wav2vec-fame-frisian / trainer_state.json
techsword's picture
Initial commit
d82e0ce
{
"best_metric": 0.31727737305643144,
"best_model_checkpoint": "./save/wietsedv/wav2vec2-large-xlsr-53-frisian/checkpoint-7086",
"epoch": 11.0,
"global_step": 12991,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.999703640982218e-05,
"loss": 8.2316,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 4.9992802709568166e-05,
"loss": 3.9359,
"step": 20
},
{
"epoch": 0.03,
"learning_rate": 4.998856900931414e-05,
"loss": 3.571,
"step": 30
},
{
"epoch": 0.03,
"learning_rate": 4.998433530906012e-05,
"loss": 3.3783,
"step": 40
},
{
"epoch": 0.04,
"learning_rate": 4.99801016088061e-05,
"loss": 3.2272,
"step": 50
},
{
"epoch": 0.05,
"learning_rate": 4.9975867908552076e-05,
"loss": 3.1061,
"step": 60
},
{
"epoch": 0.06,
"learning_rate": 4.9971634208298054e-05,
"loss": 3.0072,
"step": 70
},
{
"epoch": 0.07,
"learning_rate": 4.996740050804403e-05,
"loss": 2.8691,
"step": 80
},
{
"epoch": 0.08,
"learning_rate": 4.9963166807790016e-05,
"loss": 2.8989,
"step": 90
},
{
"epoch": 0.08,
"learning_rate": 4.995893310753599e-05,
"loss": 2.8434,
"step": 100
},
{
"epoch": 0.09,
"learning_rate": 4.995469940728197e-05,
"loss": 2.7761,
"step": 110
},
{
"epoch": 0.1,
"learning_rate": 4.995046570702794e-05,
"loss": 2.689,
"step": 120
},
{
"epoch": 0.11,
"learning_rate": 4.9946232006773926e-05,
"loss": 2.6679,
"step": 130
},
{
"epoch": 0.12,
"learning_rate": 4.99419983065199e-05,
"loss": 2.5585,
"step": 140
},
{
"epoch": 0.13,
"learning_rate": 4.993776460626588e-05,
"loss": 2.5666,
"step": 150
},
{
"epoch": 0.14,
"learning_rate": 4.993353090601185e-05,
"loss": 2.359,
"step": 160
},
{
"epoch": 0.14,
"learning_rate": 4.992929720575784e-05,
"loss": 2.2344,
"step": 170
},
{
"epoch": 0.15,
"learning_rate": 4.992506350550381e-05,
"loss": 2.107,
"step": 180
},
{
"epoch": 0.16,
"learning_rate": 4.992082980524979e-05,
"loss": 1.9212,
"step": 190
},
{
"epoch": 0.17,
"learning_rate": 4.991659610499577e-05,
"loss": 1.8697,
"step": 200
},
{
"epoch": 0.18,
"learning_rate": 4.991236240474175e-05,
"loss": 1.6646,
"step": 210
},
{
"epoch": 0.19,
"learning_rate": 4.9908128704487725e-05,
"loss": 1.5922,
"step": 220
},
{
"epoch": 0.19,
"learning_rate": 4.99038950042337e-05,
"loss": 1.6642,
"step": 230
},
{
"epoch": 0.2,
"learning_rate": 4.989966130397968e-05,
"loss": 1.5205,
"step": 240
},
{
"epoch": 0.21,
"learning_rate": 4.989542760372566e-05,
"loss": 1.6621,
"step": 250
},
{
"epoch": 0.22,
"learning_rate": 4.989119390347164e-05,
"loss": 1.3049,
"step": 260
},
{
"epoch": 0.23,
"learning_rate": 4.988696020321761e-05,
"loss": 1.2634,
"step": 270
},
{
"epoch": 0.24,
"learning_rate": 4.98827265029636e-05,
"loss": 1.2498,
"step": 280
},
{
"epoch": 0.25,
"learning_rate": 4.987849280270957e-05,
"loss": 1.3302,
"step": 290
},
{
"epoch": 0.25,
"learning_rate": 4.987425910245555e-05,
"loss": 1.3514,
"step": 300
},
{
"epoch": 0.26,
"learning_rate": 4.9870025402201523e-05,
"loss": 1.1561,
"step": 310
},
{
"epoch": 0.27,
"learning_rate": 4.986579170194751e-05,
"loss": 1.0196,
"step": 320
},
{
"epoch": 0.28,
"learning_rate": 4.986155800169348e-05,
"loss": 1.1086,
"step": 330
},
{
"epoch": 0.29,
"learning_rate": 4.985732430143946e-05,
"loss": 1.1578,
"step": 340
},
{
"epoch": 0.3,
"learning_rate": 4.9853090601185434e-05,
"loss": 1.2532,
"step": 350
},
{
"epoch": 0.3,
"learning_rate": 4.984885690093142e-05,
"loss": 0.9676,
"step": 360
},
{
"epoch": 0.31,
"learning_rate": 4.9844623200677396e-05,
"loss": 1.0227,
"step": 370
},
{
"epoch": 0.32,
"learning_rate": 4.9840389500423373e-05,
"loss": 1.0093,
"step": 380
},
{
"epoch": 0.33,
"learning_rate": 4.983615580016935e-05,
"loss": 1.0484,
"step": 390
},
{
"epoch": 0.34,
"learning_rate": 4.983192209991533e-05,
"loss": 1.1028,
"step": 400
},
{
"epoch": 0.35,
"learning_rate": 4.9827688399661306e-05,
"loss": 0.9524,
"step": 410
},
{
"epoch": 0.36,
"learning_rate": 4.9823454699407284e-05,
"loss": 0.9216,
"step": 420
},
{
"epoch": 0.36,
"learning_rate": 4.981922099915326e-05,
"loss": 0.8981,
"step": 430
},
{
"epoch": 0.37,
"learning_rate": 4.981498729889924e-05,
"loss": 0.9876,
"step": 440
},
{
"epoch": 0.38,
"learning_rate": 4.981075359864522e-05,
"loss": 1.1131,
"step": 450
},
{
"epoch": 0.39,
"learning_rate": 4.9806519898391194e-05,
"loss": 0.8188,
"step": 460
},
{
"epoch": 0.4,
"learning_rate": 4.980228619813718e-05,
"loss": 0.8327,
"step": 470
},
{
"epoch": 0.41,
"learning_rate": 4.979805249788315e-05,
"loss": 0.9766,
"step": 480
},
{
"epoch": 0.41,
"learning_rate": 4.9793818797629134e-05,
"loss": 0.9445,
"step": 490
},
{
"epoch": 0.42,
"learning_rate": 4.9789585097375105e-05,
"loss": 1.0943,
"step": 500
},
{
"epoch": 0.43,
"learning_rate": 4.978535139712109e-05,
"loss": 0.8088,
"step": 510
},
{
"epoch": 0.44,
"learning_rate": 4.978111769686706e-05,
"loss": 0.8138,
"step": 520
},
{
"epoch": 0.45,
"learning_rate": 4.9776883996613045e-05,
"loss": 0.8536,
"step": 530
},
{
"epoch": 0.46,
"learning_rate": 4.977265029635902e-05,
"loss": 0.9564,
"step": 540
},
{
"epoch": 0.47,
"learning_rate": 4.97688399661304e-05,
"loss": 0.9857,
"step": 550
},
{
"epoch": 0.47,
"learning_rate": 4.976460626587638e-05,
"loss": 0.7244,
"step": 560
},
{
"epoch": 0.48,
"learning_rate": 4.976037256562235e-05,
"loss": 0.8323,
"step": 570
},
{
"epoch": 0.49,
"learning_rate": 4.975613886536834e-05,
"loss": 0.8926,
"step": 580
},
{
"epoch": 0.5,
"learning_rate": 4.975190516511431e-05,
"loss": 0.855,
"step": 590
},
{
"epoch": 0.51,
"learning_rate": 4.974767146486029e-05,
"loss": 1.063,
"step": 600
},
{
"epoch": 0.52,
"learning_rate": 4.974343776460627e-05,
"loss": 0.6973,
"step": 610
},
{
"epoch": 0.52,
"learning_rate": 4.973920406435225e-05,
"loss": 0.8564,
"step": 620
},
{
"epoch": 0.53,
"learning_rate": 4.9734970364098226e-05,
"loss": 0.8261,
"step": 630
},
{
"epoch": 0.54,
"learning_rate": 4.97307366638442e-05,
"loss": 0.8945,
"step": 640
},
{
"epoch": 0.55,
"learning_rate": 4.972650296359018e-05,
"loss": 0.9791,
"step": 650
},
{
"epoch": 0.56,
"learning_rate": 4.972226926333616e-05,
"loss": 0.7355,
"step": 660
},
{
"epoch": 0.57,
"learning_rate": 4.9718035563082136e-05,
"loss": 0.7749,
"step": 670
},
{
"epoch": 0.58,
"learning_rate": 4.9713801862828114e-05,
"loss": 0.8765,
"step": 680
},
{
"epoch": 0.58,
"learning_rate": 4.970956816257409e-05,
"loss": 0.914,
"step": 690
},
{
"epoch": 0.59,
"learning_rate": 4.970533446232007e-05,
"loss": 1.0133,
"step": 700
},
{
"epoch": 0.6,
"learning_rate": 4.970110076206605e-05,
"loss": 0.6857,
"step": 710
},
{
"epoch": 0.61,
"learning_rate": 4.9696867061812024e-05,
"loss": 0.6903,
"step": 720
},
{
"epoch": 0.62,
"learning_rate": 4.969263336155801e-05,
"loss": 0.7509,
"step": 730
},
{
"epoch": 0.63,
"learning_rate": 4.968839966130398e-05,
"loss": 0.8627,
"step": 740
},
{
"epoch": 0.64,
"learning_rate": 4.9684165961049964e-05,
"loss": 1.0669,
"step": 750
},
{
"epoch": 0.64,
"learning_rate": 4.9679932260795935e-05,
"loss": 0.6301,
"step": 760
},
{
"epoch": 0.65,
"learning_rate": 4.967569856054192e-05,
"loss": 0.7653,
"step": 770
},
{
"epoch": 0.66,
"learning_rate": 4.967146486028789e-05,
"loss": 0.7241,
"step": 780
},
{
"epoch": 0.67,
"learning_rate": 4.9667231160033874e-05,
"loss": 0.8716,
"step": 790
},
{
"epoch": 0.68,
"learning_rate": 4.9662997459779845e-05,
"loss": 0.9682,
"step": 800
},
{
"epoch": 0.69,
"learning_rate": 4.965876375952583e-05,
"loss": 0.6231,
"step": 810
},
{
"epoch": 0.69,
"learning_rate": 4.96545300592718e-05,
"loss": 0.7077,
"step": 820
},
{
"epoch": 0.7,
"learning_rate": 4.9650296359017785e-05,
"loss": 0.6977,
"step": 830
},
{
"epoch": 0.71,
"learning_rate": 4.964606265876376e-05,
"loss": 0.77,
"step": 840
},
{
"epoch": 0.72,
"learning_rate": 4.9642252328535145e-05,
"loss": 0.8742,
"step": 850
},
{
"epoch": 0.73,
"learning_rate": 4.9638018628281116e-05,
"loss": 0.725,
"step": 860
},
{
"epoch": 0.74,
"learning_rate": 4.96337849280271e-05,
"loss": 0.7597,
"step": 870
},
{
"epoch": 0.75,
"learning_rate": 4.962955122777308e-05,
"loss": 0.7368,
"step": 880
},
{
"epoch": 0.75,
"learning_rate": 4.9625317527519055e-05,
"loss": 0.8343,
"step": 890
},
{
"epoch": 0.76,
"learning_rate": 4.962108382726503e-05,
"loss": 1.0221,
"step": 900
},
{
"epoch": 0.77,
"learning_rate": 4.961685012701101e-05,
"loss": 0.6496,
"step": 910
},
{
"epoch": 0.78,
"learning_rate": 4.961261642675699e-05,
"loss": 0.7248,
"step": 920
},
{
"epoch": 0.79,
"learning_rate": 4.9608382726502966e-05,
"loss": 0.7301,
"step": 930
},
{
"epoch": 0.8,
"learning_rate": 4.9604149026248944e-05,
"loss": 0.8114,
"step": 940
},
{
"epoch": 0.8,
"learning_rate": 4.959991532599492e-05,
"loss": 0.8109,
"step": 950
},
{
"epoch": 0.81,
"learning_rate": 4.95956816257409e-05,
"loss": 0.6532,
"step": 960
},
{
"epoch": 0.82,
"learning_rate": 4.9591447925486876e-05,
"loss": 0.7072,
"step": 970
},
{
"epoch": 0.83,
"learning_rate": 4.958721422523286e-05,
"loss": 0.7874,
"step": 980
},
{
"epoch": 0.84,
"learning_rate": 4.958298052497883e-05,
"loss": 0.7866,
"step": 990
},
{
"epoch": 0.85,
"learning_rate": 4.9578746824724816e-05,
"loss": 1.0286,
"step": 1000
},
{
"epoch": 0.86,
"learning_rate": 4.957451312447079e-05,
"loss": 0.6527,
"step": 1010
},
{
"epoch": 0.86,
"learning_rate": 4.957027942421677e-05,
"loss": 0.697,
"step": 1020
},
{
"epoch": 0.87,
"learning_rate": 4.956604572396274e-05,
"loss": 0.7519,
"step": 1030
},
{
"epoch": 0.88,
"learning_rate": 4.9561812023708727e-05,
"loss": 0.7881,
"step": 1040
},
{
"epoch": 0.89,
"learning_rate": 4.9557578323454704e-05,
"loss": 1.1226,
"step": 1050
},
{
"epoch": 0.9,
"learning_rate": 4.955334462320068e-05,
"loss": 0.6177,
"step": 1060
},
{
"epoch": 0.91,
"learning_rate": 4.954911092294666e-05,
"loss": 0.6793,
"step": 1070
},
{
"epoch": 0.91,
"learning_rate": 4.954487722269264e-05,
"loss": 0.7337,
"step": 1080
},
{
"epoch": 0.92,
"learning_rate": 4.9540643522438615e-05,
"loss": 0.7287,
"step": 1090
},
{
"epoch": 0.93,
"learning_rate": 4.953640982218459e-05,
"loss": 0.805,
"step": 1100
},
{
"epoch": 0.94,
"learning_rate": 4.953217612193057e-05,
"loss": 0.6376,
"step": 1110
},
{
"epoch": 0.95,
"learning_rate": 4.952794242167655e-05,
"loss": 0.6818,
"step": 1120
},
{
"epoch": 0.96,
"learning_rate": 4.9523708721422525e-05,
"loss": 0.7406,
"step": 1130
},
{
"epoch": 0.97,
"learning_rate": 4.95194750211685e-05,
"loss": 0.7519,
"step": 1140
},
{
"epoch": 0.97,
"learning_rate": 4.951524132091448e-05,
"loss": 0.9363,
"step": 1150
},
{
"epoch": 0.98,
"learning_rate": 4.951100762066046e-05,
"loss": 0.6559,
"step": 1160
},
{
"epoch": 0.99,
"learning_rate": 4.9506773920406436e-05,
"loss": 0.6731,
"step": 1170
},
{
"epoch": 1.0,
"learning_rate": 4.950254022015241e-05,
"loss": 0.9067,
"step": 1180
},
{
"epoch": 1.0,
"eval_cer": 0.1576126742082073,
"eval_loss": 0.6544902324676514,
"eval_mer": 0.4400347939545504,
"eval_runtime": 29.5295,
"eval_samples_per_second": 73.757,
"eval_steps_per_second": 4.639,
"step": 1181
},
{
"epoch": 1.01,
"learning_rate": 4.949830651989839e-05,
"loss": 0.5277,
"step": 1190
},
{
"epoch": 1.02,
"learning_rate": 4.949407281964437e-05,
"loss": 0.5804,
"step": 1200
},
{
"epoch": 1.02,
"learning_rate": 4.948983911939035e-05,
"loss": 0.6302,
"step": 1210
},
{
"epoch": 1.03,
"learning_rate": 4.948560541913633e-05,
"loss": 0.6828,
"step": 1220
},
{
"epoch": 1.04,
"learning_rate": 4.948137171888231e-05,
"loss": 0.8391,
"step": 1230
},
{
"epoch": 1.05,
"learning_rate": 4.9477138018628286e-05,
"loss": 0.6364,
"step": 1240
},
{
"epoch": 1.06,
"learning_rate": 4.947290431837426e-05,
"loss": 0.5684,
"step": 1250
},
{
"epoch": 1.07,
"learning_rate": 4.946867061812024e-05,
"loss": 0.6349,
"step": 1260
},
{
"epoch": 1.08,
"learning_rate": 4.946443691786622e-05,
"loss": 0.7817,
"step": 1270
},
{
"epoch": 1.08,
"learning_rate": 4.94606265876376e-05,
"loss": 0.8709,
"step": 1280
},
{
"epoch": 1.09,
"learning_rate": 4.945639288738357e-05,
"loss": 0.5876,
"step": 1290
},
{
"epoch": 1.1,
"learning_rate": 4.9452159187129556e-05,
"loss": 0.5962,
"step": 1300
},
{
"epoch": 1.11,
"learning_rate": 4.944792548687553e-05,
"loss": 0.6397,
"step": 1310
},
{
"epoch": 1.12,
"learning_rate": 4.944369178662151e-05,
"loss": 0.6253,
"step": 1320
},
{
"epoch": 1.13,
"learning_rate": 4.943945808636748e-05,
"loss": 0.7778,
"step": 1330
},
{
"epoch": 1.13,
"learning_rate": 4.943522438611347e-05,
"loss": 0.5529,
"step": 1340
},
{
"epoch": 1.14,
"learning_rate": 4.9430990685859444e-05,
"loss": 0.5595,
"step": 1350
},
{
"epoch": 1.15,
"learning_rate": 4.942675698560542e-05,
"loss": 0.6319,
"step": 1360
},
{
"epoch": 1.16,
"learning_rate": 4.94225232853514e-05,
"loss": 0.6649,
"step": 1370
},
{
"epoch": 1.17,
"learning_rate": 4.941828958509738e-05,
"loss": 0.8097,
"step": 1380
},
{
"epoch": 1.18,
"learning_rate": 4.9414055884843355e-05,
"loss": 0.6178,
"step": 1390
},
{
"epoch": 1.19,
"learning_rate": 4.940982218458933e-05,
"loss": 0.5642,
"step": 1400
},
{
"epoch": 1.19,
"learning_rate": 4.940558848433531e-05,
"loss": 0.5683,
"step": 1410
},
{
"epoch": 1.2,
"learning_rate": 4.940135478408129e-05,
"loss": 0.5921,
"step": 1420
},
{
"epoch": 1.21,
"learning_rate": 4.939712108382727e-05,
"loss": 0.757,
"step": 1430
},
{
"epoch": 1.22,
"learning_rate": 4.939288738357324e-05,
"loss": 0.6185,
"step": 1440
},
{
"epoch": 1.23,
"learning_rate": 4.938865368331923e-05,
"loss": 0.5439,
"step": 1450
},
{
"epoch": 1.24,
"learning_rate": 4.93844199830652e-05,
"loss": 0.6081,
"step": 1460
