AlienKevin's picture
Obtain 82% accuracy on more than 8000 sentences
8f1ccbe
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.3145245559038665,
"global_step": 7000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.994775339602926e-05,
"loss": 0.3717,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 4.989550679205852e-05,
"loss": 0.2368,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 4.9843260188087774e-05,
"loss": 0.1928,
"step": 30
},
{
"epoch": 0.04,
"learning_rate": 4.979101358411704e-05,
"loss": 0.1648,
"step": 40
},
{
"epoch": 0.05,
"learning_rate": 4.9738766980146295e-05,
"loss": 0.1645,
"step": 50
},
{
"epoch": 0.06,
"learning_rate": 4.968652037617555e-05,
"loss": 0.1151,
"step": 60
},
{
"epoch": 0.07,
"learning_rate": 4.963427377220481e-05,
"loss": 0.1206,
"step": 70
},
{
"epoch": 0.08,
"learning_rate": 4.958202716823407e-05,
"loss": 0.1289,
"step": 80
},
{
"epoch": 0.09,
"learning_rate": 4.9529780564263324e-05,
"loss": 0.1123,
"step": 90
},
{
"epoch": 0.1,
"learning_rate": 4.947753396029258e-05,
"loss": 0.0992,
"step": 100
},
{
"epoch": 0.11,
"learning_rate": 4.9425287356321845e-05,
"loss": 0.097,
"step": 110
},
{
"epoch": 0.13,
"learning_rate": 4.93730407523511e-05,
"loss": 0.1028,
"step": 120
},
{
"epoch": 0.14,
"learning_rate": 4.932079414838036e-05,
"loss": 0.087,
"step": 130
},
{
"epoch": 0.15,
"learning_rate": 4.9268547544409617e-05,
"loss": 0.0907,
"step": 140
},
{
"epoch": 0.16,
"learning_rate": 4.9216300940438874e-05,
"loss": 0.0866,
"step": 150
},
{
"epoch": 0.17,
"learning_rate": 4.916405433646813e-05,
"loss": 0.085,
"step": 160
},
{
"epoch": 0.18,
"learning_rate": 4.911180773249739e-05,
"loss": 0.0846,
"step": 170
},
{
"epoch": 0.19,
"learning_rate": 4.905956112852665e-05,
"loss": 0.0791,
"step": 180
},
{
"epoch": 0.2,
"learning_rate": 4.900731452455591e-05,
"loss": 0.073,
"step": 190
},
{
"epoch": 0.21,
"learning_rate": 4.895506792058516e-05,
"loss": 0.0861,
"step": 200
},
{
"epoch": 0.22,
"learning_rate": 4.8902821316614424e-05,
"loss": 0.0828,
"step": 210
},
{
"epoch": 0.23,
"learning_rate": 4.885057471264368e-05,
"loss": 0.0711,
"step": 220
},
{
"epoch": 0.24,
"learning_rate": 4.879832810867294e-05,
"loss": 0.069,
"step": 230
},
{
"epoch": 0.25,
"learning_rate": 4.8746081504702195e-05,
"loss": 0.0647,
"step": 240
},
{
"epoch": 0.26,
"learning_rate": 4.869383490073145e-05,
"loss": 0.0862,
"step": 250
},
{
"epoch": 0.27,
"learning_rate": 4.8641588296760716e-05,
"loss": 0.0628,
"step": 260
},
{
"epoch": 0.28,
"learning_rate": 4.858934169278997e-05,
"loss": 0.0697,
"step": 270
},
{
"epoch": 0.29,
"learning_rate": 4.853709508881923e-05,
"loss": 0.0685,
"step": 280
},
{
"epoch": 0.3,
"learning_rate": 4.848484848484849e-05,
"loss": 0.0602,
"step": 290
},
{
"epoch": 0.31,
"learning_rate": 4.8432601880877745e-05,
"loss": 0.0576,
"step": 300
},
{
"epoch": 0.32,
"learning_rate": 4.8380355276907e-05,
"loss": 0.068,
"step": 310
},
{
"epoch": 0.33,
"learning_rate": 4.832810867293626e-05,
"loss": 0.0645,
"step": 320
},
{
"epoch": 0.34,
"learning_rate": 4.827586206896552e-05,
"loss": 0.0785,
"step": 330
},
{
"epoch": 0.36,
"learning_rate": 4.8223615464994774e-05,
"loss": 0.0606,
"step": 340
},
{
"epoch": 0.37,
"learning_rate": 4.817136886102404e-05,
"loss": 0.0682,
"step": 350
},
{
"epoch": 0.38,
"learning_rate": 4.8119122257053295e-05,
"loss": 0.0682,
"step": 360
},
{
"epoch": 0.39,
"learning_rate": 4.806687565308255e-05,
"loss": 0.0605,
"step": 370
},
{
"epoch": 0.4,
"learning_rate": 4.801462904911181e-05,
"loss": 0.0578,
"step": 380
},
{
"epoch": 0.41,
"learning_rate": 4.7962382445141066e-05,
"loss": 0.0562,
"step": 390
},
{
"epoch": 0.42,
"learning_rate": 4.791013584117033e-05,
"loss": 0.0724,
"step": 400
},
{
"epoch": 0.43,
"learning_rate": 4.785788923719958e-05,
"loss": 0.056,
"step": 410
},
{
"epoch": 0.44,
"learning_rate": 4.7805642633228845e-05,
"loss": 0.0509,
"step": 420
},
{
"epoch": 0.45,
"learning_rate": 4.77533960292581e-05,
"loss": 0.0516,
"step": 430
},
{
"epoch": 0.46,
"learning_rate": 4.770114942528736e-05,
"loss": 0.0588,
"step": 440
},
{
"epoch": 0.47,
"learning_rate": 4.7648902821316616e-05,
"loss": 0.0527,
"step": 450
},
{
"epoch": 0.48,
"learning_rate": 4.7596656217345873e-05,
"loss": 0.0614,
"step": 460
},
{
"epoch": 0.49,
"learning_rate": 4.754440961337514e-05,
"loss": 0.0464,
"step": 470
},
{
"epoch": 0.5,
"learning_rate": 4.749216300940439e-05,
"loss": 0.0515,
"step": 480
},
{
"epoch": 0.51,
"learning_rate": 4.743991640543365e-05,
"loss": 0.0456,
"step": 490
},
{
"epoch": 0.52,
"learning_rate": 4.738766980146291e-05,
"loss": 0.0583,
"step": 500
},
{
"epoch": 0.53,
"learning_rate": 4.7335423197492166e-05,
"loss": 0.0595,
"step": 510
},
{
"epoch": 0.54,
"learning_rate": 4.728317659352142e-05,
"loss": 0.0547,
"step": 520
},
{
"epoch": 0.55,
"learning_rate": 4.723092998955068e-05,
"loss": 0.0451,
"step": 530
},
{
"epoch": 0.56,
"learning_rate": 4.7178683385579944e-05,
"loss": 0.0395,
"step": 540
},
{
"epoch": 0.57,
"learning_rate": 4.7126436781609195e-05,
"loss": 0.0493,
"step": 550
},
{
"epoch": 0.59,
"learning_rate": 4.707419017763845e-05,
"loss": 0.0664,
"step": 560
},
{
"epoch": 0.6,
"learning_rate": 4.7021943573667716e-05,
"loss": 0.0445,
"step": 570
},
{
"epoch": 0.61,
"learning_rate": 4.696969696969697e-05,
"loss": 0.0443,
"step": 580
},
{
"epoch": 0.62,
"learning_rate": 4.691745036572623e-05,
"loss": 0.0461,
"step": 590
},
{
"epoch": 0.63,
"learning_rate": 4.686520376175549e-05,
"loss": 0.0415,
"step": 600
},
{
"epoch": 0.64,
"learning_rate": 4.6812957157784745e-05,
"loss": 0.0483,
"step": 610
},
{
"epoch": 0.65,
"learning_rate": 4.6760710553814e-05,
"loss": 0.0424,
"step": 620
},
{
"epoch": 0.66,
"learning_rate": 4.670846394984326e-05,
"loss": 0.0327,
"step": 630
},
{
"epoch": 0.67,
"learning_rate": 4.665621734587252e-05,
"loss": 0.0421,
"step": 640
},
{
"epoch": 0.68,
"learning_rate": 4.660397074190178e-05,
"loss": 0.0491,
"step": 650
},
{
"epoch": 0.69,
"learning_rate": 4.655172413793104e-05,
"loss": 0.0501,
"step": 660
},
{
"epoch": 0.7,
"learning_rate": 4.6499477533960295e-05,
"loss": 0.0447,
"step": 670
},
{
"epoch": 0.71,
"learning_rate": 4.644723092998955e-05,
"loss": 0.0449,
"step": 680
},
{
"epoch": 0.72,
"learning_rate": 4.639498432601881e-05,
"loss": 0.0478,
"step": 690
},
{
"epoch": 0.73,
"learning_rate": 4.6342737722048066e-05,
"loss": 0.0505,
"step": 700
},
{
"epoch": 0.74,
"learning_rate": 4.629049111807733e-05,
"loss": 0.0394,
"step": 710
},
{
"epoch": 0.75,
"learning_rate": 4.623824451410659e-05,
"loss": 0.0406,
"step": 720
},
{
"epoch": 0.76,
"learning_rate": 4.6185997910135844e-05,
"loss": 0.0347,
"step": 730
},
{
"epoch": 0.77,
"learning_rate": 4.61337513061651e-05,
"loss": 0.0445,
"step": 740
},
{
"epoch": 0.78,
"learning_rate": 4.608150470219436e-05,
"loss": 0.0428,
"step": 750
},
{
"epoch": 0.79,
"learning_rate": 4.6029258098223616e-05,
"loss": 0.0466,
"step": 760
},
{
"epoch": 0.8,
"learning_rate": 4.597701149425287e-05,
"loss": 0.0395,
"step": 770
},
{
"epoch": 0.82,
"learning_rate": 4.592476489028214e-05,
"loss": 0.0387,
"step": 780
},
{
"epoch": 0.83,
"learning_rate": 4.5872518286311394e-05,
"loss": 0.0395,
"step": 790
},
{
"epoch": 0.84,
"learning_rate": 4.582027168234065e-05,
"loss": 0.0305,
"step": 800
},
{
"epoch": 0.85,
"learning_rate": 4.576802507836991e-05,
"loss": 0.0434,
"step": 810
},
{
"epoch": 0.86,
"learning_rate": 4.5715778474399166e-05,
"loss": 0.0395,
"step": 820
},
{
"epoch": 0.87,
"learning_rate": 4.566353187042842e-05,
"loss": 0.0374,
"step": 830
},
{
"epoch": 0.88,
"learning_rate": 4.561128526645768e-05,
"loss": 0.0321,
"step": 840
},
{
"epoch": 0.89,
"learning_rate": 4.5559038662486944e-05,
"loss": 0.0394,
"step": 850
},
{
"epoch": 0.9,
"learning_rate": 4.55067920585162e-05,
"loss": 0.0383,
"step": 860
},
{
"epoch": 0.91,
"learning_rate": 4.545454545454546e-05,
"loss": 0.0388,
"step": 870
},
{
"epoch": 0.92,
"learning_rate": 4.5402298850574716e-05,
"loss": 0.035,
"step": 880
},
{
"epoch": 0.93,
"learning_rate": 4.535005224660397e-05,
"loss": 0.0414,
"step": 890
},
{
