og-buda / trainer_state.json
AlekseyKorshuk's picture
huggingartists
a8807a1
{
"best_metric": 1.5331557989120483,
"best_model_checkpoint": "output/og-buda/checkpoint-750",
"epoch": 5.0,
"global_step": 750,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.00013683401763576845,
"loss": 2.7577,
"step": 5
},
{
"epoch": 0.07,
"learning_rate": 0.0001357399755894565,
"loss": 2.4782,
"step": 10
},
{
"epoch": 0.1,
"learning_rate": 0.00013392954733321228,
"loss": 2.3702,
"step": 15
},
{
"epoch": 0.13,
"learning_rate": 0.00013142205020853694,
"loss": 2.2696,
"step": 20
},
{
"epoch": 0.16,
"learning_rate": 0.00012824423930948275,
"loss": 2.2909,
"step": 25
},
{
"epoch": 0.2,
"learning_rate": 0.00012443002200473538,
"loss": 2.2339,
"step": 30
},
{
"epoch": 0.23,
"learning_rate": 0.00012002009614464126,
"loss": 2.1468,
"step": 35
},
{
"epoch": 0.26,
"learning_rate": 0.00011506151581352585,
"loss": 2.2085,
"step": 40
},
{
"epoch": 0.3,
"learning_rate": 0.00010960718926074231,
"loss": 2.0772,
"step": 45
},
{
"epoch": 0.33,
"learning_rate": 0.00010371531436754662,
"loss": 2.1209,
"step": 50
},
{
"epoch": 0.36,
"learning_rate": 9.744875767338973e-05,
"loss": 2.1212,
"step": 55
},
{
"epoch": 0.39,
"learning_rate": 9.08743835874413e-05,
"loss": 2.1052,
"step": 60
},
{
"epoch": 0.43,
"learning_rate": 8.40623409426843e-05,
"loss": 2.0228,
"step": 65
},
{
"epoch": 0.46,
"learning_rate": 7.708531450507727e-05,
"loss": 2.0757,
"step": 70
},
{
"epoch": 0.49,
"learning_rate": 7.001774942421023e-05,
"loss": 2.001,
"step": 75
},
{
"epoch": 0.53,
"learning_rate": 6.293505690059801e-05,
"loss": 1.9417,
"step": 80
},
{
"epoch": 0.56,
"learning_rate": 5.59128095451711e-05,
"loss": 2.0215,
"step": 85
},
{
"epoch": 0.59,
"learning_rate": 4.902593501651139e-05,
"loss": 2.0056,
"step": 90
},
{
"epoch": 0.62,
"learning_rate": 4.2347916539754844e-05,
"loss": 2.0711,
"step": 95
},
{
"epoch": 0.66,
"learning_rate": 3.5950008837656755e-05,
"loss": 2.0311,
"step": 100
},
{
"epoch": 0.69,
"learning_rate": 2.9900477839865778e-05,
"loss": 1.9157,
"step": 105
},
{
"epoch": 0.72,
"learning_rate": 2.4263872282739445e-05,
"loss": 2.0822,
"step": 110
},
{
"epoch": 0.76,
"learning_rate": 1.910033497175995e-05,
"loss": 2.0431,
"step": 115
},
{
"epoch": 0.79,
"learning_rate": 1.4464961055407408e-05,
"loss": 1.9706,
"step": 120
},
{
"epoch": 0.82,
"learning_rate": 1.040721015773445e-05,
"loss": 1.9533,
"step": 125
},
{
"epoch": 0.86,
"learning_rate": 6.97037864221025e-06,
"loss": 1.968,
"step": 130
},
{
"epoch": 0.89,
"learning_rate": 4.1911376378003306e-06,
"loss": 1.9282,
"step": 135
},
{
"epoch": 0.92,
"learning_rate": 2.0991417565619363e-06,
"loss": 1.9382,
"step": 140
},
{
"epoch": 0.95,
"learning_rate": 7.167126777538922e-07,
"loss": 1.9811,
"step": 145
},
{
"epoch": 0.99,
"learning_rate": 5.860097463115006e-08,
"loss": 1.9187,
"step": 150
},
{
"epoch": 1.0,
"eval_loss": 1.8772083520889282,