},
{
"epoch": 1.24,
"learning_rate": 4.938018628281118e-05,
"loss": 0.6448,
"step": 1470
},
{
"epoch": 1.25,
"learning_rate": 4.9375952582557153e-05,
"loss": 0.7704,
"step": 1480
},
{
"epoch": 1.26,
"learning_rate": 4.937171888230314e-05,
"loss": 0.5645,
"step": 1490
},
{
"epoch": 1.27,
"learning_rate": 4.936748518204911e-05,
"loss": 0.5591,
"step": 1500
},
{
"epoch": 1.28,
"learning_rate": 4.936325148179509e-05,
"loss": 0.6333,
"step": 1510
},
{
"epoch": 1.29,
"learning_rate": 4.9359017781541064e-05,
"loss": 0.6027,
"step": 1520
},
{
"epoch": 1.3,
"learning_rate": 4.935478408128705e-05,
"loss": 0.8533,
"step": 1530
},
{
"epoch": 1.3,
"learning_rate": 4.9350550381033026e-05,
"loss": 0.5244,
"step": 1540
},
{
"epoch": 1.31,
"learning_rate": 4.9346316680779004e-05,
"loss": 0.5265,
"step": 1550
},
{
"epoch": 1.32,
"learning_rate": 4.934208298052498e-05,
"loss": 0.605,
"step": 1560
},
{
"epoch": 1.33,
"learning_rate": 4.933784928027096e-05,
"loss": 0.6919,
"step": 1570
},
{
"epoch": 1.34,
"learning_rate": 4.9333615580016936e-05,
"loss": 0.8943,
"step": 1580
},
{
"epoch": 1.35,
"learning_rate": 4.9329381879762914e-05,
"loss": 0.5891,
"step": 1590
},
{
"epoch": 1.35,
"learning_rate": 4.93251481795089e-05,
"loss": 0.5561,
"step": 1600
},
{
"epoch": 1.36,
"learning_rate": 4.932091447925487e-05,
"loss": 0.5738,
"step": 1610
},
{
"epoch": 1.37,
"learning_rate": 4.9316680779000854e-05,
"loss": 0.7086,
"step": 1620
},
{
"epoch": 1.38,
"learning_rate": 4.9312447078746824e-05,
"loss": 0.8479,
"step": 1630
},
{
"epoch": 1.39,
"learning_rate": 4.930821337849281e-05,
"loss": 0.5043,
"step": 1640
},
{
"epoch": 1.4,
"learning_rate": 4.930397967823878e-05,
"loss": 0.5307,
"step": 1650
},
{
"epoch": 1.41,
"learning_rate": 4.9299745977984764e-05,
"loss": 0.5975,
"step": 1660
},
{
"epoch": 1.41,
"learning_rate": 4.9295512277730735e-05,
"loss": 0.6344,
"step": 1670
},
{
"epoch": 1.42,
"learning_rate": 4.929127857747672e-05,
"loss": 0.7336,
"step": 1680
},
{
"epoch": 1.43,
"learning_rate": 4.928704487722269e-05,
"loss": 0.5094,
"step": 1690
},
{
"epoch": 1.44,
"learning_rate": 4.9282811176968675e-05,
"loss": 0.5233,
"step": 1700
},
{
"epoch": 1.45,
"learning_rate": 4.927857747671465e-05,
"loss": 0.5785,
"step": 1710
},
{
"epoch": 1.46,
"learning_rate": 4.927434377646063e-05,
"loss": 0.6206,
"step": 1720
},
{
"epoch": 1.46,
"learning_rate": 4.927011007620661e-05,
"loss": 0.7518,
"step": 1730
},
{
"epoch": 1.47,
"learning_rate": 4.9265876375952585e-05,
"loss": 0.498,
"step": 1740
},
{
"epoch": 1.48,
"learning_rate": 4.926164267569856e-05,
"loss": 0.5847,
"step": 1750
},
{
"epoch": 1.49,
"learning_rate": 4.925740897544454e-05,
"loss": 0.5772,
"step": 1760
},
{
"epoch": 1.5,
"learning_rate": 4.925317527519052e-05,
"loss": 0.5796,
"step": 1770
},
{
"epoch": 1.51,
"learning_rate": 4.9248941574936496e-05,
"loss": 0.6816,
"step": 1780
},
{
"epoch": 1.52,
"learning_rate": 4.924470787468248e-05,
"loss": 0.6114,
"step": 1790
},
{
"epoch": 1.52,
"learning_rate": 4.924047417442845e-05,
"loss": 0.5652,
"step": 1800
},
{
"epoch": 1.53,
"learning_rate": 4.9236240474174435e-05,
"loss": 0.5561,
"step": 1810
},
{
"epoch": 1.54,
"learning_rate": 4.9232006773920406e-05,
"loss": 0.6507,
"step": 1820
},
{
"epoch": 1.55,
"learning_rate": 4.922777307366639e-05,
"loss": 0.8466,
"step": 1830
},
{
"epoch": 1.56,
"learning_rate": 4.922353937341236e-05,
"loss": 0.5228,
"step": 1840
},
{
"epoch": 1.57,
"learning_rate": 4.9219305673158346e-05,
"loss": 0.5455,
"step": 1850
},
{
"epoch": 1.57,
"learning_rate": 4.9215071972904316e-05,
"loss": 0.5846,
"step": 1860
},
{
"epoch": 1.58,
"learning_rate": 4.92108382726503e-05,
"loss": 0.533,
"step": 1870
},
{
"epoch": 1.59,
"learning_rate": 4.920660457239628e-05,
"loss": 0.7392,
"step": 1880
},
{
"epoch": 1.6,
"learning_rate": 4.9202370872142256e-05,
"loss": 0.4746,
"step": 1890
},
{
"epoch": 1.61,
"learning_rate": 4.9198137171888234e-05,
"loss": 0.5498,
"step": 1900
},
{
"epoch": 1.62,
"learning_rate": 4.919390347163421e-05,
"loss": 0.5825,
"step": 1910
},
{
"epoch": 1.63,
"learning_rate": 4.918966977138019e-05,
"loss": 0.6736,
"step": 1920
},
{
"epoch": 1.63,
"learning_rate": 4.9185436071126167e-05,
"loss": 0.8107,
"step": 1930
},
{
"epoch": 1.64,
"learning_rate": 4.9181202370872144e-05,
"loss": 0.4814,
"step": 1940
},
{
"epoch": 1.65,
"learning_rate": 4.917696867061812e-05,
"loss": 0.4699,
"step": 1950
},
{
"epoch": 1.66,
"learning_rate": 4.91727349703641e-05,
"loss": 0.5519,
"step": 1960
},
{
"epoch": 1.67,
"learning_rate": 4.916850127011008e-05,
"loss": 0.6432,
"step": 1970
},
{
"epoch": 1.68,
"learning_rate": 4.9164267569856055e-05,
"loss": 0.7772,
"step": 1980
},
{
"epoch": 1.69,
"learning_rate": 4.916003386960203e-05,
"loss": 0.4573,
"step": 1990
},
{
"epoch": 1.69,
"learning_rate": 4.915580016934802e-05,
"loss": 0.5186,
"step": 2000
},
{
"epoch": 1.7,
"learning_rate": 4.915156646909399e-05,
"loss": 0.5835,
"step": 2010
},
{
"epoch": 1.71,
"learning_rate": 4.914733276883997e-05,
"loss": 0.6779,
"step": 2020
},
{
"epoch": 1.72,
"learning_rate": 4.914309906858594e-05,
"loss": 0.6225,
"step": 2030
},
{
"epoch": 1.73,
"learning_rate": 4.9139288738357325e-05,
"loss": 0.3982,
"step": 2040
},
{
"epoch": 1.74,
"learning_rate": 4.91350550381033e-05,
"loss": 0.4857,
"step": 2050
},
{
"epoch": 1.74,
"learning_rate": 4.913082133784928e-05,
"loss": 0.6068,
"step": 2060
},
{
"epoch": 1.75,
"learning_rate": 4.9126587637595265e-05,
"loss": 0.5961,
"step": 2070
},
{
"epoch": 1.76,
"learning_rate": 4.9122353937341236e-05,
"loss": 0.767,
"step": 2080
},
{
"epoch": 1.77,
"learning_rate": 4.911812023708722e-05,
"loss": 0.504,
"step": 2090
},
{
"epoch": 1.78,
"learning_rate": 4.911388653683319e-05,
"loss": 0.5235,
"step": 2100
},
{
"epoch": 1.79,
"learning_rate": 4.9109652836579175e-05,
"loss": 0.5895,
"step": 2110
},
{
"epoch": 1.8,
"learning_rate": 4.9105419136325146e-05,
"loss": 0.6355,
"step": 2120
},
{
"epoch": 1.8,
"learning_rate": 4.910118543607113e-05,
"loss": 0.7225,
"step": 2130
},
{
"epoch": 1.81,
"learning_rate": 4.90969517358171e-05,
"loss": 0.5133,
"step": 2140
},
{
"epoch": 1.82,
"learning_rate": 4.9092718035563086e-05,
"loss": 0.4604,
"step": 2150
},
{
"epoch": 1.83,
"learning_rate": 4.9088484335309063e-05,
"loss": 0.5613,
"step": 2160
},
{
"epoch": 1.84,
"learning_rate": 4.908425063505504e-05,
"loss": 0.6061,
"step": 2170
},
{
"epoch": 1.85,
"learning_rate": 4.908001693480102e-05,
"loss": 0.693,
"step": 2180
},
{
"epoch": 1.85,
"learning_rate": 4.9075783234546996e-05,
"loss": 0.4893,
"step": 2190
},
{
"epoch": 1.86,
"learning_rate": 4.9071549534292974e-05,
"loss": 0.4893,
"step": 2200
},
{
"epoch": 1.87,
"learning_rate": 4.906731583403895e-05,
"loss": 0.5351,
"step": 2210
},
{
"epoch": 1.88,
"learning_rate": 4.906308213378493e-05,
"loss": 0.5383,
"step": 2220
},
{
"epoch": 1.89,
"learning_rate": 4.905884843353091e-05,
"loss": 0.7868,
"step": 2230
},
{
"epoch": 1.9,
"learning_rate": 4.905461473327689e-05,
"loss": 0.4928,
"step": 2240
},
{
"epoch": 1.91,
"learning_rate": 4.905038103302286e-05,
"loss": 0.5192,
"step": 2250
},
{
"epoch": 1.91,
"learning_rate": 4.9046147332768846e-05,
"loss": 0.5743,
"step": 2260
},
{
"epoch": 1.92,
"learning_rate": 4.904191363251482e-05,
"loss": 0.595,
"step": 2270
},
{
"epoch": 1.93,
"learning_rate": 4.90376799322608e-05,
"loss": 0.7781,
"step": 2280
},
{
"epoch": 1.94,
"learning_rate": 4.903344623200677e-05,
"loss": 0.437,
"step": 2290
},
{
"epoch": 1.95,
"learning_rate": 4.902921253175276e-05,
"loss": 0.4474,
"step": 2300
},
{
"epoch": 1.96,
"learning_rate": 4.902497883149873e-05,
"loss": 0.5298,
"step": 2310
},
{
"epoch": 1.96,
"learning_rate": 4.902074513124471e-05,
"loss": 0.628,
"step": 2320
},
{
"epoch": 1.97,
"learning_rate": 4.901651143099068e-05,
"loss": 0.5911,
"step": 2330
},
{
"epoch": 1.98,
"learning_rate": 4.901227773073667e-05,
"loss": 0.4738,
"step": 2340
},
{
"epoch": 1.99,
"learning_rate": 4.900804403048264e-05,
"loss": 0.5894,
"step": 2350
},
{
"epoch": 2.0,
"learning_rate": 4.900381033022862e-05,
"loss": 0.5701,
"step": 2360
},
{
"epoch": 2.0,
"eval_cer": 0.13196314866227435,
"eval_loss": 0.576825737953186,
"eval_mer": 0.3741437425247363,
"eval_runtime": 24.0365,
"eval_samples_per_second": 90.612,
"eval_steps_per_second": 5.7,
"step": 2362
},
{
"epoch": 2.01,
"learning_rate": 4.89995766299746e-05,
"loss": 0.5197,
"step": 2370
},
{
"epoch": 2.02,
"learning_rate": 4.899534292972058e-05,
"loss": 0.4524,
"step": 2380
},
{
"epoch": 2.02,
"learning_rate": 4.8991109229466555e-05,
"loss": 0.4972,
"step": 2390
},
{
"epoch": 2.03,
"learning_rate": 4.898687552921253e-05,
"loss": 0.4893,
"step": 2400
},
{
"epoch": 2.04,
"learning_rate": 4.898264182895852e-05,
"loss": 0.5848,
"step": 2410
},
{
"epoch": 2.05,
"learning_rate": 4.897840812870449e-05,
"loss": 0.5342,
"step": 2420
},
{
"epoch": 2.06,
"learning_rate": 4.897417442845047e-05,
"loss": 0.4377,
"step": 2430
},
{
"epoch": 2.07,
"learning_rate": 4.8969940728196444e-05,
"loss": 0.4364,
"step": 2440
},
{
"epoch": 2.07,
"learning_rate": 4.896570702794243e-05,
"loss": 0.4539,
"step": 2450
},
{
"epoch": 2.08,
"learning_rate": 4.89614733276884e-05,
"loss": 0.6036,
"step": 2460
},
{
"epoch": 2.09,
"learning_rate": 4.895723962743438e-05,
"loss": 0.5118,
"step": 2470
},
{
"epoch": 2.1,
"learning_rate": 4.8953005927180354e-05,
"loss": 0.4158,
"step": 2480
},
{
"epoch": 2.11,
"learning_rate": 4.894877222692634e-05,
"loss": 0.511,
"step": 2490
},
{
"epoch": 2.12,
"learning_rate": 4.894453852667231e-05,
"loss": 0.4624,
"step": 2500
},
{
"epoch": 2.13,
"learning_rate": 4.8940304826418294e-05,
"loss": 0.6149,
"step": 2510
},
{
"epoch": 2.13,
"learning_rate": 4.893607112616427e-05,
"loss": 0.3919,
"step": 2520
},
{
"epoch": 2.14,
"learning_rate": 4.893183742591025e-05,
"loss": 0.3736,
"step": 2530
},
{
"epoch": 2.15,
"learning_rate": 4.8927603725656227e-05,
"loss": 0.4507,
"step": 2540
},
{
"epoch": 2.16,
"learning_rate": 4.8923370025402204e-05,
"loss": 0.4847,
"step": 2550
},
{
"epoch": 2.17,
"learning_rate": 4.891913632514818e-05,
"loss": 0.6876,
"step": 2560
},
{
"epoch": 2.18,
"learning_rate": 4.891490262489416e-05,
"loss": 0.5018,
"step": 2570
},
{
"epoch": 2.18,
"learning_rate": 4.891066892464014e-05,
"loss": 0.4007,
"step": 2580
},
{
"epoch": 2.19,
"learning_rate": 4.8906435224386115e-05,
"loss": 0.4385,
"step": 2590
},
{
"epoch": 2.2,
"learning_rate": 4.89022015241321e-05,
"loss": 0.5196,
"step": 2600
},
{
"epoch": 2.21,
"learning_rate": 4.889796782387807e-05,
"loss": 0.5398,
"step": 2610
},
{
"epoch": 2.22,
"learning_rate": 4.8893734123624054e-05,
"loss": 0.4602,
"step": 2620
},
{
"epoch": 2.23,
"learning_rate": 4.8889500423370025e-05,
"loss": 0.3691,
"step": 2630
},
{
"epoch": 2.24,
"learning_rate": 4.888526672311601e-05,
"loss": 0.4319,
"step": 2640
},
{
"epoch": 2.24,
"learning_rate": 4.888103302286198e-05,
"loss": 0.5262,
"step": 2650
},
{
"epoch": 2.25,
"learning_rate": 4.8876799322607965e-05,
"loss": 0.6359,
"step": 2660
},
{
"epoch": 2.26,
"learning_rate": 4.8872565622353936e-05,
"loss": 0.4502,
"step": 2670
},
{
"epoch": 2.27,
"learning_rate": 4.886833192209992e-05,
"loss": 0.4397,
"step": 2680
},
{
"epoch": 2.28,
"learning_rate": 4.88640982218459e-05,
"loss": 0.4362,
"step": 2690
},
{
"epoch": 2.29,
"learning_rate": 4.8859864521591875e-05,
"loss": 0.4408,
"step": 2700
},
{
"epoch": 2.29,
"learning_rate": 4.885563082133785e-05,
"loss": 0.5903,
"step": 2710
},
{
"epoch": 2.3,
"learning_rate": 4.885139712108383e-05,
"loss": 0.5097,
"step": 2720
},
{
"epoch": 2.31,
"learning_rate": 4.884716342082981e-05,
"loss": 0.4245,
"step": 2730
},
{
"epoch": 2.32,
"learning_rate": 4.8842929720575786e-05,
"loss": 0.496,
"step": 2740
},
{
"epoch": 2.33,
"learning_rate": 4.883869602032176e-05,
"loss": 0.5596,
"step": 2750
},
{
"epoch": 2.34,
"learning_rate": 4.883446232006774e-05,
"loss": 0.5434,
"step": 2760
},
{
"epoch": 2.35,
"learning_rate": 4.883022861981372e-05,
"loss": 0.4734,
"step": 2770
},
{
"epoch": 2.35,
"learning_rate": 4.8825994919559696e-05,
"loss": 0.4264,
"step": 2780
},
{
"epoch": 2.36,
"learning_rate": 4.8821761219305674e-05,
"loss": 0.4084,
"step": 2790
},
{
"epoch": 2.37,
"learning_rate": 4.881752751905165e-05,
"loss": 0.4768,
"step": 2800
},
{
"epoch": 2.38,
"learning_rate": 4.8813293818797636e-05,
"loss": 0.566,
"step": 2810
},
{
"epoch": 2.39,
"learning_rate": 4.8809060118543607e-05,
"loss": 0.4692,
"step": 2820
},
{
"epoch": 2.4,
"learning_rate": 4.880482641828959e-05,
"loss": 0.428,
"step": 2830
},
{
"epoch": 2.4,
"learning_rate": 4.880059271803556e-05,
"loss": 0.4359,
"step": 2840
},
{
"epoch": 2.41,
"learning_rate": 4.8796359017781546e-05,
"loss": 0.5232,
"step": 2850
},
{
"epoch": 2.42,
"learning_rate": 4.8792125317527524e-05,
"loss": 0.6228,
"step": 2860
},
{
"epoch": 2.43,
"learning_rate": 4.87883149872989e-05,
"loss": 0.3815,
"step": 2870
},
{
"epoch": 2.44,
"learning_rate": 4.878408128704488e-05,
"loss": 0.4012,
"step": 2880
},
{
"epoch": 2.45,
"learning_rate": 4.8779847586790855e-05,
"loss": 0.3818,
"step": 2890
},
{
"epoch": 2.46,
"learning_rate": 4.877561388653684e-05,
"loss": 0.456,
"step": 2900
},
{
"epoch": 2.46,
"learning_rate": 4.877138018628281e-05,
"loss": 0.4737,
"step": 2910
},
{
"epoch": 2.47,
"learning_rate": 4.8767146486028795e-05,
"loss": 0.4481,
"step": 2920
},
{
"epoch": 2.48,
"learning_rate": 4.8762912785774765e-05,
"loss": 0.4737,
"step": 2930
},
{
"epoch": 2.49,
"learning_rate": 4.875867908552075e-05,