"epoch": 0.94,
"learning_rate": 4.529780564263323e-05,
"loss": 0.0439,
"step": 900
},
{
"epoch": 0.95,
"learning_rate": 4.524555903866249e-05,
"loss": 0.0337,
"step": 910
},
{
"epoch": 0.96,
"learning_rate": 4.519331243469175e-05,
"loss": 0.0384,
"step": 920
},
{
"epoch": 0.97,
"learning_rate": 4.514106583072101e-05,
"loss": 0.0353,
"step": 930
},
{
"epoch": 0.98,
"learning_rate": 4.508881922675026e-05,
"loss": 0.0351,
"step": 940
},
{
"epoch": 0.99,
"learning_rate": 4.503657262277952e-05,
"loss": 0.0324,
"step": 950
},
{
"epoch": 1.0,
"learning_rate": 4.498432601880878e-05,
"loss": 0.0296,
"step": 960
},
{
"epoch": 1.01,
"learning_rate": 4.493207941483804e-05,
"loss": 0.0262,
"step": 970
},
{
"epoch": 1.02,
"learning_rate": 4.4879832810867294e-05,
"loss": 0.0274,
"step": 980
},
{
"epoch": 1.03,
"learning_rate": 4.482758620689655e-05,
"loss": 0.0278,
"step": 990
},
{
"epoch": 1.04,
"learning_rate": 4.4775339602925815e-05,
"loss": 0.0205,
"step": 1000
},
{
"epoch": 1.04,
"eval_acc": 0.5359801475533992,
"eval_cer": 0.06462110646211065,
"eval_loss": 0.0336129367351532,
"eval_runtime": 135.8305,
"eval_samples_per_second": 2.967,
"eval_steps_per_second": 0.375,
"step": 1000
},
{
"epoch": 1.06,
"learning_rate": 4.4723092998955066e-05,
"loss": 0.0264,
"step": 1010
},
{
"epoch": 1.07,
"learning_rate": 4.467084639498433e-05,
"loss": 0.0218,
"step": 1020
},
{
"epoch": 1.08,
"learning_rate": 4.461859979101359e-05,
"loss": 0.0269,
"step": 1030
},
{
"epoch": 1.09,
"learning_rate": 4.4566353187042844e-05,
"loss": 0.0242,
"step": 1040
},
{
"epoch": 1.1,
"learning_rate": 4.45141065830721e-05,
"loss": 0.03,
"step": 1050
},
{
"epoch": 1.11,
"learning_rate": 4.446185997910136e-05,
"loss": 0.0214,
"step": 1060
},
{
"epoch": 1.12,
"learning_rate": 4.440961337513062e-05,
"loss": 0.0301,
"step": 1070
},
{
"epoch": 1.13,
"learning_rate": 4.435736677115987e-05,
"loss": 0.0217,
"step": 1080
},
{
"epoch": 1.14,
"learning_rate": 4.430512016718914e-05,
"loss": 0.0323,
"step": 1090
},
{
"epoch": 1.15,
"learning_rate": 4.4252873563218394e-05,
"loss": 0.0258,
"step": 1100
},
{
"epoch": 1.16,
"learning_rate": 4.420062695924765e-05,
"loss": 0.0293,
"step": 1110
},
{
"epoch": 1.17,
"learning_rate": 4.414838035527691e-05,
"loss": 0.0265,
"step": 1120
},
{
"epoch": 1.18,
"learning_rate": 4.4096133751306166e-05,
"loss": 0.0237,
"step": 1130
},
{
"epoch": 1.19,
"learning_rate": 4.404388714733543e-05,
"loss": 0.0275,
"step": 1140
},
{
"epoch": 1.2,
"learning_rate": 4.399164054336468e-05,
"loss": 0.0219,
"step": 1150
},
{
"epoch": 1.21,
"learning_rate": 4.3939393939393944e-05,
"loss": 0.0334,
"step": 1160
},
{
"epoch": 1.22,
"learning_rate": 4.38871473354232e-05,
"loss": 0.0183,
"step": 1170
},
{
"epoch": 1.23,
"learning_rate": 4.383490073145246e-05,
"loss": 0.0241,
"step": 1180
},
{
"epoch": 1.24,
"learning_rate": 4.3782654127481716e-05,
"loss": 0.0233,
"step": 1190
},
{
"epoch": 1.25,
"learning_rate": 4.373040752351097e-05,
"loss": 0.025,
"step": 1200
},
{
"epoch": 1.26,
"learning_rate": 4.367816091954024e-05,
"loss": 0.0228,
"step": 1210
},
{
"epoch": 1.27,
"learning_rate": 4.362591431556949e-05,
"loss": 0.0182,
"step": 1220
},
{
"epoch": 1.29,
"learning_rate": 4.357366771159875e-05,
"loss": 0.0234,
"step": 1230
},
{
"epoch": 1.3,
"learning_rate": 4.352142110762801e-05,
"loss": 0.019,
"step": 1240
},
{
"epoch": 1.31,
"learning_rate": 4.346917450365726e-05,
"loss": 0.0191,
"step": 1250
},
{
"epoch": 1.32,
"learning_rate": 4.341692789968652e-05,
"loss": 0.0177,
"step": 1260
},
{
"epoch": 1.33,
"learning_rate": 4.336468129571578e-05,
"loss": 0.0214,
"step": 1270
},
{
"epoch": 1.34,
"learning_rate": 4.3312434691745044e-05,
"loss": 0.0252,
"step": 1280
},
{
"epoch": 1.35,
"learning_rate": 4.3260188087774294e-05,
"loss": 0.0233,
"step": 1290
},
{
"epoch": 1.36,
"learning_rate": 4.320794148380355e-05,
"loss": 0.0184,
"step": 1300
},
{
"epoch": 1.37,
"learning_rate": 4.3155694879832815e-05,
"loss": 0.0212,
"step": 1310
},
{
"epoch": 1.38,
"learning_rate": 4.3103448275862066e-05,
"loss": 0.0186,
"step": 1320
},
{
"epoch": 1.39,
"learning_rate": 4.305120167189133e-05,
"loss": 0.0179,
"step": 1330
},
{
"epoch": 1.4,
"learning_rate": 4.299895506792059e-05,
"loss": 0.0184,
"step": 1340
},
{
"epoch": 1.41,
"learning_rate": 4.294670846394985e-05,
"loss": 0.0208,
"step": 1350
},
{
"epoch": 1.42,
"learning_rate": 4.28944618599791e-05,
"loss": 0.0218,
"step": 1360
},
{
"epoch": 1.43,
"learning_rate": 4.284221525600836e-05,
"loss": 0.0203,
"step": 1370
},
{
"epoch": 1.44,
"learning_rate": 4.278996865203762e-05,
"loss": 0.022,
"step": 1380
},
{
"epoch": 1.45,
"learning_rate": 4.273772204806687e-05,
"loss": 0.0247,
"step": 1390
},
{
"epoch": 1.46,
"learning_rate": 4.268547544409614e-05,
"loss": 0.0195,
"step": 1400
},
{
"epoch": 1.47,
"learning_rate": 4.2633228840125394e-05,
"loss": 0.0236,
"step": 1410
},
{
"epoch": 1.48,
"learning_rate": 4.258098223615465e-05,
"loss": 0.0182,
"step": 1420
},
{
"epoch": 1.49,
"learning_rate": 4.252873563218391e-05,
"loss": 0.0238,
"step": 1430
},
{
"epoch": 1.5,
"learning_rate": 4.2476489028213165e-05,
"loss": 0.0244,
"step": 1440
},
{
"epoch": 1.52,
"learning_rate": 4.242424242424243e-05,
"loss": 0.0366,
"step": 1450
},
{
"epoch": 1.53,
"learning_rate": 4.2371995820271687e-05,
"loss": 0.0159,
"step": 1460
},
{
"epoch": 1.54,
"learning_rate": 4.2319749216300944e-05,
"loss": 0.0146,
"step": 1470
},
{
"epoch": 1.55,
"learning_rate": 4.22675026123302e-05,
"loss": 0.0219,
"step": 1480
},
{
"epoch": 1.56,
"learning_rate": 4.221525600835946e-05,
"loss": 0.0176,
"step": 1490
},
{
"epoch": 1.57,
"learning_rate": 4.2163009404388715e-05,
"loss": 0.0207,
"step": 1500
},
{
"epoch": 1.58,
"learning_rate": 4.211076280041797e-05,
"loss": 0.0141,
"step": 1510
},
{
"epoch": 1.59,
"learning_rate": 4.2058516196447236e-05,
"loss": 0.023,
"step": 1520
},
{
"epoch": 1.6,
"learning_rate": 4.2006269592476494e-05,
"loss": 0.0154,
"step": 1530
},
{
"epoch": 1.61,
"learning_rate": 4.195402298850575e-05,
"loss": 0.0208,
"step": 1540
},
{
"epoch": 1.62,
"learning_rate": 4.190177638453501e-05,
"loss": 0.0283,
"step": 1550
},
{
"epoch": 1.63,
"learning_rate": 4.1849529780564265e-05,
"loss": 0.0238,
"step": 1560
},
{
"epoch": 1.64,
"learning_rate": 4.179728317659352e-05,
"loss": 0.0261,
"step": 1570
},
{
"epoch": 1.65,
"learning_rate": 4.174503657262278e-05,
"loss": 0.0206,
"step": 1580
},
{
"epoch": 1.66,
"learning_rate": 4.1692789968652043e-05,
"loss": 0.0239,
"step": 1590
},
{
"epoch": 1.67,
"learning_rate": 4.16405433646813e-05,
"loss": 0.0256,
"step": 1600
},
{
"epoch": 1.68,
"learning_rate": 4.158829676071056e-05,
"loss": 0.0216,
"step": 1610
},
{
"epoch": 1.69,
"learning_rate": 4.1536050156739815e-05,
"loss": 0.0188,
"step": 1620
},
{
"epoch": 1.7,
"learning_rate": 4.148380355276907e-05,
"loss": 0.018,
"step": 1630
},
{
"epoch": 1.71,
"learning_rate": 4.143155694879833e-05,
"loss": 0.0149,
"step": 1640
},
{
"epoch": 1.72,
"learning_rate": 4.1379310344827587e-05,
"loss": 0.0158,
"step": 1650
},
{
"epoch": 1.73,
"learning_rate": 4.132706374085685e-05,
"loss": 0.0216,
"step": 1660
},
{
"epoch": 1.75,
"learning_rate": 4.127481713688611e-05,
"loss": 0.0249,
"step": 1670
},
{
"epoch": 1.76,
"learning_rate": 4.122257053291536e-05,
"loss": 0.02,
"step": 1680
},
{
"epoch": 1.77,
"learning_rate": 4.117032392894462e-05,
"loss": 0.0188,
"step": 1690
},
{
"epoch": 1.78,
"learning_rate": 4.111807732497388e-05,
"loss": 0.0194,
"step": 1700
},
{
"epoch": 1.79,
"learning_rate": 4.1065830721003136e-05,
"loss": 0.0258,
"step": 1710
},
{
"epoch": 1.8,
"learning_rate": 4.1013584117032394e-05,
"loss": 0.0187,
"step": 1720
},
{
"epoch": 1.81,
"learning_rate": 4.096133751306165e-05,
"loss": 0.0195,
"step": 1730
},
{
"epoch": 1.82,
"learning_rate": 4.0909090909090915e-05,
"loss": 0.0218,
"step": 1740
},
{
"epoch": 1.83,
"learning_rate": 4.0856844305120165e-05,
"loss": 0.0182,
"step": 1750
},
{
"epoch": 1.84,
"learning_rate": 4.080459770114943e-05,
"loss": 0.0227,
"step": 1760
},
{
"epoch": 1.85,
"learning_rate": 4.0752351097178686e-05,
"loss": 0.0267,
