eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
47f6cea
raw
history blame
22.4 kB
{
"best_metric": 0.3754318356513977,
"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": 9.668898730225426e-06,
"loss": 0.6451,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 8.494471426523322e-06,
"loss": 0.6405,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 7.391359230778123e-06,
"loss": 0.6683,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 6.360870983745512e-06,
"loss": 0.6156,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 5.404229357984623e-06,
"loss": 0.7032,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 4.522569407168996e-06,
"loss": 0.6814,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 3.7169372193369285e-06,
"loss": 0.7428,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 2.988288675715626e-06,
"loss": 0.6695,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 2.3374883165699825e-06,
"loss": 0.6716,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 1.7653083154249965e-06,
"loss": 0.6442,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 1.2724275628843243e-06,
"loss": 0.7111,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 8.594308611261225e-07,
"loss": 0.717,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 5.26808230038689e-07,
"loss": 0.6728,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 2.7495432581104424e-07,
"loss": 0.5893,
"step": 70
},
{
"epoch": 0.16,
"learning_rate": 1.0416797267454092e-07,
"loss": 0.6088,
"step": 75
},
{
"epoch": 0.18,
"learning_rate": 1.465180834793276e-08,
"loss": 0.6029,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 6.5120436089167734e-09,
"loss": 0.6099,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 7.975833627447873e-08,
"loss": 0.5843,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 2.3430377974243327e-07,
"loss": 0.6587,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 4.699650061053238e-07,
"loss": 0.5792,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 7.864624037171852e-07,
"loss": 0.6884,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 1.1834204489510753e-06,
"loss": 0.6179,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 1.6603681517579621e-06,
"loss": 0.6588,
"step": 115
},
{
"epoch": 0.26,
"learning_rate": 2.2167396144949356e-06,
"loss": 0.5756,
"step": 120
},
{
"epoch": 0.27,
"learning_rate": 2.851874703363704e-06,
"loss": 0.7218,
"step": 125
},
{
"epoch": 0.29,
"learning_rate": 3.565019831655522e-06,
"loss": 0.6423,
"step": 130
},
{
"epoch": 0.3,
"learning_rate": 4.3553288538822644e-06,
"loss": 0.7181,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 5.221864069725448e-06,
"loss": 0.7147,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 6.163597336619036e-06,
"loss": 0.6411,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 7.179411289634524e-06,
"loss": 0.6499,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 8.268100667234415e-06,
"loss": 0.6962,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 9.428373741315982e-06,
"loss": 0.6569,
"step": 160
},
{
"epoch": 0.36,
"learning_rate": 1.0658853849838009e-05,
"loss": 0.6364,
"step": 165
},
{
"epoch": 0.37,
"learning_rate": 1.1958081030236758e-05,
"loss": 0.6697,
"step": 170
},
{
"epoch": 0.38,
"learning_rate": 1.3324513751656602e-05,
