eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
2d6dc04
raw
history blame
23.1 kB
{
"best_metric": 3.550342321395874,
"best_model_checkpoint": "output/eminem/checkpoint-920",
"epoch": 2.0,
"global_step": 920,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.00013715876234566868,
"loss": 4.4386,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 0.00013703509896122095,
"loss": 4.1948,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 0.00013682915852268886,
"loss": 4.0957,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 0.00013654118862484264,
"loss": 3.9438,
"step": 20
},
{
"epoch": 0.06,
"learning_rate": 0.00013617153548351626,
"loss": 4.058,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 0.00013572064351936462,
"loss": 4.0441,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 0.00013518905482355273,
"loss": 3.872,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 0.00013457740850601892,
"loss": 4.2427,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 0.00013388643992709594,
"loss": 4.0466,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 0.0001331169798134139,
"loss": 3.9197,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 0.00013226995325914744,
"loss": 4.0498,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 0.00013134637861380834,
"loss": 3.9624,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 0.0001303473662579206,
"loss": 3.8491,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 0.00012927411726804995,
"loss": 3.8877,
"step": 70
},
{
"epoch": 0.17,
"learning_rate": 0.00012812792197279278,
"loss": 3.9778,
"step": 75
},
{
"epoch": 0.18,
"learning_rate": 0.00012691015840146053,
"loss": 3.9312,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 0.00012562229062732468,
"loss": 3.8452,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 0.00012426586700741422,
"loss": 3.9473,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 0.00012284251832098172,
"loss": 4.0293,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 0.00012135395580887633,
"loss": 3.8128,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 0.00011980196911618039,
"loss": 4.0833,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 0.0001181884241405837,
"loss": 3.7779,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 0.00011651526078908192,
"loss": 3.8247,
"step": 115
},
{
"epoch": 0.26,
"learning_rate": 0.00011478449064569633,
"loss": 3.9986,
"step": 120
},
{
"epoch": 0.28,
"learning_rate": 0.00011299819455301873,
"loss": 4.0396,
"step": 125
},
{
"epoch": 0.29,
"learning_rate": 0.0001111585201104895,
"loss": 3.8924,
"step": 130
},
{
"epoch": 0.3,
"learning_rate": 0.000109267679092416,
"loss": 3.8117,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 0.00010732794478883606,
"loss": 3.9015,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 0.00010534164927242335,
"loss": 3.7786,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 0.0001033111805947203,
"loss": 4.0211,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 0.00010123897991506982,
"loss": 3.8728,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 9.912753856569734e-05,
"loss": 3.9266,
"step": 160
},
{
"epoch": 0.36,
"learning_rate": 9.697939505647188e-05,
"loss": 3.9463,
"step": 165
},
{
"epoch": 0.38,
"learning_rate": 9.479713202294696e-05,
"loss": 3.7152,
"step": 170
},
{
"epoch": 0.39,
"learning_rate": 9.258337312135107e-05,
"loss": 3.9496,
"step": 175
},
{