"step": 1770
},
{
"epoch": 1.86,
"learning_rate": 4.0700104493207943e-05,
"loss": 0.0187,
"step": 1780
},
{
"epoch": 1.87,
"learning_rate": 4.06478578892372e-05,
"loss": 0.0251,
"step": 1790
},
{
"epoch": 1.88,
"learning_rate": 4.059561128526646e-05,
"loss": 0.0153,
"step": 1800
},
{
"epoch": 1.89,
"learning_rate": 4.054336468129572e-05,
"loss": 0.0301,
"step": 1810
},
{
"epoch": 1.9,
"learning_rate": 4.049111807732497e-05,
"loss": 0.0157,
"step": 1820
},
{
"epoch": 1.91,
"learning_rate": 4.0438871473354236e-05,
"loss": 0.0204,
"step": 1830
},
{
"epoch": 1.92,
"learning_rate": 4.038662486938349e-05,
"loss": 0.0203,
"step": 1840
},
{
"epoch": 1.93,
"learning_rate": 4.033437826541275e-05,
"loss": 0.0138,
"step": 1850
},
{
"epoch": 1.94,
"learning_rate": 4.028213166144201e-05,
"loss": 0.0231,
"step": 1860
},
{
"epoch": 1.95,
"learning_rate": 4.0229885057471265e-05,
"loss": 0.0288,
"step": 1870
},
{
"epoch": 1.96,
"learning_rate": 4.017763845350053e-05,
"loss": 0.0189,
"step": 1880
},
{
"epoch": 1.97,
"learning_rate": 4.012539184952978e-05,
"loss": 0.0229,
"step": 1890
},
{
"epoch": 1.99,
"learning_rate": 4.007314524555904e-05,
"loss": 0.0171,
"step": 1900
},
{
"epoch": 2.0,
"learning_rate": 4.00208986415883e-05,
"loss": 0.02,
"step": 1910
},
{
"epoch": 2.01,
"learning_rate": 3.996865203761756e-05,
"loss": 0.014,
"step": 1920
},
{
"epoch": 2.02,
"learning_rate": 3.9916405433646815e-05,
"loss": 0.0164,
"step": 1930
},
{
"epoch": 2.03,
"learning_rate": 3.986415882967607e-05,
"loss": 0.0148,
"step": 1940
},
{
"epoch": 2.04,
"learning_rate": 3.9811912225705336e-05,
"loss": 0.0135,
"step": 1950
},
{
"epoch": 2.05,
"learning_rate": 3.9759665621734586e-05,
"loss": 0.015,
"step": 1960
},
{
"epoch": 2.06,
"learning_rate": 3.970741901776385e-05,
"loss": 0.0183,
"step": 1970
},
{
"epoch": 2.07,
"learning_rate": 3.965517241379311e-05,
"loss": 0.0105,
"step": 1980
},
{
"epoch": 2.08,
"learning_rate": 3.960292580982236e-05,
"loss": 0.0152,
"step": 1990
},
{
"epoch": 2.09,
"learning_rate": 3.955067920585162e-05,
"loss": 0.0063,
"step": 2000
},
{
"epoch": 2.09,
"eval_acc": 0.660049626153723,
"eval_cer": 0.04602510460251046,
"eval_loss": 0.023699596524238586,
"eval_runtime": 127.5932,
"eval_samples_per_second": 3.158,
"eval_steps_per_second": 0.4,
"step": 2000
},
{
"epoch": 2.1,
"learning_rate": 3.949843260188088e-05,
"loss": 0.0147,
"step": 2010
},
{
"epoch": 2.11,
"learning_rate": 3.944618599791014e-05,
"loss": 0.0126,
"step": 2020
},
{
"epoch": 2.12,
"learning_rate": 3.939393939393939e-05,
"loss": 0.0143,
"step": 2030
},
{
"epoch": 2.13,
"learning_rate": 3.934169278996865e-05,
"loss": 0.0147,
"step": 2040
},
{
"epoch": 2.14,
"learning_rate": 3.9289446185997915e-05,
"loss": 0.0134,
"step": 2050
},
{
"epoch": 2.15,
"learning_rate": 3.9237199582027165e-05,
"loss": 0.013,
"step": 2060
},
{
"epoch": 2.16,
"learning_rate": 3.918495297805643e-05,
"loss": 0.0217,
"step": 2070
},
{
"epoch": 2.17,
"learning_rate": 3.9132706374085686e-05,
"loss": 0.0174,
"step": 2080
},
{
"epoch": 2.18,
"learning_rate": 3.908045977011495e-05,
"loss": 0.0116,
"step": 2090
},
{
"epoch": 2.19,
"learning_rate": 3.90282131661442e-05,
"loss": 0.0169,
"step": 2100
},
{
"epoch": 2.2,
"learning_rate": 3.897596656217346e-05,
"loss": 0.0199,
"step": 2110
},
{
"epoch": 2.22,
"learning_rate": 3.892371995820272e-05,
"loss": 0.0124,
"step": 2120
},
{
"epoch": 2.23,
"learning_rate": 3.887147335423197e-05,
"loss": 0.0114,
"step": 2130
},
{
"epoch": 2.24,
"learning_rate": 3.8819226750261236e-05,
"loss": 0.0128,
"step": 2140
},
{
"epoch": 2.25,
"learning_rate": 3.876698014629049e-05,
"loss": 0.0097,
"step": 2150
},
{
"epoch": 2.26,
"learning_rate": 3.871473354231975e-05,
"loss": 0.0132,
"step": 2160
},
{
"epoch": 2.27,
"learning_rate": 3.866248693834901e-05,
"loss": 0.0133,
"step": 2170
},
{
"epoch": 2.28,
"learning_rate": 3.8610240334378265e-05,
"loss": 0.0094,
"step": 2180
},
{
"epoch": 2.29,
"learning_rate": 3.855799373040753e-05,
"loss": 0.0185,
"step": 2190
},
{
"epoch": 2.3,
"learning_rate": 3.850574712643678e-05,
"loss": 0.017,
"step": 2200
},
{
"epoch": 2.31,
"learning_rate": 3.845350052246604e-05,
"loss": 0.0128,
"step": 2210
},
{
"epoch": 2.32,
"learning_rate": 3.84012539184953e-05,
"loss": 0.0121,
"step": 2220
},
{
"epoch": 2.33,
"learning_rate": 3.834900731452456e-05,
"loss": 0.015,
"step": 2230
},
{
"epoch": 2.34,
"learning_rate": 3.8296760710553815e-05,
"loss": 0.0077,
"step": 2240
},
{
"epoch": 2.35,
"learning_rate": 3.824451410658307e-05,
"loss": 0.0118,
"step": 2250
},
{
"epoch": 2.36,
"learning_rate": 3.8192267502612336e-05,
"loss": 0.0133,
"step": 2260
},
{
"epoch": 2.37,
"learning_rate": 3.8140020898641586e-05,
"loss": 0.0156,
"step": 2270
},
{
"epoch": 2.38,
"learning_rate": 3.808777429467085e-05,
"loss": 0.0131,
"step": 2280
},
{
"epoch": 2.39,
"learning_rate": 3.803552769070011e-05,
"loss": 0.01,
"step": 2290
},
{
"epoch": 2.4,
"learning_rate": 3.7983281086729364e-05,
"loss": 0.0166,
"step": 2300
},
{
"epoch": 2.41,
"learning_rate": 3.793103448275862e-05,
"loss": 0.0125,
"step": 2310
},
{
"epoch": 2.42,
"learning_rate": 3.787878787878788e-05,
"loss": 0.0092,
"step": 2320
},
{
"epoch": 2.43,
"learning_rate": 3.782654127481714e-05,
"loss": 0.0105,
"step": 2330
},
{
"epoch": 2.45,
"learning_rate": 3.777429467084639e-05,
"loss": 0.0138,
"step": 2340
},
{
"epoch": 2.46,
"learning_rate": 3.772204806687566e-05,
"loss": 0.0136,
"step": 2350
},
{
"epoch": 2.47,
"learning_rate": 3.7669801462904914e-05,
"loss": 0.0153,
"step": 2360
},
{
"epoch": 2.48,
"learning_rate": 3.761755485893417e-05,
"loss": 0.0094,
"step": 2370
},
{
"epoch": 2.49,
"learning_rate": 3.756530825496343e-05,
"loss": 0.0081,
"step": 2380
},
{
"epoch": 2.5,
"learning_rate": 3.7513061650992686e-05,
"loss": 0.0169,
"step": 2390
},
{
"epoch": 2.51,
"learning_rate": 3.746081504702195e-05,
"loss": 0.0086,
"step": 2400
},
{
"epoch": 2.52,
"learning_rate": 3.740856844305121e-05,
"loss": 0.0124,
"step": 2410
},
{
"epoch": 2.53,
"learning_rate": 3.735632183908046e-05,
"loss": 0.0143,
"step": 2420
},
{
"epoch": 2.54,
"learning_rate": 3.730407523510972e-05,
"loss": 0.0092,
"step": 2430
},
{
"epoch": 2.55,
"learning_rate": 3.725182863113898e-05,
"loss": 0.0147,
"step": 2440
},
{
"epoch": 2.56,
"learning_rate": 3.7199582027168236e-05,
"loss": 0.009,
"step": 2450
},
{
"epoch": 2.57,
"learning_rate": 3.714733542319749e-05,
"loss": 0.0083,
"step": 2460
},
{
"epoch": 2.58,
"learning_rate": 3.709508881922675e-05,
"loss": 0.0146,
"step": 2470
},
{
"epoch": 2.59,
"learning_rate": 3.7042842215256014e-05,
"loss": 0.0142,
"step": 2480
},
{
"epoch": 2.6,
"learning_rate": 3.6990595611285264e-05,
"loss": 0.0097,
"step": 2490
},
{
"epoch": 2.61,
"learning_rate": 3.693834900731453e-05,
"loss": 0.0067,
"step": 2500
},
{
"epoch": 2.62,
"learning_rate": 3.6886102403343786e-05,
"loss": 0.0085,
"step": 2510
},
{
"epoch": 2.63,
"learning_rate": 3.683385579937304e-05,
"loss": 0.01,
"step": 2520
},
{
"epoch": 2.64,
"learning_rate": 3.67816091954023e-05,
"loss": 0.0128,
"step": 2530
},
{
"epoch": 2.65,
"learning_rate": 3.672936259143156e-05,
"loss": 0.0113,
"step": 2540
},
{
"epoch": 2.66,
"learning_rate": 3.667711598746082e-05,
"loss": 0.0098,
"step": 2550
},
{
"epoch": 2.68,
"learning_rate": 3.662486938349007e-05,
"loss": 0.0159,
"step": 2560
},
{
"epoch": 2.69,
"learning_rate": 3.6572622779519335e-05,
"loss": 0.0188,
"step": 2570
},
{
"epoch": 2.7,
"learning_rate": 3.652037617554859e-05,
"loss": 0.0149,
"step": 2580
},
{
"epoch": 2.71,
"learning_rate": 3.646812957157785e-05,
"loss": 0.0159,
"step": 2590
},
{
"epoch": 2.72,
"learning_rate": 3.641588296760711e-05,
"loss": 0.0093,
"step": 2600
},
{
"epoch": 2.73,
"learning_rate": 3.6363636363636364e-05,
"loss": 0.0115,
"step": 2610
},
{
"epoch": 2.74,
"learning_rate": 3.631138975966563e-05,
"loss": 0.0138,
"step": 2620
},
{
"epoch": 2.75,
"learning_rate": 3.625914315569488e-05,
"loss": 0.011,
"step": 2630
},
{
"epoch": 2.76,
"learning_rate": 3.620689655172414e-05,
"loss": 0.0144,
"step": 2640
},
{
"epoch": 2.77,
"learning_rate": 3.61546499477534e-05,