"loss": 0.6964,
"step": 175
},
{
"epoch": 0.39,
"learning_rate": 1.47565307439888e-05,
"loss": 0.6235,
"step": 180
},
{
"epoch": 0.41,
"learning_rate": 1.6252432921492454e-05,
"loss": 0.6185,
"step": 185
},
{
"epoch": 0.42,
"learning_rate": 1.7810445398767395e-05,
"loss": 0.7024,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 1.9428719596644185e-05,
"loss": 0.6126,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 2.1105335435523642e-05,
"loss": 0.6189,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 2.283830361355242e-05,
"loss": 0.5719,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 2.4625567966918e-05,
"loss": 0.7211,
"step": 210
},
{
"epoch": 0.47,
"learning_rate": 2.6465007909489048e-05,
"loss": 0.6735,
"step": 215
},
{
"epoch": 0.48,
"learning_rate": 2.8354440948884326e-05,
"loss": 0.5916,
"step": 220
},
{
"epoch": 0.49,
"learning_rate": 3.0291625276002506e-05,
"loss": 0.5753,
"step": 225
},
{
"epoch": 0.5,
"learning_rate": 3.2274262424915514e-05,
"loss": 0.6077,
"step": 230
},
{
"epoch": 0.52,
"learning_rate": 3.430000000000064e-05,
"loss": 0.7398,
"step": 235
},
{
"epoch": 0.53,
"learning_rate": 3.636643446705075e-05,
"loss": 0.7589,
"step": 240
},
{
"epoch": 0.54,
"learning_rate": 3.847111400507036e-05,
"loss": 0.7287,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 4.0611541415345134e-05,
"loss": 0.7015,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 4.2785177084377745e-05,
"loss": 0.6185,
"step": 255
},
{
"epoch": 0.57,
"learning_rate": 4.49894419971228e-05,
"loss": 0.6638,
"step": 260
},
{
"epoch": 0.58,
"learning_rate": 4.7221720796988937e-05,
"loss": 0.6769,
"step": 265
},
{
"epoch": 0.59,
"learning_rate": 4.947936488896066e-05,
"loss": 0.7237,
"step": 270
},
{
"epoch": 0.6,
"learning_rate": 5.1759695582140806e-05,
"loss": 0.7138,
"step": 275
},
{
"epoch": 0.61,
"learning_rate": 5.4060007268020076e-05,
"loss": 0.655,
"step": 280
},
{
"epoch": 0.62,
"learning_rate": 5.637757063067322e-05,
"loss": 0.7932,
"step": 285
},
{
"epoch": 0.64,
"learning_rate": 5.870963588509545e-05,
"loss": 0.6143,
"step": 290
},
{
"epoch": 0.65,
"learning_rate": 6.105343603980619e-05,
"loss": 0.6686,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 6.34061901798874e-05,
"loss": 0.7298,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 6.576510676653068e-05,
"loss": 0.6932,
"step": 305
},
{
"epoch": 0.68,
"learning_rate": 6.812738694920095e-05,
"loss": 0.7121,
"step": 310
},
{
"epoch": 0.69,
"learning_rate": 7.049022788645635e-05,
"loss": 0.668,
"step": 315
},
{
"epoch": 0.7,
"learning_rate": 7.285082607152067e-05,
"loss": 0.6626,
"step": 320
},
{
"epoch": 0.71,
"learning_rate": 7.52063806586452e-05,
"loss": 0.7163,
"step": 325
},
{
"epoch": 0.72,
"learning_rate": 7.755409678629386e-05,
"loss": 0.753,
"step": 330
},
{
"epoch": 0.73,
"learning_rate": 7.989118889325919e-05,
"loss": 0.6771,
"step": 335
},
{
"epoch": 0.75,
"learning_rate": 8.221488402370275e-05,
"loss": 0.6505,
"step": 340
},
{
"epoch": 0.76,
"learning_rate": 8.452242511728957e-05,
"loss": 0.6098,
"step": 345
},
{
"epoch": 0.77,
"learning_rate": 8.681107428041253e-05,
"loss": 0.6978,
"step": 350
},
{
"epoch": 0.78,
"learning_rate": 8.907811603471558e-05,
"loss": 0.6606,
"step": 355
},
{
"epoch": 0.79,
"learning_rate": 9.132086053899106e-05,
"loss": 0.788,
"step": 360
},
{
