patrickvonplaten's picture
up
daf256a
raw
history blame
25.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.484304932735426,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 2.2499999999999996e-06,
"loss": 11.7449,
"step": 10
},
{
"epoch": 0.04,
"learning_rate": 9.75e-06,
"loss": 9.9528,
"step": 20
},
{
"epoch": 0.07,
"learning_rate": 1.725e-05,
"loss": 7.8835,
"step": 30
},
{
"epoch": 0.09,
"learning_rate": 2.475e-05,
"loss": 6.9463,
"step": 40
},
{
"epoch": 0.11,
"learning_rate": 3.225e-05,
"loss": 6.3116,
"step": 50
},
{
"epoch": 0.13,
"learning_rate": 3.975e-05,
"loss": 5.8418,
"step": 60
},
{
"epoch": 0.16,
"learning_rate": 4.7249999999999997e-05,
"loss": 5.6196,
"step": 70
},
{
"epoch": 0.18,
"learning_rate": 5.4749999999999996e-05,
"loss": 5.4065,
"step": 80
},
{
"epoch": 0.2,
"learning_rate": 6.225e-05,
"loss": 5.3217,
"step": 90
},
{
"epoch": 0.22,
"learning_rate": 6.975e-05,
"loss": 5.2094,
"step": 100
},
{
"epoch": 0.25,
"learning_rate": 7.725e-05,
"loss": 5.2388,
"step": 110
},
{
"epoch": 0.27,
"learning_rate": 8.474999999999999e-05,
"loss": 5.1522,
"step": 120
},
{
"epoch": 0.29,
"learning_rate": 9.224999999999999e-05,
"loss": 5.1679,
"step": 130
},
{
"epoch": 0.31,
"learning_rate": 9.975e-05,
"loss": 5.112,
"step": 140
},
{
"epoch": 0.34,
"learning_rate": 0.00010724999999999999,
"loss": 5.1008,
"step": 150
},
{
"epoch": 0.36,
"learning_rate": 0.00011475,
"loss": 5.1024,
"step": 160
},
{
"epoch": 0.38,
"learning_rate": 0.00012225,
"loss": 5.0238,
"step": 170
},
{
"epoch": 0.4,
"learning_rate": 0.00012974999999999998,
"loss": 5.061,
"step": 180
},
{
"epoch": 0.43,
"learning_rate": 0.00013725,
"loss": 5.0293,
"step": 190
},
{
"epoch": 0.45,
"learning_rate": 0.00014475,
"loss": 5.0315,
"step": 200
},
{
"epoch": 0.47,
"learning_rate": 0.00015224999999999996,
"loss": 5.1264,
"step": 210
},
{
"epoch": 0.49,
"learning_rate": 0.00015974999999999998,
"loss": 4.9663,
"step": 220
},
{
"epoch": 0.52,
"learning_rate": 0.00016724999999999997,
"loss": 4.9525,
"step": 230
},
{
"epoch": 0.54,
"learning_rate": 0.00017475,
"loss": 4.9757,
"step": 240
},
{
"epoch": 0.56,
"learning_rate": 0.00018224999999999998,
"loss": 4.9203,
"step": 250
},
{
"epoch": 0.58,
"learning_rate": 0.00018974999999999998,
"loss": 4.908,
"step": 260
},
{
"epoch": 0.61,
"learning_rate": 0.00019724999999999997,
"loss": 4.845,
"step": 270
},
{
"epoch": 0.63,
"learning_rate": 0.00020475,
"loss": 4.8367,
"step": 280
},
{
"epoch": 0.65,
"learning_rate": 0.00021224999999999998,
"loss": 4.8034,
"step": 290
},
{
"epoch": 0.67,
"learning_rate": 0.00021975,
"loss": 4.8215,
"step": 300
},
{
"epoch": 0.7,
"learning_rate": 0.00022724999999999997,
"loss": 4.7776,