"loss": 0.5637,
"step": 2940
},
{
"epoch": 2.5,
"learning_rate": 4.875444538526673e-05,
"loss": 0.5322,
"step": 2950
},
{
"epoch": 2.51,
"learning_rate": 4.8750211685012705e-05,
"loss": 0.6033,
"step": 2960
},
{
"epoch": 2.51,
"learning_rate": 4.874597798475868e-05,
"loss": 0.4817,
"step": 2970
},
{
"epoch": 2.52,
"learning_rate": 4.874174428450466e-05,
"loss": 0.449,
"step": 2980
},
{
"epoch": 2.53,
"learning_rate": 4.873751058425064e-05,
"loss": 0.4926,
"step": 2990
},
{
"epoch": 2.54,
"learning_rate": 4.8733276883996615e-05,
"loss": 0.5486,
"step": 3000
},
{
"epoch": 2.55,
"learning_rate": 4.872904318374259e-05,
"loss": 0.6407,
"step": 3010
},
{
"epoch": 2.56,
"learning_rate": 4.872480948348857e-05,
"loss": 0.376,
"step": 3020
},
{
"epoch": 2.57,
"learning_rate": 4.872057578323455e-05,
"loss": 0.3779,
"step": 3030
},
{
"epoch": 2.57,
"learning_rate": 4.8716342082980526e-05,
"loss": 0.486,
"step": 3040
},
{
"epoch": 2.58,
"learning_rate": 4.8712108382726504e-05,
"loss": 0.5398,
"step": 3050
},
{
"epoch": 2.59,
"learning_rate": 4.870787468247248e-05,
"loss": 0.5042,
"step": 3060
},
{
"epoch": 2.6,
"learning_rate": 4.8703640982218466e-05,
"loss": 0.4882,
"step": 3070
},
{
"epoch": 2.61,
"learning_rate": 4.8699407281964436e-05,
"loss": 0.4227,
"step": 3080
},
{
"epoch": 2.62,
"learning_rate": 4.869517358171042e-05,
"loss": 0.4296,
"step": 3090
},
{
"epoch": 2.62,
"learning_rate": 4.869093988145639e-05,
"loss": 0.5495,
"step": 3100
},
{
"epoch": 2.63,
"learning_rate": 4.8686706181202376e-05,
"loss": 0.5862,
"step": 3110
},
{
"epoch": 2.64,
"learning_rate": 4.868247248094835e-05,
"loss": 0.4743,
"step": 3120
},
{
"epoch": 2.65,
"learning_rate": 4.867823878069433e-05,
"loss": 0.4568,
"step": 3130
},
{
"epoch": 2.66,
"learning_rate": 4.86740050804403e-05,
"loss": 0.4377,
"step": 3140
},
{
"epoch": 2.67,
"learning_rate": 4.8669771380186286e-05,
"loss": 0.4263,
"step": 3150
},
{
"epoch": 2.68,
"learning_rate": 4.866553767993226e-05,
"loss": 0.5528,
"step": 3160
},
{
"epoch": 2.68,
"learning_rate": 4.866130397967824e-05,
"loss": 0.4947,
"step": 3170
},
{
"epoch": 2.69,
"learning_rate": 4.865707027942422e-05,
"loss": 0.3768,
"step": 3180
},
{
"epoch": 2.7,
"learning_rate": 4.86528365791702e-05,
"loss": 0.4159,
"step": 3190
},
{
"epoch": 2.71,
"learning_rate": 4.8648602878916175e-05,
"loss": 0.464,
"step": 3200
},
{
"epoch": 2.72,
"learning_rate": 4.864436917866215e-05,
"loss": 0.6557,
"step": 3210
},
{
"epoch": 2.73,
"learning_rate": 4.8640558848433535e-05,
"loss": 0.4162,
"step": 3220
},
{
"epoch": 2.73,
"learning_rate": 4.863632514817951e-05,
"loss": 0.3844,
"step": 3230
},
{
"epoch": 2.74,
"learning_rate": 4.863209144792549e-05,
"loss": 0.5124,
"step": 3240
},
{
"epoch": 2.75,
"learning_rate": 4.862785774767147e-05,
"loss": 0.561,
"step": 3250
},
{
"epoch": 2.76,
"learning_rate": 4.8623624047417445e-05,
"loss": 0.5773,
"step": 3260
},
{
"epoch": 2.77,
"learning_rate": 4.861939034716342e-05,
"loss": 0.4254,
"step": 3270
},
{
"epoch": 2.78,
"learning_rate": 4.86151566469094e-05,
"loss": 0.4155,
"step": 3280
},
{
"epoch": 2.79,
"learning_rate": 4.861092294665538e-05,
"loss": 0.4501,
"step": 3290
},
{
"epoch": 2.79,
"learning_rate": 4.8606689246401356e-05,
"loss": 0.4316,
"step": 3300
},
{
"epoch": 2.8,
"learning_rate": 4.860245554614733e-05,
"loss": 0.5491,
"step": 3310
},
{
"epoch": 2.81,
"learning_rate": 4.859822184589332e-05,
"loss": 0.5051,
"step": 3320
},
{
"epoch": 2.82,
"learning_rate": 4.859398814563929e-05,
"loss": 0.3897,
"step": 3330
},
{
"epoch": 2.83,
"learning_rate": 4.858975444538527e-05,
"loss": 0.4812,
"step": 3340
},
{
"epoch": 2.84,
"learning_rate": 4.8585520745131244e-05,
"loss": 0.4686,
"step": 3350
},
{
"epoch": 2.85,
"learning_rate": 4.858128704487723e-05,
"loss": 0.5991,
"step": 3360
},
{
"epoch": 2.85,
"learning_rate": 4.85770533446232e-05,
"loss": 0.4779,
"step": 3370
},
{
"epoch": 2.86,
"learning_rate": 4.8572819644369183e-05,
"loss": 0.3563,
"step": 3380
},
{
"epoch": 2.87,
"learning_rate": 4.856858594411516e-05,
"loss": 0.4934,
"step": 3390
},
{
"epoch": 2.88,
"learning_rate": 4.856435224386114e-05,
"loss": 0.4977,
"step": 3400
},
{
"epoch": 2.89,
"learning_rate": 4.8560118543607116e-05,
"loss": 0.581,
"step": 3410
},
{
"epoch": 2.9,
"learning_rate": 4.8555884843353094e-05,
"loss": 0.4515,
"step": 3420
},
{
"epoch": 2.9,
"learning_rate": 4.855165114309907e-05,
"loss": 0.394,
"step": 3430
},
{
"epoch": 2.91,
"learning_rate": 4.854741744284505e-05,
"loss": 0.4212,
"step": 3440
},
{
"epoch": 2.92,
"learning_rate": 4.854318374259103e-05,
"loss": 0.4282,
"step": 3450
},
{
"epoch": 2.93,
"learning_rate": 4.8538950042337004e-05,
"loss": 0.4853,
"step": 3460
},
{
"epoch": 2.94,
"learning_rate": 4.853471634208298e-05,
"loss": 0.3892,
"step": 3470
},
{
"epoch": 2.95,
"learning_rate": 4.853048264182896e-05,
"loss": 0.3855,
"step": 3480
},
{
"epoch": 2.96,
"learning_rate": 4.852624894157494e-05,
"loss": 0.4771,
"step": 3490
},
{
"epoch": 2.96,
"learning_rate": 4.8522015241320915e-05,
"loss": 0.5244,
"step": 3500
},
{
"epoch": 2.97,
"learning_rate": 4.851778154106689e-05,
"loss": 0.5556,
"step": 3510
},
{
"epoch": 2.98,
"learning_rate": 4.851354784081287e-05,
"loss": 0.3799,
"step": 3520
},
{
"epoch": 2.99,
"learning_rate": 4.850931414055885e-05,
"loss": 0.3802,
"step": 3530
},
{
"epoch": 3.0,
"learning_rate": 4.8505080440304825e-05,
"loss": 0.5161,
"step": 3540
},
{
"epoch": 3.0,
"eval_cer": 0.12360667965027142,
"eval_loss": 0.5676856637001038,
"eval_mer": 0.3453843644666739,
"eval_runtime": 24.0168,
"eval_samples_per_second": 90.686,
"eval_steps_per_second": 5.704,
"step": 3543
},
{
"epoch": 3.01,
"learning_rate": 4.850084674005081e-05,
"loss": 0.3992,
"step": 3550
},
{
"epoch": 3.01,
"learning_rate": 4.849661303979679e-05,
"loss": 0.3129,
"step": 3560
},
{
"epoch": 3.02,
"learning_rate": 4.8492379339542765e-05,
"loss": 0.3575,
"step": 3570
},
{
"epoch": 3.03,
"learning_rate": 4.848814563928874e-05,
"loss": 0.3663,
"step": 3580
},
{
"epoch": 3.04,
"learning_rate": 4.848391193903472e-05,
"loss": 0.3684,
"step": 3590
},
{
"epoch": 3.05,
"learning_rate": 4.84796782387807e-05,
"loss": 0.4295,
"step": 3600
},
{
"epoch": 3.06,
"learning_rate": 4.8475444538526675e-05,
"loss": 0.3392,
"step": 3610
},
{
"epoch": 3.07,
"learning_rate": 4.847121083827265e-05,
"loss": 0.3861,
"step": 3620
},
{
"epoch": 3.07,
"learning_rate": 4.846697713801863e-05,
"loss": 0.4343,
"step": 3630
},
{
"epoch": 3.08,
"learning_rate": 4.846274343776461e-05,
"loss": 0.4075,
"step": 3640
},
{
"epoch": 3.09,
"learning_rate": 4.8458509737510586e-05,
"loss": 0.4533,
"step": 3650
},
{
"epoch": 3.1,
"learning_rate": 4.8454276037256564e-05,
"loss": 0.3654,
"step": 3660
},
{
"epoch": 3.11,
"learning_rate": 4.845004233700254e-05,
"loss": 0.3483,
"step": 3670
},
{
"epoch": 3.12,
"learning_rate": 4.844580863674852e-05,
"loss": 0.3918,
"step": 3680
},
{
"epoch": 3.12,
"learning_rate": 4.8441574936494496e-05,
"loss": 0.4442,
"step": 3690
},
{
"epoch": 3.13,
"learning_rate": 4.8437341236240474e-05,
"loss": 0.3351,
"step": 3700
},
{
"epoch": 3.14,
"learning_rate": 4.843310753598646e-05,
"loss": 0.3112,
"step": 3710
},
{
"epoch": 3.15,
"learning_rate": 4.842887383573243e-05,
"loss": 0.3461,
"step": 3720
},
{
"epoch": 3.16,
"learning_rate": 4.8424640135478414e-05,
"loss": 0.3967,
"step": 3730
},
{
"epoch": 3.17,
"learning_rate": 4.8420406435224384e-05,
"loss": 0.5231,
"step": 3740
},
{
"epoch": 3.18,
"learning_rate": 4.841617273497037e-05,
"loss": 0.3954,
"step": 3750
},
{
"epoch": 3.18,
"learning_rate": 4.8411939034716346e-05,
"loss": 0.3153,
"step": 3760
},
{
"epoch": 3.19,
"learning_rate": 4.8407705334462324e-05,
"loss": 0.3667,
"step": 3770
},
{
"epoch": 3.2,
"learning_rate": 4.84034716342083e-05,
"loss": 0.4518,
"step": 3780
},
{
"epoch": 3.21,
"learning_rate": 4.839923793395428e-05,
"loss": 0.5369,
"step": 3790
},
{
"epoch": 3.22,
"learning_rate": 4.839500423370026e-05,
"loss": 0.3811,
"step": 3800
},
{
"epoch": 3.23,
"learning_rate": 4.8390770533446235e-05,
"loss": 0.3276,
"step": 3810
},
{
"epoch": 3.23,
"learning_rate": 4.838653683319221e-05,
"loss": 0.3428,
"step": 3820
},
{
"epoch": 3.24,
"learning_rate": 4.838230313293819e-05,
"loss": 0.4002,
"step": 3830
},
{
"epoch": 3.25,
"learning_rate": 4.837806943268417e-05,
"loss": 0.512,
"step": 3840
},
{
"epoch": 3.26,
"learning_rate": 4.8373835732430145e-05,
"loss": 0.4051,
"step": 3850
},
{
"epoch": 3.27,
"learning_rate": 4.836960203217612e-05,
"loss": 0.3545,
"step": 3860
},
{
"epoch": 3.28,
"learning_rate": 4.83653683319221e-05,
"loss": 0.4291,
"step": 3870
},
{
"epoch": 3.29,
"learning_rate": 4.8361134631668085e-05,
"loss": 0.3827,
"step": 3880
},
{
"epoch": 3.29,
"learning_rate": 4.8356900931414055e-05,
"loss": 0.4711,
"step": 3890
},
{
"epoch": 3.3,
"learning_rate": 4.835266723116004e-05,
"loss": 0.4549,
"step": 3900
},
{
"epoch": 3.31,
"learning_rate": 4.834843353090601e-05,
"loss": 0.3589,
"step": 3910
},
{
"epoch": 3.32,
"learning_rate": 4.8344199830651995e-05,
"loss": 0.3354,
"step": 3920
},
{
"epoch": 3.33,
"learning_rate": 4.8339966130397966e-05,
"loss": 0.4284,
"step": 3930
},
{
"epoch": 3.34,
"learning_rate": 4.833573243014395e-05,
"loss": 0.5216,
"step": 3940
},
{
"epoch": 3.34,
"learning_rate": 4.833149872988992e-05,
"loss": 0.4689,
"step": 3950
},
{
"epoch": 3.35,
"learning_rate": 4.8327265029635906e-05,
"loss": 0.3412,
"step": 3960
},
{
"epoch": 3.36,
"learning_rate": 4.832303132938188e-05,
"loss": 0.3283,
"step": 3970
},
{
"epoch": 3.37,
"learning_rate": 4.831879762912786e-05,
"loss": 0.3946,
"step": 3980
},
{
"epoch": 3.38,
"learning_rate": 4.831456392887384e-05,
"loss": 0.504,
"step": 3990
},
{
"epoch": 3.39,
"learning_rate": 4.8310330228619816e-05,
"loss": 0.4394,
"step": 4000
},
{
"epoch": 3.4,
"learning_rate": 4.8306096528365794e-05,
"loss": 0.3218,
"step": 4010
},
{
"epoch": 3.4,
"learning_rate": 4.830186282811177e-05,
"loss": 0.4287,
"step": 4020
},
{
"epoch": 3.41,
"learning_rate": 4.829762912785775e-05,
"loss": 0.3832,
"step": 4030
},
{
"epoch": 3.42,
"learning_rate": 4.8293395427603727e-05,
"loss": 0.5022,
"step": 4040
},
{
"epoch": 3.43,
"learning_rate": 4.828916172734971e-05,
"loss": 0.4091,
"step": 4050
},
{
"epoch": 3.44,
"learning_rate": 4.828492802709568e-05,
"loss": 0.2818,
"step": 4060
},
{
"epoch": 3.45,
"learning_rate": 4.8280694326841666e-05,
"loss": 0.4073,
"step": 4070
},
{
"epoch": 3.45,
"learning_rate": 4.827646062658764e-05,
"loss": 0.3177,
"step": 4080
},
{
"epoch": 3.46,
"learning_rate": 4.827222692633362e-05,
"loss": 0.4037,
"step": 4090
},
{
"epoch": 3.47,
"learning_rate": 4.826799322607959e-05,
"loss": 0.4102,
"step": 4100
},
{
"epoch": 3.48,
"learning_rate": 4.826375952582558e-05,
"loss": 0.3053,
"step": 4110
},
{
"epoch": 3.49,
"learning_rate": 4.825952582557155e-05,
"loss": 0.4242,
"step": 4120
},
{
"epoch": 3.5,
"learning_rate": 4.825529212531753e-05,
"loss": 0.4177,
"step": 4130
},
{
"epoch": 3.51,
"learning_rate": 4.82510584250635e-05,
"loss": 0.4545,
"step": 4140
},
{
"epoch": 3.51,
"learning_rate": 4.824682472480949e-05,
"loss": 0.358,
"step": 4150
},
{
"epoch": 3.52,
"learning_rate": 4.824259102455546e-05,
"loss": 0.3987,
"step": 4160
},
{
"epoch": 3.53,
"learning_rate": 4.823835732430144e-05,
"loss": 0.3892,
"step": 4170
},
{
"epoch": 3.54,
"learning_rate": 4.823412362404742e-05,
"loss": 0.4229,
"step": 4180
},
{
"epoch": 3.55,
"learning_rate": 4.82298899237934e-05,
"loss": 0.4626,
"step": 4190
},
{
"epoch": 3.56,
"learning_rate": 4.822607959356478e-05,
"loss": 0.3527,
"step": 4200
},
{
"epoch": 3.56,
"learning_rate": 4.822184589331076e-05,
"loss": 0.3111,
"step": 4210
},
{
"epoch": 3.57,
"learning_rate": 4.8217612193056735e-05,
"loss": 0.3451,
"step": 4220
},
{
"epoch": 3.58,
"learning_rate": 4.821337849280271e-05,
"loss": 0.3782,
"step": 4230
},
{
"epoch": 3.59,
"learning_rate": 4.820914479254869e-05,
"loss": 0.4518,
"step": 4240
},
{
"epoch": 3.6,
"learning_rate": 4.820491109229467e-05,
"loss": 0.375,
"step": 4250
},
{
"epoch": 3.61,
"learning_rate": 4.8200677392040646e-05,
"loss": 0.3501,
"step": 4260
},
{
"epoch": 3.62,
"learning_rate": 4.8196443691786623e-05,
"loss": 0.3453,
"step": 4270
},
{
"epoch": 3.62,
"learning_rate": 4.81922099915326e-05,
"loss": 0.3315,
"step": 4280
},
{
"epoch": 3.63,
"learning_rate": 4.818797629127858e-05,
"loss": 0.5497,
"step": 4290
},
{
"epoch": 3.64,
"learning_rate": 4.8183742591024556e-05,
"loss": 0.3749,
"step": 4300
},
{
"epoch": 3.65,
"learning_rate": 4.8179508890770534e-05,
"loss": 0.312,
"step": 4310
},
{
"epoch": 3.66,
"learning_rate": 4.817527519051651e-05,
"loss": 0.3664,
"step": 4320
},
{
"epoch": 3.67,
"learning_rate": 4.817104149026249e-05,
"loss": 0.4353,
"step": 4330
},
{
"epoch": 3.67,
"learning_rate": 4.8166807790008474e-05,
"loss": 0.4016,
"step": 4340
},
{
"epoch": 3.68,
"learning_rate": 4.8162574089754444e-05,
"loss": 0.3658,
"step": 4350
},
{
"epoch": 3.69,
"learning_rate": 4.815834038950043e-05,
"loss": 0.2703,
"step": 4360
},
{
"epoch": 3.7,
"learning_rate": 4.8154106689246406e-05,
"loss": 0.3408,
"step": 4370
},
{
"epoch": 3.71,
"learning_rate": 4.8149872988992384e-05,
"loss": 0.3901,
"step": 4380
},
{
"epoch": 3.72,
"learning_rate": 4.814563928873836e-05,
"loss": 0.4561,
"step": 4390
},
{
"epoch": 3.73,
"learning_rate": 4.814140558848434e-05,
"loss": 0.4306,
"step": 4400
},
{
"epoch": 3.73,
"learning_rate": 4.813717188823032e-05,
"loss": 0.3023,
"step": 4410
},
{
"epoch": 3.74,