"loss": 0.0098,
"step": 2650
},
{
"epoch": 2.78,
"learning_rate": 3.610240334378266e-05,
"loss": 0.0132,
"step": 2660
},
{
"epoch": 2.79,
"learning_rate": 3.6050156739811914e-05,
"loss": 0.0125,
"step": 2670
},
{
"epoch": 2.8,
"learning_rate": 3.599791013584117e-05,
"loss": 0.0118,
"step": 2680
},
{
"epoch": 2.81,
"learning_rate": 3.5945663531870435e-05,
"loss": 0.0131,
"step": 2690
},
{
"epoch": 2.82,
"learning_rate": 3.5893416927899686e-05,
"loss": 0.0081,
"step": 2700
},
{
"epoch": 2.83,
"learning_rate": 3.584117032392895e-05,
"loss": 0.0115,
"step": 2710
},
{
"epoch": 2.84,
"learning_rate": 3.578892371995821e-05,
"loss": 0.0109,
"step": 2720
},
{
"epoch": 2.85,
"learning_rate": 3.573667711598746e-05,
"loss": 0.0149,
"step": 2730
},
{
"epoch": 2.86,
"learning_rate": 3.568443051201672e-05,
"loss": 0.0093,
"step": 2740
},
{
"epoch": 2.87,
"learning_rate": 3.563218390804598e-05,
"loss": 0.0124,
"step": 2750
},
{
"epoch": 2.88,
"learning_rate": 3.557993730407524e-05,
"loss": 0.0094,
"step": 2760
},
{
"epoch": 2.89,
"learning_rate": 3.552769070010449e-05,
"loss": 0.0105,
"step": 2770
},
{
"epoch": 2.9,
"learning_rate": 3.547544409613375e-05,
"loss": 0.0221,
"step": 2780
},
{
"epoch": 2.92,
"learning_rate": 3.5423197492163014e-05,
"loss": 0.0088,
"step": 2790
},
{
"epoch": 2.93,
"learning_rate": 3.5370950888192264e-05,
"loss": 0.0077,
"step": 2800
},
{
"epoch": 2.94,
"learning_rate": 3.531870428422153e-05,
"loss": 0.0106,
"step": 2810
},
{
"epoch": 2.95,
"learning_rate": 3.5266457680250785e-05,
"loss": 0.0113,
"step": 2820
},
{
"epoch": 2.96,
"learning_rate": 3.521421107628005e-05,
"loss": 0.012,
"step": 2830
},
{
"epoch": 2.97,
"learning_rate": 3.51619644723093e-05,
"loss": 0.0094,
"step": 2840
},
{
"epoch": 2.98,
"learning_rate": 3.510971786833856e-05,
"loss": 0.0088,
"step": 2850
},
{
"epoch": 2.99,
"learning_rate": 3.505747126436782e-05,
"loss": 0.0162,
"step": 2860
},
{
"epoch": 3.0,
"learning_rate": 3.500522466039707e-05,
"loss": 0.0104,
"step": 2870
},
{
"epoch": 3.01,
"learning_rate": 3.4952978056426335e-05,
"loss": 0.0045,
"step": 2880
},
{
"epoch": 3.02,
"learning_rate": 3.490073145245559e-05,
"loss": 0.0076,
"step": 2890
},
{
"epoch": 3.03,
"learning_rate": 3.484848484848485e-05,
"loss": 0.0092,
"step": 2900
},
{
"epoch": 3.04,
"learning_rate": 3.479623824451411e-05,
"loss": 0.0088,
"step": 2910
},
{
"epoch": 3.05,
"learning_rate": 3.4743991640543364e-05,
"loss": 0.0068,
"step": 2920
},
{
"epoch": 3.06,
"learning_rate": 3.469174503657263e-05,
"loss": 0.0037,
"step": 2930
},
{
"epoch": 3.07,
"learning_rate": 3.463949843260188e-05,
"loss": 0.008,
"step": 2940
},
{
"epoch": 3.08,
"learning_rate": 3.458725182863114e-05,
"loss": 0.0053,
"step": 2950
},
{
"epoch": 3.09,
"learning_rate": 3.45350052246604e-05,
"loss": 0.0085,
"step": 2960
},
{
"epoch": 3.1,
"learning_rate": 3.4482758620689657e-05,
"loss": 0.0122,
"step": 2970
},
{
"epoch": 3.11,
"learning_rate": 3.4430512016718914e-05,
"loss": 0.0096,
"step": 2980
},
{
"epoch": 3.12,
"learning_rate": 3.437826541274817e-05,
"loss": 0.0079,
"step": 2990
},
{
"epoch": 3.13,
"learning_rate": 3.4326018808777435e-05,
"loss": 0.01,
"step": 3000
},
{
"epoch": 3.13,
"eval_acc": 0.7444168716019433,
"eval_cer": 0.031845653184565316,
"eval_loss": 0.017891723662614822,
"eval_runtime": 127.5496,
"eval_samples_per_second": 3.16,
"eval_steps_per_second": 0.4,
"step": 3000
},
{
"epoch": 3.15,
"learning_rate": 3.4273772204806685e-05,
"loss": 0.0105,
"step": 3010
},
{
"epoch": 3.16,
"learning_rate": 3.422152560083595e-05,
"loss": 0.0091,
"step": 3020
},
{
"epoch": 3.17,
"learning_rate": 3.4169278996865206e-05,
"loss": 0.005,
"step": 3030
},
{
"epoch": 3.18,
"learning_rate": 3.4117032392894464e-05,
"loss": 0.0069,
"step": 3040
},
{
"epoch": 3.19,
"learning_rate": 3.406478578892372e-05,
"loss": 0.0066,
"step": 3050
},
{
"epoch": 3.2,
"learning_rate": 3.401253918495298e-05,
"loss": 0.0065,
"step": 3060
},
{
"epoch": 3.21,
"learning_rate": 3.396029258098224e-05,
"loss": 0.0098,
"step": 3070
},
{
"epoch": 3.22,
"learning_rate": 3.390804597701149e-05,
"loss": 0.0089,
"step": 3080
},
{
"epoch": 3.23,
"learning_rate": 3.3855799373040756e-05,
"loss": 0.0082,
"step": 3090
},
{
"epoch": 3.24,
"learning_rate": 3.3803552769070014e-05,
"loss": 0.0089,
"step": 3100
},
{
"epoch": 3.25,
"learning_rate": 3.375130616509927e-05,
"loss": 0.0097,
"step": 3110
},
{
"epoch": 3.26,
"learning_rate": 3.369905956112853e-05,
"loss": 0.0079,
"step": 3120
},
{
"epoch": 3.27,
"learning_rate": 3.3646812957157785e-05,
"loss": 0.0079,
"step": 3130
},
{
"epoch": 3.28,
"learning_rate": 3.359456635318705e-05,
"loss": 0.0062,
"step": 3140
},
{
"epoch": 3.29,
"learning_rate": 3.35423197492163e-05,
"loss": 0.0095,
"step": 3150
},
{
"epoch": 3.3,
"learning_rate": 3.3490073145245557e-05,
"loss": 0.0089,
"step": 3160
},
{
"epoch": 3.31,
"learning_rate": 3.343782654127482e-05,
"loss": 0.0107,
"step": 3170
},
{
"epoch": 3.32,
"learning_rate": 3.338557993730408e-05,
"loss": 0.0135,
"step": 3180
},
{
"epoch": 3.33,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.006,
"step": 3190
},
{
"epoch": 3.34,
"learning_rate": 3.328108672936259e-05,
"loss": 0.0078,
"step": 3200
},
{
"epoch": 3.35,
"learning_rate": 3.322884012539185e-05,
"loss": 0.005,
"step": 3210
},
{
"epoch": 3.36,
"learning_rate": 3.3176593521421106e-05,
"loss": 0.0073,
"step": 3220
},
{
"epoch": 3.38,
"learning_rate": 3.3124346917450364e-05,
"loss": 0.0084,
"step": 3230
},
{
"epoch": 3.39,
"learning_rate": 3.307210031347963e-05,
"loss": 0.0077,
"step": 3240
},
{
"epoch": 3.4,
"learning_rate": 3.3019853709508885e-05,
"loss": 0.0065,
"step": 3250
},
{
"epoch": 3.41,
"learning_rate": 3.296760710553814e-05,
"loss": 0.0052,
"step": 3260
},
{
"epoch": 3.42,
"learning_rate": 3.29153605015674e-05,
"loss": 0.0078,
"step": 3270
},
{
"epoch": 3.43,
"learning_rate": 3.2863113897596656e-05,
"loss": 0.0056,
"step": 3280
},
{
"epoch": 3.44,
"learning_rate": 3.2810867293625914e-05,
"loss": 0.0084,
"step": 3290
},
{
"epoch": 3.45,
"learning_rate": 3.275862068965517e-05,
"loss": 0.0077,
"step": 3300
},
{
"epoch": 3.46,
"learning_rate": 3.2706374085684435e-05,
"loss": 0.0071,
"step": 3310
},
{
"epoch": 3.47,
"learning_rate": 3.265412748171369e-05,
"loss": 0.0084,
"step": 3320
},
{
"epoch": 3.48,
"learning_rate": 3.260188087774295e-05,
"loss": 0.0079,
"step": 3330
},
{
"epoch": 3.49,
"learning_rate": 3.2549634273772206e-05,
"loss": 0.0058,
"step": 3340
},
{
"epoch": 3.5,
"learning_rate": 3.2497387669801463e-05,
"loss": 0.0074,
"step": 3350
},
{
"epoch": 3.51,
"learning_rate": 3.244514106583073e-05,
"loss": 0.007,
"step": 3360
},
{
"epoch": 3.52,
"learning_rate": 3.239289446185998e-05,
"loss": 0.0089,
"step": 3370
},
{
"epoch": 3.53,
"learning_rate": 3.234064785788924e-05,
"loss": 0.0059,
"step": 3380
},
{
"epoch": 3.54,
"learning_rate": 3.22884012539185e-05,
"loss": 0.0082,
"step": 3390
},
{
"epoch": 3.55,
"learning_rate": 3.2236154649947756e-05,
"loss": 0.0107,
"step": 3400
},
{
"epoch": 3.56,
"learning_rate": 3.218390804597701e-05,
"loss": 0.0063,
"step": 3410
},
{
"epoch": 3.57,
"learning_rate": 3.213166144200627e-05,
"loss": 0.0059,
"step": 3420
},
{
"epoch": 3.58,
"learning_rate": 3.2079414838035534e-05,
"loss": 0.0074,
"step": 3430
},
{
"epoch": 3.59,
"learning_rate": 3.2027168234064785e-05,
"loss": 0.011,
"step": 3440
},
{
"epoch": 3.61,
"learning_rate": 3.197492163009405e-05,
"loss": 0.0107,
"step": 3450
},
{
"epoch": 3.62,
"learning_rate": 3.1922675026123306e-05,
"loss": 0.0071,
"step": 3460
},
{
"epoch": 3.63,
"learning_rate": 3.1870428422152556e-05,
"loss": 0.0073,
"step": 3470
},
{
"epoch": 3.64,
"learning_rate": 3.181818181818182e-05,
"loss": 0.0079,
"step": 3480
},
{
"epoch": 3.65,
"learning_rate": 3.176593521421108e-05,
"loss": 0.0068,
"step": 3490
},
{
"epoch": 3.66,
"learning_rate": 3.171368861024034e-05,
"loss": 0.0085,
"step": 3500
},
{
"epoch": 3.67,
"learning_rate": 3.166144200626959e-05,
"loss": 0.0087,
"step": 3510
},
{
"epoch": 3.68,
"learning_rate": 3.160919540229885e-05,
"loss": 0.005,
"step": 3520
},
{
"epoch": 3.69,