"step": 310
},
{
"epoch": 0.72,
"learning_rate": 0.00023474999999999996,
"loss": 4.7166,
"step": 320
},
{
"epoch": 0.74,
"learning_rate": 0.00024224999999999998,
"loss": 4.7696,
"step": 330
},
{
"epoch": 0.76,
"learning_rate": 0.000249,
"loss": 4.6817,
"step": 340
},
{
"epoch": 0.78,
"learning_rate": 0.00025649999999999995,
"loss": 4.7147,
"step": 350
},
{
"epoch": 0.81,
"learning_rate": 0.00026399999999999997,
"loss": 4.6688,
"step": 360
},
{
"epoch": 0.83,
"learning_rate": 0.0002715,
"loss": 4.5991,
"step": 370
},
{
"epoch": 0.85,
"learning_rate": 0.000279,
"loss": 4.6401,
"step": 380
},
{
"epoch": 0.87,
"learning_rate": 0.00028649999999999997,
"loss": 4.6029,
"step": 390
},
{
"epoch": 0.9,
"learning_rate": 0.000294,
"loss": 4.5118,
"step": 400
},
{
"epoch": 0.9,
"eval_loss": 4.247889041900635,
"eval_runtime": 51.3593,
"eval_samples_per_second": 51.442,
"eval_steps_per_second": 0.818,
"eval_wer": 0.9033518445061484,
"step": 400
},
{
"epoch": 0.92,
"learning_rate": 0.000299672131147541,
"loss": 4.5483,
"step": 410
},
{
"epoch": 0.94,
"learning_rate": 0.00029803278688524587,
"loss": 4.4005,
"step": 420
},
{
"epoch": 0.96,
"learning_rate": 0.00029639344262295076,
"loss": 4.327,
"step": 430
},
{
"epoch": 0.99,
"learning_rate": 0.0002947540983606557,
"loss": 4.2717,
"step": 440
},
{
"epoch": 1.01,
"learning_rate": 0.0002931147540983606,
"loss": 4.0139,
"step": 450
},
{
"epoch": 1.03,
"learning_rate": 0.00029147540983606555,
"loss": 3.8304,
"step": 460
},
{
"epoch": 1.05,
"learning_rate": 0.0002898360655737705,
"loss": 3.5825,
"step": 470
},
{
"epoch": 1.08,
"learning_rate": 0.0002881967213114754,
"loss": 3.4968,
"step": 480
},
{
"epoch": 1.1,
"learning_rate": 0.0002865573770491803,
"loss": 3.2832,
"step": 490
},
{
"epoch": 1.12,
"learning_rate": 0.00028491803278688524,
"loss": 3.1368,
"step": 500
},
{
"epoch": 1.14,
"learning_rate": 0.00028327868852459013,
"loss": 2.899,
"step": 510
},
{
"epoch": 1.17,
"learning_rate": 0.0002816393442622951,
"loss": 2.6293,
"step": 520
},
{
"epoch": 1.19,
"learning_rate": 0.00028,
"loss": 2.7137,
"step": 530
},
{
"epoch": 1.21,
"learning_rate": 0.00027836065573770487,
"loss": 2.4069,
"step": 540
},
{
"epoch": 1.23,
"learning_rate": 0.0002767213114754098,
"loss": 2.3349,
"step": 550
},
{
"epoch": 1.26,
"learning_rate": 0.0002750819672131147,
"loss": 2.1064,
"step": 560
},
{
"epoch": 1.28,
"learning_rate": 0.00027344262295081966,
"loss": 1.9375,
"step": 570
},
{
"epoch": 1.3,
"learning_rate": 0.00027180327868852455,
"loss": 1.9395,
"step": 580
},
{
"epoch": 1.32,
"learning_rate": 0.0002701639344262295,
"loss": 1.7861,
"step": 590
},
{
"epoch": 1.35,
"learning_rate": 0.0002685245901639344,
"loss": 1.6795,