"learning_rate": 4.8132938187976295e-05,
"loss": 0.3275,
"step": 4420
},
{
"epoch": 3.75,
"learning_rate": 4.812870448772227e-05,
"loss": 0.3832,
"step": 4430
},
{
"epoch": 3.76,
"learning_rate": 4.812447078746825e-05,
"loss": 0.452,
"step": 4440
},
{
"epoch": 3.77,
"learning_rate": 4.812066045723963e-05,
"loss": 0.453,
"step": 4450
},
{
"epoch": 3.78,
"learning_rate": 4.81164267569856e-05,
"loss": 0.3208,
"step": 4460
},
{
"epoch": 3.78,
"learning_rate": 4.811219305673159e-05,
"loss": 0.3845,
"step": 4470
},
{
"epoch": 3.79,
"learning_rate": 4.810795935647756e-05,
"loss": 0.4085,
"step": 4480
},
{
"epoch": 3.8,
"learning_rate": 4.810372565622354e-05,
"loss": 0.4681,
"step": 4490
},
{
"epoch": 3.81,
"learning_rate": 4.809949195596952e-05,
"loss": 0.3528,
"step": 4500
},
{
"epoch": 3.82,
"learning_rate": 4.80952582557155e-05,
"loss": 0.3221,
"step": 4510
},
{
"epoch": 3.83,
"learning_rate": 4.8091024555461476e-05,
"loss": 0.3381,
"step": 4520
},
{
"epoch": 3.84,
"learning_rate": 4.808679085520745e-05,
"loss": 0.3763,
"step": 4530
},
{
"epoch": 3.84,
"learning_rate": 4.808255715495343e-05,
"loss": 0.5083,
"step": 4540
},
{
"epoch": 3.85,
"learning_rate": 4.807832345469941e-05,
"loss": 0.4185,
"step": 4550
},
{
"epoch": 3.86,
"learning_rate": 4.8074089754445386e-05,
"loss": 0.3625,
"step": 4560
},
{
"epoch": 3.87,
"learning_rate": 4.8069856054191364e-05,
"loss": 0.3564,
"step": 4570
},
{
"epoch": 3.88,
"learning_rate": 4.806562235393735e-05,
"loss": 0.3963,
"step": 4580
},
{
"epoch": 3.89,
"learning_rate": 4.806138865368332e-05,
"loss": 0.4047,
"step": 4590
},
{
"epoch": 3.9,
"learning_rate": 4.80571549534293e-05,
"loss": 0.4063,
"step": 4600
},
{
"epoch": 3.9,
"learning_rate": 4.8052921253175274e-05,
"loss": 0.3512,
"step": 4610
},
{
"epoch": 3.91,
"learning_rate": 4.804868755292126e-05,
"loss": 0.3988,
"step": 4620
},
{
"epoch": 3.92,
"learning_rate": 4.804445385266723e-05,
"loss": 0.4998,
"step": 4630
},
{
"epoch": 3.93,
"learning_rate": 4.8040220152413214e-05,
"loss": 0.4857,
"step": 4640
},
{
"epoch": 3.94,
"learning_rate": 4.8035986452159185e-05,
"loss": 0.4727,
"step": 4650
},
{
"epoch": 3.95,
"learning_rate": 4.803175275190517e-05,
"loss": 0.3506,
"step": 4660
},
{
"epoch": 3.95,
"learning_rate": 4.802751905165114e-05,
"loss": 0.3694,
"step": 4670
},
{
"epoch": 3.96,
"learning_rate": 4.8023285351397124e-05,
"loss": 0.3642,
"step": 4680
},
{
"epoch": 3.97,
"learning_rate": 4.80190516511431e-05,
"loss": 0.4493,
"step": 4690
},
{
"epoch": 3.98,
"learning_rate": 4.801481795088908e-05,
"loss": 0.4068,
"step": 4700
},
{
"epoch": 3.99,
"learning_rate": 4.801058425063506e-05,
"loss": 0.3905,
"step": 4710
},
{
"epoch": 4.0,
"learning_rate": 4.8006350550381035e-05,
"loss": 0.4209,
"step": 4720
},
{
"epoch": 4.0,
"eval_cer": 0.12201825992071713,
"eval_loss": 0.5889235138893127,
"eval_mer": 0.3426117212134392,
"eval_runtime": 27.1757,
"eval_samples_per_second": 80.145,
"eval_steps_per_second": 5.041,
"step": 4724
},
{
"epoch": 4.01,
"learning_rate": 4.800211685012701e-05,
"loss": 0.404,
"step": 4730
},
{
"epoch": 4.01,
"learning_rate": 4.799788314987299e-05,
"loss": 0.2647,
"step": 4740
},
{
"epoch": 4.02,
"learning_rate": 4.7993649449618974e-05,
"loss": 0.2844,
"step": 4750
},
{
"epoch": 4.03,
"learning_rate": 4.7989415749364945e-05,
"loss": 0.2902,
"step": 4760
},
{
"epoch": 4.04,
"learning_rate": 4.798518204911093e-05,
"loss": 0.416,
"step": 4770
},
{
"epoch": 4.05,
"learning_rate": 4.79809483488569e-05,
"loss": 0.2932,
"step": 4780
},
{
"epoch": 4.06,
"learning_rate": 4.7976714648602885e-05,
"loss": 0.3144,
"step": 4790
},
{
"epoch": 4.06,
"learning_rate": 4.7972480948348856e-05,
"loss": 0.2721,
"step": 4800
},
{
"epoch": 4.07,
"learning_rate": 4.796824724809484e-05,
"loss": 0.3158,
"step": 4810
},
{
"epoch": 4.08,
"learning_rate": 4.796401354784081e-05,
"loss": 0.4105,
"step": 4820
},
{
"epoch": 4.09,
"learning_rate": 4.7959779847586795e-05,
"loss": 0.43,
"step": 4830
},
{
"epoch": 4.1,
"learning_rate": 4.7955546147332766e-05,
"loss": 0.2806,
"step": 4840
},
{
"epoch": 4.11,
"learning_rate": 4.795131244707875e-05,
"loss": 0.2901,
"step": 4850
},
{
"epoch": 4.12,
"learning_rate": 4.794707874682473e-05,
"loss": 0.3427,
"step": 4860
},
{
"epoch": 4.12,
"learning_rate": 4.7942845046570706e-05,
"loss": 0.3866,
"step": 4870
},
{
"epoch": 4.13,
"learning_rate": 4.7938611346316683e-05,
"loss": 0.338,
"step": 4880
},
{
"epoch": 4.14,
"learning_rate": 4.793437764606266e-05,
"loss": 0.2824,
"step": 4890
},
{
"epoch": 4.15,
"learning_rate": 4.793014394580864e-05,
"loss": 0.3001,
"step": 4900
},
{
"epoch": 4.16,
"learning_rate": 4.7925910245554616e-05,
"loss": 0.3255,
"step": 4910
},
{
"epoch": 4.17,
"learning_rate": 4.7921676545300594e-05,
"loss": 0.3247,
"step": 4920
},
{
"epoch": 4.17,
"learning_rate": 4.791744284504657e-05,
"loss": 0.3804,
"step": 4930
},
{
"epoch": 4.18,
"learning_rate": 4.7913209144792556e-05,
"loss": 0.2978,
"step": 4940
},
{
"epoch": 4.19,
"learning_rate": 4.790897544453853e-05,
"loss": 0.279,
"step": 4950
},
{
"epoch": 4.2,
"learning_rate": 4.790474174428451e-05,
"loss": 0.3198,
"step": 4960
},
{
"epoch": 4.21,
"learning_rate": 4.790050804403048e-05,
"loss": 0.3622,
"step": 4970
},
{
"epoch": 4.22,
"learning_rate": 4.7896274343776466e-05,
"loss": 0.4556,
"step": 4980
},
{
"epoch": 4.23,
"learning_rate": 4.789204064352244e-05,
"loss": 0.275,
"step": 4990
},
{
"epoch": 4.23,
"learning_rate": 4.788780694326842e-05,
"loss": 0.339,
"step": 5000
},
{
"epoch": 4.24,
"learning_rate": 4.788357324301439e-05,
"loss": 0.3579,
"step": 5010
},
{
"epoch": 4.25,
"learning_rate": 4.787933954276038e-05,
"loss": 0.3492,
"step": 5020
},
{
"epoch": 4.26,
"learning_rate": 4.787552921253175e-05,
"loss": 0.5051,
"step": 5030
},
{
"epoch": 4.27,
"learning_rate": 4.787129551227773e-05,
"loss": 0.2782,
"step": 5040
},
{
"epoch": 4.28,
"learning_rate": 4.7867061812023715e-05,
"loss": 0.3414,
"step": 5050
},
{
"epoch": 4.28,
"learning_rate": 4.7862828111769686e-05,
"loss": 0.3327,
"step": 5060
},
{
"epoch": 4.29,
"learning_rate": 4.785859441151567e-05,
"loss": 0.3754,
"step": 5070
},
{
"epoch": 4.3,
"learning_rate": 4.785436071126165e-05,
"loss": 0.3647,
"step": 5080
},
{
"epoch": 4.31,
"learning_rate": 4.7850127011007625e-05,
"loss": 0.2644,
"step": 5090
},
{
"epoch": 4.32,
"learning_rate": 4.78458933107536e-05,
"loss": 0.2835,
"step": 5100
},
{
"epoch": 4.33,
"learning_rate": 4.784165961049958e-05,
"loss": 0.3444,
"step": 5110
},
{
"epoch": 4.34,
"learning_rate": 4.783742591024556e-05,
"loss": 0.3817,
"step": 5120
},
{
"epoch": 4.34,
"learning_rate": 4.7833192209991536e-05,
"loss": 0.3676,
"step": 5130
},
{
"epoch": 4.35,
"learning_rate": 4.782895850973751e-05,
"loss": 0.2442,
"step": 5140
},
{
"epoch": 4.36,
"learning_rate": 4.782472480948349e-05,
"loss": 0.2907,
"step": 5150
},
{
"epoch": 4.37,
"learning_rate": 4.782049110922947e-05,
"loss": 0.3163,
"step": 5160
},
{
"epoch": 4.38,
"learning_rate": 4.7816257408975446e-05,
"loss": 0.338,
"step": 5170
},
{
"epoch": 4.39,
"learning_rate": 4.7812023708721424e-05,
"loss": 0.4044,
"step": 5180
},
{
"epoch": 4.39,
"learning_rate": 4.78077900084674e-05,
"loss": 0.3073,
"step": 5190
},
{
"epoch": 4.4,
"learning_rate": 4.780355630821338e-05,
"loss": 0.2753,
"step": 5200
},
{
"epoch": 4.41,
"learning_rate": 4.7799322607959357e-05,
"loss": 0.3416,
"step": 5210
},
{
"epoch": 4.42,
"learning_rate": 4.779508890770534e-05,
"loss": 0.3171,
"step": 5220
},
{
"epoch": 4.43,
"learning_rate": 4.779085520745131e-05,
"loss": 0.4188,
"step": 5230
},
{
"epoch": 4.44,
"learning_rate": 4.7786621507197296e-05,
"loss": 0.2775,
"step": 5240
},
{
"epoch": 4.45,
"learning_rate": 4.778238780694327e-05,
"loss": 0.3707,
"step": 5250
},
{
"epoch": 4.45,
"learning_rate": 4.777815410668925e-05,
"loss": 0.3064,
"step": 5260
},
{
"epoch": 4.46,
"learning_rate": 4.777392040643522e-05,
"loss": 0.3798,
"step": 5270
},
{
"epoch": 4.47,
"learning_rate": 4.776968670618121e-05,
"loss": 0.3791,
"step": 5280
},
{
"epoch": 4.48,
"learning_rate": 4.7765453005927184e-05,
"loss": 0.2691,
"step": 5290
},
{
"epoch": 4.49,
"learning_rate": 4.776121930567316e-05,
"loss": 0.2965,
"step": 5300
},
{
"epoch": 4.5,
"learning_rate": 4.775698560541914e-05,
"loss": 0.3141,
"step": 5310
},
{
"epoch": 4.5,
"learning_rate": 4.775275190516512e-05,
"loss": 0.4067,
"step": 5320
},
{
"epoch": 4.51,
"learning_rate": 4.7748518204911095e-05,
"loss": 0.3959,
"step": 5330
},
{
"epoch": 4.52,
"learning_rate": 4.774428450465707e-05,
"loss": 0.2693,
"step": 5340
},
{
"epoch": 4.53,
"learning_rate": 4.774005080440305e-05,
"loss": 0.2673,
"step": 5350
},
{
"epoch": 4.54,
"learning_rate": 4.773581710414903e-05,
"loss": 0.2952,
"step": 5360
},
{
"epoch": 4.55,
"learning_rate": 4.7731583403895005e-05,
"loss": 0.3988,
"step": 5370
},
{
"epoch": 4.56,
"learning_rate": 4.772734970364098e-05,
"loss": 0.309,
"step": 5380
},
{
"epoch": 4.56,
"learning_rate": 4.772311600338697e-05,
"loss": 0.2741,
"step": 5390
},
{
"epoch": 4.57,
"learning_rate": 4.771888230313294e-05,
"loss": 0.3079,
"step": 5400
},
{
"epoch": 4.58,
"learning_rate": 4.771464860287892e-05,
"loss": 0.3819,
"step": 5410
},
{
"epoch": 4.59,
"learning_rate": 4.771041490262489e-05,
"loss": 0.3842,
"step": 5420
},
{
"epoch": 4.6,
"learning_rate": 4.770618120237088e-05,
"loss": 0.4219,
"step": 5430
},
{
"epoch": 4.61,
"learning_rate": 4.770194750211685e-05,
"loss": 0.2819,
"step": 5440
},
{
"epoch": 4.61,
"learning_rate": 4.769771380186283e-05,
"loss": 0.2979,
"step": 5450
},
{
"epoch": 4.62,
"learning_rate": 4.7693480101608804e-05,
"loss": 0.2718,
"step": 5460
},
{
"epoch": 4.63,
"learning_rate": 4.768924640135479e-05,
"loss": 0.3435,
"step": 5470
},
{
"epoch": 4.64,
"learning_rate": 4.768501270110076e-05,
"loss": 0.3944,
"step": 5480
},
{
"epoch": 4.65,
"learning_rate": 4.7680779000846743e-05,
"loss": 0.2934,
"step": 5490
},
{
"epoch": 4.66,
"learning_rate": 4.7676545300592714e-05,
"loss": 0.3293,
"step": 5500
},
{
"epoch": 4.67,
"learning_rate": 4.76723116003387e-05,
"loss": 0.3617,
"step": 5510
},
{
"epoch": 4.67,
"learning_rate": 4.7668077900084676e-05,
"loss": 0.3764,
"step": 5520
},
{
"epoch": 4.68,
"learning_rate": 4.7663844199830654e-05,
"loss": 0.4426,
"step": 5530
},
{
"epoch": 4.69,
"learning_rate": 4.765961049957663e-05,
"loss": 0.255,
"step": 5540
},
{
"epoch": 4.7,
"learning_rate": 4.765537679932261e-05,
"loss": 0.2891,
"step": 5550
},
{
"epoch": 4.71,
"learning_rate": 4.7651143099068594e-05,
"loss": 0.311,
"step": 5560
},
{
"epoch": 4.72,
"learning_rate": 4.7646909398814564e-05,
"loss": 0.3547,
"step": 5570
},
{
"epoch": 4.72,
"learning_rate": 4.764267569856055e-05,
"loss": 0.3187,
"step": 5580
},
{
"epoch": 4.73,
"learning_rate": 4.763844199830652e-05,
"loss": 0.2431,
"step": 5590
},
{
"epoch": 4.74,
"learning_rate": 4.7634208298052504e-05,
"loss": 0.2886,
"step": 5600
},
{
"epoch": 4.75,
"learning_rate": 4.7629974597798475e-05,
"loss": 0.3226,
"step": 5610
},
{
"epoch": 4.76,
"learning_rate": 4.762574089754446e-05,
"loss": 0.3693,
"step": 5620
},
{
"epoch": 4.77,
"learning_rate": 4.762150719729043e-05,
"loss": 0.3709,
"step": 5630
},
{
"epoch": 4.78,
"learning_rate": 4.7617273497036414e-05,
"loss": 0.297,
"step": 5640
},
{
"epoch": 4.78,
"learning_rate": 4.7613039796782385e-05,
"loss": 0.2794,
"step": 5650
},
{
"epoch": 4.79,
"learning_rate": 4.760880609652837e-05,
"loss": 0.3495,
"step": 5660
},
{
"epoch": 4.8,
"learning_rate": 4.760457239627435e-05,
"loss": 0.3761,
"step": 5670
},
{
"epoch": 4.81,
"learning_rate": 4.7600338696020325e-05,
"loss": 0.3361,
"step": 5680
},
{
"epoch": 4.82,
"learning_rate": 4.75961049957663e-05,
"loss": 0.2881,
"step": 5690
},
{
"epoch": 4.83,
"learning_rate": 4.759187129551228e-05,
"loss": 0.3044,
"step": 5700
},
{
"epoch": 4.83,
"learning_rate": 4.758763759525826e-05,
"loss": 0.3407,
"step": 5710
},
{
"epoch": 4.84,
"learning_rate": 4.7583403895004235e-05,
"loss": 0.3124,
"step": 5720
},
{
"epoch": 4.85,
"learning_rate": 4.757917019475022e-05,
"loss": 0.3587,
"step": 5730
},
{
"epoch": 4.86,
"learning_rate": 4.757493649449619e-05,
"loss": 0.2854,
"step": 5740
},
{
"epoch": 4.87,
"learning_rate": 4.7570702794242175e-05,
"loss": 0.2995,
"step": 5750
},
{
"epoch": 4.88,
"learning_rate": 4.7566469093988146e-05,
"loss": 0.3462,
"step": 5760
},
{
"epoch": 4.89,
"learning_rate": 4.756223539373413e-05,
"loss": 0.3729,
"step": 5770
},
{
"epoch": 4.89,
"learning_rate": 4.75580016934801e-05,
"loss": 0.3514,
"step": 5780
},
{
"epoch": 4.9,
"learning_rate": 4.7553767993226086e-05,
"loss": 0.2361,
"step": 5790
},
{
"epoch": 4.91,
"learning_rate": 4.7549534292972056e-05,
"loss": 0.2972,
"step": 5800
},
{
"epoch": 4.92,
"learning_rate": 4.754530059271804e-05,
"loss": 0.3605,
"step": 5810
},
{
"epoch": 4.93,
"learning_rate": 4.754106689246401e-05,
"loss": 0.3621,
"step": 5820
},
{
"epoch": 4.94,
"learning_rate": 4.7536833192209996e-05,
"loss": 0.3325,
"step": 5830
},
{
"epoch": 4.94,
"learning_rate": 4.7532599491955974e-05,
"loss": 0.2995,
"step": 5840
},
{
"epoch": 4.95,
"learning_rate": 4.752836579170195e-05,
"loss": 0.3203,
"step": 5850
},
{
"epoch": 4.96,
"learning_rate": 4.752413209144793e-05,
"loss": 0.3237,
"step": 5860
},
{
"epoch": 4.97,
"learning_rate": 4.7519898391193906e-05,
"loss": 0.4474,
"step": 5870
},
{
"epoch": 4.98,
"learning_rate": 4.7515664690939884e-05,
"loss": 0.2895,
"step": 5880
},
{
"epoch": 4.99,
"learning_rate": 4.751143099068586e-05,
"loss": 0.3015,
"step": 5890