"eval_runtime": 9.7424,
"eval_samples_per_second": 22.274,
"eval_steps_per_second": 2.874,
"step": 152
},
{
"epoch": 1.02,
"learning_rate": 1.3182872524144475e-07,
"loss": 1.9052,
"step": 155
},
{
"epoch": 1.05,
"learning_rate": 9.35614586573241e-07,
"loss": 1.9644,
"step": 160
},
{
"epoch": 1.09,
"learning_rate": 2.461382131516746e-06,
"loss": 1.9317,
"step": 165
},
{
"epoch": 1.12,
"learning_rate": 4.692851359682261e-06,
"loss": 2.0608,
"step": 170
},
{
"epoch": 1.15,
"learning_rate": 7.606212405651899e-06,
"loss": 1.8731,
"step": 175
},
{
"epoch": 1.18,
"learning_rate": 1.1170379591190527e-05,
"loss": 1.8801,
"step": 180
},
{
"epoch": 1.22,
"learning_rate": 1.5347323110669033e-05,
"loss": 1.8913,
"step": 185
},
{
"epoch": 1.25,
"learning_rate": 2.009247481060283e-05,
"loss": 1.9579,
"step": 190
},
{
"epoch": 1.28,
"learning_rate": 2.5355203733622272e-05,
"loss": 1.9638,
"step": 195
},
{
"epoch": 1.32,
"learning_rate": 3.1079356352801514e-05,
"loss": 1.9192,
"step": 200
},
{
"epoch": 1.35,
"learning_rate": 3.7203855732024184e-05,
"loss": 1.9342,
"step": 205
},
{
"epoch": 1.38,
"learning_rate": 4.3663353219321276e-05,
"loss": 1.9085,
"step": 210
},
{
"epoch": 1.41,
"learning_rate": 5.038892571958626e-05,
"loss": 1.9788,
"step": 215
},
{
"epoch": 1.45,
"learning_rate": 5.7308811106741675e-05,
"loss": 1.9669,
"step": 220
},
{
"epoch": 1.48,
"learning_rate": 6.434917392847842e-05,
"loss": 1.927,
"step": 225
},
{
"epoch": 1.51,
"learning_rate": 7.143489323346855e-05,
"loss": 1.9403,
"step": 230
},
{
"epoch": 1.55,
"learning_rate": 7.849036411490396e-05,
"loss": 1.8539,
"step": 235
},
{
"epoch": 1.58,
"learning_rate": 8.54403044178588e-05,
"loss": 1.9165,
"step": 240
},
{
"epoch": 1.61,
"learning_rate": 9.221055800287699e-05,
"loss": 1.906,
"step": 245
},
{
"epoch": 1.64,
"learning_rate": 9.87288859949296e-05,
"loss": 1.852,
"step": 250
},
{
"epoch": 1.68,
"learning_rate": 0.00010492573757508462,
"loss": 1.8733,
"step": 255
},
{
"epoch": 1.71,
"learning_rate": 0.00011073499209051121,
"loss": 1.874,
"step": 260
},
{
"epoch": 1.74,
"learning_rate": 0.00011609466456447531,
"loss": 1.8853,
"step": 265
},
{
"epoch": 1.78,
"learning_rate": 0.00012094756707850673,
"loss": 1.9022,
"step": 270
},
{
"epoch": 1.81,
"learning_rate": 0.0001252419189697637,
"loss": 1.8298,
"step": 275
},
{
"epoch": 1.84,
"learning_rate": 0.00012893189933276512,
"loss": 1.9105,
"step": 280
},
{
"epoch": 1.88,
"learning_rate": 0.00013197813593027427,
"loss": 1.8641,
"step": 285
},
{
"epoch": 1.91,
"learning_rate": 0.00013434812529663614,
"loss": 1.8622,
"step": 290
},
{
"epoch": 1.94,
"learning_rate": 0.00013601657955104887,
"loss": 1.906,
"step": 295
},
{
"epoch": 1.97,
"learning_rate": 0.00013696569622025754,
"loss": 1.9195,
"step": 300
},
{
"epoch": 2.0,
"eval_loss": 1.8195273876190186,
"eval_runtime": 9.7405,
"eval_samples_per_second": 22.278,