"step": 600
},
{
"epoch": 1.37,
"learning_rate": 0.0002668852459016393,
"loss": 1.5166,
"step": 610
},
{
"epoch": 1.39,
"learning_rate": 0.00026524590163934424,
"loss": 1.4776,
"step": 620
},
{
"epoch": 1.41,
"learning_rate": 0.00026360655737704913,
"loss": 1.4286,
"step": 630
},
{
"epoch": 1.43,
"learning_rate": 0.0002619672131147541,
"loss": 1.3071,
"step": 640
},
{
"epoch": 1.46,
"learning_rate": 0.00026032786885245903,
"loss": 1.277,
"step": 650
},
{
"epoch": 1.48,
"learning_rate": 0.0002586885245901639,
"loss": 1.2741,
"step": 660
},
{
"epoch": 1.5,
"learning_rate": 0.0002570491803278688,
"loss": 1.1661,
"step": 670
},
{
"epoch": 1.52,
"learning_rate": 0.00025540983606557377,
"loss": 1.1585,
"step": 680
},
{
"epoch": 1.55,
"learning_rate": 0.00025377049180327866,
"loss": 1.1015,
"step": 690
},
{
"epoch": 1.57,
"learning_rate": 0.0002521311475409836,
"loss": 1.0403,
"step": 700
},
{
"epoch": 1.59,
"learning_rate": 0.0002504918032786885,
"loss": 1.0001,
"step": 710
},
{
"epoch": 1.61,
"learning_rate": 0.0002488524590163934,
"loss": 0.9915,
"step": 720
},
{
"epoch": 1.64,
"learning_rate": 0.00024721311475409835,
"loss": 0.9481,
"step": 730
},
{
"epoch": 1.66,
"learning_rate": 0.00024557377049180324,
"loss": 0.9522,
"step": 740
},
{
"epoch": 1.68,
"learning_rate": 0.00024393442622950816,
"loss": 0.9063,
"step": 750
},
{
"epoch": 1.7,
"learning_rate": 0.00024229508196721308,
"loss": 0.8868,
"step": 760
},
{
"epoch": 1.73,
"learning_rate": 0.000240655737704918,
"loss": 0.818,
"step": 770
},
{
"epoch": 1.75,
"learning_rate": 0.00023901639344262293,
"loss": 0.8436,
"step": 780
},
{
"epoch": 1.77,
"learning_rate": 0.00023737704918032785,
"loss": 0.7755,
"step": 790
},
{
"epoch": 1.79,
"learning_rate": 0.00023573770491803274,
"loss": 0.8016,
"step": 800
},
{
"epoch": 1.79,
"eval_loss": 0.8116478323936462,
"eval_runtime": 61.0806,
"eval_samples_per_second": 43.254,
"eval_steps_per_second": 0.688,
"eval_wer": 0.2206862356207854,
"step": 800
},
{
"epoch": 1.82,
"learning_rate": 0.00023409836065573766,
"loss": 0.7623,
"step": 810
},
{
"epoch": 1.84,
"learning_rate": 0.00023245901639344258,
"loss": 0.7884,
"step": 820
},
{
"epoch": 1.86,
"learning_rate": 0.00023081967213114753,
"loss": 0.7791,
"step": 830
},
{
"epoch": 1.88,
"learning_rate": 0.00022918032786885245,
"loss": 0.7111,
"step": 840
},
{
"epoch": 1.91,
"learning_rate": 0.00022754098360655738,
"loss": 0.7063,
"step": 850
},
{
"epoch": 1.93,
"learning_rate": 0.0002259016393442623,
"loss": 0.7307,
"step": 860
},
{
"epoch": 1.95,
"learning_rate": 0.0002242622950819672,
"loss": 0.6792,
"step": 870
},
{
"epoch": 1.97,
"learning_rate": 0.0002226229508196721,
"loss": 0.6496,