},
{
"epoch": 5.0,
"learning_rate": 4.750719729043184e-05,
"loss": 0.3663,
"step": 5900
},
{
"epoch": 5.0,
"eval_cer": 0.11584414149366704,
"eval_loss": 0.5903986096382141,
"eval_mer": 0.32532347504621073,
"eval_runtime": 24.1423,
"eval_samples_per_second": 90.215,
"eval_steps_per_second": 5.675,
"step": 5905
},
{
"epoch": 5.0,
"learning_rate": 4.750338696020322e-05,
"loss": 0.2998,
"step": 5910
},
{
"epoch": 5.01,
"learning_rate": 4.74991532599492e-05,
"loss": 0.222,
"step": 5920
},
{
"epoch": 5.02,
"learning_rate": 4.749491955969518e-05,
"loss": 0.2321,
"step": 5930
},
{
"epoch": 5.03,
"learning_rate": 4.7490685859441155e-05,
"loss": 0.2734,
"step": 5940
},
{
"epoch": 5.04,
"learning_rate": 4.748645215918713e-05,
"loss": 0.2485,
"step": 5950
},
{
"epoch": 5.05,
"learning_rate": 4.748221845893311e-05,
"loss": 0.3122,
"step": 5960
},
{
"epoch": 5.06,
"learning_rate": 4.747798475867909e-05,
"loss": 0.2434,
"step": 5970
},
{
"epoch": 5.06,
"learning_rate": 4.7473751058425065e-05,
"loss": 0.2556,
"step": 5980
},
{
"epoch": 5.07,
"learning_rate": 4.746951735817104e-05,
"loss": 0.3138,
"step": 5990
},
{
"epoch": 5.08,
"learning_rate": 4.746528365791702e-05,
"loss": 0.2655,
"step": 6000
},
{
"epoch": 5.09,
"learning_rate": 4.7461049957663e-05,
"loss": 0.3957,
"step": 6010
},
{
"epoch": 5.1,
"learning_rate": 4.7456816257408976e-05,
"loss": 0.2179,
"step": 6020
},
{
"epoch": 5.11,
"learning_rate": 4.745258255715495e-05,
"loss": 0.2319,
"step": 6030
},
{
"epoch": 5.11,
"learning_rate": 4.744834885690093e-05,
"loss": 0.2649,
"step": 6040
},
{
"epoch": 5.12,
"learning_rate": 4.7444115156646915e-05,
"loss": 0.2678,
"step": 6050
},
{
"epoch": 5.13,
"learning_rate": 4.7439881456392886e-05,
"loss": 0.2765,
"step": 6060
},
{
"epoch": 5.14,
"learning_rate": 4.743564775613887e-05,
"loss": 0.221,
"step": 6070
},
{
"epoch": 5.15,
"learning_rate": 4.743141405588484e-05,
"loss": 0.3079,
"step": 6080
},
{
"epoch": 5.16,
"learning_rate": 4.7427180355630826e-05,
"loss": 0.2536,
"step": 6090
},
{
"epoch": 5.17,
"learning_rate": 4.7422946655376803e-05,
"loss": 0.3262,
"step": 6100
},
{
"epoch": 5.17,
"learning_rate": 4.741871295512278e-05,
"loss": 0.3531,
"step": 6110
},
{
"epoch": 5.18,
"learning_rate": 4.741447925486876e-05,
"loss": 0.2218,
"step": 6120
},
{
"epoch": 5.19,
"learning_rate": 4.7410245554614736e-05,
"loss": 0.2826,
"step": 6130
},
{
"epoch": 5.2,
"learning_rate": 4.7406011854360714e-05,
"loss": 0.3085,
"step": 6140
},
{
"epoch": 5.21,
"learning_rate": 4.740177815410669e-05,
"loss": 0.2716,
"step": 6150
},
{
"epoch": 5.22,
"learning_rate": 4.739754445385267e-05,
"loss": 0.3887,
"step": 6160
},
{
"epoch": 5.22,
"learning_rate": 4.739331075359865e-05,
"loss": 0.2004,
"step": 6170
},
{
"epoch": 5.23,
"learning_rate": 4.7389077053344624e-05,
"loss": 0.2359,
"step": 6180
},
{
"epoch": 5.24,
"learning_rate": 4.73848433530906e-05,
"loss": 0.2551,
"step": 6190
},
{
"epoch": 5.25,
"learning_rate": 4.738060965283658e-05,
"loss": 0.3348,
"step": 6200
},
{
"epoch": 5.26,
"learning_rate": 4.737637595258256e-05,
"loss": 0.2479,
"step": 6210
},
{
"epoch": 5.27,
"learning_rate": 4.737214225232854e-05,
"loss": 0.2186,
"step": 6220
},
{
"epoch": 5.28,
"learning_rate": 4.736790855207451e-05,
"loss": 0.2655,
"step": 6230
},
{
"epoch": 5.28,
"learning_rate": 4.73636748518205e-05,
"loss": 0.2573,
"step": 6240
},
{
"epoch": 5.29,
"learning_rate": 4.735944115156647e-05,
"loss": 0.3231,
"step": 6250
},
{
"epoch": 5.3,
"learning_rate": 4.735520745131245e-05,
"loss": 0.4099,
"step": 6260
},
{
"epoch": 5.31,
"learning_rate": 4.735097375105842e-05,
"loss": 0.227,
"step": 6270
},
{
"epoch": 5.32,
"learning_rate": 4.734674005080441e-05,
"loss": 0.2678,
"step": 6280
},
{
"epoch": 5.33,
"learning_rate": 4.734250635055038e-05,
"loss": 0.266,
"step": 6290
},
{
"epoch": 5.33,
"learning_rate": 4.733827265029636e-05,
"loss": 0.342,
"step": 6300
},
{
"epoch": 5.34,
"learning_rate": 4.733403895004234e-05,
"loss": 0.3919,
"step": 6310
},
{
"epoch": 5.35,
"learning_rate": 4.732980524978832e-05,
"loss": 0.2578,
"step": 6320
},
{
"epoch": 5.36,
"learning_rate": 4.7325571549534295e-05,
"loss": 0.2261,
"step": 6330
},
{
"epoch": 5.37,
"learning_rate": 4.732133784928027e-05,
"loss": 0.305,
"step": 6340
},
{
"epoch": 5.38,
"learning_rate": 4.731710414902625e-05,
"loss": 0.2893,
"step": 6350
},
{
"epoch": 5.39,
"learning_rate": 4.731287044877223e-05,
"loss": 0.3134,
"step": 6360
},
{
"epoch": 5.39,
"learning_rate": 4.7308636748518206e-05,
"loss": 0.2319,
"step": 6370
},
{
"epoch": 5.4,
"learning_rate": 4.7304403048264183e-05,
"loss": 0.2853,
"step": 6380
},
{
"epoch": 5.41,
"learning_rate": 4.730016934801017e-05,
"loss": 0.2884,
"step": 6390
},
{
"epoch": 5.42,
"learning_rate": 4.729593564775614e-05,
"loss": 0.3469,
"step": 6400
},
{
"epoch": 5.43,
"learning_rate": 4.729170194750212e-05,
"loss": 0.34,
"step": 6410
},
{
"epoch": 5.44,
"learning_rate": 4.7287468247248094e-05,
"loss": 0.2632,
"step": 6420
},
{
"epoch": 5.44,
"learning_rate": 4.728323454699408e-05,
"loss": 0.2544,
"step": 6430
},
{
"epoch": 5.45,
"learning_rate": 4.727900084674005e-05,
"loss": 0.3574,
"step": 6440
},
{
"epoch": 5.46,
"learning_rate": 4.7274767146486034e-05,
"loss": 0.3068,
"step": 6450
},
{
"epoch": 5.47,
"learning_rate": 4.7270533446232004e-05,
"loss": 0.3268,
"step": 6460
},
{
"epoch": 5.48,
"learning_rate": 4.726629974597799e-05,
"loss": 0.2279,
"step": 6470
},
{
"epoch": 5.49,
"learning_rate": 4.726206604572396e-05,
"loss": 0.2484,
"step": 6480
},
{
"epoch": 5.5,
"learning_rate": 4.7257832345469944e-05,
"loss": 0.2426,
"step": 6490
},
{
"epoch": 5.5,
"learning_rate": 4.725359864521592e-05,
"loss": 0.2924,
"step": 6500
},
{
"epoch": 5.51,
"learning_rate": 4.72493649449619e-05,
"loss": 0.3111,
"step": 6510
},
{
"epoch": 5.52,
"learning_rate": 4.724513124470788e-05,
"loss": 0.2546,
"step": 6520
},
{
"epoch": 5.53,
"learning_rate": 4.7240897544453855e-05,
"loss": 0.2674,
"step": 6530
},
{
"epoch": 5.54,
"learning_rate": 4.723666384419983e-05,
"loss": 0.3439,
"step": 6540
},
{
"epoch": 5.55,
"learning_rate": 4.723243014394581e-05,
"loss": 0.3073,
"step": 6550
},
{
"epoch": 5.55,
"learning_rate": 4.7228196443691794e-05,
"loss": 0.3552,
"step": 6560
},
{
"epoch": 5.56,
"learning_rate": 4.7223962743437765e-05,
"loss": 0.2658,
"step": 6570
},
{
"epoch": 5.57,
"learning_rate": 4.721972904318375e-05,
"loss": 0.2772,
"step": 6580
},
{
"epoch": 5.58,
"learning_rate": 4.721549534292972e-05,
"loss": 0.2463,
"step": 6590
},
{
"epoch": 5.59,
"learning_rate": 4.7211261642675705e-05,
"loss": 0.2514,
"step": 6600
},
{
"epoch": 5.6,
"learning_rate": 4.7207027942421675e-05,
"loss": 0.286,
"step": 6610
},
{
"epoch": 5.61,
"learning_rate": 4.720279424216766e-05,
"loss": 0.2621,
"step": 6620
},
{
"epoch": 5.61,
"learning_rate": 4.719856054191363e-05,
"loss": 0.209,
"step": 6630
},
{
"epoch": 5.62,
"learning_rate": 4.7194326841659615e-05,
"loss": 0.296,
"step": 6640
},
{
"epoch": 5.63,
"learning_rate": 4.7190093141405586e-05,
"loss": 0.3293,
"step": 6650
},
{
"epoch": 5.64,
"learning_rate": 4.718585944115157e-05,
"loss": 0.3592,
"step": 6660
},
{
"epoch": 5.65,
"learning_rate": 4.718162574089755e-05,
"loss": 0.2537,
"step": 6670
},
{
"epoch": 5.66,
"learning_rate": 4.7177392040643526e-05,
"loss": 0.224,
"step": 6680
},
{
"epoch": 5.66,
"learning_rate": 4.71731583403895e-05,
"loss": 0.2843,
"step": 6690
},
{
"epoch": 5.67,
"learning_rate": 4.716892464013548e-05,
"loss": 0.3668,
"step": 6700
},
{
"epoch": 5.68,
"learning_rate": 4.716511430990686e-05,
"loss": 0.2834,
"step": 6710
},
{
"epoch": 5.69,
"learning_rate": 4.716088060965284e-05,
"loss": 0.217,
"step": 6720
},
{
"epoch": 5.7,
"learning_rate": 4.715664690939882e-05,
"loss": 0.2405,
"step": 6730
},
{
"epoch": 5.71,
"learning_rate": 4.7152413209144796e-05,
"loss": 0.3263,
"step": 6740
},
{
"epoch": 5.72,
"learning_rate": 4.7148179508890774e-05,
"loss": 0.3204,
"step": 6750
},
{
"epoch": 5.72,
"learning_rate": 4.714394580863675e-05,
"loss": 0.3493,
"step": 6760
},
{
"epoch": 5.73,
"learning_rate": 4.713971210838273e-05,
"loss": 0.2501,
"step": 6770
},
{
"epoch": 5.74,
"learning_rate": 4.713547840812871e-05,
"loss": 0.2726,
"step": 6780
},
{
"epoch": 5.75,
"learning_rate": 4.7131244707874684e-05,
"loss": 0.2663,
"step": 6790
},
{
"epoch": 5.76,
"learning_rate": 4.712701100762066e-05,
"loss": 0.3557,
"step": 6800
},
{
"epoch": 5.77,
"learning_rate": 4.712277730736664e-05,
"loss": 0.329,
"step": 6810
},
{
"epoch": 5.77,
"learning_rate": 4.711854360711262e-05,
"loss": 0.221,
"step": 6820
},
{
"epoch": 5.78,
"learning_rate": 4.7114309906858595e-05,
"loss": 0.2512,
"step": 6830
},
{
"epoch": 5.79,
"learning_rate": 4.711007620660457e-05,
"loss": 0.3038,
"step": 6840
},
{
"epoch": 5.8,
"learning_rate": 4.710584250635055e-05,
"loss": 0.3058,
"step": 6850
},
{
"epoch": 5.81,
"learning_rate": 4.7101608806096534e-05,
"loss": 0.2861,
"step": 6860
},
{
"epoch": 5.82,
"learning_rate": 4.7097375105842505e-05,
"loss": 0.2352,
"step": 6870
},
{
"epoch": 5.83,
"learning_rate": 4.709314140558849e-05,
"loss": 0.2317,
"step": 6880
},
{
"epoch": 5.83,
"learning_rate": 4.708890770533446e-05,
"loss": 0.3027,
"step": 6890
},
{
"epoch": 5.84,
"learning_rate": 4.7084674005080445e-05,
"loss": 0.3641,
"step": 6900
},
{
"epoch": 5.85,
"learning_rate": 4.708086367485182e-05,
"loss": 0.2341,
"step": 6910
},
{
"epoch": 5.86,
"learning_rate": 4.7076629974597805e-05,
"loss": 0.2346,
"step": 6920
},
{
"epoch": 5.87,
"learning_rate": 4.7072396274343776e-05,
"loss": 0.2628,
"step": 6930
},
{
"epoch": 5.88,
"learning_rate": 4.706816257408976e-05,
"loss": 0.281,
"step": 6940
},
{
"epoch": 5.88,
"learning_rate": 4.706392887383573e-05,
"loss": 0.2478,
"step": 6950
},
{
"epoch": 5.89,
"learning_rate": 4.7059695173581716e-05,
"loss": 0.3587,
"step": 6960
},
{
"epoch": 5.9,
"learning_rate": 4.7055461473327686e-05,
"loss": 0.2311,
"step": 6970
},
{
"epoch": 5.91,
"learning_rate": 4.705122777307367e-05,
"loss": 0.2168,
"step": 6980
},
{
"epoch": 5.92,
"learning_rate": 4.704699407281964e-05,
"loss": 0.2599,
"step": 6990
},
{
"epoch": 5.93,
"learning_rate": 4.7042760372565626e-05,
"loss": 0.3293,
"step": 7000
},
{
"epoch": 5.94,
"learning_rate": 4.7038526672311604e-05,
"loss": 0.4742,
"step": 7010
},
{
"epoch": 5.94,
"learning_rate": 4.703429297205758e-05,
"loss": 0.2366,
"step": 7020
},
{
"epoch": 5.95,
"learning_rate": 4.703005927180356e-05,
"loss": 0.2503,
"step": 7030
},
{
"epoch": 5.96,
"learning_rate": 4.7025825571549536e-05,
"loss": 0.3153,
"step": 7040
},
{
"epoch": 5.97,
"learning_rate": 4.7021591871295514e-05,
"loss": 0.2815,
"step": 7050
},
{
"epoch": 5.98,
"learning_rate": 4.701735817104149e-05,
"loss": 0.3547,
"step": 7060
},
{
"epoch": 5.99,
"learning_rate": 4.7013124470787476e-05,
"loss": 0.2523,
"step": 7070
},
{
"epoch": 5.99,
"learning_rate": 4.700889077053345e-05,
"loss": 0.3156,
"step": 7080
},
{
"epoch": 6.0,
"eval_cer": 0.11519496125637095,
"eval_loss": 0.619243323802948,
"eval_mer": 0.31727737305643144,
"eval_runtime": 24.0991,
"eval_samples_per_second": 90.377,
"eval_steps_per_second": 5.685,
"step": 7086
},
{
"epoch": 6.0,
"learning_rate": 4.700465707027943e-05,
"loss": 0.346,
"step": 7090
},
{
"epoch": 6.01,
"learning_rate": 4.70004233700254e-05,
"loss": 0.1886,
"step": 7100
},
{
"epoch": 6.02,
"learning_rate": 4.6996189669771387e-05,
"loss": 0.2036,
"step": 7110
},
{
"epoch": 6.03,
"learning_rate": 4.699195596951736e-05,
"loss": 0.2303,
"step": 7120
},
{
"epoch": 6.04,
"learning_rate": 4.698772226926334e-05,
"loss": 0.2681,
"step": 7130
},
{
"epoch": 6.05,
"learning_rate": 4.698348856900931e-05,
"loss": 0.2554,
"step": 7140
},
{
"epoch": 6.05,
"learning_rate": 4.69792548687553e-05,
"loss": 0.2108,
"step": 7150
},
{
"epoch": 6.06,
"learning_rate": 4.697502116850127e-05,
"loss": 0.2299,
"step": 7160
},
{
"epoch": 6.07,
"learning_rate": 4.697078746824725e-05,
"loss": 0.2759,
"step": 7170
},
{
"epoch": 6.08,
"learning_rate": 4.696655376799323e-05,
"loss": 0.2758,
"step": 7180
},
{
"epoch": 6.09,
"learning_rate": 4.696232006773921e-05,
"loss": 0.2705,
"step": 7190
},
{
"epoch": 6.1,
"learning_rate": 4.6958086367485185e-05,
"loss": 0.2431,
"step": 7200
},
{
"epoch": 6.1,
"learning_rate": 4.695385266723116e-05,
"loss": 0.2274,
"step": 7210
},
{
"epoch": 6.11,
"learning_rate": 4.694961896697714e-05,
"loss": 0.2633,
"step": 7220
},
{
"epoch": 6.12,
"learning_rate": 4.694538526672312e-05,
"loss": 0.2357,
"step": 7230
},
{
"epoch": 6.13,
"learning_rate": 4.6941151566469096e-05,
"loss": 0.3212,
"step": 7240
},
{
"epoch": 6.14,
"learning_rate": 4.693691786621507e-05,
"loss": 0.2035,
"step": 7250
},
{
"epoch": 6.15,
"learning_rate": 4.693268416596105e-05,
"loss": 0.2362,
"step": 7260
},
{
"epoch": 6.16,
"learning_rate": 4.692845046570703e-05,
"loss": 0.2695,
"step": 7270
},
{
"epoch": 6.16,
"learning_rate": 4.692421676545301e-05,
"loss": 0.2789,
"step": 7280
},
{
"epoch": 6.17,
"learning_rate": 4.6919983065198984e-05,
"loss": 0.2722,
"step": 7290
},
{
"epoch": 6.18,
"learning_rate": 4.691574936494497e-05,
"loss": 0.2175,
"step": 7300
},
{
"epoch": 6.19,
"learning_rate": 4.691151566469094e-05,
"loss": 0.2006,
"step": 7310
},
{
"epoch": 6.2,
"learning_rate": 4.690728196443692e-05,
"loss": 0.2645,
"step": 7320
},
{
"epoch": 6.21,
"learning_rate": 4.6903048264182894e-05,
"loss": 0.2411,
"step": 7330
},
{
"epoch": 6.22,
"learning_rate": 4.689881456392888e-05,
"loss": 0.286,
"step": 7340
},
{
"epoch": 6.22,