"eval_steps_per_second": 2.875,
"step": 304
},
{
"epoch": 2.01,
"learning_rate": 0.0001371853481916521,
"loss": 1.7633,
"step": 305
},
{
"epoch": 2.04,
"learning_rate": 0.00013667319176996132,
"loss": 1.8122,
"step": 310
},
{
"epoch": 2.07,
"learning_rate": 0.00013543469168457496,
"loss": 1.8677,
"step": 315
},
{
"epoch": 2.11,
"learning_rate": 0.00013348306278066356,
"loss": 1.8531,
"step": 320
},
{
"epoch": 2.14,
"learning_rate": 0.000130839129016255,
"loss": 1.8181,
"step": 325
},
{
"epoch": 2.17,
"learning_rate": 0.00012753110126977412,
"loss": 1.8011,
"step": 330
},
{
"epoch": 2.2,
"learning_rate": 0.0001235942763288477,
"loss": 1.8196,
"step": 335
},
{
"epoch": 2.24,
"learning_rate": 0.00011907066027217684,
"loss": 1.8093,
"step": 340
},
{
"epoch": 2.27,
"learning_rate": 0.00011400852026300626,
"loss": 1.8397,
"step": 345
},
{
"epoch": 2.3,
"learning_rate": 0.00010846186953657296,
"loss": 1.7362,
"step": 350
},
{
"epoch": 2.34,
"learning_rate": 0.00010248989107673776,
"loss": 1.6683,
"step": 355
},
{
"epoch": 2.37,
"learning_rate": 9.615630613119375e-05,
"loss": 1.8356,
"step": 360
},
{
"epoch": 2.4,
"learning_rate": 8.952869430321868e-05,
"loss": 1.8106,
"step": 365
},
{
"epoch": 2.43,
"learning_rate": 8.267777247462053e-05,
"loss": 1.8027,
"step": 370
},
{
"epoch": 2.47,
"learning_rate": 7.567664025379571e-05,
"loss": 1.7443,
"step": 375
},
{
"epoch": 2.5,
"learning_rate": 6.860000000000001e-05,
"loss": 1.7579,
"step": 380
},
{
"epoch": 2.53,
"learning_rate": 6.152335974620434e-05,
"loss": 1.7847,
"step": 385
},
{
"epoch": 2.57,
"learning_rate": 5.45222275253795e-05,
"loss": 1.7549,
"step": 390
},
{
"epoch": 2.6,
"learning_rate": 4.76713056967813e-05,
"loss": 1.7157,
"step": 395
},
{
"epoch": 2.63,
"learning_rate": 4.1043693868806304e-05,
"loss": 1.7905,
"step": 400
},
{
"epoch": 2.66,
"learning_rate": 3.471010892326228e-05,
"loss": 1.7419,
"step": 405
},
{
"epoch": 2.7,
"learning_rate": 2.873813046342713e-05,
"loss": 1.7189,
"step": 410
},
{
"epoch": 2.73,
"learning_rate": 2.3191479736993775e-05,
"loss": 1.7705,
"step": 415
},
{
"epoch": 2.76,
"learning_rate": 1.812933972782316e-05,
"loss": 1.6987,
"step": 420
},
{
"epoch": 2.8,
"learning_rate": 1.3605723671152311e-05,
"loss": 1.7951,
"step": 425
},
{
"epoch": 2.83,
"learning_rate": 9.668898730225861e-06,
"loss": 1.7204,
"step": 430
},
{
"epoch": 2.86,
"learning_rate": 6.360870983745002e-06,
"loss": 1.8063,
"step": 435
},
{
"epoch": 2.89,
"learning_rate": 3.7169372193364486e-06,
"loss": 1.6961,
"step": 440
},
{
"epoch": 2.93,
"learning_rate": 1.7653083154250801e-06,
"loss": 1.6979,
"step": 445
},
{
"epoch": 2.96,
"learning_rate": 5.268082300387043e-07,
"loss": 1.7594,
"step": 450
},
{
"epoch": 2.99,
"learning_rate": 1.465180834793276e-08,
"loss": 1.7887,
"step": 455
},
{
"epoch": 3.0,
"eval_loss": 1.7267966270446777,
"eval_runtime": 9.7478,
"eval_samples_per_second": 22.262,
