eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
ead4f2b
raw
history blame
22.4 kB
{
"best_metric": 0.16843362152576447,
"best_model_checkpoint": "output/eminem/checkpoint-904",
"epoch": 2.0,
"global_step": 904,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 8.139815125844835e-05,
"loss": 0.3572,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 7.904875437627615e-05,
"loss": 0.332,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 7.6686739692467e-05,
"loss": 0.3325,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 7.431495955015108e-05,
"loss": 0.3373,
"step": 20
},
{
"epoch": 0.06,
"learning_rate": 7.193627808511377e-05,
"loss": 0.3129,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 6.955356776710818e-05,
"loss": 0.3354,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 6.716970593108238e-05,
"loss": 0.3096,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 6.478757130255084e-05,
"loss": 0.3554,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 6.24100405212653e-05,
"loss": 0.3456,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 6.0039984667425774e-05,
"loss": 0.3283,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 5.768026579458568e-05,
"loss": 0.3464,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 5.5333733473478955e-05,
"loss": 0.3187,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 5.3003221350899904e-05,
"loss": 0.3377,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 5.069154372783178e-05,
"loss": 0.3172,
"step": 70
},
{
"epoch": 0.17,
"learning_rate": 4.8401492160918125e-05,
"loss": 0.3366,
"step": 75
},
{
"epoch": 0.18,
"learning_rate": 4.6135832091409655e-05,
"loss": 0.3463,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 4.389729950565688e-05,
"loss": 0.3223,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 4.168859763115363e-05,
"loss": 0.3103,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 3.9512393672158115e-05,
"loss": 0.321,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 3.737131558879536e-05,
"loss": 0.341,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 3.526794892357027e-05,
"loss": 0.316,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 3.320483367908598e-05,
"loss": 0.3275,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 3.118446125077491e-05,
"loss": 0.3246,
"step": 115
},
{
"epoch": 0.27,
"learning_rate": 2.9209271418308243e-05,
"loss": 0.3166,
"step": 120
},
{
"epoch": 0.28,
"learning_rate": 2.7281649399352843e-05,
"loss": 0.3269,
"step": 125
},
{
"epoch": 0.29,
"learning_rate": 2.540392296919879e-05,
"loss": 0.3522,
"step": 130
},
{
"epoch": 0.3,
"learning_rate": 2.3578359649771312e-05,
"loss": 0.3318,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 2.1807163971388526e-05,
"loss": 0.3207,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 2.009247481060366e-05,
"loss": 0.3127,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 1.843636280731375e-05,
"loss": 0.2845,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 1.6840827864284427e-05,
"loss": 0.3354,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 1.530779673208225e-05,
"loss": 0.3237,
"step": 160
},
{
"epoch": 0.37,
"learning_rate": 1.3839120682352292e-05,
"loss": 0.3352,
"step": 165
},
{
"epoch": 0.38,
"learning_rate": 1.243657327224889e-05,
"loss": 0.3299,
"step": 170
},
{