"learning_rate": 4.6894580863674856e-05,
"loss": 0.2099,
"step": 7350
},
{
"epoch": 6.23,
"learning_rate": 4.6890347163420834e-05,
"loss": 0.2343,
"step": 7360
},
{
"epoch": 6.24,
"learning_rate": 4.688611346316681e-05,
"loss": 0.2238,
"step": 7370
},
{
"epoch": 6.25,
"learning_rate": 4.688187976291279e-05,
"loss": 0.2739,
"step": 7380
},
{
"epoch": 6.26,
"learning_rate": 4.687764606265877e-05,
"loss": 0.2951,
"step": 7390
},
{
"epoch": 6.27,
"learning_rate": 4.6873412362404744e-05,
"loss": 0.1936,
"step": 7400
},
{
"epoch": 6.27,
"learning_rate": 4.686917866215072e-05,
"loss": 0.2228,
"step": 7410
},
{
"epoch": 6.28,
"learning_rate": 4.68649449618967e-05,
"loss": 0.2278,
"step": 7420
},
{
"epoch": 6.29,
"learning_rate": 4.686071126164268e-05,
"loss": 0.2854,
"step": 7430
},
{
"epoch": 6.3,
"learning_rate": 4.685690093141406e-05,
"loss": 0.3059,
"step": 7440
},
{
"epoch": 6.31,
"learning_rate": 4.685266723116004e-05,
"loss": 0.193,
"step": 7450
},
{
"epoch": 6.32,
"learning_rate": 4.6848433530906015e-05,
"loss": 0.2073,
"step": 7460
},
{
"epoch": 6.33,
"learning_rate": 4.684419983065199e-05,
"loss": 0.2896,
"step": 7470
},
{
"epoch": 6.33,
"learning_rate": 4.683996613039797e-05,
"loss": 0.2773,
"step": 7480
},
{
"epoch": 6.34,
"learning_rate": 4.683573243014395e-05,
"loss": 0.22,
"step": 7490
},
{
"epoch": 6.35,
"learning_rate": 4.6831498729889925e-05,
"loss": 0.2059,
"step": 7500
},
{
"epoch": 6.36,
"learning_rate": 4.68272650296359e-05,
"loss": 0.2365,
"step": 7510
},
{
"epoch": 6.37,
"learning_rate": 4.682303132938188e-05,
"loss": 0.2368,
"step": 7520
},
{
"epoch": 6.38,
"learning_rate": 4.681879762912786e-05,
"loss": 0.2624,
"step": 7530
},
{
"epoch": 6.38,
"learning_rate": 4.681498729889924e-05,
"loss": 0.2982,
"step": 7540
},
{
"epoch": 6.39,
"learning_rate": 4.681075359864522e-05,
"loss": 0.2593,
"step": 7550
},
{
"epoch": 6.4,
"learning_rate": 4.6806519898391196e-05,
"loss": 0.2347,
"step": 7560
},
{
"epoch": 6.41,
"learning_rate": 4.6802286198137174e-05,
"loss": 0.253,
"step": 7570
},
{
"epoch": 6.42,
"learning_rate": 4.679805249788316e-05,
"loss": 0.2896,
"step": 7580
},
{
"epoch": 6.43,
"learning_rate": 4.679381879762913e-05,
"loss": 0.3142,
"step": 7590
},
{
"epoch": 6.44,
"learning_rate": 4.678958509737511e-05,
"loss": 0.2309,
"step": 7600
},
{
"epoch": 6.44,
"learning_rate": 4.6785351397121084e-05,
"loss": 0.2358,
"step": 7610
},
{
"epoch": 6.45,
"learning_rate": 4.678111769686707e-05,
"loss": 0.2623,
"step": 7620
},
{
"epoch": 6.46,
"learning_rate": 4.677688399661304e-05,
"loss": 0.335,
"step": 7630
},
{
"epoch": 6.47,
"learning_rate": 4.6772650296359024e-05,
"loss": 0.317,
"step": 7640
},
{
"epoch": 6.48,
"learning_rate": 4.6768416596104995e-05,
"loss": 0.2199,
"step": 7650
},
{
"epoch": 6.49,
"learning_rate": 4.676418289585098e-05,
"loss": 0.2316,
"step": 7660
},
{
"epoch": 6.49,
"learning_rate": 4.675994919559695e-05,
"loss": 0.2344,
"step": 7670
},
{
"epoch": 6.5,
"learning_rate": 4.6755715495342934e-05,
"loss": 0.2517,
"step": 7680
},
{
"epoch": 6.51,
"learning_rate": 4.6751481795088905e-05,
"loss": 0.2638,
"step": 7690
},
{
"epoch": 6.52,
"learning_rate": 4.674724809483489e-05,
"loss": 0.2674,
"step": 7700
},
{
"epoch": 6.53,
"learning_rate": 4.674301439458087e-05,
"loss": 0.2267,
"step": 7710
},
{
"epoch": 6.54,
"learning_rate": 4.6738780694326845e-05,
"loss": 0.2819,
"step": 7720
},
{
"epoch": 6.55,
"learning_rate": 4.673454699407282e-05,
"loss": 0.2611,
"step": 7730
},
{
"epoch": 6.55,
"learning_rate": 4.67303132938188e-05,
"loss": 0.4463,
"step": 7740
},
{
"epoch": 6.56,
"learning_rate": 4.672607959356478e-05,
"loss": 0.2092,
"step": 7750
},
{
"epoch": 6.57,
"learning_rate": 4.6721845893310755e-05,
"loss": 0.2554,
"step": 7760
},
{
"epoch": 6.58,
"learning_rate": 4.671761219305673e-05,
"loss": 0.2662,
"step": 7770
},
{
"epoch": 6.59,
"learning_rate": 4.671337849280271e-05,
"loss": 0.3036,
"step": 7780
},
{
"epoch": 6.6,
"learning_rate": 4.6709144792548695e-05,
"loss": 0.3882,
"step": 7790
},
{
"epoch": 6.6,
"learning_rate": 4.6704911092294666e-05,
"loss": 0.2478,
"step": 7800
},
{
"epoch": 6.61,
"learning_rate": 4.670067739204065e-05,
"loss": 0.276,
"step": 7810
},
{
"epoch": 6.62,
"learning_rate": 4.669644369178662e-05,
"loss": 0.2807,
"step": 7820
},
{
"epoch": 6.63,
"learning_rate": 4.6692209991532605e-05,
"loss": 0.3415,
"step": 7830
},
{
"epoch": 6.64,
"learning_rate": 4.6687976291278576e-05,
"loss": 0.3372,
"step": 7840
},
{
"epoch": 6.65,
"learning_rate": 4.668374259102456e-05,
"loss": 0.2488,
"step": 7850
},
{
"epoch": 6.66,
"learning_rate": 4.667950889077053e-05,
"loss": 0.2932,
"step": 7860
},
{
"epoch": 6.66,
"learning_rate": 4.6675275190516516e-05,
"loss": 0.3033,
"step": 7870
},
{
"epoch": 6.67,
"learning_rate": 4.667104149026249e-05,
"loss": 0.2999,
"step": 7880
},
{
"epoch": 6.68,
"learning_rate": 4.666680779000847e-05,
"loss": 0.4221,
"step": 7890
},
{
"epoch": 6.69,
"learning_rate": 4.666257408975445e-05,
"loss": 0.2158,
"step": 7900
},
{
"epoch": 6.7,
"learning_rate": 4.6658340389500426e-05,
"loss": 0.2835,
"step": 7910
},
{
"epoch": 6.71,
"learning_rate": 4.6654106689246404e-05,
"loss": 0.2663,
"step": 7920
},
{
"epoch": 6.71,
"learning_rate": 4.664987298899238e-05,
"loss": 0.2763,
"step": 7930
},
{
"epoch": 6.72,
"learning_rate": 4.664563928873836e-05,
"loss": 0.3322,
"step": 7940
},
{
"epoch": 6.73,
"learning_rate": 4.664140558848434e-05,
"loss": 0.2148,
"step": 7950
},
{
"epoch": 6.74,
"learning_rate": 4.6637171888230314e-05,
"loss": 0.232,
"step": 7960
},
{
"epoch": 6.75,
"learning_rate": 4.663293818797629e-05,
"loss": 0.3275,
"step": 7970
},
{
"epoch": 6.76,
"learning_rate": 4.662870448772227e-05,
"loss": 0.318,
"step": 7980
},
{
"epoch": 6.77,
"learning_rate": 4.662447078746825e-05,
"loss": 0.3615,
"step": 7990
},
{
"epoch": 6.77,
"learning_rate": 4.662023708721423e-05,
"loss": 0.2483,
"step": 8000
},
{
"epoch": 6.78,
"learning_rate": 4.66160033869602e-05,
"loss": 0.232,
"step": 8010
},
{
"epoch": 6.79,
"learning_rate": 4.661176968670619e-05,
"loss": 0.3508,
"step": 8020
},
{
"epoch": 6.8,
"learning_rate": 4.660753598645216e-05,
"loss": 0.3212,
"step": 8030
},
{
"epoch": 6.81,
"learning_rate": 4.660330228619814e-05,
"loss": 0.3302,
"step": 8040
},
{
"epoch": 6.82,
"learning_rate": 4.659906858594412e-05,
"loss": 0.2584,
"step": 8050
},
{
"epoch": 6.82,
"learning_rate": 4.65948348856901e-05,
"loss": 0.2511,
"step": 8060
},
{
"epoch": 6.83,
"learning_rate": 4.6590601185436075e-05,
"loss": 0.2861,
"step": 8070
},
{
"epoch": 6.84,
"learning_rate": 4.658636748518205e-05,
"loss": 0.24,
"step": 8080
},
{
"epoch": 6.85,
"learning_rate": 4.658213378492803e-05,
"loss": 0.3855,
"step": 8090
},
{
"epoch": 6.86,
"learning_rate": 4.657790008467401e-05,
"loss": 0.2516,
"step": 8100
},
{
"epoch": 6.87,
"learning_rate": 4.6573666384419985e-05,
"loss": 0.243,
"step": 8110
},
{
"epoch": 6.88,
"learning_rate": 4.656943268416596e-05,
"loss": 0.264,
"step": 8120
},
{
"epoch": 6.88,
"learning_rate": 4.656519898391194e-05,
"loss": 0.2724,
"step": 8130
},
{
"epoch": 6.89,
"learning_rate": 4.656096528365792e-05,
"loss": 0.3747,
"step": 8140
},
{
"epoch": 6.9,
"learning_rate": 4.6556731583403896e-05,
"loss": 0.2474,
"step": 8150
},
{
"epoch": 6.91,
"learning_rate": 4.6552497883149873e-05,
"loss": 0.2911,
"step": 8160
},
{
"epoch": 6.92,
"learning_rate": 4.654826418289585e-05,
"loss": 0.3067,
"step": 8170
},
{
"epoch": 6.93,
"learning_rate": 4.654403048264183e-05,
"loss": 0.2923,
"step": 8180
},
{
"epoch": 6.93,
"learning_rate": 4.6539796782387806e-05,
"loss": 0.3718,
"step": 8190
},
{
"epoch": 6.94,
"learning_rate": 4.653556308213379e-05,
"loss": 0.2335,
"step": 8200
},
{
"epoch": 6.95,
"learning_rate": 4.653132938187976e-05,
"loss": 0.2691,
"step": 8210
},
{
"epoch": 6.96,
"learning_rate": 4.6527095681625746e-05,
"loss": 0.2714,
"step": 8220
},
{
"epoch": 6.97,
"learning_rate": 4.6522861981371724e-05,
"loss": 0.318,
"step": 8230
},
{
"epoch": 6.98,
"learning_rate": 4.65186282811177e-05,
"loss": 0.2996,
"step": 8240
},
{
"epoch": 6.99,
"learning_rate": 4.651439458086368e-05,
"loss": 0.2504,
"step": 8250
},
{
"epoch": 6.99,
"learning_rate": 4.6510160880609656e-05,
"loss": 0.3216,
"step": 8260
},
{
"epoch": 7.0,
"eval_cer": 0.12025027969999585,
"eval_loss": 0.6333013772964478,
"eval_mer": 0.3283679460693704,
"eval_runtime": 28.1996,
"eval_samples_per_second": 77.235,
"eval_steps_per_second": 4.858,
"step": 8267
},
{
"epoch": 7.0,
"learning_rate": 4.6505927180355634e-05,
"loss": 0.4237,
"step": 8270
},
{
"epoch": 7.01,
"learning_rate": 4.650169348010161e-05,
"loss": 0.211,
"step": 8280
},
{
"epoch": 7.02,
"learning_rate": 4.649745977984759e-05,
"loss": 0.2188,
"step": 8290
},
{
"epoch": 7.03,
"learning_rate": 4.649322607959357e-05,
"loss": 0.2625,
"step": 8300
},
{
"epoch": 7.04,
"learning_rate": 4.6488992379339545e-05,
"loss": 0.2426,
"step": 8310
},
{
"epoch": 7.04,
"learning_rate": 4.648475867908552e-05,
"loss": 0.3566,
"step": 8320
},
{
"epoch": 7.05,
"learning_rate": 4.64805249788315e-05,
"loss": 0.2016,
"step": 8330
},
{
"epoch": 7.06,
"learning_rate": 4.647629127857748e-05,
"loss": 0.2261,
"step": 8340
},
{
"epoch": 7.07,
"learning_rate": 4.6472057578323455e-05,
"loss": 0.2851,
"step": 8350
},
{
"epoch": 7.08,
"learning_rate": 4.646782387806943e-05,
"loss": 0.2439,
"step": 8360
},
{
"epoch": 7.09,
"learning_rate": 4.646359017781542e-05,
"loss": 0.2758,
"step": 8370
},
{
"epoch": 7.1,
"learning_rate": 4.645935647756139e-05,
"loss": 0.2368,
"step": 8380
},
{
"epoch": 7.1,
"learning_rate": 4.645512277730737e-05,
"loss": 0.2133,
"step": 8390
},
{
"epoch": 7.11,
"learning_rate": 4.645088907705334e-05,
"loss": 0.2877,
"step": 8400
},
{
"epoch": 7.12,
"learning_rate": 4.644665537679933e-05,
"loss": 0.3397,
"step": 8410
},
{
"epoch": 7.13,
"learning_rate": 4.64424216765453e-05,
"loss": 0.2874,
"step": 8420
},
{
"epoch": 7.14,
"learning_rate": 4.643818797629128e-05,
"loss": 0.2254,
"step": 8430
},
{
"epoch": 7.15,
"learning_rate": 4.643395427603726e-05,
"loss": 0.2248,
"step": 8440
},
{
"epoch": 7.15,
"learning_rate": 4.642972057578324e-05,
"loss": 0.3419,
"step": 8450
},
{
"epoch": 7.16,
"learning_rate": 4.6425486875529216e-05,
"loss": 0.298,
"step": 8460
},
{
"epoch": 7.17,
"learning_rate": 4.642125317527519e-05,
"loss": 0.3014,
"step": 8470
},
{
"epoch": 7.18,
"learning_rate": 4.641701947502117e-05,
"loss": 0.2305,
"step": 8480
},
{
"epoch": 7.19,
"learning_rate": 4.641278577476715e-05,
"loss": 0.2746,
"step": 8490
},
{
"epoch": 7.2,
"learning_rate": 4.6408552074513126e-05,
"loss": 0.2775,
"step": 8500
},
{
"epoch": 7.21,
"learning_rate": 4.6404318374259104e-05,
"loss": 0.2395,
"step": 8510
},
{
"epoch": 7.21,
"learning_rate": 4.6400508044030486e-05,
"loss": 0.3842,
"step": 8520
},
{
"epoch": 7.22,
"learning_rate": 4.6396274343776464e-05,
"loss": 0.2504,
"step": 8530
},
{
"epoch": 7.23,
"learning_rate": 4.639204064352244e-05,
"loss": 0.2445,
"step": 8540
},
{
"epoch": 7.24,
"learning_rate": 4.638780694326842e-05,
"loss": 0.3471,
"step": 8550
},
{
"epoch": 7.25,
"learning_rate": 4.63835732430144e-05,
"loss": 0.2972,
"step": 8560
},
{
"epoch": 7.26,
"learning_rate": 4.6379339542760374e-05,
"loss": 0.3146,
"step": 8570
},
{
"epoch": 7.27,
"learning_rate": 4.637510584250635e-05,
"loss": 0.2575,
"step": 8580
},
{
"epoch": 7.27,
"learning_rate": 4.637087214225233e-05,
"loss": 0.2692,
"step": 8590
},
{
"epoch": 7.28,
"learning_rate": 4.6366638441998314e-05,
"loss": 0.3045,
"step": 8600
},
{
"epoch": 7.29,
"learning_rate": 4.6362404741744285e-05,
"loss": 0.3537,
"step": 8610
},
{
"epoch": 7.3,
"learning_rate": 4.635817104149027e-05,
"loss": 0.3779,
"step": 8620
},
{
"epoch": 7.31,
"learning_rate": 4.635393734123624e-05,
"loss": 0.2749,
"step": 8630
},
{
"epoch": 7.32,
"learning_rate": 4.6349703640982224e-05,
"loss": 0.2731,
"step": 8640
},
{
"epoch": 7.32,
"learning_rate": 4.6345469940728195e-05,
"loss": 0.3241,
"step": 8650
},
{
"epoch": 7.33,
"learning_rate": 4.634123624047418e-05,
"loss": 0.3026,
"step": 8660
},
{
"epoch": 7.34,
"learning_rate": 4.633700254022015e-05,
"loss": 0.3854,
"step": 8670
},
{
"epoch": 7.35,
"learning_rate": 4.6332768839966135e-05,
"loss": 0.3231,
"step": 8680
},
{
"epoch": 7.36,
"learning_rate": 4.632853513971211e-05,
"loss": 0.3154,
"step": 8690
},
{
"epoch": 7.37,
"learning_rate": 4.632430143945809e-05,
"loss": 0.3299,
"step": 8700
},
{
"epoch": 7.38,
"learning_rate": 4.632006773920407e-05,
"loss": 0.3534,
"step": 8710
},
{
"epoch": 7.38,
"learning_rate": 4.6316257408975444e-05,
"loss": 0.3238,
"step": 8720
},
{
"epoch": 7.39,
"learning_rate": 4.631202370872143e-05,
"loss": 0.2916,
"step": 8730
},
{
"epoch": 7.4,
"learning_rate": 4.6307790008467406e-05,
"loss": 0.2876,
"step": 8740
},
{
"epoch": 7.41,
"learning_rate": 4.630355630821338e-05,
"loss": 0.3361,
"step": 8750
},
{
"epoch": 7.42,
"learning_rate": 4.629932260795936e-05,
"loss": 0.412,
"step": 8760
},
{
"epoch": 7.43,
"learning_rate": 4.629508890770534e-05,
"loss": 0.4966,
"step": 8770
},
{
"epoch": 7.43,
"learning_rate": 4.6290855207451316e-05,
"loss": 0.3582,
"step": 8780
},
{
"epoch": 7.44,
"learning_rate": 4.6286621507197294e-05,
"loss": 0.3851,
"step": 8790
},
{
"epoch": 7.45,
"learning_rate": 4.628238780694327e-05,
"loss": 0.4106,
"step": 8800
},
{
"epoch": 7.46,
"learning_rate": 4.627815410668925e-05,
"loss": 0.3664,
"step": 8810
},
{
"epoch": 7.47,
"learning_rate": 4.6273920406435226e-05,
"loss": 0.411,
