eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
d9d24b5
raw
history blame
23.3 kB
{
"best_metric": 0.37163448333740234,
"best_model_checkpoint": "output/eminem/checkpoint-924",
"epoch": 2.0,
"global_step": 924,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.0001144556130966329,
"loss": 0.7526,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 0.00011618189922507932,
"loss": 0.7708,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 0.00011785197628750708,
"loss": 0.869,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 0.00011946387140192834,
"loss": 0.8428,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 0.00012101568041738691,
"loss": 0.7731,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 0.00012250557016335326,
"loss": 0.8297,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 0.00012393178061527061,
"loss": 0.8621,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 0.00012529262697369198,
"loss": 0.7284,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 0.00012658650165455924,
"loss": 0.7879,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 0.00012781187618824956,
"loss": 0.9203,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 0.00012896730302519073,
"loss": 0.8996,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 0.0001300514172458524,
"loss": 0.949,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 0.0001310629381731565,
"loss": 0.8166,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 0.0001320006708853459,
"loss": 0.8125,
"step": 70
},
{
"epoch": 0.16,
"learning_rate": 0.00013286350762756457,
"loss": 0.8287,
"step": 75
},
{
"epoch": 0.18,
"learning_rate": 0.0001336504291204605,
"loss": 0.8779,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 0.00013436050576427267,
"loss": 0.8684,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 0.00013499289873697794,
"loss": 0.8235,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 0.00013554686098520387,
"loss": 0.9249,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 0.00013602173810672657,
"loss": 0.8379,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 0.0001364169691235302,
"loss": 0.9086,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 0.0001367320871444924,
"loss": 0.8637,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 0.0001369667199169302,
"loss": 0.8941,
"step": 115
},
{
"epoch": 0.26,
"learning_rate": 0.00013712059026634634,
"loss": 0.8238,
"step": 120
},
{
"epoch": 0.27,
"learning_rate": 0.00013719351642385878,
"loss": 0.8578,
"step": 125
},
{
"epoch": 0.28,
"learning_rate": 0.0001371854122409264,
"loss": 0.8231,
"step": 130
},
{
"epoch": 0.3,
"learning_rate": 0.0001370962872911174,
"loss": 0.815,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 0.00013692624685879965,
"loss": 0.726,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 0.00013667549181476735,
"loss": 0.8906,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 0.00013634431837894983,
"loss": 0.7687,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 0.0001359331177704834,
"loss": 0.8484,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 0.00013544237574555944,
"loss": 0.9342,
"step": 160
},
{
"epoch": 0.36,
"learning_rate": 0.00013487267202359548,
"loss": 0.887,
"step": 165
},
{
"epoch": 0.37,
"learning_rate": 0.00013422467960240306,
"loss": 0.9549,
"step": 170
},
{
"epoch": 0.38,
"learning_rate": 0.00013349916396317312,
"loss": 1.0232,
"step": 175
},
{
"epoch": 0.39,
"learning_rate": 0.00013269698216619567,