"eval_steps_per_second": 2.872,
"step": 456
},
{
"epoch": 2.99,
"learning_rate": 5.708896496161406e-08,
"loss": 1.777,
"step": 460
},
{
"epoch": 3.0,
"eval_loss": 1.6688779592514038,
"eval_runtime": 9.6717,
"eval_samples_per_second": 20.576,
"eval_steps_per_second": 2.585,
"step": 462
},
{
"epoch": 3.1,
"learning_rate": 3.35752298215246e-06,
"loss": 1.6773,
"step": 465
},
{
"epoch": 3.13,
"learning_rate": 5.930781605717588e-06,
"loss": 1.7743,
"step": 470
},
{
"epoch": 3.17,
"learning_rate": 9.190657300387474e-06,
"loss": 1.6782,
"step": 475
},
{
"epoch": 3.2,
"learning_rate": 1.310143418587859e-05,
"loss": 1.7254,
"step": 480
},
{
"epoch": 3.23,
"learning_rate": 1.7620264972250762e-05,
"loss": 1.7116,
"step": 485
},
{
"epoch": 3.27,
"learning_rate": 2.269764040378228e-05,
"loss": 1.737,
"step": 490
},
{
"epoch": 3.3,
"learning_rate": 2.827793169273632e-05,
"loss": 1.6298,
"step": 495
},
{
"epoch": 3.33,
"learning_rate": 3.4300000000000014e-05,
"loss": 1.784,
"step": 500
},
{
"epoch": 3.37,
"learning_rate": 4.069786628500004e-05,
"loss": 1.7561,
"step": 505
},
{
"epoch": 3.4,
"learning_rate": 4.740143418587858e-05,
"loss": 1.7454,
"step": 510
},
{
"epoch": 3.43,
"learning_rate": 5.4337258009901596e-05,
"loss": 1.8009,
"step": 515
},
{
"epoch": 3.47,
"learning_rate": 6.142934741983902e-05,
"loss": 1.7686,
"step": 520
},
{
"epoch": 3.5,
"learning_rate": 6.859999999999997e-05,
"loss": 1.697,
"step": 525
},
{
"epoch": 3.53,
"learning_rate": 7.577065258016093e-05,
"loss": 1.6804,
"step": 530
},
{
"epoch": 3.57,
"learning_rate": 8.286274199009834e-05,
"loss": 1.7485,
"step": 535
},
{
"epoch": 3.6,
"learning_rate": 8.979856581412137e-05,
"loss": 1.7389,
"step": 540
},
{
"epoch": 3.63,
"learning_rate": 9.650213371499991e-05,
"loss": 1.6993,
"step": 545
},
{
"epoch": 3.67,
"learning_rate": 0.00010289999999999994,
"loss": 1.7811,
"step": 550
},
{
"epoch": 3.7,
"learning_rate": 0.00010892206830726364,
"loss": 1.6838,
"step": 555
},
{
"epoch": 3.73,
"learning_rate": 0.00011450235959621768,
"loss": 1.76,
"step": 560
},
{
"epoch": 3.77,
"learning_rate": 0.0001195797350277492,
"loss": 1.7419,
"step": 565
},
{
"epoch": 3.8,
"learning_rate": 0.00012409856581412136,
"loss": 1.7739,
"step": 570
},
{
"epoch": 3.83,
"learning_rate": 0.0001280093426996125,
"loss": 1.7108,
"step": 575
},
{
"epoch": 3.87,
"learning_rate": 0.0001312692183942824,
"loss": 1.7828,
"step": 580
},
{
"epoch": 3.9,
"learning_rate": 0.00013384247701784751,
"loss": 1.7588,
"step": 585
},
{
"epoch": 3.93,
"learning_rate": 0.00013570092541033904,
"loss": 1.6845,
"step": 590
},
{
"epoch": 3.97,
"learning_rate": 0.00013682420202226357,
"loss": 1.7817,
"step": 595
},
{
"epoch": 4.0,
"learning_rate": 0.0001372,
"loss": 1.7611,
"step": 600
},
{
"epoch": 4.0,
"eval_loss": 1.6418354511260986,
"eval_runtime": 10.0255,
"eval_samples_per_second": 22.942,
"eval_steps_per_second": 2.893,
"step": 600
},
{