"step": 880
},
{
"epoch": 2.0,
"learning_rate": 0.00022098360655737703,
"loss": 0.6929,
"step": 890
},
{
"epoch": 2.02,
"learning_rate": 0.00021934426229508195,
"loss": 0.5503,
"step": 900
},
{
"epoch": 2.04,
"learning_rate": 0.00021770491803278688,
"loss": 0.4833,
"step": 910
},
{
"epoch": 2.06,
"learning_rate": 0.0002160655737704918,
"loss": 0.5107,
"step": 920
},
{
"epoch": 2.09,
"learning_rate": 0.0002144262295081967,
"loss": 0.4637,
"step": 930
},
{
"epoch": 2.11,
"learning_rate": 0.0002127868852459016,
"loss": 0.4764,
"step": 940
},
{
"epoch": 2.13,
"learning_rate": 0.00021114754098360653,
"loss": 0.4332,
"step": 950
},
{
"epoch": 2.15,
"learning_rate": 0.00020950819672131146,
"loss": 0.4599,
"step": 960
},
{
"epoch": 2.17,
"learning_rate": 0.00020786885245901638,
"loss": 0.4582,
"step": 970
},
{
"epoch": 2.2,
"learning_rate": 0.00020622950819672127,
"loss": 0.4112,
"step": 980
},
{
"epoch": 2.22,
"learning_rate": 0.0002045901639344262,
"loss": 0.4441,
"step": 990
},
{
"epoch": 2.24,
"learning_rate": 0.0002029508196721311,
"loss": 0.4575,
"step": 1000
},
{
"epoch": 2.26,
"learning_rate": 0.00020131147540983603,
"loss": 0.4177,
"step": 1010
},
{
"epoch": 2.29,
"learning_rate": 0.00019967213114754098,
"loss": 0.4323,
"step": 1020
},
{
"epoch": 2.31,
"learning_rate": 0.0001980327868852459,
"loss": 0.4295,
"step": 1030
},
{
"epoch": 2.33,
"learning_rate": 0.00019639344262295083,
"loss": 0.4595,
"step": 1040
},
{
"epoch": 2.35,
"learning_rate": 0.00019475409836065572,
"loss": 0.4338,
"step": 1050
},
{
"epoch": 2.38,
"learning_rate": 0.00019311475409836064,
"loss": 0.4104,
"step": 1060
},
{
"epoch": 2.4,
"learning_rate": 0.00019147540983606556,
"loss": 0.4216,
"step": 1070
},
{
"epoch": 2.42,
"learning_rate": 0.00018983606557377048,
"loss": 0.3729,
"step": 1080
},
{
"epoch": 2.44,
"learning_rate": 0.0001881967213114754,
"loss": 0.3998,
"step": 1090
},
{
"epoch": 2.47,
"learning_rate": 0.0001865573770491803,
"loss": 0.4082,
"step": 1100
},
{
"epoch": 2.49,
"learning_rate": 0.00018491803278688522,
"loss": 0.38,
"step": 1110
},
{
"epoch": 2.51,
"learning_rate": 0.00018327868852459014,
"loss": 0.4157,
"step": 1120
},
{
"epoch": 2.53,
"learning_rate": 0.00018163934426229506,
"loss": 0.3937,
"step": 1130
},
{
"epoch": 2.56,
"learning_rate": 0.00017999999999999998,
"loss": 0.3772,
"step": 1140
},
{
"epoch": 2.58,
"learning_rate": 0.0001783606557377049,
"loss": 0.3767,
"step": 1150
},
{
"epoch": 2.6,
"learning_rate": 0.0001767213114754098,
"loss": 0.3602,
"step": 1160
},
{
"epoch": 2.62,
"learning_rate": 0.00017508196721311472,
"loss": 0.3876,
"step": 1170
},
{
"epoch": 2.65,
"learning_rate": 0.00017344262295081964,
"loss": 0.3776,
"step": 1180