"epoch": 0.4,
"learning_rate": 9.034077987426021e-05,
"loss": 3.7439,
"step": 180
},
{
"epoch": 0.41,
"learning_rate": 8.807204847074523e-05,
"loss": 3.7879,
"step": 185
},
{
"epoch": 0.42,
"learning_rate": 8.577990652484077e-05,
"loss": 3.7768,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 8.34671097962332e-05,
"loss": 3.6851,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 8.113643887711011e-05,
"loss": 3.7911,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 7.879069584915438e-05,
"loss": 3.6839,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 7.643270091470234e-05,
"loss": 3.7319,
"step": 210
},
{
"epoch": 0.47,
"learning_rate": 7.406528900611617e-05,
"loss": 3.6703,
"step": 215
},
{
"epoch": 0.49,
"learning_rate": 7.169130637744674e-05,
"loss": 3.7467,
"step": 220
},
{
"epoch": 0.5,
"learning_rate": 6.931360718248504e-05,
"loss": 3.8521,
"step": 225
},
{
"epoch": 0.51,
"learning_rate": 6.693505004331577e-05,
"loss": 3.6709,
"step": 230
},
{
"epoch": 0.52,
"learning_rate": 6.455849461349907e-05,
"loss": 3.6242,
"step": 235
},
{
"epoch": 0.53,
"learning_rate": 6.218679814001198e-05,
"loss": 3.6225,
"step": 240
},
{
"epoch": 0.54,
"learning_rate": 5.9822812028083505e-05,
"loss": 3.8016,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 5.746937841305257e-05,
"loss": 3.6923,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 5.512932674337138e-05,
"loss": 3.894,
"step": 255
},
{
"epoch": 0.57,
"learning_rate": 5.280547037886122e-05,
"loss": 3.7461,
"step": 260
},
{
"epoch": 0.58,
"learning_rate": 5.050060320831149e-05,
"loss": 3.5966,
"step": 265
},
{
"epoch": 0.6,
"learning_rate": 4.821749629048772e-05,
"loss": 3.7551,
"step": 270
},
{
"epoch": 0.61,
"learning_rate": 4.595889452258756e-05,
"loss": 3.8694,
"step": 275
},
{
"epoch": 0.62,
"learning_rate": 4.372751334014969e-05,
"loss": 3.7224,
"step": 280
},
{
"epoch": 0.63,
"learning_rate": 4.1526035452383523e-05,
"loss": 3.915,
"step": 285
},
{
"epoch": 0.64,
"learning_rate": 3.935710761684453e-05,
"loss": 3.8484,
"step": 290
},
{
"epoch": 0.65,
"learning_rate": 3.722333745733311e-05,
"loss": 3.7062,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 3.512729032884219e-05,
"loss": 3.676,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 3.3071486233323674e-05,
"loss": 3.6448,
"step": 305
},
{
"epoch": 0.68,
"learning_rate": 3.105839678998049e-05,
"loss": 3.9731,
"step": 310
},
{
"epoch": 0.7,
"learning_rate": 2.9090442263728265e-05,
"loss": 3.6278,
"step": 315
},
{
"epoch": 0.71,
"learning_rate": 2.716998865539764e-05,
"loss": 3.6389,
"step": 320
},
{
"epoch": 0.72,
"learning_rate": 2.5299344857176957e-05,
"loss": 3.6411,
"step": 325
},
{
"epoch": 0.73,
"learning_rate": 2.3480759876714295e-05,
"loss": 3.7764,
"step": 330
},
{
"epoch": 0.74,
"learning_rate": 2.1716420133216482e-05,
"loss": 3.7769,
"step": 335
},
{
"epoch": 0.75,
"learning_rate": 2.0008446828796293e-05,
"loss": 3.528,
"step": 340
},
{
"epoch": 0.76,
"learning_rate": 1.8358893398227267e-05,
"loss": 3.819,
"step": 345
},
{
"epoch": 0.77,
"learning_rate": 1.6769743040173313e-05,
"loss": 3.6014,
"step": 350
},
{
"epoch": 0.78,
"learning_rate": 1.5242906332860249e-05,
"loss": 3.6993,
"step": 355
},
{
"epoch": 0.79,
"learning_rate": 1.3780218937056495e-05,
"loss": 3.7919,
"step": 360
},
{
"epoch": 0.81,
"learning_rate": 1.2383439389124231e-05,
"loss": 3.6903,
"step": 365
},
{
"epoch": 0.82,
"learning_rate": 1.105424698679451e-05,