"epoch": 4.03,
"learning_rate": 0.00013682420202226357,
"loss": 1.7067,
"step": 605
},
{
"epoch": 4.07,
"learning_rate": 0.0001357009254103391,
"loss": 1.7111,
"step": 610
},
{
"epoch": 4.1,
"learning_rate": 0.0001338424770178476,
"loss": 1.6736,
"step": 615
},
{
"epoch": 4.13,
"learning_rate": 0.00013126921839428241,
"loss": 1.7152,
"step": 620
},
{
"epoch": 4.17,
"learning_rate": 0.00012800934269961248,
"loss": 1.7257,
"step": 625
},
{
"epoch": 4.2,
"learning_rate": 0.00012409856581412142,
"loss": 1.7207,
"step": 630
},
{
"epoch": 4.23,
"learning_rate": 0.00011957973502774922,
"loss": 1.6457,
"step": 635
},
{
"epoch": 4.27,
"learning_rate": 0.00011450235959621773,
"loss": 1.713,
"step": 640
},
{
"epoch": 4.3,
"learning_rate": 0.00010892206830726369,
"loss": 1.7242,
"step": 645
},
{
"epoch": 4.33,
"learning_rate": 0.00010290000000000009,
"loss": 1.6386,
"step": 650
},
{
"epoch": 4.37,
"learning_rate": 9.650213371499996e-05,
"loss": 1.6539,
"step": 655
},
{
"epoch": 4.4,
"learning_rate": 8.97985658141213e-05,
"loss": 1.7034,
"step": 660
},
{
"epoch": 4.43,
"learning_rate": 8.286274199009828e-05,
"loss": 1.6681,
"step": 665
},
{
"epoch": 4.47,
"learning_rate": 7.577065258016099e-05,
"loss": 1.6742,
"step": 670
},
{
"epoch": 4.5,
"learning_rate": 6.860000000000003e-05,
"loss": 1.687,
"step": 675
},
{
"epoch": 4.53,
"learning_rate": 6.14293474198391e-05,
"loss": 1.7066,
"step": 680
},
{
"epoch": 4.57,
"learning_rate": 5.433725800990179e-05,
"loss": 1.6622,
"step": 685
},
{
"epoch": 4.6,
"learning_rate": 4.740143418587876e-05,
"loss": 1.6647,
"step": 690
},
{
"epoch": 4.63,
"learning_rate": 4.069786628500011e-05,
"loss": 1.6251,
"step": 695
},
{
"epoch": 4.67,
"learning_rate": 3.429999999999996e-05,
"loss": 1.6732,
"step": 700
},
{
"epoch": 4.7,
"learning_rate": 2.8277931692736372e-05,
"loss": 1.5917,
"step": 705
},
{
"epoch": 4.73,
"learning_rate": 2.2697640403782324e-05,
"loss": 1.5938,
"step": 710
},
{
"epoch": 4.77,
"learning_rate": 1.762026497225081e-05,
"loss": 1.7145,
"step": 715
},
{
"epoch": 4.8,
"learning_rate": 1.3101434185878628e-05,
"loss": 1.6437,
"step": 720
},
{
"epoch": 4.83,
"learning_rate": 9.190657300387574e-06,
"loss": 1.5361,
"step": 725
},
{
"epoch": 4.87,
"learning_rate": 5.930781605717611e-06,
"loss": 1.5239,
"step": 730
},
{
"epoch": 4.9,
"learning_rate": 3.3575229821524373e-06,
"loss": 1.656,
"step": 735
},
{
"epoch": 4.93,
"learning_rate": 1.4990745896609297e-06,
"loss": 1.6033,
"step": 740
},
{
"epoch": 4.97,
"learning_rate": 3.757979777364447e-07,
"loss": 1.6277,
"step": 745
},
{
"epoch": 5.0,
"learning_rate": 0.0,
"loss": 1.5745,
"step": 750
},
{
"epoch": 5.0,
"eval_loss": 1.5331557989120483,
"eval_runtime": 11.0656,
"eval_samples_per_second": 21.327,
"eval_steps_per_second": 2.711,
"step": 750
}
],
"max_steps": 750,
"num_train_epochs": 5,
"total_flos": 781001883648000.0,
"trial_name": null,
"trial_params": null
}