"loss": 0.856,
"step": 180
},
{
"epoch": 0.4,
"learning_rate": 0.00013181908183841387,
"loss": 0.8467,
"step": 185
},
{
"epoch": 0.42,
"learning_rate": 0.0001308665000539712,
"loss": 0.8928,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 0.00012984036210911092,
"loss": 0.9901,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 0.00012874188019284667,
"loss": 0.79,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 0.0001275723519549898,
"loss": 0.8893,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 0.00012633315897322012,
"loss": 0.9893,
"step": 210
},
{
"epoch": 0.47,
"learning_rate": 0.0001250257651210129,
"loss": 0.8199,
"step": 215
},
{
"epoch": 0.48,
"learning_rate": 0.00012365171483834196,
"loss": 0.8304,
"step": 220
},
{
"epoch": 0.49,
"learning_rate": 0.00012221263130722672,
"loss": 0.924,
"step": 225
},
{
"epoch": 0.5,
"learning_rate": 0.00012071021453423055,
"loss": 0.8617,
"step": 230
},
{
"epoch": 0.51,
"learning_rate": 0.00011914623934224188,
"loss": 0.8712,
"step": 235
},
{
"epoch": 0.53,
"learning_rate": 0.00011752255327383619,
"loss": 1.0037,
"step": 240
},
{
"epoch": 0.54,
"learning_rate": 0.00011584107440876567,
"loss": 0.8259,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 0.0001141037890981024,
"loss": 0.8382,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 0.00011231274961773941,
"loss": 1.0107,
"step": 255
},
{
"epoch": 0.57,
"learning_rate": 0.00011047007174401259,
"loss": 0.9122,
"step": 260
},
{
"epoch": 0.58,
"learning_rate": 0.00010857793225431036,
"loss": 0.9962,
"step": 265
},
{
"epoch": 0.59,
"learning_rate": 0.00010663856635561276,
"loss": 0.7623,
"step": 270
},
{
"epoch": 0.6,
"learning_rate": 0.0001046542650440321,
"loss": 0.9404,
"step": 275
},
{
"epoch": 0.61,
"learning_rate": 0.00010262737239840785,
"loss": 0.9339,
"step": 280
},
{
"epoch": 0.62,
"learning_rate": 0.00010056028281124445,
"loss": 0.9049,
"step": 285
},
{
"epoch": 0.63,
"learning_rate": 9.845543816016187e-05,
"loss": 1.0607,
"step": 290
},
{
"epoch": 0.65,
"learning_rate": 9.63153249232949e-05,
"loss": 0.8714,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 9.414247124197839e-05,
"loss": 0.9058,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 9.19394439342254e-05,
"loss": 0.9844,
"step": 305
},
{
"epoch": 0.68,
"learning_rate": 8.970884546251429e-05,
"loss": 0.8605,
"step": 310
},
{
"epoch": 0.69,
"learning_rate": 8.74533108594708e-05,
"loss": 1.0247,
"step": 315
},
{
"epoch": 0.7,
"learning_rate": 8.517550461506486e-05,
"loss": 0.8763,
"step": 320
},
{
"epoch": 0.71,
"learning_rate": 8.287811752902797e-05,
"loss": 0.9301,
"step": 325
},
{
"epoch": 0.72,
"learning_rate": 8.056386353217202e-05,
"loss": 0.8877,
"step": 330
},
{
"epoch": 0.73,
"learning_rate": 7.823547648039207e-05,
"loss": 0.7894,
"step": 335
},
{
"epoch": 0.74,
"learning_rate": 7.589570692512989e-05,
"loss": 1.0017,
"step": 340
},
{
"epoch": 0.75,
"learning_rate": 7.354731886411453e-05,
"loss": 0.9258,
"step": 345
},
{
"epoch": 0.77,
"learning_rate": 7.119308647622086e-05,
"loss": 0.8299,
"step": 350
},
{
"epoch": 0.78,
"learning_rate": 6.883579084428906e-05,
"loss": 0.936,
"step": 355
},
{
"epoch": 0.79,
"learning_rate": 6.64782166698198e-05,
"loss": 0.8412,
"step": 360
},
{
"epoch": 0.8,
"learning_rate": 6.412314898334569e-05,
"loss": 0.9055,
"step": 365
},
{
"epoch": 0.81,
"learning_rate": 6.17733698544629e-05,
"loss": 0.8395,
"step": 370
},
{
"epoch": 0.82,
"learning_rate": 5.943165510532945e-05,
