284319684507726e-05,
"loss": 0.4414,
"step": 410
},
{
"epoch": 0.9,
"learning_rate": 1.782978760723747e-05,
"loss": 0.4684,
"step": 415
},
{
"epoch": 0.91,
"learning_rate": 1.9434451326547662e-05,
"loss": 0.5108,
"step": 420
},
{
"epoch": 0.92,
"learning_rate": 2.109643987632069e-05,
"loss": 0.4894,
"step": 425
},
{
"epoch": 0.93,
"learning_rate": 2.2813815452250717e-05,
"loss": 0.5156,
"step": 430
},
{
"epoch": 0.95,
"learning_rate": 2.4584575671241315e-05,
"loss": 0.4643,
"step": 435
},
{
"epoch": 0.96,
"learning_rate": 2.640665590608444e-05,
"loss": 0.457,
"step": 440
},
{
"epoch": 0.97,
"learning_rate": 2.8277931692735254e-05,
"loss": 0.4539,
"step": 445
},
{
"epoch": 0.98,
"learning_rate": 3.0196221207331556e-05,
"loss": 0.4085,
"step": 450
},
{
"epoch": 0.99,
"learning_rate": 3.215928781010894e-05,
"loss": 0.4488,
"step": 455
},
{
"epoch": 1.0,
"learning_rate": 3.416484265322113e-05,
"loss": 0.4469,
"step": 460
},
{
"epoch": 1.0,
"eval_loss": 0.2455403357744217,
"eval_runtime": 13.8337,
"eval_samples_per_second": 43.661,
"eval_steps_per_second": 5.494,
"step": 460
},
{
"epoch": 1.02,
"learning_rate": 0.0001310364026633813,
"loss": 0.4239,
"step": 465
},
{
"epoch": 1.03,
"learning_rate": 0.0001319781359302741,
"loss": 0.5115,
"step": 470
},
{
"epoch": 1.04,
"learning_rate": 0.00013284467114611735,
"loss": 0.4756,
"step": 475
},
{
"epoch": 1.05,
"learning_rate": 0.00013363498016834413,
"loss": 0.4892,
"step": 480
},
{
"epoch": 1.06,
"learning_rate": 0.00013434812529663595,
"loss": 0.5002,
"step": 485
},
{
"epoch": 1.07,
"learning_rate": 0.00013498326038550478,
"loss": 0.5714,
"step": 490
},
{
"epoch": 1.09,
"learning_rate": 0.00013553963184824178,
"loss": 0.5845,
"step": 495
},
{
"epoch": 1.1,
"learning_rate": 0.0001360165795510487,
"loss": 0.4911,
"step": 500
},
{
"epoch": 1.11,
"learning_rate": 0.00013641353759628265,
"loss": 0.5275,
"step": 505
},
{
"epoch": 1.12,
"learning_rate": 0.00013673003499389456,
"loss": 0.5292,
"step": 510
},
{
"epoch": 1.13,
"learning_rate": 0.00013696569622025749,
"loss": 0.5506,
"step": 515
},
{
"epoch": 1.14,
"learning_rate": 0.00013712024166372548,
"loss": 0.5552,
"step": 520
},
{
"epoch": 1.15,
"learning_rate": 0.0001371934879563911,
"loss": 0.5404,
"step": 525
},
{
"epoch": 1.16,
"learning_rate": 0.0001371853481916521,
"loss": 0.5364,
"step": 530
},
{
"epoch": 1.17,
"learning_rate": 0.0001370958320273255,
"loss": 0.4954,
"step": 535
},
{
"epoch": 1.18,
"learning_rate": 0.00013692504567418905,
"loss": 0.6414,
"step": 540
},
{
"epoch": 1.2,
"learning_rate": 0.0001366731917699612,
"loss": 0.5806,
"step": 545
},
{
"epoch": 1.21,
"learning_rate": 0.00013634056913887406,
"loss": 0.5283,
"step": 550
},
{
"epoch": 1.22,
"learning_rate": 0.0001359275724371159,
"loss": 0.5036,
"step": 555
},
{
"epoch": 1.23,
"learning_rate": 0.00013543469168457526,
"loss": 0.583,
"step": 560
},
{
"epoch": 1.24,
"learning_rate": 0.00013486251168343032,
"loss": 0.5384,
"step": 565
},
{
"epoch": 1.25,
"learning_rate": 0.0001342117113242847,
"loss": 0.5136,
"step": 570
},
{
"epoch": 1.26,
"learning_rate": 0.00013348306278066345,