"learning_rate": 3.155694879832811e-05,
"loss": 0.0051,
"step": 3530
},
{
"epoch": 3.7,
"learning_rate": 3.1504702194357363e-05,
"loss": 0.0076,
"step": 3540
},
{
"epoch": 3.71,
"learning_rate": 3.145245559038663e-05,
"loss": 0.0054,
"step": 3550
},
{
"epoch": 3.72,
"learning_rate": 3.1400208986415885e-05,
"loss": 0.0083,
"step": 3560
},
{
"epoch": 3.73,
"learning_rate": 3.134796238244515e-05,
"loss": 0.0076,
"step": 3570
},
{
"epoch": 3.74,
"learning_rate": 3.12957157784744e-05,
"loss": 0.0094,
"step": 3580
},
{
"epoch": 3.75,
"learning_rate": 3.1243469174503656e-05,
"loss": 0.0083,
"step": 3590
},
{
"epoch": 3.76,
"learning_rate": 3.119122257053292e-05,
"loss": 0.0073,
"step": 3600
},
{
"epoch": 3.77,
"learning_rate": 3.113897596656217e-05,
"loss": 0.0063,
"step": 3610
},
{
"epoch": 3.78,
"learning_rate": 3.1086729362591434e-05,
"loss": 0.0086,
"step": 3620
},
{
"epoch": 3.79,
"learning_rate": 3.103448275862069e-05,
"loss": 0.0118,
"step": 3630
},
{
"epoch": 3.8,
"learning_rate": 3.098223615464995e-05,
"loss": 0.0088,
"step": 3640
},
{
"epoch": 3.81,
"learning_rate": 3.0929989550679206e-05,
"loss": 0.0064,
"step": 3650
},
{
"epoch": 3.82,
"learning_rate": 3.087774294670846e-05,
"loss": 0.0065,
"step": 3660
},
{
"epoch": 3.83,
"learning_rate": 3.082549634273773e-05,
"loss": 0.0057,
"step": 3670
},
{
"epoch": 3.85,
"learning_rate": 3.077324973876698e-05,
"loss": 0.0129,
"step": 3680
},
{
"epoch": 3.86,
"learning_rate": 3.072100313479624e-05,
"loss": 0.0119,
"step": 3690
},
{
"epoch": 3.87,
"learning_rate": 3.06687565308255e-05,
"loss": 0.0074,
"step": 3700
},
{
"epoch": 3.88,
"learning_rate": 3.0616509926854756e-05,
"loss": 0.0058,
"step": 3710
},
{
"epoch": 3.89,
"learning_rate": 3.056426332288401e-05,
"loss": 0.0049,
"step": 3720
},
{
"epoch": 3.9,
"learning_rate": 3.0512016718913274e-05,
"loss": 0.0075,
"step": 3730
},
{
"epoch": 3.91,
"learning_rate": 3.045977011494253e-05,
"loss": 0.0087,
"step": 3740
},
{
"epoch": 3.92,
"learning_rate": 3.0407523510971785e-05,
"loss": 0.0071,
"step": 3750
},
{
"epoch": 3.93,
"learning_rate": 3.0355276907001045e-05,
"loss": 0.0101,
"step": 3760
},
{
"epoch": 3.94,
"learning_rate": 3.0303030303030306e-05,
"loss": 0.008,
"step": 3770
},
{
"epoch": 3.95,
"learning_rate": 3.0250783699059566e-05,
"loss": 0.0077,
"step": 3780
},
{
"epoch": 3.96,
"learning_rate": 3.019853709508882e-05,
"loss": 0.0114,
"step": 3790
},
{
"epoch": 3.97,
"learning_rate": 3.0146290491118077e-05,
"loss": 0.0067,
"step": 3800
},
{
"epoch": 3.98,
"learning_rate": 3.0094043887147338e-05,
"loss": 0.0067,
"step": 3810
},
{
"epoch": 3.99,
"learning_rate": 3.004179728317659e-05,
"loss": 0.0099,
"step": 3820
},
{
"epoch": 4.0,
"learning_rate": 2.9989550679205852e-05,
"loss": 0.0078,
"step": 3830
},
{
"epoch": 4.01,
"learning_rate": 2.9937304075235113e-05,
"loss": 0.0042,
"step": 3840
},
{
"epoch": 4.02,
"learning_rate": 2.988505747126437e-05,
"loss": 0.0056,
"step": 3850
},
{
"epoch": 4.03,
"learning_rate": 2.9832810867293627e-05,
"loss": 0.0043,
"step": 3860
},
{
"epoch": 4.04,
"learning_rate": 2.9780564263322884e-05,
"loss": 0.0042,
"step": 3870
},
{
"epoch": 4.05,
"learning_rate": 2.9728317659352145e-05,
"loss": 0.0062,
"step": 3880
},
{
"epoch": 4.06,
"learning_rate": 2.96760710553814e-05,
"loss": 0.0057,
"step": 3890
},
{
"epoch": 4.08,
"learning_rate": 2.962382445141066e-05,
"loss": 0.0069,
"step": 3900
},
{
"epoch": 4.09,
"learning_rate": 2.957157784743992e-05,
"loss": 0.0072,
"step": 3910
},
{
"epoch": 4.1,
"learning_rate": 2.9519331243469177e-05,
"loss": 0.0072,
"step": 3920
},
{
"epoch": 4.11,
"learning_rate": 2.946708463949843e-05,
"loss": 0.0032,
"step": 3930
},
{
"epoch": 4.12,
"learning_rate": 2.941483803552769e-05,
"loss": 0.0089,
"step": 3940
},
{
"epoch": 4.13,
"learning_rate": 2.9362591431556952e-05,
"loss": 0.0043,
"step": 3950
},
{
"epoch": 4.14,
"learning_rate": 2.9310344827586206e-05,
"loss": 0.0063,
"step": 3960
},
{
"epoch": 4.15,
"learning_rate": 2.9258098223615466e-05,
"loss": 0.0045,
"step": 3970
},
{
"epoch": 4.16,
"learning_rate": 2.9205851619644723e-05,
"loss": 0.0058,
"step": 3980
},
{
"epoch": 4.17,
"learning_rate": 2.9153605015673984e-05,
"loss": 0.005,
"step": 3990
},
{
"epoch": 4.18,
"learning_rate": 2.9101358411703238e-05,
"loss": 0.0044,
"step": 4000
},
{
"epoch": 4.18,
"eval_acc": 0.769230767322008,
"eval_cer": 0.0299860529986053,
"eval_loss": 0.01652107760310173,
"eval_runtime": 126.6833,
"eval_samples_per_second": 3.181,
"eval_steps_per_second": 0.403,
"step": 4000
},
{
"epoch": 4.19,
"learning_rate": 2.90491118077325e-05,
"loss": 0.0068,
"step": 4010
},
{
"epoch": 4.2,
"learning_rate": 2.899686520376176e-05,
"loss": 0.0039,
"step": 4020
},
{
"epoch": 4.21,
"learning_rate": 2.8944618599791013e-05,
"loss": 0.01,
"step": 4030
},
{
"epoch": 4.22,
"learning_rate": 2.8892371995820273e-05,
"loss": 0.0046,
"step": 4040
},
{
"epoch": 4.23,
"learning_rate": 2.884012539184953e-05,
"loss": 0.0096,
"step": 4050
},
{
"epoch": 4.24,
"learning_rate": 2.878787878787879e-05,
"loss": 0.006,
"step": 4060
},
{
"epoch": 4.25,
"learning_rate": 2.8735632183908045e-05,
"loss": 0.0052,
"step": 4070
},
{
"epoch": 4.26,
"learning_rate": 2.8683385579937305e-05,
"loss": 0.005,
"step": 4080
},
{
"epoch": 4.27,
"learning_rate": 2.8631138975966566e-05,
"loss": 0.004,
"step": 4090
},
{
"epoch": 4.28,
"learning_rate": 2.857889237199582e-05,
"loss": 0.0045,
"step": 4100
},
{
"epoch": 4.29,
"learning_rate": 2.8526645768025077e-05,
"loss": 0.0037,
"step": 4110
},
{
"epoch": 4.31,
"learning_rate": 2.8474399164054338e-05,
"loss": 0.0125,
"step": 4120
},
{
"epoch": 4.32,
"learning_rate": 2.8422152560083598e-05,
"loss": 0.0057,
"step": 4130
},
{
"epoch": 4.33,
"learning_rate": 2.8369905956112852e-05,
"loss": 0.0057,
"step": 4140
},
{
"epoch": 4.34,
"learning_rate": 2.8317659352142113e-05,
"loss": 0.0062,
"step": 4150
},
{
"epoch": 4.35,
"learning_rate": 2.8265412748171373e-05,
"loss": 0.0074,
"step": 4160
},
{
"epoch": 4.36,
"learning_rate": 2.8213166144200627e-05,
"loss": 0.0034,
"step": 4170
},
{
"epoch": 4.37,
"learning_rate": 2.8160919540229884e-05,
"loss": 0.0055,
"step": 4180
},
{
"epoch": 4.38,
"learning_rate": 2.8108672936259145e-05,
"loss": 0.0055,
"step": 4190
},
{
"epoch": 4.39,
"learning_rate": 2.8056426332288405e-05,
"loss": 0.0041,
"step": 4200
},
{
"epoch": 4.4,
"learning_rate": 2.800417972831766e-05,
"loss": 0.0089,
"step": 4210
},
{
"epoch": 4.41,
"learning_rate": 2.795193312434692e-05,
"loss": 0.0045,
"step": 4220
},
{
"epoch": 4.42,
"learning_rate": 2.7899686520376177e-05,
"loss": 0.0056,
"step": 4230
},
{
"epoch": 4.43,
"learning_rate": 2.7847439916405434e-05,
"loss": 0.0036,
"step": 4240
},
{
"epoch": 4.44,
"learning_rate": 2.779519331243469e-05,
"loss": 0.0036,
"step": 4250
},
{
"epoch": 4.45,
"learning_rate": 2.774294670846395e-05,
"loss": 0.004,
"step": 4260
},
{
"epoch": 4.46,
"learning_rate": 2.7690700104493212e-05,
"loss": 0.0049,
"step": 4270
},
{
"epoch": 4.47,
"learning_rate": 2.7638453500522466e-05,
"loss": 0.0042,
"step": 4280
},
{
"epoch": 4.48,
"learning_rate": 2.7586206896551727e-05,
"loss": 0.0066,
"step": 4290
},
{
"epoch": 4.49,
"learning_rate": 2.7533960292580984e-05,
"loss": 0.0054,
"step": 4300
},
{
"epoch": 4.5,
"learning_rate": 2.7481713688610244e-05,
"loss": 0.0041,
"step": 4310
},
{
"epoch": 4.51,
"learning_rate": 2.7429467084639498e-05,
"loss": 0.0034,
"step": 4320
},
{
"epoch": 4.52,
"learning_rate": 2.737722048066876e-05,
"loss": 0.0065,
"step": 4330
},
{
"epoch": 4.54,
"learning_rate": 2.732497387669802e-05,
"loss": 0.0062,
"step": 4340
},
{
"epoch": 4.55,
"learning_rate": 2.7272727272727273e-05,
"loss": 0.0068,
"step": 4350
},
{
"epoch": 4.56,
"learning_rate": 2.722048066875653e-05,
"loss": 0.0045,
"step": 4360
},
{
"epoch": 4.57,
"learning_rate": 2.716823406478579e-05,
"loss": 0.0089,
"step": 4370
},
{
"epoch": 4.58,
"learning_rate": 2.711598746081505e-05,
"loss": 0.0046,
"step": 4380
},
{
"epoch": 4.59,
"learning_rate": 2.7063740856844305e-05,
"loss": 0.0061,
"step": 4390
},
{
"epoch": 4.6,
"learning_rate": 2.7011494252873566e-05,