"epoch": 0.39,
"learning_rate": 1.1101848202702286e-05,
"loss": 0.3208,
"step": 175
},
{
"epoch": 0.4,
"learning_rate": 9.836557273124208e-06,
"loss": 0.2988,
"step": 180
},
{
"epoch": 0.41,
"learning_rate": 8.642228435020553e-06,
"loss": 0.3097,
"step": 185
},
{
"epoch": 0.42,
"learning_rate": 7.520303946848552e-06,
"loss": 0.3237,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 6.472138632361784e-06,
"loss": 0.3487,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 5.498998244544231e-06,
"loss": 0.3417,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 4.602057937097127e-06,
"loss": 0.312,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 3.7824008453364313e-06,
"loss": 0.3209,
"step": 210
},
{
"epoch": 0.48,
"learning_rate": 3.041016778213118e-06,
"loss": 0.3107,
"step": 215
},
{
"epoch": 0.49,
"learning_rate": 2.378801023027184e-06,
"loss": 0.3087,
"step": 220
},
{
"epoch": 0.5,
"learning_rate": 1.796553264289097e-06,
"loss": 0.2947,
"step": 225
},
{
"epoch": 0.51,
"learning_rate": 1.29497661803209e-06,
"loss": 0.3013,
"step": 230
},
{
"epoch": 0.52,
"learning_rate": 8.746767827346987e-07,
"loss": 0.2997,
"step": 235
},
{
"epoch": 0.53,
"learning_rate": 5.361613078862018e-07,
"loss": 0.2932,
"step": 240
},
{
"epoch": 0.54,
"learning_rate": 2.7983898107679704e-07,
"loss": 0.3453,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 1.0601933434688134e-07,
"loss": 0.3002,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 1.4912270403722228e-08,
"loss": 0.3528,
"step": 255
},
{
"epoch": 0.58,
"learning_rate": 6.627809142250918e-09,
"loss": 0.3299,
"step": 260
},
{
"epoch": 0.59,
"learning_rate": 8.117595478739506e-08,
"loss": 0.3005,
"step": 265
},
{
"epoch": 0.6,
"learning_rate": 2.384666838130307e-07,
"loss": 0.313,
"step": 270
},
{
"epoch": 0.61,
"learning_rate": 4.783100536523972e-07,
"loss": 0.2978,
"step": 275
},
{
"epoch": 0.62,
"learning_rate": 8.004164320734335e-07,
"loss": 0.3066,
"step": 280
},
{
"epoch": 0.63,
"learning_rate": 1.2043968469315794e-06,
"loss": 0.3224,
"step": 285
},
{
"epoch": 0.64,
"learning_rate": 1.689763455891368e-06,
"loss": 0.3433,
"step": 290
},
{
"epoch": 0.65,
"learning_rate": 2.255930135538237e-06,
"loss": 0.3279,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 2.9022131891755135e-06,
"loss": 0.3128,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 3.6278321724456592e-06,
"loss": 0.3398,
"step": 305
},
{
"epoch": 0.69,
"learning_rate": 4.431910835795094e-06,
"loss": 0.346,
"step": 310
},
{
"epoch": 0.7,
"learning_rate": 5.313478182611713e-06,
"loss": 0.3162,
"step": 315
},
{
"epoch": 0.71,
"learning_rate": 6.271469641798839e-06,
"loss": 0.2976,
"step": 320
},
{
"epoch": 0.72,
"learning_rate": 7.304728353335969e-06,
"loss": 0.3161,
"step": 325
},
{
"epoch": 0.73,
"learning_rate": 8.412006565290002e-06,
"loss": 0.3494,
"step": 330
},
{
"epoch": 0.74,
"learning_rate": 9.591967140590498e-06,
"loss": 0.3082,
"step": 335
},
{
"epoch": 0.75,
"learning_rate": 1.0843185171737421e-05,
"loss": 0.3302,
"step": 340
},
{
"epoch": 0.76,
"learning_rate": 1.2164149701503293e-05,
"loss": 0.3535,
"step": 345
},
{
"epoch": 0.77,
"learning_rate": 1.3553265547553905e-05,
"loss": 0.2955,
"step": 350
},
{
"epoch": 0.79,
"learning_rate": 1.500885522876901e-05,
"loss": 0.3486,
"step": 355
},
{
"epoch": 0.8,
"learning_rate": 1.6529160990949952e-05,