"step": 8820
},
{
"epoch": 7.48,
"learning_rate": 4.6269686706181204e-05,
"loss": 0.3432,
"step": 8830
},
{
"epoch": 7.49,
"learning_rate": 4.626545300592718e-05,
"loss": 0.3722,
"step": 8840
},
{
"epoch": 7.49,
"learning_rate": 4.626121930567316e-05,
"loss": 0.3891,
"step": 8850
},
{
"epoch": 7.5,
"learning_rate": 4.625698560541914e-05,
"loss": 0.4339,
"step": 8860
},
{
"epoch": 7.51,
"learning_rate": 4.6252751905165115e-05,
"loss": 0.4677,
"step": 8870
},
{
"epoch": 7.52,
"learning_rate": 4.624851820491109e-05,
"loss": 0.4089,
"step": 8880
},
{
"epoch": 7.53,
"learning_rate": 4.624428450465707e-05,
"loss": 0.3883,
"step": 8890
},
{
"epoch": 7.54,
"learning_rate": 4.6240050804403054e-05,
"loss": 0.4075,
"step": 8900
},
{
"epoch": 7.54,
"learning_rate": 4.6235817104149025e-05,
"loss": 0.4336,
"step": 8910
},
{
"epoch": 7.55,
"learning_rate": 4.623158340389501e-05,
"loss": 0.404,
"step": 8920
},
{
"epoch": 7.56,
"learning_rate": 4.622734970364098e-05,
"loss": 0.3651,
"step": 8930
},
{
"epoch": 7.57,
"learning_rate": 4.6223116003386965e-05,
"loss": 0.3829,
"step": 8940
},
{
"epoch": 7.58,
"learning_rate": 4.6218882303132936e-05,
"loss": 0.4203,
"step": 8950
},
{
"epoch": 7.59,
"learning_rate": 4.621464860287892e-05,
"loss": 0.4489,
"step": 8960
},
{
"epoch": 7.6,
"learning_rate": 4.62104149026249e-05,
"loss": 0.5394,
"step": 8970
},
{
"epoch": 7.6,
"learning_rate": 4.6206181202370875e-05,
"loss": 0.4269,
"step": 8980
},
{
"epoch": 7.61,
"learning_rate": 4.620194750211685e-05,
"loss": 0.3924,
"step": 8990
},
{
"epoch": 7.62,
"learning_rate": 4.619771380186283e-05,
"loss": 0.3954,
"step": 9000
},
{
"epoch": 7.63,
"learning_rate": 4.619348010160881e-05,
"loss": 0.4603,
"step": 9010
},
{
"epoch": 7.64,
"learning_rate": 4.6189246401354786e-05,
"loss": 0.5261,
"step": 9020
},
{
"epoch": 7.65,
"learning_rate": 4.618501270110076e-05,
"loss": 0.38,
"step": 9030
},
{
"epoch": 7.65,
"learning_rate": 4.618077900084674e-05,
"loss": 0.411,
"step": 9040
},
{
"epoch": 7.66,
"learning_rate": 4.617654530059272e-05,
"loss": 0.4202,
"step": 9050
},
{
"epoch": 7.67,
"learning_rate": 4.6172311600338696e-05,
"loss": 0.4901,
"step": 9060
},
{
"epoch": 7.68,
"learning_rate": 4.616807790008468e-05,
"loss": 0.515,
"step": 9070
},
{
"epoch": 7.69,
"learning_rate": 4.616384419983065e-05,
"loss": 0.3756,
"step": 9080
},
{
"epoch": 7.7,
"learning_rate": 4.6159610499576636e-05,
"loss": 0.4191,
"step": 9090
},
{
"epoch": 7.71,
"learning_rate": 4.6155376799322607e-05,
"loss": 0.4276,
"step": 9100
},
{
"epoch": 7.71,
"learning_rate": 4.615114309906859e-05,
"loss": 0.4326,
"step": 9110
},
{
"epoch": 7.72,
"learning_rate": 4.614690939881456e-05,
"loss": 0.6135,
"step": 9120
},
{
"epoch": 7.73,
"learning_rate": 4.6142675698560546e-05,
"loss": 0.4903,
"step": 9130
},
{
"epoch": 7.74,
"learning_rate": 4.613844199830652e-05,
"loss": 0.4487,
"step": 9140
},
{
"epoch": 7.75,
"learning_rate": 4.61342082980525e-05,
"loss": 0.4621,
"step": 9150
},
{
"epoch": 7.76,
"learning_rate": 4.612997459779847e-05,
"loss": 0.4555,
"step": 9160
},
{
"epoch": 7.76,
"learning_rate": 4.612574089754446e-05,
"loss": 0.4764,
"step": 9170
},
{
"epoch": 7.77,
"learning_rate": 4.6121507197290434e-05,
"loss": 0.3884,
"step": 9180
},
{
"epoch": 7.78,
"learning_rate": 4.611727349703641e-05,
"loss": 0.4318,
"step": 9190
},
{
"epoch": 7.79,
"learning_rate": 4.611303979678239e-05,
"loss": 0.4756,
"step": 9200
},
{
"epoch": 7.8,
"learning_rate": 4.610880609652837e-05,
"loss": 0.4576,
"step": 9210
},
{
"epoch": 7.81,
"learning_rate": 4.6104572396274345e-05,
"loss": 0.5186,
"step": 9220
},
{
"epoch": 7.82,
"learning_rate": 4.610033869602032e-05,
"loss": 0.4117,
"step": 9230
},
{
"epoch": 7.82,
"learning_rate": 4.609610499576631e-05,
"loss": 0.4868,
"step": 9240
},
{
"epoch": 7.83,
"learning_rate": 4.609187129551228e-05,
"loss": 0.5135,
"step": 9250
},
{
"epoch": 7.84,
"learning_rate": 4.608763759525826e-05,
"loss": 0.5431,
"step": 9260
},
{
"epoch": 7.85,
"learning_rate": 4.608340389500423e-05,
"loss": 0.5298,
"step": 9270
},
{
"epoch": 7.86,
"learning_rate": 4.607917019475022e-05,
"loss": 0.464,
"step": 9280
},
{
"epoch": 7.87,
"learning_rate": 4.607493649449619e-05,
"loss": 0.4526,
"step": 9290
},
{
"epoch": 7.87,
"learning_rate": 4.607070279424217e-05,
"loss": 0.4326,
"step": 9300
},
{
"epoch": 7.88,
"learning_rate": 4.606646909398814e-05,
"loss": 0.5202,
"step": 9310
},
{
"epoch": 7.89,
"learning_rate": 4.606223539373413e-05,
"loss": 0.5257,
"step": 9320
},
{
"epoch": 7.9,
"learning_rate": 4.60580016934801e-05,
"loss": 0.398,
"step": 9330
},
{
"epoch": 7.91,
"learning_rate": 4.605376799322608e-05,
"loss": 0.4207,
"step": 9340
},
{
"epoch": 7.92,
"learning_rate": 4.604953429297206e-05,
"loss": 0.4808,
"step": 9350
},
{
"epoch": 7.93,
"learning_rate": 4.604530059271804e-05,
"loss": 0.4916,
"step": 9360
},
{
"epoch": 7.93,
"learning_rate": 4.6041066892464016e-05,
"loss": 0.514,
"step": 9370
},
{
"epoch": 7.94,
"learning_rate": 4.6036833192209993e-05,
"loss": 0.4566,
"step": 9380
},
{
"epoch": 7.95,
"learning_rate": 4.603259949195598e-05,
"loss": 0.4996,
"step": 9390
},
{
"epoch": 7.96,
"learning_rate": 4.602836579170195e-05,
"loss": 0.5469,
"step": 9400
},
{
"epoch": 7.97,
"learning_rate": 4.602413209144793e-05,
"loss": 0.5833,
"step": 9410
},
{
"epoch": 7.98,
"learning_rate": 4.6019898391193904e-05,
"loss": 0.5203,
"step": 9420
},
{
"epoch": 7.98,
"learning_rate": 4.601566469093989e-05,
"loss": 0.4557,
"step": 9430
},
{
"epoch": 7.99,
"learning_rate": 4.601143099068586e-05,
"loss": 0.4486,
"step": 9440
},
{
"epoch": 8.0,
"eval_cer": 0.12729457589193222,
"eval_loss": 0.6445448398590088,
"eval_mer": 0.3521256931608133,
"eval_runtime": 24.3275,
"eval_samples_per_second": 89.528,
"eval_steps_per_second": 5.631,
"step": 9448
},
{
"epoch": 8.0,
"learning_rate": 4.6007197290431844e-05,
"loss": 0.4439,
"step": 9450
},
{
"epoch": 8.01,
"learning_rate": 4.6002963590177814e-05,
"loss": 0.416,
"step": 9460
},
{
"epoch": 8.02,
"learning_rate": 4.59987298899238e-05,
"loss": 0.4454,
"step": 9470
},
{
"epoch": 8.03,
"learning_rate": 4.599449618966977e-05,
"loss": 0.4962,
"step": 9480
},
{
"epoch": 8.04,
"learning_rate": 4.5990262489415754e-05,
"loss": 0.5252,
"step": 9490
},
{
"epoch": 8.04,
"learning_rate": 4.5986028789161725e-05,
"loss": 0.5922,
"step": 9500
},
{
"epoch": 8.05,
"learning_rate": 4.598179508890771e-05,
"loss": 0.4395,
"step": 9510
},
{
"epoch": 8.06,
"learning_rate": 4.597756138865369e-05,
"loss": 0.4726,
"step": 9520
},
{
"epoch": 8.07,
"learning_rate": 4.5973327688399664e-05,
"loss": 0.4738,
"step": 9530
},
{
"epoch": 8.08,
"learning_rate": 4.596909398814564e-05,
"loss": 0.5014,
"step": 9540
},
{
"epoch": 8.09,
"learning_rate": 4.5965283657917025e-05,
"loss": 0.4696,
"step": 9550
},
{
"epoch": 8.09,
"learning_rate": 4.5961049957663e-05,
"loss": 0.4771,
"step": 9560
},
{
"epoch": 8.1,
"learning_rate": 4.595681625740898e-05,
"loss": 0.4974,
"step": 9570
},
{
"epoch": 8.11,
"learning_rate": 4.595258255715496e-05,
"loss": 0.4569,
"step": 9580
},
{
"epoch": 8.12,
"learning_rate": 4.5948348856900935e-05,
"loss": 0.4682,
"step": 9590
},
{
"epoch": 8.13,
"learning_rate": 4.594411515664691e-05,
"loss": 0.5194,
"step": 9600
},
{
"epoch": 8.14,
"learning_rate": 4.593988145639289e-05,
"loss": 0.5246,
"step": 9610
},
{
"epoch": 8.15,
"learning_rate": 4.593564775613887e-05,
"loss": 0.4881,
"step": 9620
},
{
"epoch": 8.15,
"learning_rate": 4.5931414055884846e-05,
"loss": 0.5594,
"step": 9630
},
{
"epoch": 8.16,
"learning_rate": 4.592718035563082e-05,
"loss": 0.5379,
"step": 9640
},
{
"epoch": 8.17,
"learning_rate": 4.59229466553768e-05,
"loss": 0.5215,
"step": 9650
},
{
"epoch": 8.18,
"learning_rate": 4.591871295512278e-05,
"loss": 0.5857,
"step": 9660
},
{
"epoch": 8.19,
"learning_rate": 4.5914479254868756e-05,
"loss": 0.6261,
"step": 9670
},
{
"epoch": 8.2,
"learning_rate": 4.5910245554614734e-05,
"loss": 0.653,
"step": 9680
},
{
"epoch": 8.2,
"learning_rate": 4.590601185436071e-05,
"loss": 0.6165,
"step": 9690
},
{
"epoch": 8.21,
"learning_rate": 4.590177815410669e-05,
"loss": 0.6577,
"step": 9700
},
{
"epoch": 8.22,
"learning_rate": 4.589754445385267e-05,
"loss": 0.6304,
"step": 9710
},
{
"epoch": 8.23,
"learning_rate": 4.5893310753598644e-05,
"loss": 0.6166,
"step": 9720
},
{
"epoch": 8.24,
"learning_rate": 4.588907705334463e-05,
"loss": 0.6116,
"step": 9730
},
{
"epoch": 8.25,
"learning_rate": 4.58848433530906e-05,
"loss": 0.6094,
"step": 9740
},
{
"epoch": 8.26,
"learning_rate": 4.5880609652836584e-05,
"loss": 0.6384,
"step": 9750
},
{
"epoch": 8.26,
"learning_rate": 4.587637595258256e-05,
"loss": 0.6939,
"step": 9760
},
{
"epoch": 8.27,
"learning_rate": 4.587214225232854e-05,
"loss": 0.6348,
"step": 9770
},
{
"epoch": 8.28,
"learning_rate": 4.586790855207452e-05,
"loss": 0.6529,
"step": 9780
},
{
"epoch": 8.29,
"learning_rate": 4.5863674851820494e-05,
"loss": 0.6509,
"step": 9790
},
{
"epoch": 8.3,
"learning_rate": 4.585944115156647e-05,
"loss": 0.7135,
"step": 9800
},
{
"epoch": 8.31,
"learning_rate": 4.585520745131245e-05,
"loss": 0.6971,
"step": 9810
},
{
"epoch": 8.31,
"learning_rate": 4.585097375105843e-05,
"loss": 0.6923,
"step": 9820
},
{
"epoch": 8.32,
"learning_rate": 4.5846740050804405e-05,
"loss": 0.703,
"step": 9830
},
{
"epoch": 8.33,
"learning_rate": 4.584250635055038e-05,
"loss": 0.7819,
"step": 9840
},
{
"epoch": 8.34,
"learning_rate": 4.583827265029636e-05,
"loss": 0.8499,
"step": 9850
},
{
"epoch": 8.35,
"learning_rate": 4.583403895004234e-05,
"loss": 0.7978,
"step": 9860
},
{
"epoch": 8.36,
"learning_rate": 4.5829805249788315e-05,
"loss": 0.8724,
"step": 9870
},
{
"epoch": 8.37,
"learning_rate": 4.58255715495343e-05,
"loss": 0.8589,
"step": 9880
},
{
"epoch": 8.37,
"learning_rate": 4.582133784928027e-05,
"loss": 0.8798,
"step": 9890
},
{
"epoch": 8.38,
"learning_rate": 4.5817104149026255e-05,
"loss": 0.8424,
"step": 9900
},
{
"epoch": 8.39,
"learning_rate": 4.5812870448772226e-05,
"loss": 0.8478,
"step": 9910
},
{
"epoch": 8.4,
"learning_rate": 4.580863674851821e-05,
"loss": 0.8032,
"step": 9920
},
{
"epoch": 8.41,
"learning_rate": 4.580440304826418e-05,
"loss": 0.8084,
"step": 9930
},
{
"epoch": 8.42,
"learning_rate": 4.5800169348010165e-05,
"loss": 0.8081,
"step": 9940
},
{
"epoch": 8.43,
"learning_rate": 4.5795935647756136e-05,
"loss": 0.8007,
"step": 9950
},
{
"epoch": 8.43,
"learning_rate": 4.579170194750212e-05,
"loss": 0.838,
"step": 9960
},
{
"epoch": 8.44,
"learning_rate": 4.57874682472481e-05,
"loss": 0.8209,
"step": 9970
},
{
"epoch": 8.45,
"learning_rate": 4.5783234546994076e-05,
"loss": 0.8775,
"step": 9980
},
{
"epoch": 8.46,
"learning_rate": 4.577900084674005e-05,
"loss": 0.8605,
"step": 9990
},
{
"epoch": 8.47,
"learning_rate": 4.577476714648603e-05,
"loss": 0.8787,
"step": 10000
},
{
"epoch": 8.48,
"learning_rate": 4.577053344623201e-05,
"loss": 0.8102,
"step": 10010
},
{
"epoch": 8.48,
"learning_rate": 4.5766299745977986e-05,
"loss": 0.8055,
"step": 10020
},
{
"epoch": 8.49,
"learning_rate": 4.5762066045723964e-05,
"loss": 0.8361,
"step": 10030
},
{
"epoch": 8.5,
"learning_rate": 4.575783234546994e-05,
"loss": 0.8053,
"step": 10040
},
{
"epoch": 8.51,
"learning_rate": 4.5753598645215926e-05,
"loss": 0.8641,
"step": 10050
},
{
"epoch": 8.52,
"learning_rate": 4.57493649449619e-05,
"loss": 0.8012,
"step": 10060
},
{
"epoch": 8.53,
"learning_rate": 4.574513124470788e-05,
"loss": 0.7983,
"step": 10070
},
{
"epoch": 8.54,
"learning_rate": 4.574089754445385e-05,
"loss": 0.8566,
"step": 10080
},
{
"epoch": 8.54,
"learning_rate": 4.5736663844199836e-05,
"loss": 0.8913,
"step": 10090
},
{
"epoch": 8.55,
"learning_rate": 4.573243014394581e-05,
"loss": 0.9211,
"step": 10100
},
{
"epoch": 8.56,
"learning_rate": 4.572819644369179e-05,
"loss": 0.8904,
"step": 10110
},
{
"epoch": 8.57,
"learning_rate": 4.572396274343776e-05,
"loss": 0.9624,
"step": 10120
},
{
"epoch": 8.58,
"learning_rate": 4.571972904318375e-05,
"loss": 0.9416,
"step": 10130
},
{
"epoch": 8.59,
"learning_rate": 4.571549534292972e-05,
"loss": 0.9605,
"step": 10140
},
{
"epoch": 8.59,
"learning_rate": 4.57112616426757e-05,
"loss": 0.9472,
"step": 10150
},
{
"epoch": 8.6,
"learning_rate": 4.570702794242168e-05,
"loss": 0.9249,
"step": 10160
},
{
"epoch": 8.61,
"learning_rate": 4.570279424216766e-05,
"loss": 0.9371,
"step": 10170
},
{
"epoch": 8.62,
"learning_rate": 4.5698560541913635e-05,
"loss": 1.0134,
"step": 10180
},
{
"epoch": 8.63,
"learning_rate": 4.569432684165961e-05,
"loss": 1.0054,
"step": 10190
},
{
"epoch": 8.64,
"learning_rate": 4.569009314140559e-05,
"loss": 0.9591,
"step": 10200
},
{
"epoch": 8.65,
"learning_rate": 4.568585944115157e-05,
"loss": 1.004,
"step": 10210
},
{
"epoch": 8.65,
"learning_rate": 4.568162574089755e-05,
"loss": 0.9699,
"step": 10220
},
{
"epoch": 8.66,
"learning_rate": 4.567739204064352e-05,
"loss": 0.9819,
"step": 10230
},
{
"epoch": 8.67,
"learning_rate": 4.567315834038951e-05,
"loss": 0.9193,
"step": 10240
},
{
"epoch": 8.68,
"learning_rate": 4.566892464013548e-05,
"loss": 1.008,
"step": 10250
},
{
"epoch": 8.69,
"learning_rate": 4.566469093988146e-05,
"loss": 1.0062,
"step": 10260
},
{
"epoch": 8.7,
"learning_rate": 4.5660457239627433e-05,
"loss": 0.9635,
"step": 10270
},
{
"epoch": 8.7,
"learning_rate": 4.565622353937342e-05,
"loss": 0.956,
"step": 10280
},
{
"epoch": 8.71,
"learning_rate": 4.565198983911939e-05,
"loss": 0.9055,
"step": 10290
},
{
"epoch": 8.72,