"epoch": 0.8,
"learning_rate": 9.353664678067919e-05,
"loss": 0.7524,
"step": 365
},
{
"epoch": 0.81,
"learning_rate": 9.572284573315139e-05,
"loss": 0.6861,
"step": 370
},
{
"epoch": 0.82,
"learning_rate": 9.787686347505305e-05,
"loss": 0.6812,
"step": 375
},
{
"epoch": 0.83,
"learning_rate": 9.999614426797605e-05,
"loss": 0.6453,
"step": 380
},
{
"epoch": 0.84,
"learning_rate": 0.00010207817358884356,
"loss": 0.7003,
"step": 385
},
{
"epoch": 0.86,
"learning_rate": 0.00010412048111339246,
"loss": 0.714,
"step": 390
},
{
"epoch": 0.87,
"learning_rate": 0.00010612064364719859,
"loss": 0.7008,
"step": 395
},
{
"epoch": 0.88,
"learning_rate": 0.00010807628800079975,
"loss": 0.6857,
"step": 400
},
{
"epoch": 0.89,
"learning_rate": 0.00010998509380547761,
"loss": 0.7609,
"step": 405
},
{
"epoch": 0.9,
"learning_rate": 0.00011184479626637769,
"loss": 0.7942,
"step": 410
},
{
"epoch": 0.91,
"learning_rate": 0.00011365318884967352,
"loss": 0.6518,
"step": 415
},
{
"epoch": 0.92,
"learning_rate": 0.00011540812590062915,
"loss": 0.7253,
"step": 420
},
{
"epoch": 0.93,
"learning_rate": 0.000117107525189397,
"loss": 0.7335,
"step": 425
},
{
"epoch": 0.94,
"learning_rate": 0.00011874937038159925,
"loss": 0.7592,
"step": 430
},
{
"epoch": 0.95,
"learning_rate": 0.00012033171343069444,
"loss": 0.7893,
"step": 435
},
{
"epoch": 0.96,
"learning_rate": 0.00012185267688933072,
"loss": 0.7095,
"step": 440
},
{
"epoch": 0.98,
"learning_rate": 0.00012331045613693775,
"loss": 0.7761,
"step": 445
},
{
"epoch": 0.99,
"learning_rate": 0.0001247033215209036,
"loss": 0.6766,
"step": 450
},
{
"epoch": 1.0,
"learning_rate": 0.00012602962040880915,
"loss": 0.7222,
"step": 455
},
{
"epoch": 1.0,
"eval_loss": 0.41102197766304016,
"eval_runtime": 15.2365,
"eval_samples_per_second": 41.742,
"eval_steps_per_second": 5.251,
"step": 456
},
{
"epoch": 1.02,
"learning_rate": 0.00011710752518939759,
"loss": 0.6694,
"step": 460
},
{
"epoch": 1.03,
"learning_rate": 0.00011539283602861273,
"loss": 0.6712,
"step": 465
},
{
"epoch": 1.04,
"learning_rate": 0.00011362164035023001,
"loss": 0.7083,
"step": 470
},
{
"epoch": 1.05,
"learning_rate": 0.00011179607703080105,
"loss": 0.6899,
"step": 475
},
{
"epoch": 1.06,
"learning_rate": 0.000109918350600647,
"loss": 0.7508,
"step": 480
},
{
"epoch": 1.07,
"learning_rate": 0.0001079907285816916,
"loss": 0.6351,
"step": 485
},
{
"epoch": 1.08,
"learning_rate": 0.00010601553874922655,
"loss": 0.7873,
"step": 490
},
{
"epoch": 1.1,
"learning_rate": 0.0001039951663209155,
"loss": 0.7405,
"step": 495
},
{
"epoch": 1.11,
"learning_rate": 0.00010193205107643125,
"loss": 0.8115,
"step": 500
},
{
"epoch": 1.12,
"learning_rate": 9.982868441120446e-05,
"loss": 0.8309,
"step": 505
},
{
"epoch": 1.13,
"learning_rate": 9.768760632784171e-05,
"loss": 0.7548,
"step": 510
},
{
"epoch": 1.14,
"learning_rate": 9.551140236884618e-05,
"loss": 0.7107,
"step": 515
},
{
"epoch": 1.15,
"learning_rate": 9.330270049434476e-05,
"loss": 0.6035,
"step": 520
},
{
"epoch": 1.16,
"learning_rate": 9.1064167908592e-05,
"loss": 0.724,
"step": 525
},
{
"epoch": 1.17,
"learning_rate": 8.879850783908354e-05,
"loss": 0.6642,
"step": 530
},
{
"epoch": 1.18,
"learning_rate": 8.650845627216803e-05,
"loss": 0.6811,
"step": 535
},
{
"epoch": 1.19,
"learning_rate": 8.41967786490999e-05,
"loss": 0.6505,