"loss": 0.3031,
"step": 360
},
{
"epoch": 0.81,
"learning_rate": 1.8112346929469448e-05,
"loss": 0.3208,
"step": 365
},
{
"epoch": 0.82,
"learning_rate": 1.975650120628388e-05,
"loss": 0.302,
"step": 370
},
{
"epoch": 0.83,
"learning_rate": 2.1459638358647496e-05,
"loss": 0.3412,
"step": 375
},
{
"epoch": 0.84,
"learning_rate": 2.3219701696742428e-05,
"loss": 0.3244,
"step": 380
},
{
"epoch": 0.85,
"learning_rate": 2.5034565787309464e-05,
"loss": 0.3438,
"step": 385
},
{
"epoch": 0.86,
"learning_rate": 2.6902039020297848e-05,
"loss": 0.3173,
"step": 390
},
{
"epoch": 0.87,
"learning_rate": 2.881986625543576e-05,
"loss": 0.3234,
"step": 395
},
{
"epoch": 0.88,
"learning_rate": 3.078573154550556e-05,
"loss": 0.3361,
"step": 400
},
{
"epoch": 0.9,
"learning_rate": 3.279726093305591e-05,
"loss": 0.3391,
"step": 405
},
{
"epoch": 0.91,
"learning_rate": 3.485202531717502e-05,
"loss": 0.3205,
"step": 410
},
{
"epoch": 0.92,
"learning_rate": 3.6947543386838276e-05,
"loss": 0.3197,
"step": 415
},
{
"epoch": 0.93,
"learning_rate": 3.9081284617320727e-05,
"loss": 0.3272,
"step": 420
},
{
"epoch": 0.94,
"learning_rate": 4.125067232602091e-05,
"loss": 0.3092,
"step": 425
},
{
"epoch": 0.95,
"learning_rate": 4.345308678404316e-05,
"loss": 0.3419,
"step": 430
},
{
"epoch": 0.96,
"learning_rate": 4.568586837974475e-05,
"loss": 0.3065,
"step": 435
},
{
"epoch": 0.97,
"learning_rate": 4.794632083046458e-05,
"loss": 0.3082,
"step": 440
},
{
"epoch": 0.98,
"learning_rate": 5.023171443851573e-05,
"loss": 0.3379,
"step": 445
},
{
"epoch": 1.0,
"learning_rate": 5.2539289387547704e-05,
"loss": 0.3392,
"step": 450
},
{
"epoch": 1.0,
"eval_loss": 0.182983860373497,
"eval_runtime": 15.9367,
"eval_samples_per_second": 41.79,
"eval_steps_per_second": 5.271,
"step": 452
},
{
"epoch": 1.01,
"learning_rate": 5.486625907525938e-05,
"loss": 0.3706,
"step": 455
},
{
"epoch": 1.02,
"learning_rate": 5.7209813478478894e-05,
"loss": 0.3441,
"step": 460
},
{
"epoch": 1.03,
"learning_rate": 5.956712254650734e-05,
"loss": 0.336,
"step": 465
},
{
"epoch": 1.04,
"learning_rate": 6.193533961866664e-05,
"loss": 0.3351,
"step": 470
},
{
"epoch": 1.05,
"learning_rate": 6.431160486189602e-05,
"loss": 0.379,
"step": 475
},
{
"epoch": 1.06,
"learning_rate": 6.669304872424543e-05,
"loss": 0.348,
"step": 480
},
{
"epoch": 1.07,
"learning_rate": 6.907679540012287e-05,
"loss": 0.3403,
"step": 485
},
{
"epoch": 1.08,
"learning_rate": 7.145996630308584e-05,
"loss": 0.3578,
"step": 490
},
{
"epoch": 1.1,
"learning_rate": 7.383968354198309e-05,
"loss": 0.313,
"step": 495
},
{
"epoch": 1.11,
"learning_rate": 7.621307339627422e-05,
"loss": 0.3456,
"step": 500
},
{
"epoch": 1.12,
"learning_rate": 7.857726978630327e-05,
"loss": 0.3309,
"step": 505
},
{
"epoch": 1.13,
"learning_rate": 8.09294177343355e-05,
"loss": 0.3781,
"step": 510
},
{
"epoch": 1.14,
"learning_rate": 8.326667681220494e-05,
"loss": 0.3312,
"step": 515
},
{
"epoch": 1.15,
"learning_rate": 8.558622457138458e-05,
"loss": 0.3715,
"step": 520
},
{
"epoch": 1.16,
"learning_rate": 8.78852599513366e-05,
"loss": 0.3699,
"step": 525
},
{
"epoch": 1.17,
"learning_rate": 9.016100666205213e-05,
"loss": 0.3742,
"step": 530
},
{
"epoch": 1.18,
"learning_rate": 9.241071653666853e-05,
"loss": 0.3531,
"step": 535
},
{
"epoch": 1.19,
"learning_rate": 9.46316728501164e-05,