},
{
"epoch": 2.67,
"learning_rate": 0.00017180327868852456,
"loss": 0.3617,
"step": 1190
},
{
"epoch": 2.69,
"learning_rate": 0.0001701639344262295,
"loss": 0.3627,
"step": 1200
},
{
"epoch": 2.69,
"eval_loss": 0.5665903091430664,
"eval_runtime": 61.1317,
"eval_samples_per_second": 43.218,
"eval_steps_per_second": 0.687,
"eval_wer": 0.1506743355811186,
"step": 1200
},
{
"epoch": 2.71,
"learning_rate": 0.00016852459016393443,
"loss": 0.3478,
"step": 1210
},
{
"epoch": 2.74,
"learning_rate": 0.00016688524590163935,
"loss": 0.3861,
"step": 1220
},
{
"epoch": 2.76,
"learning_rate": 0.00016524590163934425,
"loss": 0.3283,
"step": 1230
},
{
"epoch": 2.78,
"learning_rate": 0.00016360655737704917,
"loss": 0.3455,
"step": 1240
},
{
"epoch": 2.8,
"learning_rate": 0.0001619672131147541,
"loss": 0.3457,
"step": 1250
},
{
"epoch": 2.83,
"learning_rate": 0.000160327868852459,
"loss": 0.3178,
"step": 1260
},
{
"epoch": 2.85,
"learning_rate": 0.00015868852459016393,
"loss": 0.3429,
"step": 1270
},
{
"epoch": 2.87,
"learning_rate": 0.00015704918032786883,
"loss": 0.3253,
"step": 1280
},
{
"epoch": 2.89,
"learning_rate": 0.00015540983606557375,
"loss": 0.3469,
"step": 1290
},
{
"epoch": 2.91,
"learning_rate": 0.00015377049180327867,
"loss": 0.334,
"step": 1300
},
{
"epoch": 2.94,
"learning_rate": 0.0001521311475409836,
"loss": 0.3246,
"step": 1310
},
{
"epoch": 2.96,
"learning_rate": 0.0001504918032786885,
"loss": 0.3243,
"step": 1320
},
{
"epoch": 2.98,
"learning_rate": 0.00014885245901639343,
"loss": 0.3219,
"step": 1330
},
{
"epoch": 3.0,
"learning_rate": 0.00014721311475409836,
"loss": 0.3022,
"step": 1340
},
{
"epoch": 3.03,
"learning_rate": 0.00014557377049180328,
"loss": 0.1935,
"step": 1350
},
{
"epoch": 3.05,
"learning_rate": 0.0001439344262295082,
"loss": 0.2093,
"step": 1360
},
{
"epoch": 3.07,
"learning_rate": 0.0001422950819672131,
"loss": 0.1961,
"step": 1370
},
{
"epoch": 3.09,
"learning_rate": 0.00014065573770491801,
"loss": 0.1788,
"step": 1380
},
{
"epoch": 3.12,
"learning_rate": 0.00013901639344262294,
"loss": 0.2245,
"step": 1390
},
{
"epoch": 3.14,
"learning_rate": 0.00013737704918032786,
"loss": 0.1815,
"step": 1400
},
{
"epoch": 3.16,
"learning_rate": 0.00013573770491803278,
"loss": 0.2018,
"step": 1410
},
{
"epoch": 3.18,
"learning_rate": 0.0001340983606557377,
"loss": 0.1909,
"step": 1420
},
{
"epoch": 3.21,
"learning_rate": 0.00013245901639344262,
"loss": 0.2025,
"step": 1430
},
{
"epoch": 3.23,
"learning_rate": 0.00013081967213114754,
"loss": 0.1927,
"step": 1440
},
{
"epoch": 3.25,
"learning_rate": 0.00012918032786885246,
"loss": 0.1816,
"step": 1450
},
{
"epoch": 3.27,
"learning_rate": 0.00012754098360655736,