"loss": 0.5162,
"step": 575
},
{
"epoch": 1.27,
"learning_rate": 0.0001326774305928314,
"loss": 0.4983,
"step": 580
},
{
"epoch": 1.28,
"learning_rate": 0.00013179577064201505,
"loss": 0.5491,
"step": 585
},
{
"epoch": 1.29,
"learning_rate": 0.00013083912901625498,
"loss": 0.4611,
"step": 590
},
{
"epoch": 1.3,
"learning_rate": 0.0001298086407692224,
"loss": 0.5359,
"step": 595
},
{
"epoch": 1.32,
"learning_rate": 0.00012870552857347723,
"loss": 0.5226,
"step": 600
},
{
"epoch": 1.33,
"learning_rate": 0.00012753110126977415,
"loss": 0.5316,
"step": 605
},
{
"epoch": 1.34,
"learning_rate": 0.00012628675231412197,
"loss": 0.5227,
"step": 610
},
{
"epoch": 1.35,
"learning_rate": 0.0001249739581244523,
"loss": 0.4961,
"step": 615
},
{
"epoch": 1.36,
"learning_rate": 0.00012359427632884786,
"loss": 0.5487,
"step": 620
},
{
"epoch": 1.37,
"learning_rate": 0.0001221493439174271,
"loss": 0.5994,
"step": 625
},
{
"epoch": 1.38,
"learning_rate": 0.00012064087530004776,
"loss": 0.5401,
"step": 630
},
{
"epoch": 1.39,
"learning_rate": 0.00011907066027217711,
"loss": 0.5615,
"step": 635
},
{
"epoch": 1.4,
"learning_rate": 0.00011744056189128525,
"loss": 0.5705,
"step": 640
},
{
"epoch": 1.41,
"learning_rate": 0.00011575251426634572,
"loss": 0.5663,
"step": 645
},
{
"epoch": 1.43,
"learning_rate": 0.00011400852026300672,
"loss": 0.5623,
"step": 650
},
{
"epoch": 1.44,
"learning_rate": 0.00011221064912720382,
"loss": 0.5898,
"step": 655
},
{
"epoch": 1.45,
"learning_rate": 0.00011036103402999931,
"loss": 0.5708,
"step": 660
},
{
"epoch": 1.46,
"learning_rate": 0.00010846186953657361,
"loss": 0.5299,
"step": 665
},
{
"epoch": 1.47,
"learning_rate": 0.00010651540900238634,
"loss": 0.6369,
"step": 670
},
{
"epoch": 1.48,
"learning_rate": 0.00010452396189956808,
"loss": 0.532,
"step": 675
},
{
"epoch": 1.49,
"learning_rate": 0.0001024898910767386,
"loss": 0.5957,
"step": 680
},
{
"epoch": 1.5,
"learning_rate": 0.00010041560995548433,
"loss": 0.4907,
"step": 685
},
{
"epoch": 1.51,
"learning_rate": 9.830357966684779e-05,
"loss": 0.5726,
"step": 690
},
{
"epoch": 1.52,
"learning_rate": 9.615630613119297e-05,
"loss": 0.5582,
"step": 695
},
{
"epoch": 1.54,
"learning_rate": 9.397633708493872e-05,
"loss": 0.6085,
"step": 700
},
{
"epoch": 1.55,
"learning_rate": 9.17662590576683e-05,
"loss": 0.6075,
"step": 705
},
{
"epoch": 1.56,
"learning_rate": 8.95286943032199e-05,
"loss": 0.5199,
"step": 710
},
{
"epoch": 1.57,
"learning_rate": 8.726629768839068e-05,
"loss": 0.565,
"step": 715
},
{
"epoch": 1.58,
"learning_rate": 8.49817535429363e-05,
"loss": 0.5779,
"step": 720
},
{
"epoch": 1.59,
"learning_rate": 8.267777247462197e-05,
"loss": 0.514,
"step": 725
},
{
"epoch": 1.6,
"learning_rate": 8.035708815308325e-05,
"loss": 0.6287,
"step": 730
},
{
"epoch": 1.61,
"learning_rate": 7.802245406634311e-05,
"loss": 0.5617,
"step": 735
},
{
"epoch": 1.62,
"learning_rate": 7.56766402537954e-05,
"loss": 0.6135,
"step": 740
},
{
"epoch": 1.63,
"learning_rate": 7.332243001956172e-05,
"loss": 0.5151,
"step": 745
},
{
"epoch": 1.64,
"learning_rate": 7.096261663009792e-05,
"loss": 0.5282,
"step": 750