"loss": 3.4355,
"step": 370
},
{
"epoch": 0.83,
"learning_rate": 9.794239770208025e-06,
"loss": 3.7661,
"step": 375
},
{
"epoch": 0.84,
"learning_rate": 8.604932600649156e-06,
"loss": 4.0533,
"step": 380
},
{
"epoch": 0.85,
"learning_rate": 7.487755339282637e-06,
"loss": 3.7796,
"step": 385
},
{
"epoch": 0.86,
"learning_rate": 6.444051128083183e-06,
"loss": 3.7312,
"step": 390
},
{
"epoch": 0.87,
"learning_rate": 5.4750747750241e-06,
"loss": 3.7231,
"step": 395
},
{
"epoch": 0.88,
"learning_rate": 4.581991245466992e-06,
"loss": 3.7788,
"step": 400
},
{
"epoch": 0.89,
"learning_rate": 3.7658742615658213e-06,
"loss": 3.7838,
"step": 405
},
{
"epoch": 0.91,
"learning_rate": 3.027705011369445e-06,
"loss": 3.7598,
"step": 410
},
{
"epoch": 0.92,
"learning_rate": 2.3683709691745994e-06,
"loss": 3.954,
"step": 415
},
{
"epoch": 0.93,
"learning_rate": 1.7886648285474887e-06,
"loss": 3.7673,
"step": 420
},
{
"epoch": 0.94,
"learning_rate": 1.289283549296875e-06,
"loss": 3.8811,
"step": 425
},
{
"epoch": 0.95,
"learning_rate": 8.708275195444353e-07,
"loss": 3.6489,
"step": 430
},
{
"epoch": 0.96,
"learning_rate": 5.337998338997259e-07,
"loss": 3.6089,
"step": 435
},
{
"epoch": 0.97,
"learning_rate": 2.786056886076668e-07,
"loss": 3.5491,
"step": 440
},
{
"epoch": 0.98,
"learning_rate": 1.0555189439568316e-07,
"loss": 3.7324,
"step": 445
},
{
"epoch": 0.99,
"learning_rate": 1.484650760624615e-08,
"loss": 3.7782,
"step": 450
},
{
"epoch": 1.0,
"eval_loss": 3.7048158645629883,
"eval_runtime": 29.3019,
"eval_samples_per_second": 22.388,
"eval_steps_per_second": 2.798,
"step": 453
},
{
"epoch": 0.99,
"learning_rate": 3.9992265680461966e-08,
"loss": 3.6362,
"step": 455
},
{
"epoch": 1.0,
"learning_rate": 0.0,
"loss": 3.3994,
"step": 460
},
{
"epoch": 1.0,
"eval_loss": 3.5555710792541504,
"eval_runtime": 13.5198,
"eval_samples_per_second": 44.675,
"eval_steps_per_second": 5.621,
"step": 460
},
{
"epoch": 1.01,
"learning_rate": 3.9992265680461966e-08,
"loss": 3.5614,
"step": 465
},
{
"epoch": 1.02,
"learning_rate": 1.5992243352901425e-07,
"loss": 3.7249,
"step": 470
},
{
"epoch": 1.03,
"learning_rate": 3.5965067033469397e-07,
"loss": 3.4991,
"step": 475
},
{
"epoch": 1.04,
"learning_rate": 6.389441019077102e-07,
"loss": 3.439,
"step": 480
},
{
"epoch": 1.05,
"learning_rate": 9.97477084600295e-07,
"loss": 3.6568,
"step": 485
},
{
"epoch": 1.07,
"learning_rate": 1.4348315849926483e-06,
"loss": 3.5489,
"step": 490
},
{
"epoch": 1.08,
"learning_rate": 1.9504976673012086e-06,
"loss": 3.6268,
"step": 495
},
{
"epoch": 1.09,
"learning_rate": 2.5438740879409643e-06,
"loss": 3.8246,
"step": 500
},
{
"epoch": 1.1,
"learning_rate": 3.2142689965485674e-06,
"loss": 3.5789,
"step": 505
},
{
"epoch": 1.11,
"learning_rate": 3.960900742648913e-06,
"loss": 3.618,
"step": 510
},
{
"epoch": 1.12,
"learning_rate": 4.782898787024646e-06,
"loss": 3.6802,
"step": 515
},
{
"epoch": 1.13,
"learning_rate": 5.679304716725914e-06,
"loss": 3.6302,
"step": 520
},
{
"epoch": 1.14,
"learning_rate": 6.649073362537048e-06,
"loss": 3.5926,
"step": 525
},
{
"epoch": 1.15,
"learning_rate": 7.691074017597052e-06,
"loss": 3.5005,
"step": 530
},
{
"epoch": 1.16,
"learning_rate": 8.804091755753263e-06,
"loss": 3.6247,
"step": 535
},
{
"epoch": 1.17,
"learning_rate": 9.986828848110884e-06,
"loss": 3.5699,
"step": 540
},
{
"epoch": 1.18,
"learning_rate": 1.1237906276126821e-05,
"loss": 3.5435,
"step": 545
},
{
"epoch": 1.2,