"loss": 0.8795,
"step": 375
},
{
"epoch": 0.83,
"learning_rate": 5.710077103155516e-05,
"loss": 0.9003,
"step": 380
},
{
"epoch": 0.84,
"learning_rate": 5.4783471134343704e-05,
"loss": 0.7688,
"step": 385
},
{
"epoch": 0.85,
"learning_rate": 5.248249286775053e-05,
"loss": 0.8768,
"step": 390
},
{
"epoch": 0.86,
"learning_rate": 5.02005544048863e-05,
"loss": 0.8473,
"step": 395
},
{
"epoch": 0.88,
"learning_rate": 4.7940351426927284e-05,
"loss": 0.9321,
"step": 400
},
{
"epoch": 0.89,
"learning_rate": 4.5704553938646724e-05,
"loss": 0.7829,
"step": 405
},
{
"epoch": 0.9,
"learning_rate": 4.349580311433621e-05,
"loss": 0.9309,
"step": 410
},
{
"epoch": 0.91,
"learning_rate": 4.1316708177727135e-05,
"loss": 0.8959,
"step": 415
},
{
"epoch": 0.92,
"learning_rate": 3.9169843319703834e-05,
"loss": 1.0492,
"step": 420
},
{
"epoch": 0.93,
"learning_rate": 3.7057744657371865e-05,
"loss": 0.8489,
"step": 425
},
{
"epoch": 0.94,
"learning_rate": 3.4982907238114506e-05,
"loss": 0.8533,
"step": 430
},
{
"epoch": 0.95,
"learning_rate": 3.294778209216339e-05,
"loss": 0.8134,
"step": 435
},
{
"epoch": 0.96,
"learning_rate": 3.0954773337168915e-05,
"loss": 0.715,
"step": 440
},
{
"epoch": 0.97,
"learning_rate": 2.9006235338179113e-05,
"loss": 0.8985,
"step": 445
},
{
"epoch": 0.98,
"learning_rate": 2.7104469926418013e-05,
"loss": 0.7858,
"step": 450
},
{
"epoch": 1.0,
"learning_rate": 2.5251723680080823e-05,
"loss": 0.8067,
"step": 455
},
{
"epoch": 1.0,
"eval_loss": 0.5359192490577698,
"eval_runtime": 35.4879,
"eval_samples_per_second": 17.527,
"eval_steps_per_second": 2.198,
"step": 457
},
{
"epoch": 1.02,
"learning_rate": 7.411387289257788e-05,
"loss": 0.7428,
"step": 460
},
{
"epoch": 1.04,
"learning_rate": 7.650217107773121e-05,
"loss": 0.7294,
"step": 465
},
{
"epoch": 1.05,
"learning_rate": 7.888079875674826e-05,
"loss": 0.8298,
"step": 470
},
{
"epoch": 1.06,
"learning_rate": 8.1246845016499e-05,
"loss": 0.7537,
"step": 475
},
{
"epoch": 1.07,
"learning_rate": 8.359741434072031e-05,
"loss": 0.6783,
"step": 480
},
{
"epoch": 1.08,
"learning_rate": 8.59296301534902e-05,
"loss": 0.7367,
"step": 485
},
{
"epoch": 1.09,
"learning_rate": 8.824063833951144e-05,
"loss": 0.8382,
"step": 490
},
{
"epoch": 1.1,
"learning_rate": 9.052761073692769e-05,
"loss": 0.6501,
"step": 495
},
{
"epoch": 1.11,
"learning_rate": 9.2787748598361e-05,
"loss": 0.7697,
"step": 500
},
{
"epoch": 1.12,
"learning_rate": 9.501828601595296e-05,
"loss": 0.7842,
"step": 505
},
{
"epoch": 1.14,
"learning_rate": 9.72164933062357e-05,
"loss": 0.8064,
"step": 510
},
{
"epoch": 1.15,
"learning_rate": 9.937968035064501e-05,
"loss": 0.8361,
"step": 515
},
{
"epoch": 1.16,
"learning_rate": 0.00010150519988764197,
"loss": 0.7419,
"step": 520
},
{
"epoch": 1.17,
"learning_rate": 0.00010359045075236705,
"loss": 0.8181,
"step": 525
},
{
"epoch": 1.18,
"learning_rate": 0.00010563288105989297,
"loss": 0.7946,
"step": 530
},
{
"epoch": 1.19,
"learning_rate": 0.00010762999132815968,
"loss": 0.7522,
"step": 535
},
{
"epoch": 1.2,
"learning_rate": 0.00010957933753679646,
"loss": 0.721,
"step": 540
},
{
"epoch": 1.21,
"learning_rate": 0.00011147853411805452,
"loss": 0.7965,
"step": 545
},
{
"epoch": 1.22,
"learning_rate": 0.00011332525687621633,
"loss": 0.8731,
"step": 550
},
{
"epoch": 1.24,
"learning_rate": 0.00011511724583189029,
"loss": 0.7461,
"step": 555
},
{
"epoch": 1.25,
"learning_rate": 0.00011685230798773347,
"loss": 0.8536,
"step": 560
},
{