"loss": 0.2217,
"step": 1460
},
{
"epoch": 3.3,
"learning_rate": 0.00012590163934426228,
"loss": 0.2035,
"step": 1470
},
{
"epoch": 3.32,
"learning_rate": 0.0001242622950819672,
"loss": 0.1904,
"step": 1480
},
{
"epoch": 3.34,
"learning_rate": 0.00012262295081967212,
"loss": 0.1945,
"step": 1490
},
{
"epoch": 3.36,
"learning_rate": 0.00012098360655737703,
"loss": 0.1724,
"step": 1500
},
{
"epoch": 3.39,
"learning_rate": 0.00011934426229508195,
"loss": 0.1704,
"step": 1510
},
{
"epoch": 3.41,
"learning_rate": 0.00011770491803278688,
"loss": 0.1745,
"step": 1520
},
{
"epoch": 3.43,
"learning_rate": 0.00011606557377049179,
"loss": 0.1756,
"step": 1530
},
{
"epoch": 3.45,
"learning_rate": 0.00011442622950819671,
"loss": 0.1864,
"step": 1540
},
{
"epoch": 3.48,
"learning_rate": 0.00011278688524590164,
"loss": 0.1799,
"step": 1550
},
{
"epoch": 3.5,
"learning_rate": 0.00011114754098360654,
"loss": 0.1974,
"step": 1560
},
{
"epoch": 3.52,
"learning_rate": 0.00010950819672131146,
"loss": 0.1751,
"step": 1570
},
{
"epoch": 3.54,
"learning_rate": 0.00010786885245901639,
"loss": 0.1792,
"step": 1580
},
{
"epoch": 3.57,
"learning_rate": 0.00010622950819672129,
"loss": 0.1887,
"step": 1590
},
{
"epoch": 3.59,
"learning_rate": 0.00010459016393442621,
"loss": 0.1763,
"step": 1600
},
{
"epoch": 3.59,
"eval_loss": 0.47305428981781006,
"eval_runtime": 60.8349,
"eval_samples_per_second": 43.429,
"eval_steps_per_second": 0.69,
"eval_wer": 0.12973026576755256,
"step": 1600
},
{
"epoch": 3.61,
"learning_rate": 0.00010295081967213115,
"loss": 0.1767,
"step": 1610
},
{
"epoch": 3.63,
"learning_rate": 0.00010131147540983606,
"loss": 0.1769,
"step": 1620
},
{
"epoch": 3.65,
"learning_rate": 9.967213114754098e-05,
"loss": 0.1691,
"step": 1630
},
{
"epoch": 3.68,
"learning_rate": 9.80327868852459e-05,
"loss": 0.1762,
"step": 1640
},
{
"epoch": 3.7,
"learning_rate": 9.639344262295081e-05,
"loss": 0.1636,
"step": 1650
},
{
"epoch": 3.72,
"learning_rate": 9.475409836065573e-05,
"loss": 0.1834,
"step": 1660
},
{
"epoch": 3.74,
"learning_rate": 9.311475409836065e-05,
"loss": 0.1781,
"step": 1670
},
{
"epoch": 3.77,
"learning_rate": 9.147540983606556e-05,
"loss": 0.1686,
"step": 1680
},
{
"epoch": 3.79,
"learning_rate": 8.983606557377048e-05,
"loss": 0.1657,
"step": 1690
},
{
"epoch": 3.81,
"learning_rate": 8.819672131147541e-05,
"loss": 0.1468,
"step": 1700
},
{
"epoch": 3.83,
"learning_rate": 8.655737704918032e-05,
"loss": 0.1686,
"step": 1710
},
{
"epoch": 3.86,
"learning_rate": 8.491803278688524e-05,
"loss": 0.1583,
"step": 1720
},
{
"epoch": 3.88,
"learning_rate": 8.327868852459016e-05,
"loss": 0.1656,
"step": 1730
},
{
"epoch": 3.9,
"learning_rate": 8.163934426229507e-05,