"learning_rate": 1.2555865339483672e-05,
"loss": 3.5681,
"step": 550
},
{
"epoch": 1.21,
"learning_rate": 1.3939169356868945e-05,
"loss": 3.5438,
"step": 555
},
{
"epoch": 1.22,
"learning_rate": 1.5386205457676803e-05,
"loss": 3.6615,
"step": 560
},
{
"epoch": 1.23,
"learning_rate": 1.6895286462543014e-05,
"loss": 3.5334,
"step": 565
},
{
"epoch": 1.24,
"learning_rate": 1.8464652850520678e-05,
"loss": 3.5256,
"step": 570
},
{
"epoch": 1.25,
"learning_rate": 2.009247481060283e-05,
"loss": 3.5908,
"step": 575
},
{
"epoch": 1.26,
"learning_rate": 2.1776854375200328e-05,
"loss": 3.6817,
"step": 580
},
{
"epoch": 1.27,
"learning_rate": 2.351582763308709e-05,
"loss": 3.6254,
"step": 585
},
{
"epoch": 1.28,
"learning_rate": 2.5307367019232758e-05,
"loss": 3.5371,
"step": 590
},
{
"epoch": 1.29,
"learning_rate": 2.714938367885288e-05,
"loss": 3.6406,
"step": 595
},
{
"epoch": 1.3,
"learning_rate": 2.9039729902920112e-05,
"loss": 3.5787,
"step": 600
},
{
"epoch": 1.32,
"learning_rate": 3.097620163229676e-05,
"loss": 3.5732,
"step": 605
},
{
"epoch": 1.33,
"learning_rate": 3.295654102756921e-05,
"loss": 3.5091,
"step": 610
},
{
"epoch": 1.34,
"learning_rate": 3.4978439101588006e-05,
"loss": 3.617,
"step": 615
},
{
"epoch": 1.35,
"learning_rate": 3.703953841164292e-05,
"loss": 3.582,
"step": 620
},
{
"epoch": 1.36,
"learning_rate": 3.913743580813637e-05,
"loss": 3.6594,
"step": 625
},
{
"epoch": 1.37,
"learning_rate": 4.126968523654786e-05,
"loss": 3.6063,
"step": 630
},
{
"epoch": 1.38,
"learning_rate": 4.343380058942427e-05,
"loss": 3.6137,
"step": 635
},
{
"epoch": 1.39,
"learning_rate": 4.562725860507034e-05,
"loss": 3.5565,
"step": 640
},
{
"epoch": 1.4,
"learning_rate": 4.784750180955822e-05,
"loss": 3.6099,
"step": 645
},
{
"epoch": 1.41,
"learning_rate": 5.009194149862813e-05,
"loss": 3.6417,
"step": 650
},
{
"epoch": 1.42,
"learning_rate": 5.235796075600178e-05,
"loss": 3.7049,
"step": 655
},
{
"epoch": 1.43,
"learning_rate": 5.4642917504589275e-05,
"loss": 3.7319,
"step": 660
},
{
"epoch": 1.45,
"learning_rate": 5.694414758703346e-05,
"loss": 3.554,
"step": 665
},
{
"epoch": 1.46,
"learning_rate": 5.92589678719975e-05,
"loss": 3.5522,
"step": 670
},
{
"epoch": 1.47,
"learning_rate": 6.158467938257645e-05,
"loss": 3.6115,
"step": 675
},
{
"epoch": 1.48,
"learning_rate": 6.391857044318355e-05,
"loss": 3.6475,
"step": 680
},
{
"epoch": 1.49,
"learning_rate": 6.625791984124255e-05,
"loss": 3.5034,
"step": 685
},
{
"epoch": 1.5,
"learning_rate": 6.859999999999999e-05,
"loss": 3.5767,
"step": 690
},
{
"epoch": 1.51,
"learning_rate": 7.094208015875743e-05,
"loss": 3.6471,
"step": 695
},
{
"epoch": 1.52,
"learning_rate": 7.328142955681643e-05,
"loss": 3.6515,
"step": 700
},
{
"epoch": 1.53,
"learning_rate": 7.561532061742353e-05,
"loss": 3.6649,
"step": 705
},
{
"epoch": 1.54,
"learning_rate": 7.794103212800247e-05,
"loss": 3.4103,
"step": 710
},
{
"epoch": 1.55,
"learning_rate": 8.025585241296653e-05,
"loss": 3.5488,
"step": 715
},
{
"epoch": 1.57,
"learning_rate": 8.25570824954107e-05,
"loss": 3.5076,
"step": 720
},
{
"epoch": 1.58,
"learning_rate": 8.484203924399819e-05,
"loss": 3.5708,
"step": 725
},
{
"epoch": 1.59,
"learning_rate": 8.710805850137184e-05,
"loss": 3.7272,
"step": 730
},
{
"epoch": 1.6,
"learning_rate": 8.935249819044176e-05,
"loss": 3.6548,
"step": 735
},
{
"epoch": 1.61,
"learning_rate": 9.157274139492964e-05,
"loss": 3.6463,
"step": 740
},
{
"epoch": 1.62,