"epoch": 1.26,
"learning_rate": 0.00011852832001218846,
"loss": 0.8069,
"step": 565
},
{
"epoch": 1.27,
"learning_rate": 0.00012014323083797337,
"loss": 0.7363,
"step": 570
},
{
"epoch": 1.28,
"learning_rate": 0.00012169506417212839,
"loss": 0.8555,
"step": 575
},
{
"epoch": 1.29,
"learning_rate": 0.0001231819209145695,
"loss": 0.8844,
"step": 580
},
{
"epoch": 1.3,
"learning_rate": 0.00012460198148215605,
"loss": 0.8991,
"step": 585
},
{
"epoch": 1.31,
"learning_rate": 0.00012595350803546582,
"loss": 0.7508,
"step": 590
},
{
"epoch": 1.33,
"learning_rate": 0.0001272348466055214,
"loss": 0.8352,
"step": 595
},
{
"epoch": 1.34,
"learning_rate": 0.00012844442911787884,
"loss": 0.8305,
"step": 600
},
{
"epoch": 1.35,
"learning_rate": 0.00012958077531160963,
"loss": 0.8743,
"step": 605
},
{
"epoch": 1.36,
"learning_rate": 0.0001306424945508076,
"loss": 0.9164,
"step": 610
},
{
"epoch": 1.37,
"learning_rate": 0.00013162828752642103,
"loss": 0.8642,
"step": 615
},
{
"epoch": 1.38,
"learning_rate": 0.0001325369478463158,
"loss": 0.845,
"step": 620
},
{
"epoch": 1.39,
"learning_rate": 0.00013336736351163685,
"loss": 0.8869,
"step": 625
},
{
"epoch": 1.4,
"learning_rate": 0.00013411851827764502,
"loss": 0.8367,
"step": 630
},
{
"epoch": 1.41,
"learning_rate": 0.00013478949289737682,
"loss": 0.8495,
"step": 635
},
{
"epoch": 1.43,
"learning_rate": 0.00013537946624659676,
"loss": 0.9375,
"step": 640
},
{
"epoch": 1.44,
"learning_rate": 0.00013588771632867518,
"loss": 0.9639,
"step": 645
},
{
"epoch": 1.45,
"learning_rate": 0.00013631362115814792,
"loss": 0.8622,
"step": 650
},
{
"epoch": 1.46,
"learning_rate": 0.00013665665952189337,
"loss": 0.7462,
"step": 655
},
{
"epoch": 1.47,
"learning_rate": 0.0001369164116169769,
"loss": 0.8065,
"step": 660
},
{
"epoch": 1.48,
"learning_rate": 0.0001370925595644003,
"loss": 0.7944,
"step": 665
},
{
"epoch": 1.49,
"learning_rate": 0.00013718488779811326,
"loss": 0.9673,
"step": 670
},
{
"epoch": 1.5,
"learning_rate": 0.00013719328332881955,
"loss": 0.8319,
"step": 675
},
{
"epoch": 1.51,
"learning_rate": 0.00013711773588225001,
"loss": 0.7709,
"step": 680
},
{
"epoch": 1.53,
"learning_rate": 0.00013695833791173674,
"loss": 0.8833,
"step": 685
},
{
"epoch": 1.54,
"learning_rate": 0.00013671528448507009,
"loss": 0.8851,
"step": 690
},
{
"epoch": 1.55,
"learning_rate": 0.00013638887304577923,
"loss": 0.8425,
"step": 695
},
{
"epoch": 1.56,
"learning_rate": 0.00013597950304912684,
"loss": 0.9205,
"step": 700
},
{
"epoch": 1.57,
"learning_rate": 0.0001354876754732656,
"loss": 0.9954,
"step": 705
},
{
"epoch": 1.58,
"learning_rate": 0.00013491399220615015,
"loss": 0.8622,
"step": 710
},
{
"epoch": 1.59,
"learning_rate": 0.00013425915530896078,
"loss": 0.8284,
"step": 715
},
{
"epoch": 1.6,
"learning_rate": 0.00013352396615693527,
"loss": 0.8501,
"step": 720
},
{
"epoch": 1.61,
"learning_rate": 0.00013270932445866562,
"loss": 0.7884,
"step": 725
},
{
"epoch": 1.63,
"learning_rate": 0.00013181622715504977,
"loss": 0.9186,
"step": 730
},
{
"epoch": 1.64,
"learning_rate": 0.000130845767199263,
"loss": 0.8647,
"step": 735
},
{
"epoch": 1.65,
"learning_rate": 0.0001297991322192165,
"loss": 0.8587,
"step": 740
},
{
"epoch": 1.66,
"learning_rate": 0.00012867760306417496,
"loss": 0.8789,
"step": 745
},
{
"epoch": 1.67,
"learning_rate": 0.00012748255223727105,
"loss": 0.7334,
"step": 750
},
{
"epoch": 1.68,
"learning_rate": 0.00012621544221587383,
"loss": 0.854,
"step": 755
},
{
"epoch": 1.69,