"loss": 0.1595,
"step": 1740
},
{
"epoch": 3.92,
"learning_rate": 7.999999999999999e-05,
"loss": 0.1563,
"step": 1750
},
{
"epoch": 3.95,
"learning_rate": 7.836065573770491e-05,
"loss": 0.1625,
"step": 1760
},
{
"epoch": 3.97,
"learning_rate": 7.672131147540982e-05,
"loss": 0.1439,
"step": 1770
},
{
"epoch": 3.99,
"learning_rate": 7.508196721311474e-05,
"loss": 0.1593,
"step": 1780
},
{
"epoch": 4.01,
"learning_rate": 7.344262295081966e-05,
"loss": 0.1042,
"step": 1790
},
{
"epoch": 4.04,
"learning_rate": 7.180327868852459e-05,
"loss": 0.0931,
"step": 1800
},
{
"epoch": 4.06,
"learning_rate": 7.01639344262295e-05,
"loss": 0.0909,
"step": 1810
},
{
"epoch": 4.08,
"learning_rate": 6.852459016393443e-05,
"loss": 0.0955,
"step": 1820
},
{
"epoch": 4.1,
"learning_rate": 6.688524590163934e-05,
"loss": 0.102,
"step": 1830
},
{
"epoch": 4.13,
"learning_rate": 6.524590163934426e-05,
"loss": 0.0889,
"step": 1840
},
{
"epoch": 4.15,
"learning_rate": 6.360655737704918e-05,
"loss": 0.0968,
"step": 1850
},
{
"epoch": 4.17,
"learning_rate": 6.19672131147541e-05,
"loss": 0.0809,
"step": 1860
},
{
"epoch": 4.19,
"learning_rate": 6.0327868852459015e-05,
"loss": 0.0869,
"step": 1870
},
{
"epoch": 4.22,
"learning_rate": 5.868852459016393e-05,
"loss": 0.0913,
"step": 1880
},
{
"epoch": 4.24,
"learning_rate": 5.7049180327868844e-05,
"loss": 0.078,
"step": 1890
},
{
"epoch": 4.26,
"learning_rate": 5.5409836065573765e-05,
"loss": 0.0924,
"step": 1900
},
{
"epoch": 4.28,
"learning_rate": 5.3770491803278686e-05,
"loss": 0.0879,
"step": 1910
},
{
"epoch": 4.3,
"learning_rate": 5.21311475409836e-05,
"loss": 0.0869,
"step": 1920
},
{
"epoch": 4.33,
"learning_rate": 5.049180327868852e-05,
"loss": 0.0832,
"step": 1930
},
{
"epoch": 4.35,
"learning_rate": 4.8852459016393436e-05,
"loss": 0.0777,
"step": 1940
},
{
"epoch": 4.37,
"learning_rate": 4.721311475409836e-05,
"loss": 0.0905,
"step": 1950
},
{
"epoch": 4.39,
"learning_rate": 4.557377049180328e-05,
"loss": 0.0818,
"step": 1960
},
{
"epoch": 4.42,
"learning_rate": 4.3934426229508194e-05,
"loss": 0.0829,
"step": 1970
},
{
"epoch": 4.44,
"learning_rate": 4.229508196721311e-05,
"loss": 0.0865,
"step": 1980
},
{
"epoch": 4.46,
"learning_rate": 4.065573770491803e-05,
"loss": 0.0895,
"step": 1990
},
{
"epoch": 4.48,
"learning_rate": 3.901639344262295e-05,
"loss": 0.089,
"step": 2000
},
{
"epoch": 4.48,
"eval_loss": 0.41652488708496094,
"eval_runtime": 61.3777,
"eval_samples_per_second": 43.045,
"eval_steps_per_second": 0.684,
"eval_wer": 0.11800872669575566,
"step": 2000
}
],
"max_steps": 2230,
"num_train_epochs": 5,
"total_flos": 2.644259324881902e+19,
"trial_name": null,
"trial_params": null
}