"step": 540
},
{
"epoch": 1.21,
"learning_rate": 8.186626652652275e-05,
"loss": 0.7759,
"step": 545
},
{
"epoch": 1.22,
"learning_rate": 7.951973420541605e-05,
"loss": 0.6233,
"step": 550
},
{
"epoch": 1.23,
"learning_rate": 7.716001533257597e-05,
"loss": 0.7224,
"step": 555
},
{
"epoch": 1.24,
"learning_rate": 7.478995947873449e-05,
"loss": 0.7252,
"step": 560
},
{
"epoch": 1.25,
"learning_rate": 7.241242869744897e-05,
"loss": 0.6465,
"step": 565
},
{
"epoch": 1.26,
"learning_rate": 7.003029406891743e-05,
"loss": 0.7505,
"step": 570
},
{
"epoch": 1.27,
"learning_rate": 6.764643223289357e-05,
"loss": 0.7117,
"step": 575
},
{
"epoch": 1.28,
"learning_rate": 6.526372191488798e-05,
"loss": 0.7289,
"step": 580
},
{
"epoch": 1.29,
"learning_rate": 6.288504044985065e-05,
"loss": 0.7069,
"step": 585
},
{
"epoch": 1.31,
"learning_rate": 6.05132603075328e-05,
"loss": 0.7925,
"step": 590
},
{
"epoch": 1.32,
"learning_rate": 5.8151245623725575e-05,
"loss": 0.7614,
"step": 595
},
{
"epoch": 1.33,
"learning_rate": 5.5801848741551456e-05,
"loss": 0.6241,
"step": 600
},
{
"epoch": 1.34,
"learning_rate": 5.3467906767025034e-05,
"loss": 0.6841,
"step": 605
},
{
"epoch": 1.35,
"learning_rate": 5.115223814297748e-05,
"loss": 0.7002,
"step": 610
},
{
"epoch": 1.36,
"learning_rate": 4.885763924553591e-05,
"loss": 0.7048,
"step": 615
},
{
"epoch": 1.37,
"learning_rate": 4.65868810072671e-05,
"loss": 0.6762,
"step": 620
},
{
"epoch": 1.38,
"learning_rate": 4.434270557101253e-05,
"loss": 0.6848,
"step": 625
},
{
"epoch": 1.39,
"learning_rate": 4.2127822978506955e-05,
"loss": 0.7911,
"step": 630
},
{
"epoch": 1.4,
"learning_rate": 3.9944907897778e-05,
"loss": 0.767,
"step": 635
},
{
"epoch": 1.42,
"learning_rate": 3.779659639322971e-05,
"loss": 0.6907,
"step": 640
},
{
"epoch": 1.43,
"learning_rate": 3.568548274236045e-05,
"loss": 0.7518,
"step": 645
},
{
"epoch": 1.44,
"learning_rate": 3.361411630295728e-05,
"loss": 0.6918,
"step": 650
},
{
"epoch": 1.45,
"learning_rate": 3.158499843450359e-05,
"loss": 0.7341,
"step": 655
},
{
"epoch": 1.46,
"learning_rate": 2.9600579477565238e-05,
"loss": 0.6637,
"step": 660
},
{
"epoch": 1.47,
"learning_rate": 2.7663255794802226e-05,
"loss": 0.7897,
"step": 665
},
{
"epoch": 1.48,
"learning_rate": 2.5775366877124698e-05,
"loss": 0.5947,
"step": 670
},
{
"epoch": 1.49,
"learning_rate": 2.393919251856543e-05,
"loss": 0.6711,
"step": 675
},
{
"epoch": 1.5,
"learning_rate": 2.21569500632249e-05,
"loss": 0.789,
"step": 680
},
{
"epoch": 1.52,
"learning_rate": 2.043079172761756e-05,
"loss": 0.6784,
"step": 685
},
{
"epoch": 1.53,
"learning_rate": 1.876280200169214e-05,
"loss": 0.648,
"step": 690
},
{
"epoch": 1.54,
"learning_rate": 1.7154995131623676e-05,
"loss": 0.8539,
"step": 695
},
{
"epoch": 1.55,
"learning_rate": 1.5609312687419874e-05,
"loss": 0.8157,
"step": 700
},
{
"epoch": 1.56,
"learning_rate": 1.4127621218314657e-05,
"loss": 0.6428,
"step": 705
},
{
"epoch": 1.57,
"learning_rate": 1.2711709998742864e-05,
"loss": 0.6836,
"step": 710
},
{
"epoch": 1.58,
"learning_rate": 1.1363288867621515e-05,
"loss": 0.6848,
"step": 715
},
{
"epoch": 1.59,
"learning_rate": 1.0083986163577838e-05,
"loss": 0.687,
"step": 720
},
{
"epoch": 1.6,
"learning_rate": 8.875346758584498e-06,
"loss": 0.6913,
"step": 725
},
{
"epoch": 1.62,
"learning_rate": 7.738830192380085e-06,