},
{
"epoch": 1.66,
"learning_rate": 6.85999999999999e-05,
"loss": 0.5488,
"step": 755
},
{
"epoch": 1.67,
"learning_rate": 6.623738336990383e-05,
"loss": 0.5678,
"step": 760
},
{
"epoch": 1.68,
"learning_rate": 6.387756998044003e-05,
"loss": 0.5534,
"step": 765
},
{
"epoch": 1.69,
"learning_rate": 6.152335974620634e-05,
"loss": 0.6249,
"step": 770
},
{
"epoch": 1.7,
"learning_rate": 5.917754593365669e-05,
"loss": 0.5588,
"step": 775
},
{
"epoch": 1.71,
"learning_rate": 5.684291184691846e-05,
"loss": 0.4985,
"step": 780
},
{
"epoch": 1.72,
"learning_rate": 5.452222752537974e-05,
"loss": 0.545,
"step": 785
},
{
"epoch": 1.73,
"learning_rate": 5.2218246457065396e-05,
"loss": 0.5905,
"step": 790
},
{
"epoch": 1.74,
"learning_rate": 4.993370231160913e-05,
"loss": 0.4741,
"step": 795
},
{
"epoch": 1.75,
"learning_rate": 4.767130569678176e-05,
"loss": 0.5473,
"step": 800
},
{
"epoch": 1.77,
"learning_rate": 4.543374094233335e-05,
"loss": 0.525,
"step": 805
},
{
"epoch": 1.78,
"learning_rate": 4.3223662915062914e-05,
"loss": 0.5134,
"step": 810
},
{
"epoch": 1.79,
"learning_rate": 4.104369386880686e-05,
"loss": 0.5726,
"step": 815
},
{
"epoch": 1.8,
"learning_rate": 3.8896420333152035e-05,
"loss": 0.4985,
"step": 820
},
{
"epoch": 1.81,
"learning_rate": 3.6784390044517205e-05,
"loss": 0.5251,
"step": 825
},
{
"epoch": 1.82,
"learning_rate": 3.471010892326291e-05,
"loss": 0.6068,
"step": 830
},
{
"epoch": 1.83,
"learning_rate": 3.267603810043175e-05,
"loss": 0.5227,
"step": 835
},
{
"epoch": 1.84,
"learning_rate": 3.068459099761349e-05,
"loss": 0.5061,
"step": 840
},
{
"epoch": 1.85,
"learning_rate": 2.8738130463427823e-05,
"loss": 0.5539,
"step": 845
},
{
"epoch": 1.86,
"learning_rate": 2.683896597000207e-05,
"loss": 0.5975,
"step": 850
},
{
"epoch": 1.88,
"learning_rate": 2.4989350872796035e-05,
"loss": 0.5843,
"step": 855
},
{
"epoch": 1.89,
"learning_rate": 2.3191479736993138e-05,
"loss": 0.571,
"step": 860
},
{
"epoch": 1.9,
"learning_rate": 2.1447485733655544e-05,
"loss": 0.5282,
"step": 865
},
{
"epoch": 1.91,
"learning_rate": 1.975943810871599e-05,
"loss": 0.5626,
"step": 870
},
{
"epoch": 1.92,
"learning_rate": 1.8129339727824064e-05,
"loss": 0.5394,
"step": 875
},
{
"epoch": 1.93,
"learning_rate": 1.6559124699952118e-05,
"loss": 0.4747,
"step": 880
},
{
"epoch": 1.94,
"learning_rate": 1.5050656082574002e-05,
"loss": 0.5601,
"step": 885
},
{
"epoch": 1.95,
"learning_rate": 1.3605723671153187e-05,
"loss": 0.4725,
"step": 890
},
{
"epoch": 1.96,
"learning_rate": 1.22260418755487e-05,
"loss": 0.5171,
"step": 895
},
{
"epoch": 1.97,
"learning_rate": 1.0913247685877927e-05,
"loss": 0.5425,
"step": 900
},
{
"epoch": 1.98,
"learning_rate": 9.66889873022574e-06,
"loss": 0.4969,
"step": 905
},
{
"epoch": 2.0,
"learning_rate": 8.494471426523612e-06,
"loss": 0.5235,
"step": 910
},
{
"epoch": 2.0,
"eval_loss": 0.21034620702266693,
"eval_runtime": 15.3022,
"eval_samples_per_second": 41.432,
"eval_steps_per_second": 5.228,
"step": 912
}
],
"max_steps": 2736,
"num_train_epochs": 6,
"total_flos": 951886872576000.0,
"trial_name": null,
"trial_params": null
}