"loss": 0.3666,
"step": 540
},
{
"epoch": 1.21,
"learning_rate": 9.682119359981451e-05,
"loss": 0.3454,
"step": 545
},
{
"epoch": 1.22,
"learning_rate": 9.897663474442683e-05,
"loss": 0.3345,
"step": 550
},
{
"epoch": 1.23,
"learning_rate": 0.00010109539339676733,
"loss": 0.3631,
"step": 555
},
{
"epoch": 1.24,
"learning_rate": 0.00010317491096704431,
"loss": 0.3796,
"step": 560
},
{
"epoch": 1.25,
"learning_rate": 0.00010521267625255079,
"loss": 0.3308,
"step": 565
},
{
"epoch": 1.26,
"learning_rate": 0.00010720622847019136,
"loss": 0.3771,
"step": 570
},
{
"epoch": 1.27,
"learning_rate": 0.00010915316022808663,
"loss": 0.3416,
"step": 575
},
{
"epoch": 1.28,
"learning_rate": 0.00011105112043271663,
"loss": 0.3921,
"step": 580
},
{
"epoch": 1.29,
"learning_rate": 0.00011289781712806653,
"loss": 0.3541,
"step": 585
},
{
"epoch": 1.31,
"learning_rate": 0.00011469102026339036,
"loss": 0.3366,
"step": 590
},
{
"epoch": 1.32,
"learning_rate": 0.00011642856438616586,
"loss": 0.348,
"step": 595
},
{
"epoch": 1.33,
"learning_rate": 0.00011810835125709373,
"loss": 0.3918,
"step": 600
},
{
"epoch": 1.34,
"learning_rate": 0.00011972835238389949,
"loss": 0.3821,
"step": 605
},
{
"epoch": 1.35,
"learning_rate": 0.00012128661147092148,
"loss": 0.3542,
"step": 610
},
{
"epoch": 1.36,
"learning_rate": 0.0001227812467815045,
"loss": 0.3587,
"step": 615
},
{
"epoch": 1.37,
"learning_rate": 0.00012421045341038303,
"loss": 0.4155,
"step": 620
},
{
"epoch": 1.38,
"learning_rate": 0.00012557250546324066,
"loss": 0.4249,
"step": 625
},
{
"epoch": 1.39,
"learning_rate": 0.00012686575814089933,
"loss": 0.4011,
"step": 630
},
{
"epoch": 1.4,
"learning_rate": 0.00012808864972555366,
"loss": 0.4206,
"step": 635
},
{
"epoch": 1.42,
"learning_rate": 0.00012923970346668468,
"loss": 0.3698,
"step": 640
},
{
"epoch": 1.43,
"learning_rate": 0.00013031752936437118,
"loss": 0.3907,
"step": 645
},
{
"epoch": 1.44,
"learning_rate": 0.0001313208258478337,
"loss": 0.3931,
"step": 650
},
{
"epoch": 1.45,
"learning_rate": 0.00013224838134719776,
"loss": 0.3725,
"step": 655
},
{
"epoch": 1.46,
"learning_rate": 0.00013309907575657624,
"loss": 0.4056,
"step": 660
},
{
"epoch": 1.47,
"learning_rate": 0.00013387188178669445,
"loss": 0.3753,
"step": 665
},
{
"epoch": 1.48,
"learning_rate": 0.0001345658662054306,
"loss": 0.4041,
"step": 670
},
{
"epoch": 1.49,
"learning_rate": 0.0001351801909647851,
"loss": 0.421,
"step": 675
},
{
"epoch": 1.5,
"learning_rate": 0.0001357141142128889,
"loss": 0.3516,
"step": 680
},
{
"epoch": 1.52,
"learning_rate": 0.0001361669911898596,
"loss": 0.4309,
"step": 685
},
{
"epoch": 1.53,
"learning_rate": 0.00013653827500640774,
"loss": 0.3797,
"step": 690
},
{
"epoch": 1.54,
"learning_rate": 0.0001368275173042528,
"loss": 0.4128,
"step": 695
},
{
"epoch": 1.55,
"learning_rate": 0.00013703436879755498,
"loss": 0.4107,
"step": 700
},
{
"epoch": 1.56,
"learning_rate": 0.00013715857969471166,
"loss": 0.4123,
"step": 705
},
{
"epoch": 1.57,
"learning_rate": 0.0001372,
"loss": 0.4361,
"step": 710
},
{
"epoch": 1.58,
"learning_rate": 0.00013715857969471174,
"loss": 0.3861,
"step": 715
},
{
"epoch": 1.59,
"learning_rate": 0.00013703436879755515,
"loss": 0.4283,
"step": 720
},
{
"epoch": 1.6,
"learning_rate": 0.00013682751730425283,
"loss": 0.4119,
"step": 725
},
{
"epoch": 1.62,
"learning_rate": 0.00013653827500640804,