"loss": 0.7138,
"step": 730
},
{
"epoch": 1.63,
"learning_rate": 6.675808909954194e-06,
"loss": 0.7293,
"step": 735
},
{
"epoch": 1.64,
"learning_rate": 5.687566604196989e-06,
"loss": 0.7797,
"step": 740
},
{
"epoch": 1.65,
"learning_rate": 4.7752966657182915e-06,
"loss": 0.6821,
"step": 745
},
{
"epoch": 1.66,
"learning_rate": 3.940100741730516e-06,
"loss": 0.5752,
"step": 750
},
{
"epoch": 1.67,
"learning_rate": 3.182987405707336e-06,
"loss": 0.7456,
"step": 755
},
{
"epoch": 1.68,
"learning_rate": 2.5048709394422254e-06,
"loss": 0.694,
"step": 760
},
{
"epoch": 1.69,
"learning_rate": 1.9065702289717339e-06,
"loss": 0.6621,
"step": 765
},
{
"epoch": 1.7,
"learning_rate": 1.3888077756976535e-06,
"loss": 0.706,
"step": 770
},
{
"epoch": 1.71,
"learning_rate": 9.522088239022507e-07,
"loss": 0.7326,
"step": 775
},
{
"epoch": 1.73,
"learning_rate": 5.973006057100785e-07,
"loss": 0.6361,
"step": 780
},
{
"epoch": 1.74,
"learning_rate": 3.245117044082019e-07,
"loss": 0.647,
"step": 785
},
{
"epoch": 1.75,
"learning_rate": 1.3417153689374065e-07,
"loss": 0.7488,
"step": 790
},
{
"epoch": 1.76,
"learning_rate": 2.6509955873614998e-08,
"loss": 0.7149,
"step": 795
},
{
"epoch": 1.77,
"learning_rate": 1.6569722969137058e-09,
"loss": 0.6775,
"step": 800
},
{
"epoch": 1.78,
"learning_rate": 5.964259835508488e-08,
"loss": 0.7746,
"step": 805
},
{
"epoch": 1.79,
"learning_rate": 2.003968112395687e-07,
"loss": 0.6598,
"step": 810
},
{
"epoch": 1.8,
"learning_rate": 4.2374963770057325e-07,
"loss": 0.6891,
"step": 815
},
{
"epoch": 1.81,
"learning_rate": 7.294313593051582e-07,
"loss": 0.6988,
"step": 820
},
{
"epoch": 1.83,
"learning_rate": 1.1170728381449087e-06,
"loss": 0.6761,
"step": 825
},
{
"epoch": 1.84,
"learning_rate": 1.5862059626056298e-06,
"loss": 0.7306,
"step": 830
},
{
"epoch": 1.85,
"learning_rate": 2.1362642126509973e-06,
"loss": 0.6297,
"step": 835
},
{
"epoch": 1.86,
"learning_rate": 2.766583343947401e-06,
"loss": 0.7385,
"step": 840
},
{
"epoch": 1.87,
"learning_rate": 3.4764021899978206e-06,
"loss": 0.7045,
"step": 845
},
{
"epoch": 1.88,
"learning_rate": 4.264863581318223e-06,
"loss": 0.6952,
"step": 850
},
{
"epoch": 1.89,
"learning_rate": 5.131015380546021e-06,
"loss": 0.7021,
"step": 855
},
{
"epoch": 1.9,
"learning_rate": 6.073811632230782e-06,
"loss": 0.6392,
"step": 860
},
{
"epoch": 1.91,
"learning_rate": 7.09211382591858e-06,
"loss": 0.694,
"step": 865
},
{
"epoch": 1.92,
"learning_rate": 8.18469227100481e-06,
"loss": 0.6864,
"step": 870
},
{
"epoch": 1.94,
"learning_rate": 9.350227581695092e-06,
"loss": 0.7318,
"step": 875
},
{
"epoch": 1.95,
"learning_rate": 1.0587312270281189e-05,
"loss": 0.6326,
"step": 880
},
{
"epoch": 1.96,
"learning_rate": 1.1894452446807735e-05,
"loss": 0.7141,
"step": 885
},
{
"epoch": 1.97,
"learning_rate": 1.3270069623077462e-05,
"loss": 0.713,
"step": 890
},
{
"epoch": 1.98,
"learning_rate": 1.47125026188163e-05,
"loss": 0.7505,
"step": 895
},
{
"epoch": 1.99,
"learning_rate": 1.6220009567696506e-05,
"loss": 0.7118,
"step": 900
},
{
"epoch": 2.0,
"eval_loss": 0.3754318356513977,
"eval_runtime": 8.7593,
"eval_samples_per_second": 76.147,
"eval_steps_per_second": 9.59,
"step": 904
}
],
"max_steps": 904,
"num_train_epochs": 2,
"total_flos": 943264235520000.0,
"trial_name": null,
"trial_params": null
}