"loss": 0.0036,
"step": 4400
},
{
"epoch": 4.61,
"learning_rate": 2.6959247648902823e-05,
"loss": 0.0048,
"step": 4410
},
{
"epoch": 4.62,
"learning_rate": 2.690700104493208e-05,
"loss": 0.0028,
"step": 4420
},
{
"epoch": 4.63,
"learning_rate": 2.6854754440961337e-05,
"loss": 0.0037,
"step": 4430
},
{
"epoch": 4.64,
"learning_rate": 2.6802507836990598e-05,
"loss": 0.0039,
"step": 4440
},
{
"epoch": 4.65,
"learning_rate": 2.675026123301986e-05,
"loss": 0.0077,
"step": 4450
},
{
"epoch": 4.66,
"learning_rate": 2.6698014629049112e-05,
"loss": 0.0035,
"step": 4460
},
{
"epoch": 4.67,
"learning_rate": 2.6645768025078373e-05,
"loss": 0.008,
"step": 4470
},
{
"epoch": 4.68,
"learning_rate": 2.659352142110763e-05,
"loss": 0.0053,
"step": 4480
},
{
"epoch": 4.69,
"learning_rate": 2.6541274817136884e-05,
"loss": 0.0031,
"step": 4490
},
{
"epoch": 4.7,
"learning_rate": 2.6489028213166144e-05,
"loss": 0.0038,
"step": 4500
},
{
"epoch": 4.71,
"learning_rate": 2.6436781609195405e-05,
"loss": 0.0042,
"step": 4510
},
{
"epoch": 4.72,
"learning_rate": 2.6384535005224666e-05,
"loss": 0.0073,
"step": 4520
},
{
"epoch": 4.73,
"learning_rate": 2.633228840125392e-05,
"loss": 0.0056,
"step": 4530
},
{
"epoch": 4.74,
"learning_rate": 2.6280041797283177e-05,
"loss": 0.0039,
"step": 4540
},
{
"epoch": 4.75,
"learning_rate": 2.6227795193312437e-05,
"loss": 0.0045,
"step": 4550
},
{
"epoch": 4.76,
"learning_rate": 2.617554858934169e-05,
"loss": 0.0065,
"step": 4560
},
{
"epoch": 4.78,
"learning_rate": 2.612330198537095e-05,
"loss": 0.0039,
"step": 4570
},
{
"epoch": 4.79,
"learning_rate": 2.6071055381400212e-05,
"loss": 0.0043,
"step": 4580
},
{
"epoch": 4.8,
"learning_rate": 2.601880877742947e-05,
"loss": 0.0033,
"step": 4590
},
{
"epoch": 4.81,
"learning_rate": 2.5966562173458726e-05,
"loss": 0.0035,
"step": 4600
},
{
"epoch": 4.82,
"learning_rate": 2.5914315569487984e-05,
"loss": 0.0046,
"step": 4610
},
{
"epoch": 4.83,
"learning_rate": 2.5862068965517244e-05,
"loss": 0.0026,
"step": 4620
},
{
"epoch": 4.84,
"learning_rate": 2.5809822361546498e-05,
"loss": 0.0031,
"step": 4630
},
{
"epoch": 4.85,
"learning_rate": 2.575757575757576e-05,
"loss": 0.0042,
"step": 4640
},
{
"epoch": 4.86,
"learning_rate": 2.570532915360502e-05,
"loss": 0.0037,
"step": 4650
},
{
"epoch": 4.87,
"learning_rate": 2.5653082549634276e-05,
"loss": 0.0034,
"step": 4660
},
{
"epoch": 4.88,
"learning_rate": 2.560083594566353e-05,
"loss": 0.006,
"step": 4670
},
{
"epoch": 4.89,
"learning_rate": 2.554858934169279e-05,
"loss": 0.005,
"step": 4680
},
{
"epoch": 4.9,
"learning_rate": 2.549634273772205e-05,
"loss": 0.0033,
"step": 4690
},
{
"epoch": 4.91,
"learning_rate": 2.5444096133751305e-05,
"loss": 0.0061,
"step": 4700
},
{
"epoch": 4.92,
"learning_rate": 2.5391849529780566e-05,
"loss": 0.0059,
"step": 4710
},
{
"epoch": 4.93,
"learning_rate": 2.5339602925809823e-05,
"loss": 0.0046,
"step": 4720
},
{
"epoch": 4.94,
"learning_rate": 2.5287356321839083e-05,
"loss": 0.0062,
"step": 4730
},
{
"epoch": 4.95,
"learning_rate": 2.5235109717868337e-05,
"loss": 0.0033,
"step": 4740
},
{
"epoch": 4.96,
"learning_rate": 2.5182863113897598e-05,
"loss": 0.0065,
"step": 4750
},
{
"epoch": 4.97,
"learning_rate": 2.5130616509926858e-05,
"loss": 0.0048,
"step": 4760
},
{
"epoch": 4.98,
"learning_rate": 2.5078369905956112e-05,
"loss": 0.0043,
"step": 4770
},
{
"epoch": 4.99,
"learning_rate": 2.5026123301985373e-05,
"loss": 0.0055,
"step": 4780
},
{
"epoch": 5.01,
"learning_rate": 2.497387669801463e-05,
"loss": 0.0039,
"step": 4790
},
{
"epoch": 5.02,
"learning_rate": 2.4921630094043887e-05,
"loss": 0.0025,
"step": 4800
},
{
"epoch": 5.03,
"learning_rate": 2.4869383490073148e-05,
"loss": 0.0039,
"step": 4810
},
{
"epoch": 5.04,
"learning_rate": 2.4817136886102405e-05,
"loss": 0.0048,
"step": 4820
},
{
"epoch": 5.05,
"learning_rate": 2.4764890282131662e-05,
"loss": 0.005,
"step": 4830
},
{
"epoch": 5.06,
"learning_rate": 2.4712643678160922e-05,
"loss": 0.0039,
"step": 4840
},
{
"epoch": 5.07,
"learning_rate": 2.466039707419018e-05,
"loss": 0.0096,
"step": 4850
},
{
"epoch": 5.08,
"learning_rate": 2.4608150470219437e-05,
"loss": 0.0046,
"step": 4860
},
{
"epoch": 5.09,
"learning_rate": 2.4555903866248694e-05,
"loss": 0.0026,
"step": 4870
},
{
"epoch": 5.1,
"learning_rate": 2.4503657262277955e-05,
"loss": 0.0043,
"step": 4880
},
{
"epoch": 5.11,
"learning_rate": 2.4451410658307212e-05,
"loss": 0.0031,
"step": 4890
},
{
"epoch": 5.12,
"learning_rate": 2.439916405433647e-05,
"loss": 0.005,
"step": 4900
},
{
"epoch": 5.13,
"learning_rate": 2.4346917450365726e-05,
"loss": 0.002,
"step": 4910
},
{
"epoch": 5.14,
"learning_rate": 2.4294670846394983e-05,
"loss": 0.0021,
"step": 4920
},
{
"epoch": 5.15,
"learning_rate": 2.4242424242424244e-05,
"loss": 0.0052,
"step": 4930
},
{
"epoch": 5.16,
"learning_rate": 2.41901776384535e-05,
"loss": 0.006,
"step": 4940
},
{
"epoch": 5.17,
"learning_rate": 2.413793103448276e-05,
"loss": 0.0028,
"step": 4950
},
{
"epoch": 5.18,
"learning_rate": 2.408568443051202e-05,
"loss": 0.0026,
"step": 4960
},
{
"epoch": 5.19,
"learning_rate": 2.4033437826541276e-05,
"loss": 0.0033,
"step": 4970
},
{
"epoch": 5.2,
"learning_rate": 2.3981191222570533e-05,
"loss": 0.003,
"step": 4980
},
{
"epoch": 5.21,
"learning_rate": 2.392894461859979e-05,
"loss": 0.0051,
"step": 4990
},
{
"epoch": 5.22,
"learning_rate": 2.387669801462905e-05,
"loss": 0.0062,
"step": 5000
},
{
"epoch": 5.22,
"eval_acc": 0.7915632734700663,
"eval_cer": 0.026731752673175267,
"eval_loss": 0.01478270161896944,
"eval_runtime": 128.2616,
"eval_samples_per_second": 3.142,
"eval_steps_per_second": 0.398,
"step": 5000
},
{
"epoch": 5.24,
"learning_rate": 2.3824451410658308e-05,
"loss": 0.0038,
"step": 5010
},
{
"epoch": 5.25,
"learning_rate": 2.377220480668757e-05,
"loss": 0.0061,
"step": 5020
},
{
"epoch": 5.26,
"learning_rate": 2.3719958202716826e-05,
"loss": 0.0035,
"step": 5030
},
{
"epoch": 5.27,
"learning_rate": 2.3667711598746083e-05,
"loss": 0.0034,
"step": 5040
},
{
"epoch": 5.28,
"learning_rate": 2.361546499477534e-05,
"loss": 0.0033,
"step": 5050
},
{
"epoch": 5.29,
"learning_rate": 2.3563218390804597e-05,
"loss": 0.0036,
"step": 5060
},
{
"epoch": 5.3,
"learning_rate": 2.3510971786833858e-05,
"loss": 0.004,
"step": 5070
},
{
"epoch": 5.31,
"learning_rate": 2.3458725182863115e-05,
"loss": 0.0023,
"step": 5080
},
{
"epoch": 5.32,
"learning_rate": 2.3406478578892372e-05,
"loss": 0.0024,
"step": 5090
},
{
"epoch": 5.33,
"learning_rate": 2.335423197492163e-05,
"loss": 0.0026,
"step": 5100
},
{
"epoch": 5.34,
"learning_rate": 2.330198537095089e-05,
"loss": 0.0031,
"step": 5110
},
{
"epoch": 5.35,
"learning_rate": 2.3249738766980147e-05,
"loss": 0.0032,
"step": 5120
},
{
"epoch": 5.36,
"learning_rate": 2.3197492163009404e-05,
"loss": 0.0032,
"step": 5130
},
{
"epoch": 5.37,
"learning_rate": 2.3145245559038665e-05,
"loss": 0.0043,
"step": 5140
},
{
"epoch": 5.38,
"learning_rate": 2.3092998955067922e-05,
"loss": 0.0039,
"step": 5150
},
{
"epoch": 5.39,
"learning_rate": 2.304075235109718e-05,
"loss": 0.0026,
"step": 5160
},
{
"epoch": 5.4,
"learning_rate": 2.2988505747126437e-05,
"loss": 0.0032,
"step": 5170
},
{
"epoch": 5.41,
"learning_rate": 2.2936259143155697e-05,
"loss": 0.0043,
"step": 5180
},
{
"epoch": 5.42,
"learning_rate": 2.2884012539184954e-05,
"loss": 0.0037,
"step": 5190
},
{
"epoch": 5.43,
"learning_rate": 2.283176593521421e-05,
"loss": 0.0038,
"step": 5200
},
{
"epoch": 5.44,
"learning_rate": 2.2779519331243472e-05,
"loss": 0.0038,
"step": 5210
},
{
"epoch": 5.45,
"learning_rate": 2.272727272727273e-05,
"loss": 0.004,
"step": 5220
},
{
"epoch": 5.46,
"learning_rate": 2.2675026123301986e-05,
"loss": 0.0025,
"step": 5230
},
{
"epoch": 5.48,
"learning_rate": 2.2622779519331244e-05,
"loss": 0.0023,
"step": 5240
},
{
"epoch": 5.49,
"learning_rate": 2.2570532915360504e-05,
"loss": 0.0024,
"step": 5250
},
{
"epoch": 5.5,
"learning_rate": 2.251828631138976e-05,
"loss": 0.0021,
"step": 5260
},
{
"epoch": 5.51,
"learning_rate": 2.246603970741902e-05,
"loss": 0.0045,
"step": 5270