"learning_rate": 9.376619941057571e-05,
"loss": 3.484,
"step": 745
},
{
"epoch": 1.63,
"learning_rate": 9.593031476345212e-05,
"loss": 3.5106,
"step": 750
},
{
"epoch": 1.64,
"learning_rate": 9.80625641918636e-05,
"loss": 3.6845,
"step": 755
},
{
"epoch": 1.65,
"learning_rate": 0.00010016046158835706,
"loss": 3.5078,
"step": 760
},
{
"epoch": 1.66,
"learning_rate": 0.00010222156089841198,
"loss": 3.62,
"step": 765
},
{
"epoch": 1.67,
"learning_rate": 0.00010424345897243078,
"loss": 3.644,
"step": 770
},
{
"epoch": 1.68,
"learning_rate": 0.00010622379836770322,
"loss": 3.7684,
"step": 775
},
{
"epoch": 1.7,
"learning_rate": 0.00010816027009707987,
"loss": 3.5978,
"step": 780
},
{
"epoch": 1.71,
"learning_rate": 0.0001100506163211471,
"loss": 3.5937,
"step": 785
},
{
"epoch": 1.72,
"learning_rate": 0.00011189263298076723,
"loss": 3.6434,
"step": 790
},
{
"epoch": 1.73,
"learning_rate": 0.00011368417236691289,
"loss": 3.6893,
"step": 795
},
{
"epoch": 1.74,
"learning_rate": 0.00011542314562479964,
"loss": 3.6739,
"step": 800
},
{
"epoch": 1.75,
"learning_rate": 0.00011710752518939715,
"loss": 3.5122,
"step": 805
},
{
"epoch": 1.76,
"learning_rate": 0.00011873534714947934,
"loss": 3.6633,
"step": 810
},
{
"epoch": 1.77,
"learning_rate": 0.00012030471353745696,
"loss": 3.5101,
"step": 815
},
{
"epoch": 1.78,
"learning_rate": 0.00012181379454232318,
"loss": 3.6736,
"step": 820
},
{
"epoch": 1.79,
"learning_rate": 0.00012326083064313103,
"loss": 3.4271,
"step": 825
},
{
"epoch": 1.8,
"learning_rate": 0.0001246441346605163,
"loss": 3.5088,
"step": 830
},
{
"epoch": 1.82,
"learning_rate": 0.00012596209372387317,
"loss": 3.5845,
"step": 835
},
{
"epoch": 1.83,
"learning_rate": 0.0001272131711518891,
"loss": 3.6172,
"step": 840
},
{
"epoch": 1.84,
"learning_rate": 0.00012839590824424672,
"loss": 3.5649,
"step": 845
},
{
"epoch": 1.85,
"learning_rate": 0.00012950892598240292,
"loss": 3.5506,
"step": 850
},
{
"epoch": 1.86,
"learning_rate": 0.00013055092663746294,
"loss": 3.4779,
"step": 855
},
{
"epoch": 1.87,
"learning_rate": 0.00013152069528327408,
"loss": 3.5707,
"step": 860
},
{
"epoch": 1.88,
"learning_rate": 0.00013241710121297533,
"loss": 3.497,
"step": 865
},
{
"epoch": 1.89,
"learning_rate": 0.0001332390992573511,
"loss": 3.6343,
"step": 870
},
{
"epoch": 1.9,
"learning_rate": 0.00013398573100345144,
"loss": 3.5251,
"step": 875
},
{
"epoch": 1.91,
"learning_rate": 0.00013465612591205902,
"loss": 3.7421,
"step": 880
},
{
"epoch": 1.92,
"learning_rate": 0.00013524950233269879,
"loss": 3.5967,
"step": 885
},
{
"epoch": 1.93,
"learning_rate": 0.00013576516841500732,
"loss": 3.5853,
"step": 890
},
{
"epoch": 1.95,
"learning_rate": 0.0001362025229153997,
"loss": 3.7016,
"step": 895
},
{
"epoch": 1.96,
"learning_rate": 0.00013656105589809228,
"loss": 3.7649,
"step": 900
},
{
"epoch": 1.97,
"learning_rate": 0.0001368403493296653,
"loss": 3.6488,
"step": 905
},
{
"epoch": 1.98,
"learning_rate": 0.000137040077566471,
"loss": 3.5113,
"step": 910
},
{
"epoch": 1.99,
"learning_rate": 0.00013716000773431953,
"loss": 3.4688,
"step": 915
},
{
"epoch": 2.0,
"learning_rate": 0.0001372,
"loss": 3.5363,
"step": 920
},
{
"epoch": 2.0,
"eval_loss": 3.550342321395874,
"eval_runtime": 15.5302,
"eval_samples_per_second": 38.892,
"eval_steps_per_second": 4.894,
"step": 920
}
],
"max_steps": 920,
"num_train_epochs": 2,
"total_flos": 959594987520000.0,
"trial_name": null,
"trial_params": null
}