"learning_rate": 0.000124877823661834,
"loss": 0.8584,
"step": 760
},
{
"epoch": 1.7,
"learning_rate": 0.0001234713335238199,
"loss": 0.8898,
"step": 765
},
{
"epoch": 1.71,
"learning_rate": 0.00012199769303405894,
"loss": 0.842,
"step": 770
},
{
"epoch": 1.73,
"learning_rate": 0.00012045870560192401,
"loss": 0.8657,
"step": 775
},
{
"epoch": 1.74,
"learning_rate": 0.00011885625460696435,
"loss": 0.8646,
"step": 780
},
{
"epoch": 1.75,
"learning_rate": 0.00011719230109406423,
"loss": 0.974,
"step": 785
},
{
"epoch": 1.76,
"learning_rate": 0.00011546888137356373,
"loss": 0.892,
"step": 790
},
{
"epoch": 1.77,
"learning_rate": 0.00011368810452925682,
"loss": 0.8279,
"step": 795
},
{
"epoch": 1.78,
"learning_rate": 0.00011185214983735281,
"loss": 0.9267,
"step": 800
},
{
"epoch": 1.79,
"learning_rate": 0.0001099632640995077,
"loss": 0.8753,
"step": 805
},
{
"epoch": 1.8,
"learning_rate": 0.00010802375889325772,
"loss": 0.8341,
"step": 810
},
{
"epoch": 1.82,
"learning_rate": 0.00010603600774314254,
"loss": 0.9575,
"step": 815
},
{
"epoch": 1.83,
"learning_rate": 0.00010400244321605133,
"loss": 0.851,
"step": 820
},
{
"epoch": 1.84,
"learning_rate": 0.0001019255539442888,
"loss": 0.7728,
"step": 825
},
{
"epoch": 1.85,
"learning_rate": 9.980788158004715e-05,
"loss": 0.8267,
"step": 830
},
{
"epoch": 1.86,
"learning_rate": 9.765201768498375e-05,
"loss": 0.8464,
"step": 835
},
{
"epoch": 1.87,
"learning_rate": 9.546060055873103e-05,
"loss": 0.8036,
"step": 840
},
{
"epoch": 1.88,
"learning_rate": 9.32363120101919e-05,
"loss": 0.9536,
"step": 845
},
{
"epoch": 1.89,
"learning_rate": 9.098187407560697e-05,
"loss": 0.8601,
"step": 850
},
{
"epoch": 1.9,
"learning_rate": 8.870004568738188e-05,
"loss": 0.78,
"step": 855
},
{
"epoch": 1.92,
"learning_rate": 8.639361929777101e-05,
"loss": 0.8761,
"step": 860
},
{
"epoch": 1.93,
"learning_rate": 8.406541746153297e-05,
"loss": 0.8593,
"step": 865
},
{
"epoch": 1.94,
"learning_rate": 8.171828938173595e-05,
"loss": 0.7247,
"step": 870
},
{
"epoch": 1.95,
"learning_rate": 7.935510742298902e-05,
"loss": 0.8669,
"step": 875
},
{
"epoch": 1.96,
"learning_rate": 7.697876359628178e-05,
"loss": 0.9187,
"step": 880
},
{
"epoch": 1.97,
"learning_rate": 7.459216601982097e-05,
"loss": 0.8965,
"step": 885
},
{
"epoch": 1.98,
"learning_rate": 7.219823536012484e-05,
"loss": 0.9159,
"step": 890
},
{
"epoch": 1.99,
"learning_rate": 6.97999012577823e-05,
"loss": 1.0147,
"step": 895
},
{
"epoch": 2.0,
"eval_loss": 0.5308113694190979,
"eval_runtime": 33.6128,
"eval_samples_per_second": 20.587,
"eval_steps_per_second": 2.588,
"step": 898
},
{
"epoch": 1.98,
"learning_rate": 0.00013331119038948074,
"loss": 0.7724,
"step": 900
},
{
"epoch": 1.99,
"learning_rate": 0.0001340585253974795,
"loss": 0.7496,
"step": 905
},
{
"epoch": 2.0,
"learning_rate": 0.0001347278522437355,
"loss": 0.8119,
"step": 910
},
{
"epoch": 2.0,
"eval_loss": 0.39583703875541687,
"eval_runtime": 13.6672,
"eval_samples_per_second": 47.047,
"eval_steps_per_second": 5.927,
"step": 910
},
{
"epoch": 1.98,
"learning_rate": 1.7062254198703994e-05,
"loss": 0.8052,
"step": 915
},
{
"epoch": 1.99,
"learning_rate": 1.8631088017841603e-05,
"loss": 0.7911,
"step": 920
},
{
"epoch": 2.0,
"eval_loss": 0.37163448333740234,
"eval_runtime": 28.6365,
"eval_samples_per_second": 20.498,
"eval_steps_per_second": 2.584,
"step": 924
}
],
"max_steps": 1386,
"num_train_epochs": 3,
"total_flos": 963383721984000.0,
"trial_name": null,
"trial_params": null
}