},
{
"epoch": 5.52,
"learning_rate": 2.2413793103448276e-05,
"loss": 0.0026,
"step": 5280
},
{
"epoch": 5.53,
"learning_rate": 2.2361546499477533e-05,
"loss": 0.0044,
"step": 5290
},
{
"epoch": 5.54,
"learning_rate": 2.2309299895506794e-05,
"loss": 0.0046,
"step": 5300
},
{
"epoch": 5.55,
"learning_rate": 2.225705329153605e-05,
"loss": 0.0038,
"step": 5310
},
{
"epoch": 5.56,
"learning_rate": 2.220480668756531e-05,
"loss": 0.009,
"step": 5320
},
{
"epoch": 5.57,
"learning_rate": 2.215256008359457e-05,
"loss": 0.0048,
"step": 5330
},
{
"epoch": 5.58,
"learning_rate": 2.2100313479623826e-05,
"loss": 0.0033,
"step": 5340
},
{
"epoch": 5.59,
"learning_rate": 2.2048066875653083e-05,
"loss": 0.0024,
"step": 5350
},
{
"epoch": 5.6,
"learning_rate": 2.199582027168234e-05,
"loss": 0.002,
"step": 5360
},
{
"epoch": 5.61,
"learning_rate": 2.19435736677116e-05,
"loss": 0.0032,
"step": 5370
},
{
"epoch": 5.62,
"learning_rate": 2.1891327063740858e-05,
"loss": 0.0055,
"step": 5380
},
{
"epoch": 5.63,
"learning_rate": 2.183908045977012e-05,
"loss": 0.0059,
"step": 5390
},
{
"epoch": 5.64,
"learning_rate": 2.1786833855799376e-05,
"loss": 0.0024,
"step": 5400
},
{
"epoch": 5.65,
"learning_rate": 2.173458725182863e-05,
"loss": 0.0039,
"step": 5410
},
{
"epoch": 5.66,
"learning_rate": 2.168234064785789e-05,
"loss": 0.002,
"step": 5420
},
{
"epoch": 5.67,
"learning_rate": 2.1630094043887147e-05,
"loss": 0.0061,
"step": 5430
},
{
"epoch": 5.68,
"learning_rate": 2.1577847439916408e-05,
"loss": 0.0056,
"step": 5440
},
{
"epoch": 5.69,
"learning_rate": 2.1525600835945665e-05,
"loss": 0.0032,
"step": 5450
},
{
"epoch": 5.71,
"learning_rate": 2.1473354231974925e-05,
"loss": 0.0029,
"step": 5460
},
{
"epoch": 5.72,
"learning_rate": 2.142110762800418e-05,
"loss": 0.0017,
"step": 5470
},
{
"epoch": 5.73,
"learning_rate": 2.1368861024033436e-05,
"loss": 0.005,
"step": 5480
},
{
"epoch": 5.74,
"learning_rate": 2.1316614420062697e-05,
"loss": 0.0051,
"step": 5490
},
{
"epoch": 5.75,
"learning_rate": 2.1264367816091954e-05,
"loss": 0.0047,
"step": 5500
},
{
"epoch": 5.76,
"learning_rate": 2.1212121212121215e-05,
"loss": 0.0032,
"step": 5510
},
{
"epoch": 5.77,
"learning_rate": 2.1159874608150472e-05,
"loss": 0.004,
"step": 5520
},
{
"epoch": 5.78,
"learning_rate": 2.110762800417973e-05,
"loss": 0.0059,
"step": 5530
},
{
"epoch": 5.79,
"learning_rate": 2.1055381400208986e-05,
"loss": 0.0045,
"step": 5540
},
{
"epoch": 5.8,
"learning_rate": 2.1003134796238247e-05,
"loss": 0.0051,
"step": 5550
},
{
"epoch": 5.81,
"learning_rate": 2.0950888192267504e-05,
"loss": 0.005,
"step": 5560
},
{
"epoch": 5.82,
"learning_rate": 2.089864158829676e-05,
"loss": 0.0039,
"step": 5570
},
{
"epoch": 5.83,
"learning_rate": 2.0846394984326022e-05,
"loss": 0.0036,
"step": 5580
},
{
"epoch": 5.84,
"learning_rate": 2.079414838035528e-05,
"loss": 0.0041,
"step": 5590
},
{
"epoch": 5.85,
"learning_rate": 2.0741901776384536e-05,
"loss": 0.002,
"step": 5600
},
{
"epoch": 5.86,
"learning_rate": 2.0689655172413793e-05,
"loss": 0.0067,
"step": 5610
},
{
"epoch": 5.87,
"learning_rate": 2.0637408568443054e-05,
"loss": 0.0061,
"step": 5620
},
{
"epoch": 5.88,
"learning_rate": 2.058516196447231e-05,
"loss": 0.0041,
"step": 5630
},
{
"epoch": 5.89,
"learning_rate": 2.0532915360501568e-05,
"loss": 0.0063,
"step": 5640
},
{
"epoch": 5.9,
"learning_rate": 2.0480668756530825e-05,
"loss": 0.0036,
"step": 5650
},
{
"epoch": 5.91,
"learning_rate": 2.0428422152560083e-05,
"loss": 0.0034,
"step": 5660
},
{
"epoch": 5.92,
"learning_rate": 2.0376175548589343e-05,
"loss": 0.0036,
"step": 5670
},
{
"epoch": 5.94,
"learning_rate": 2.03239289446186e-05,
"loss": 0.0021,
"step": 5680
},
{
"epoch": 5.95,
"learning_rate": 2.027168234064786e-05,
"loss": 0.0027,
"step": 5690
},
{
"epoch": 5.96,
"learning_rate": 2.0219435736677118e-05,
"loss": 0.004,
"step": 5700
},
{
"epoch": 5.97,
"learning_rate": 2.0167189132706375e-05,
"loss": 0.0031,
"step": 5710
},
{
"epoch": 5.98,
"learning_rate": 2.0114942528735632e-05,
"loss": 0.0022,
"step": 5720
},
{
"epoch": 5.99,
"learning_rate": 2.006269592476489e-05,
"loss": 0.0029,
"step": 5730
},
{
"epoch": 6.0,
"learning_rate": 2.001044932079415e-05,
"loss": 0.0062,
"step": 5740
},
{
"epoch": 6.01,
"learning_rate": 1.9958202716823407e-05,
"loss": 0.0016,
"step": 5750
},
{
"epoch": 6.02,
"learning_rate": 1.9905956112852668e-05,
"loss": 0.0019,
"step": 5760
},
{
"epoch": 6.03,
"learning_rate": 1.9853709508881925e-05,
"loss": 0.003,
"step": 5770
},
{
"epoch": 6.04,
"learning_rate": 1.980146290491118e-05,
"loss": 0.002,
"step": 5780
},
{
"epoch": 6.05,
"learning_rate": 1.974921630094044e-05,
"loss": 0.003,
"step": 5790
},
{
"epoch": 6.06,
"learning_rate": 1.9696969696969697e-05,
"loss": 0.0033,
"step": 5800
},
{
"epoch": 6.07,
"learning_rate": 1.9644723092998957e-05,
"loss": 0.0016,
"step": 5810
},
{
"epoch": 6.08,
"learning_rate": 1.9592476489028214e-05,
"loss": 0.004,
"step": 5820
},
{
"epoch": 6.09,
"learning_rate": 1.9540229885057475e-05,
"loss": 0.0025,
"step": 5830
},
{
"epoch": 6.1,
"learning_rate": 1.948798328108673e-05,
"loss": 0.0025,
"step": 5840
},
{
"epoch": 6.11,
"learning_rate": 1.9435736677115986e-05,
"loss": 0.0031,
"step": 5850
},
{
"epoch": 6.12,
"learning_rate": 1.9383490073145247e-05,
"loss": 0.002,
"step": 5860
},
{
"epoch": 6.13,
"learning_rate": 1.9331243469174504e-05,
"loss": 0.0027,
"step": 5870
},
{
"epoch": 6.14,
"learning_rate": 1.9278996865203764e-05,
"loss": 0.0019,
"step": 5880
},
{
"epoch": 6.15,
"learning_rate": 1.922675026123302e-05,
"loss": 0.0033,
"step": 5890
},
{
"epoch": 6.17,
"learning_rate": 1.917450365726228e-05,
"loss": 0.0027,
"step": 5900
},
{
"epoch": 6.18,
"learning_rate": 1.9122257053291536e-05,
"loss": 0.0048,
"step": 5910
},
{
"epoch": 6.19,
"learning_rate": 1.9070010449320793e-05,
"loss": 0.0016,
"step": 5920
},
{
"epoch": 6.2,
"learning_rate": 1.9017763845350054e-05,
"loss": 0.002,
"step": 5930
},
{
"epoch": 6.21,
"learning_rate": 1.896551724137931e-05,
"loss": 0.0027,
"step": 5940
},
{
"epoch": 6.22,
"learning_rate": 1.891327063740857e-05,
"loss": 0.0021,
"step": 5950
},
{
"epoch": 6.23,
"learning_rate": 1.886102403343783e-05,
"loss": 0.0068,
"step": 5960
},
{
"epoch": 6.24,
"learning_rate": 1.8808777429467086e-05,
"loss": 0.0017,
"step": 5970
},
{
"epoch": 6.25,
"learning_rate": 1.8756530825496343e-05,
"loss": 0.0013,
"step": 5980
},
{
"epoch": 6.26,
"learning_rate": 1.8704284221525603e-05,
"loss": 0.0033,
"step": 5990
},
{
"epoch": 6.27,
"learning_rate": 1.865203761755486e-05,
"loss": 0.0024,
"step": 6000
},
{
"epoch": 6.27,
"eval_acc": 0.8064516109021052,
"eval_cer": 0.02487215248721525,
"eval_loss": 0.014062131755053997,
"eval_runtime": 122.3207,
"eval_samples_per_second": 3.295,
"eval_steps_per_second": 0.417,
"step": 6000
},
{
"epoch": 6.28,
"learning_rate": 1.8599791013584118e-05,
"loss": 0.006,
"step": 6010
},
{
"epoch": 6.29,
"learning_rate": 1.8547544409613375e-05,
"loss": 0.0014,
"step": 6020
},
{
"epoch": 6.3,
"learning_rate": 1.8495297805642632e-05,
"loss": 0.0016,
"step": 6030
},
{
"epoch": 6.31,
"learning_rate": 1.8443051201671893e-05,
"loss": 0.0038,
"step": 6040
},
{
"epoch": 6.32,
"learning_rate": 1.839080459770115e-05,
"loss": 0.0032,
"step": 6050
},
{
"epoch": 6.33,
"learning_rate": 1.833855799373041e-05,
"loss": 0.0036,
"step": 6060
},
{
"epoch": 6.34,
"learning_rate": 1.8286311389759668e-05,
"loss": 0.0016,
"step": 6070
},
{
"epoch": 6.35,
"learning_rate": 1.8234064785788925e-05,
"loss": 0.0038,
"step": 6080
},
{
"epoch": 6.36,
"learning_rate": 1.8181818181818182e-05,
"loss": 0.0042,
"step": 6090
},
{
"epoch": 6.37,
"learning_rate": 1.812957157784744e-05,
"loss": 0.003,
"step": 6100
},
{
"epoch": 6.38,
"learning_rate": 1.80773249738767e-05,
"loss": 0.0035,
"step": 6110
},
{
"epoch": 6.39,
"learning_rate": 1.8025078369905957e-05,
"loss": 0.0025,
"step": 6120
},
{
"epoch": 6.41,
"learning_rate": 1.7972831765935218e-05,
"loss": 0.0022,
"step": 6130
},
{
"epoch": 6.42,
"learning_rate": 1.7920585161964475e-05,
"loss": 0.0021,
"step": 6140
},
{
"epoch": 6.43,