"loss": 0.3837,
"step": 730
},
{
"epoch": 1.63,
"learning_rate": 0.00013616699118985964,
"loss": 0.409,
"step": 735
},
{
"epoch": 1.64,
"learning_rate": 0.00013571411421288895,
"loss": 0.4209,
"step": 740
},
{
"epoch": 1.65,
"learning_rate": 0.00013518019096478564,
"loss": 0.4686,
"step": 745
},
{
"epoch": 1.66,
"learning_rate": 0.00013456586620543123,
"loss": 0.4259,
"step": 750
},
{
"epoch": 1.67,
"learning_rate": 0.00013387188178669453,
"loss": 0.4255,
"step": 755
},
{
"epoch": 1.68,
"learning_rate": 0.000133099075756577,
"loss": 0.3998,
"step": 760
},
{
"epoch": 1.69,
"learning_rate": 0.00013224838134719786,
"loss": 0.4879,
"step": 765
},
{
"epoch": 1.7,
"learning_rate": 0.00013132082584783384,
"loss": 0.4157,
"step": 770
},
{
"epoch": 1.71,
"learning_rate": 0.00013031752936437215,
"loss": 0.4395,
"step": 775
},
{
"epoch": 1.73,
"learning_rate": 0.00012923970346668571,
"loss": 0.3997,
"step": 780
},
{
"epoch": 1.74,
"learning_rate": 0.00012808864972555477,
"loss": 0.3993,
"step": 785
},
{
"epoch": 1.75,
"learning_rate": 0.0001268657581409005,
"loss": 0.4962,
"step": 790
},
{
"epoch": 1.76,
"learning_rate": 0.00012557250546324083,
"loss": 0.3908,
"step": 795
},
{
"epoch": 1.77,
"learning_rate": 0.00012421045341038322,
"loss": 0.447,
"step": 800
},
{
"epoch": 1.78,
"learning_rate": 0.00012278124678150585,
"loss": 0.4202,
"step": 805
},
{
"epoch": 1.79,
"learning_rate": 0.00012128661147092164,
"loss": 0.4339,
"step": 810
},
{
"epoch": 1.8,
"learning_rate": 0.00011972835238390097,
"loss": 0.4362,
"step": 815
},
{
"epoch": 1.81,
"learning_rate": 0.00011810835125709524,
"loss": 0.462,
"step": 820
},
{
"epoch": 1.83,
"learning_rate": 0.00011642856438616606,
"loss": 0.3672,
"step": 825
},
{
"epoch": 1.84,
"learning_rate": 0.00011469102026339056,
"loss": 0.4118,
"step": 830
},
{
"epoch": 1.85,
"learning_rate": 0.00011289781712806821,
"loss": 0.3731,
"step": 835
},
{
"epoch": 1.86,
"learning_rate": 0.00011105112043271685,
"loss": 0.4515,
"step": 840
},
{
"epoch": 1.87,
"learning_rate": 0.00010915316022808842,
"loss": 0.4245,
"step": 845
},
{
"epoch": 1.88,
"learning_rate": 0.00010720622847019321,
"loss": 0.4166,
"step": 850
},
{
"epoch": 1.89,
"learning_rate": 0.00010521267625255102,
"loss": 0.3902,
"step": 855
},
{
"epoch": 1.9,
"learning_rate": 0.00010317491096704455,
"loss": 0.442,
"step": 860
},
{
"epoch": 1.91,
"learning_rate": 0.00010109539339676927,
"loss": 0.3854,
"step": 865
},
{
"epoch": 1.92,
"learning_rate": 9.897663474442706e-05,
"loss": 0.4236,
"step": 870
},
{
"epoch": 1.94,
"learning_rate": 9.682119359981653e-05,
"loss": 0.4067,
"step": 875
},
{
"epoch": 1.95,
"learning_rate": 9.463167285011846e-05,
"loss": 0.4098,
"step": 880
},
{
"epoch": 1.96,
"learning_rate": 9.241071653666879e-05,
"loss": 0.4841,
"step": 885
},
{
"epoch": 1.97,
"learning_rate": 9.016100666205239e-05,
"loss": 0.4236,
"step": 890
},
{
"epoch": 1.98,
"learning_rate": 8.788525995133685e-05,
"loss": 0.4312,
"step": 895
},
{
"epoch": 1.99,
"learning_rate": 8.558622457138483e-05,
"loss": 0.4461,
"step": 900
},
{
"epoch": 2.0,
"eval_loss": 0.16843362152576447,
"eval_runtime": 14.4984,
"eval_samples_per_second": 46.005,
"eval_steps_per_second": 5.794,
"step": 904
}
],
"max_steps": 904,
"num_train_epochs": 2,
"total_flos": 943525527552000.0,
"trial_name": null,
"trial_params": null
}