"learning_rate": 1.786833855799373e-05,
"loss": 0.0037,
"step": 6150
},
{
"epoch": 6.44,
"learning_rate": 1.781609195402299e-05,
"loss": 0.0037,
"step": 6160
},
{
"epoch": 6.45,
"learning_rate": 1.7763845350052246e-05,
"loss": 0.0023,
"step": 6170
},
{
"epoch": 6.46,
"learning_rate": 1.7711598746081507e-05,
"loss": 0.0041,
"step": 6180
},
{
"epoch": 6.47,
"learning_rate": 1.7659352142110764e-05,
"loss": 0.0063,
"step": 6190
},
{
"epoch": 6.48,
"learning_rate": 1.7607105538140025e-05,
"loss": 0.0024,
"step": 6200
},
{
"epoch": 6.49,
"learning_rate": 1.755485893416928e-05,
"loss": 0.0033,
"step": 6210
},
{
"epoch": 6.5,
"learning_rate": 1.7502612330198536e-05,
"loss": 0.0025,
"step": 6220
},
{
"epoch": 6.51,
"learning_rate": 1.7450365726227796e-05,
"loss": 0.0019,
"step": 6230
},
{
"epoch": 6.52,
"learning_rate": 1.7398119122257053e-05,
"loss": 0.0015,
"step": 6240
},
{
"epoch": 6.53,
"learning_rate": 1.7345872518286314e-05,
"loss": 0.0026,
"step": 6250
},
{
"epoch": 6.54,
"learning_rate": 1.729362591431557e-05,
"loss": 0.0024,
"step": 6260
},
{
"epoch": 6.55,
"learning_rate": 1.7241379310344828e-05,
"loss": 0.0018,
"step": 6270
},
{
"epoch": 6.56,
"learning_rate": 1.7189132706374085e-05,
"loss": 0.0029,
"step": 6280
},
{
"epoch": 6.57,
"learning_rate": 1.7136886102403343e-05,
"loss": 0.0031,
"step": 6290
},
{
"epoch": 6.58,
"learning_rate": 1.7084639498432603e-05,
"loss": 0.0036,
"step": 6300
},
{
"epoch": 6.59,
"learning_rate": 1.703239289446186e-05,
"loss": 0.0028,
"step": 6310
},
{
"epoch": 6.6,
"learning_rate": 1.698014629049112e-05,
"loss": 0.0021,
"step": 6320
},
{
"epoch": 6.61,
"learning_rate": 1.6927899686520378e-05,
"loss": 0.0029,
"step": 6330
},
{
"epoch": 6.62,
"learning_rate": 1.6875653082549635e-05,
"loss": 0.0023,
"step": 6340
},
{
"epoch": 6.64,
"learning_rate": 1.6823406478578893e-05,
"loss": 0.0013,
"step": 6350
},
{
"epoch": 6.65,
"learning_rate": 1.677115987460815e-05,
"loss": 0.0023,
"step": 6360
},
{
"epoch": 6.66,
"learning_rate": 1.671891327063741e-05,
"loss": 0.0016,
"step": 6370
},
{
"epoch": 6.67,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.0019,
"step": 6380
},
{
"epoch": 6.68,
"learning_rate": 1.6614420062695925e-05,
"loss": 0.003,
"step": 6390
},
{
"epoch": 6.69,
"learning_rate": 1.6562173458725182e-05,
"loss": 0.0018,
"step": 6400
},
{
"epoch": 6.7,
"learning_rate": 1.6509926854754442e-05,
"loss": 0.0029,
"step": 6410
},
{
"epoch": 6.71,
"learning_rate": 1.64576802507837e-05,
"loss": 0.0039,
"step": 6420
},
{
"epoch": 6.72,
"learning_rate": 1.6405433646812957e-05,
"loss": 0.0011,
"step": 6430
},
{
"epoch": 6.73,
"learning_rate": 1.6353187042842217e-05,
"loss": 0.0019,
"step": 6440
},
{
"epoch": 6.74,
"learning_rate": 1.6300940438871475e-05,
"loss": 0.0028,
"step": 6450
},
{
"epoch": 6.75,
"learning_rate": 1.6248693834900732e-05,
"loss": 0.004,
"step": 6460
},
{
"epoch": 6.76,
"learning_rate": 1.619644723092999e-05,
"loss": 0.0021,
"step": 6470
},
{
"epoch": 6.77,
"learning_rate": 1.614420062695925e-05,
"loss": 0.0026,
"step": 6480
},
{
"epoch": 6.78,
"learning_rate": 1.6091954022988507e-05,
"loss": 0.0024,
"step": 6490
},
{
"epoch": 6.79,
"learning_rate": 1.6039707419017767e-05,
"loss": 0.003,
"step": 6500
},
{
"epoch": 6.8,
"learning_rate": 1.5987460815047024e-05,
"loss": 0.0029,
"step": 6510
},
{
"epoch": 6.81,
"learning_rate": 1.5935214211076278e-05,
"loss": 0.0012,
"step": 6520
},
{
"epoch": 6.82,
"learning_rate": 1.588296760710554e-05,
"loss": 0.0023,
"step": 6530
},
{
"epoch": 6.83,
"learning_rate": 1.5830721003134796e-05,
"loss": 0.0023,
"step": 6540
},
{
"epoch": 6.84,
"learning_rate": 1.5778474399164057e-05,
"loss": 0.0025,
"step": 6550
},
{
"epoch": 6.85,
"learning_rate": 1.5726227795193314e-05,
"loss": 0.0029,
"step": 6560
},
{
"epoch": 6.87,
"learning_rate": 1.5673981191222574e-05,
"loss": 0.0036,
"step": 6570
},
{
"epoch": 6.88,
"learning_rate": 1.5621734587251828e-05,
"loss": 0.0033,
"step": 6580
},
{
"epoch": 6.89,
"learning_rate": 1.5569487983281085e-05,
"loss": 0.0028,
"step": 6590
},
{
"epoch": 6.9,
"learning_rate": 1.5517241379310346e-05,
"loss": 0.003,
"step": 6600
},
{
"epoch": 6.91,
"learning_rate": 1.5464994775339603e-05,
"loss": 0.0014,
"step": 6610
},
{
"epoch": 6.92,
"learning_rate": 1.5412748171368864e-05,
"loss": 0.0019,
"step": 6620
},
{
"epoch": 6.93,
"learning_rate": 1.536050156739812e-05,
"loss": 0.0031,
"step": 6630
},
{
"epoch": 6.94,
"learning_rate": 1.5308254963427378e-05,
"loss": 0.0025,
"step": 6640
},
{
"epoch": 6.95,
"learning_rate": 1.5256008359456637e-05,
"loss": 0.0016,
"step": 6650
},
{
"epoch": 6.96,
"learning_rate": 1.5203761755485892e-05,
"loss": 0.0028,
"step": 6660
},
{
"epoch": 6.97,
"learning_rate": 1.5151515151515153e-05,
"loss": 0.0034,
"step": 6670
},
{
"epoch": 6.98,
"learning_rate": 1.509926854754441e-05,
"loss": 0.0016,
"step": 6680
},
{
"epoch": 6.99,
"learning_rate": 1.5047021943573669e-05,
"loss": 0.0017,
"step": 6690
},
{
"epoch": 7.0,
"learning_rate": 1.4994775339602926e-05,
"loss": 0.0032,
"step": 6700
},
{
"epoch": 7.01,
"learning_rate": 1.4942528735632185e-05,
"loss": 0.0037,
"step": 6710
},
{
"epoch": 7.02,
"learning_rate": 1.4890282131661442e-05,
"loss": 0.0037,
"step": 6720
},
{
"epoch": 7.03,
"learning_rate": 1.48380355276907e-05,
"loss": 0.0025,
"step": 6730
},
{
"epoch": 7.04,
"learning_rate": 1.478578892371996e-05,
"loss": 0.0017,
"step": 6740
},
{
"epoch": 7.05,
"learning_rate": 1.4733542319749215e-05,
"loss": 0.003,
"step": 6750
},
{
"epoch": 7.06,
"learning_rate": 1.4681295715778476e-05,
"loss": 0.001,
"step": 6760
},
{
"epoch": 7.07,
"learning_rate": 1.4629049111807733e-05,
"loss": 0.0026,
"step": 6770
},
{
"epoch": 7.08,
"learning_rate": 1.4576802507836992e-05,
"loss": 0.003,
"step": 6780
},
{
"epoch": 7.1,
"learning_rate": 1.452455590386625e-05,
"loss": 0.0018,
"step": 6790
},
{
"epoch": 7.11,
"learning_rate": 1.4472309299895506e-05,
"loss": 0.0013,
"step": 6800
},
{
"epoch": 7.12,
"learning_rate": 1.4420062695924765e-05,
"loss": 0.002,
"step": 6810
},
{
"epoch": 7.13,
"learning_rate": 1.4367816091954022e-05,
"loss": 0.0011,
"step": 6820
},
{
"epoch": 7.14,
"learning_rate": 1.4315569487983283e-05,
"loss": 0.0015,
"step": 6830
},
{
"epoch": 7.15,
"learning_rate": 1.4263322884012539e-05,
"loss": 0.0013,
"step": 6840
},
{
"epoch": 7.16,
"learning_rate": 1.4211076280041799e-05,
"loss": 0.0035,
"step": 6850
},
{
"epoch": 7.17,
"learning_rate": 1.4158829676071056e-05,
"loss": 0.0022,
"step": 6860
},
{
"epoch": 7.18,
"learning_rate": 1.4106583072100313e-05,
"loss": 0.0021,
"step": 6870
},
{
"epoch": 7.19,
"learning_rate": 1.4054336468129572e-05,
"loss": 0.002,
"step": 6880
},
{
"epoch": 7.2,
"learning_rate": 1.400208986415883e-05,
"loss": 0.0012,
"step": 6890
},
{
"epoch": 7.21,
"learning_rate": 1.3949843260188088e-05,
"loss": 0.0018,
"step": 6900
},
{
"epoch": 7.22,
"learning_rate": 1.3897596656217346e-05,
"loss": 0.0026,
"step": 6910
},
{
"epoch": 7.23,
"learning_rate": 1.3845350052246606e-05,
"loss": 0.0016,
"step": 6920
},
{
"epoch": 7.24,
"learning_rate": 1.3793103448275863e-05,
"loss": 0.0022,
"step": 6930
},
{
"epoch": 7.25,
"learning_rate": 1.3740856844305122e-05,
"loss": 0.0018,
"step": 6940
},
{
"epoch": 7.26,
"learning_rate": 1.368861024033438e-05,
"loss": 0.001,
"step": 6950
},
{
"epoch": 7.27,
"learning_rate": 1.3636363636363637e-05,
"loss": 0.0019,
"step": 6960
},
{
"epoch": 7.28,
"learning_rate": 1.3584117032392895e-05,
"loss": 0.0014,
"step": 6970
},
{
"epoch": 7.29,
"learning_rate": 1.3531870428422153e-05,
"loss": 0.0019,
"step": 6980
},
{
"epoch": 7.3,
"learning_rate": 1.3479623824451411e-05,
"loss": 0.0032,
"step": 6990
},
{
"epoch": 7.31,
"learning_rate": 1.3427377220480669e-05,
"loss": 0.0017,
"step": 7000
},
{
"epoch": 7.31,
"eval_acc": 0.821339948334144,
"eval_cer": 0.021385402138540215,
"eval_loss": 0.013037490658462048,
"eval_runtime": 122.7957,
"eval_samples_per_second": 3.282,
"eval_steps_per_second": 0.415,
"step": 7000
}
],
"max_steps": 9570,
"num_train_epochs": 10,
"total_flos": 4.3872550778956677e+18,
"trial_name": null,
"trial_params": null
}