"learning_rate": 4.564775613886537e-05,
"loss": 0.9496,
"step": 10300
},
{
"epoch": 8.73,
"learning_rate": 4.5643522438611344e-05,
"loss": 0.9319,
"step": 10310
},
{
"epoch": 8.74,
"learning_rate": 4.563928873835733e-05,
"loss": 0.874,
"step": 10320
},
{
"epoch": 8.75,
"learning_rate": 4.5635055038103306e-05,
"loss": 0.9066,
"step": 10330
},
{
"epoch": 8.76,
"learning_rate": 4.5630821337849284e-05,
"loss": 0.9554,
"step": 10340
},
{
"epoch": 8.76,
"learning_rate": 4.562658763759526e-05,
"loss": 0.9359,
"step": 10350
},
{
"epoch": 8.77,
"learning_rate": 4.562235393734124e-05,
"loss": 0.9179,
"step": 10360
},
{
"epoch": 8.78,
"learning_rate": 4.5618120237087216e-05,
"loss": 0.8902,
"step": 10370
},
{
"epoch": 8.79,
"learning_rate": 4.5613886536833194e-05,
"loss": 0.9614,
"step": 10380
},
{
"epoch": 8.8,
"learning_rate": 4.560965283657917e-05,
"loss": 0.8877,
"step": 10390
},
{
"epoch": 8.81,
"learning_rate": 4.560541913632515e-05,
"loss": 0.8941,
"step": 10400
},
{
"epoch": 8.81,
"learning_rate": 4.5601185436071134e-05,
"loss": 0.9478,
"step": 10410
},
{
"epoch": 8.82,
"learning_rate": 4.5596951735817104e-05,
"loss": 0.9337,
"step": 10420
},
{
"epoch": 8.83,
"learning_rate": 4.559271803556309e-05,
"loss": 0.9109,
"step": 10430
},
{
"epoch": 8.84,
"learning_rate": 4.558848433530906e-05,
"loss": 0.9358,
"step": 10440
},
{
"epoch": 8.85,
"learning_rate": 4.558467400508044e-05,
"loss": 0.9414,
"step": 10450
},
{
"epoch": 8.86,
"learning_rate": 4.558044030482642e-05,
"loss": 0.9808,
"step": 10460
},
{
"epoch": 8.87,
"learning_rate": 4.55762066045724e-05,
"loss": 0.9936,
"step": 10470
},
{
"epoch": 8.87,
"learning_rate": 4.5571972904318375e-05,
"loss": 0.9651,
"step": 10480
},
{
"epoch": 8.88,
"learning_rate": 4.556773920406435e-05,
"loss": 1.0022,
"step": 10490
},
{
"epoch": 8.89,
"learning_rate": 4.556350550381033e-05,
"loss": 0.9491,
"step": 10500
},
{
"epoch": 8.9,
"learning_rate": 4.555927180355631e-05,
"loss": 1.0564,
"step": 10510
},
{
"epoch": 8.91,
"learning_rate": 4.5555038103302286e-05,
"loss": 0.9787,
"step": 10520
},
{
"epoch": 8.92,
"learning_rate": 4.555080440304826e-05,
"loss": 0.9995,
"step": 10530
},
{
"epoch": 8.92,
"learning_rate": 4.554657070279425e-05,
"loss": 0.9776,
"step": 10540
},
{
"epoch": 8.93,
"learning_rate": 4.554233700254022e-05,
"loss": 0.8444,
"step": 10550
},
{
"epoch": 8.94,
"learning_rate": 4.55381033022862e-05,
"loss": 0.9355,
"step": 10560
},
{
"epoch": 8.95,
"learning_rate": 4.553386960203218e-05,
"loss": 0.9401,
"step": 10570
},
{
"epoch": 8.96,
"learning_rate": 4.552963590177816e-05,
"loss": 0.9758,
"step": 10580
},
{
"epoch": 8.97,
"learning_rate": 4.5525402201524136e-05,
"loss": 0.9573,
"step": 10590
},
{
"epoch": 8.98,
"learning_rate": 4.552116850127011e-05,
"loss": 0.9769,
"step": 10600
},
{
"epoch": 8.98,
"learning_rate": 4.551693480101609e-05,
"loss": 1.0239,
"step": 10610
},
{
"epoch": 8.99,
"learning_rate": 4.551270110076207e-05,
"loss": 0.959,
"step": 10620
},
{
"epoch": 9.0,
"eval_cer": 0.1727233801571845,
"eval_loss": 0.9587426781654358,
"eval_mer": 0.48635424594976623,
"eval_runtime": 23.7449,
"eval_samples_per_second": 91.725,
"eval_steps_per_second": 5.77,
"step": 10629
},
{
"epoch": 9.0,
"learning_rate": 4.5508467400508046e-05,
"loss": 1.0417,
"step": 10630
},
{
"epoch": 9.01,
"learning_rate": 4.5504233700254024e-05,
"loss": 1.0691,
"step": 10640
},
{
"epoch": 9.02,
"learning_rate": 4.55e-05,
"loss": 1.0545,
"step": 10650
},
{
"epoch": 9.03,
"learning_rate": 4.549576629974598e-05,
"loss": 1.169,
"step": 10660
},
{
"epoch": 9.03,
"learning_rate": 4.549153259949196e-05,
"loss": 1.1724,
"step": 10670
},
{
"epoch": 9.04,
"learning_rate": 4.5487298899237934e-05,
"loss": 1.1514,
"step": 10680
},
{
"epoch": 9.05,
"learning_rate": 4.548306519898391e-05,
"loss": 1.2339,
"step": 10690
},
{
"epoch": 9.06,
"learning_rate": 4.547883149872989e-05,
"loss": 1.2019,
"step": 10700
},
{
"epoch": 9.07,
"learning_rate": 4.5474597798475874e-05,
"loss": 1.2867,
"step": 10710
},
{
"epoch": 9.08,
"learning_rate": 4.5470364098221845e-05,
"loss": 1.2397,
"step": 10720
},
{
"epoch": 9.09,
"learning_rate": 4.546613039796783e-05,
"loss": 1.2359,
"step": 10730
},
{
"epoch": 9.09,
"learning_rate": 4.54618966977138e-05,
"loss": 1.2958,
"step": 10740
},
{
"epoch": 9.1,
"learning_rate": 4.5457662997459784e-05,
"loss": 1.2138,
"step": 10750
},
{
"epoch": 9.11,
"learning_rate": 4.5453429297205755e-05,
"loss": 1.2546,
"step": 10760
},
{
"epoch": 9.12,
"learning_rate": 4.544919559695174e-05,
"loss": 1.2149,
"step": 10770
},
{
"epoch": 9.13,
"learning_rate": 4.544496189669772e-05,
"loss": 1.2046,
"step": 10780
},
{
"epoch": 9.14,
"learning_rate": 4.5440728196443695e-05,
"loss": 1.2594,
"step": 10790
},
{
"epoch": 9.14,
"learning_rate": 4.543649449618967e-05,
"loss": 1.2634,
"step": 10800
},
{
"epoch": 9.15,
"learning_rate": 4.543226079593565e-05,
"loss": 1.2014,
"step": 10810
},
{
"epoch": 9.16,
"learning_rate": 4.542802709568163e-05,
"loss": 1.1851,
"step": 10820
},
{
"epoch": 9.17,
"learning_rate": 4.5423793395427605e-05,
"loss": 1.1358,
"step": 10830
},
{
"epoch": 9.18,
"learning_rate": 4.541955969517358e-05,
"loss": 1.2123,
"step": 10840
},
{
"epoch": 9.19,
"learning_rate": 4.541532599491956e-05,
"loss": 1.1453,
"step": 10850
},
{
"epoch": 9.2,
"learning_rate": 4.541109229466554e-05,
"loss": 1.1274,
"step": 10860
},
{
"epoch": 9.2,
"learning_rate": 4.5406858594411516e-05,
"loss": 1.0398,
"step": 10870
},
{
"epoch": 9.21,
"learning_rate": 4.54026248941575e-05,
"loss": 1.0025,
"step": 10880
},
{
"epoch": 9.22,
"learning_rate": 4.539839119390347e-05,
"loss": 1.0871,
"step": 10890
},
{
"epoch": 9.23,
"learning_rate": 4.5394157493649455e-05,
"loss": 1.1312,
"step": 10900
},
{
"epoch": 9.24,
"learning_rate": 4.5389923793395426e-05,
"loss": 1.0552,
"step": 10910
},
{
"epoch": 9.25,
"learning_rate": 4.538569009314141e-05,
"loss": 1.0676,
"step": 10920
},
{
"epoch": 9.25,
"learning_rate": 4.538145639288738e-05,
"loss": 0.9928,
"step": 10930
},
{
"epoch": 9.26,
"learning_rate": 4.5377222692633366e-05,
"loss": 1.1595,
"step": 10940
},
{
"epoch": 9.27,
"learning_rate": 4.537298899237934e-05,
"loss": 1.187,
"step": 10950
},
{
"epoch": 9.28,
"learning_rate": 4.536875529212532e-05,
"loss": 1.1794,
"step": 10960
},
{
"epoch": 9.29,
"learning_rate": 4.536452159187129e-05,
"loss": 1.2095,
"step": 10970
},
{
"epoch": 9.3,
"learning_rate": 4.5360287891617276e-05,
"loss": 1.1446,
"step": 10980
},
{
"epoch": 9.31,
"learning_rate": 4.5356054191363254e-05,
"loss": 1.2508,
"step": 10990
},
{
"epoch": 9.31,
"learning_rate": 4.535182049110923e-05,
"loss": 1.2618,
"step": 11000
},
{
"epoch": 9.32,
"learning_rate": 4.534758679085521e-05,
"loss": 1.2381,
"step": 11010
},
{
"epoch": 9.33,
"learning_rate": 4.534335309060119e-05,
"loss": 1.2213,
"step": 11020
},
{
"epoch": 9.34,
"learning_rate": 4.5339119390347164e-05,
"loss": 1.0845,
"step": 11030
},
{
"epoch": 9.35,
"learning_rate": 4.533488569009314e-05,
"loss": 1.2409,
"step": 11040
},
{
"epoch": 9.36,
"learning_rate": 4.5330651989839126e-05,
"loss": 1.2389,
"step": 11050
},
{
"epoch": 9.36,
"learning_rate": 4.53264182895851e-05,
"loss": 1.2307,
"step": 11060
},
{
"epoch": 9.37,
"learning_rate": 4.532218458933108e-05,
"loss": 1.223,
"step": 11070
},
{
"epoch": 9.38,
"learning_rate": 4.531795088907705e-05,
"loss": 1.1984,
"step": 11080
},
{
"epoch": 9.39,
"learning_rate": 4.531371718882304e-05,
"loss": 1.2973,
"step": 11090
},
{
"epoch": 9.4,
"learning_rate": 4.530948348856901e-05,
"loss": 1.2701,
"step": 11100
},
{
"epoch": 9.41,
"learning_rate": 4.530524978831499e-05,
"loss": 1.2675,
"step": 11110
},
{
"epoch": 9.42,
"learning_rate": 4.530101608806096e-05,
"loss": 1.228,
"step": 11120
},
{
"epoch": 9.42,
"learning_rate": 4.529678238780695e-05,
"loss": 1.1648,
"step": 11130
},
{
"epoch": 9.43,
"learning_rate": 4.529254868755292e-05,
"loss": 1.2387,
"step": 11140
},
{
"epoch": 9.44,
"learning_rate": 4.52883149872989e-05,
"loss": 1.2762,
"step": 11150
},
{
"epoch": 9.45,
"learning_rate": 4.528408128704488e-05,
"loss": 1.2244,
"step": 11160
},
{
"epoch": 9.46,
"learning_rate": 4.527984758679086e-05,
"loss": 1.2044,
"step": 11170
},
{
"epoch": 9.47,
"learning_rate": 4.5275613886536836e-05,
"loss": 1.2404,
"step": 11180
},
{
"epoch": 9.48,
"learning_rate": 4.527138018628281e-05,
"loss": 1.3162,
"step": 11190
},
{
"epoch": 9.48,
"learning_rate": 4.526714648602879e-05,
"loss": 1.3218,
"step": 11200
},
{
"epoch": 9.49,
"learning_rate": 4.526291278577477e-05,
"loss": 1.3126,
"step": 11210
},
{
"epoch": 9.5,
"learning_rate": 4.525867908552075e-05,
"loss": 1.322,
"step": 11220
},
{
"epoch": 9.51,
"learning_rate": 4.5254445385266724e-05,
"loss": 1.2755,
"step": 11230
},
{
"epoch": 9.52,
"learning_rate": 4.525021168501271e-05,
"loss": 1.3486,
"step": 11240
},
{
"epoch": 9.53,
"learning_rate": 4.524597798475868e-05,
"loss": 1.3322,
"step": 11250
},
{
"epoch": 9.53,
"learning_rate": 4.524174428450466e-05,
"loss": 1.3325,
"step": 11260
},
{
"epoch": 9.54,
"learning_rate": 4.5237510584250634e-05,
"loss": 1.2634,
"step": 11270
},
{
"epoch": 9.55,
"learning_rate": 4.523327688399662e-05,
"loss": 1.2853,
"step": 11280
},
{
"epoch": 9.56,
"learning_rate": 4.522904318374259e-05,
"loss": 1.3415,
"step": 11290
},
{
"epoch": 9.57,
"learning_rate": 4.5224809483488574e-05,
"loss": 1.3399,
"step": 11300
},
{
"epoch": 9.58,
"learning_rate": 4.5220575783234545e-05,
"loss": 1.364,
"step": 11310
},
{
"epoch": 9.59,
"learning_rate": 4.521634208298053e-05,
"loss": 1.3059,
"step": 11320
},
{
"epoch": 9.59,
"learning_rate": 4.5212108382726507e-05,
"loss": 1.281,
"step": 11330
},
{
"epoch": 9.6,
"learning_rate": 4.5207874682472484e-05,
"loss": 1.3479,
"step": 11340
},
{
"epoch": 9.61,
"learning_rate": 4.520364098221846e-05,
"loss": 1.3463,
"step": 11350
},
{
"epoch": 9.62,
"learning_rate": 4.519940728196444e-05,
"loss": 1.3277,
"step": 11360
},
{
"epoch": 9.63,
"learning_rate": 4.519517358171042e-05,
"loss": 1.3,
"step": 11370
},
{
"epoch": 9.64,
"learning_rate": 4.5190939881456395e-05,
"loss": 1.2584,
"step": 11380
},
{
"epoch": 9.64,
"learning_rate": 4.518670618120237e-05,
"loss": 1.3805,
"step": 11390
},
{
"epoch": 9.65,
"learning_rate": 4.518247248094835e-05,
"loss": 1.3249,
"step": 11400
},
{
"epoch": 9.66,
"learning_rate": 4.517823878069433e-05,
"loss": 1.3617,
"step": 11410
},
{
"epoch": 9.67,
"learning_rate": 4.5174005080440305e-05,
"loss": 1.4079,
"step": 11420
},
{
"epoch": 9.68,
"learning_rate": 4.516977138018629e-05,
"loss": 1.2564,
"step": 11430
},
{
"epoch": 9.69,
"learning_rate": 4.516553767993226e-05,
"loss": 1.357,
"step": 11440
},
{
"epoch": 9.7,
"learning_rate": 4.5161303979678245e-05,
"loss": 1.3412,
"step": 11450
},
{
"epoch": 9.7,
"learning_rate": 4.5157070279424216e-05,
"loss": 1.339,
"step": 11460
},
{
"epoch": 9.71,
"learning_rate": 4.51528365791702e-05,
"loss": 1.355,
"step": 11470
},
{
"epoch": 9.72,
"learning_rate": 4.514860287891617e-05,
"loss": 1.3243,
"step": 11480
},
{
"epoch": 9.73,
"learning_rate": 4.5144369178662155e-05,
"loss": 1.3956,
"step": 11490
},
{
"epoch": 9.74,
"learning_rate": 4.514013547840813e-05,
"loss": 1.4206,
"step": 11500
},
{
"epoch": 9.75,
"learning_rate": 4.513590177815411e-05,
"loss": 1.3762,
"step": 11510
},
{
"epoch": 9.75,
"learning_rate": 4.513166807790009e-05,
"loss": 1.3001,
"step": 11520
},
{
"epoch": 9.76,
"learning_rate": 4.5127434377646066e-05,
"loss": 1.3047,
"step": 11530
},
{
"epoch": 9.77,
"learning_rate": 4.512320067739204e-05,
"loss": 1.44,
"step": 11540
},
{
"epoch": 9.78,
"learning_rate": 4.511896697713802e-05,
"loss": 1.4547,
"step": 11550
},
{
"epoch": 9.79,
"learning_rate": 4.5114733276884e-05,
"loss": 1.3994,
"step": 11560
},
{
"epoch": 9.8,
"learning_rate": 4.5110499576629976e-05,
"loss": 1.4244,
"step": 11570
},
{
"epoch": 9.81,
"learning_rate": 4.510668924640136e-05,
"loss": 1.2827,
"step": 11580
},
{
"epoch": 9.81,
"learning_rate": 4.5102455546147336e-05,
"loss": 1.5738,
"step": 11590
},
{
"epoch": 9.82,
"learning_rate": 4.5098221845893314e-05,
"loss": 1.5026,
"step": 11600
},
{
"epoch": 9.83,
"learning_rate": 4.509398814563929e-05,
"loss": 1.5072,
"step": 11610
},
{
"epoch": 9.84,
"learning_rate": 4.508975444538527e-05,
"loss": 1.5236,
"step": 11620
},
{
"epoch": 9.85,
"learning_rate": 4.508552074513125e-05,
"loss": 1.4025,
"step": 11630
},
{
"epoch": 9.86,
"learning_rate": 4.5081287044877224e-05,
"loss": 1.4741,
"step": 11640
},
{
"epoch": 9.86,
"learning_rate": 4.50770533446232e-05,
"loss": 1.4503,
"step": 11650
},
{
"epoch": 9.87,
"learning_rate": 4.507281964436918e-05,
"loss": 1.4235,
"step": 11660
},
{
"epoch": 9.88,
"learning_rate": 4.506858594411516e-05,
"loss": 1.397,
"step": 11670
},
{
"epoch": 9.89,
"learning_rate": 4.5064352243861135e-05,
"loss": 1.3583,
"step": 11680
},
{
"epoch": 9.9,
"learning_rate": 4.506011854360712e-05,
"loss": 1.4387,
"step": 11690
},
{
"epoch": 9.91,
"learning_rate": 4.505588484335309e-05,
"loss": 1.4338,
"step": 11700
},
{
"epoch": 9.92,
"learning_rate": 4.5051651143099075e-05,
"loss": 1.4273,
"step": 11710
},
{
"epoch": 9.92,
"learning_rate": 4.5047417442845045e-05,
"loss": 1.4278,
"step": 11720
},
{
"epoch": 9.93,
"learning_rate": 4.504318374259103e-05,
"loss": 1.4686,
"step": 11730
},
{
"epoch": 9.94,
"learning_rate": 4.5038950042337e-05,
"loss": 1.4621,
"step": 11740
},
{
"epoch": 9.95,
"learning_rate": 4.5034716342082985e-05,
"loss": 1.5398,
"step": 11750
},
{
"epoch": 9.96,
"learning_rate": 4.5030482641828956e-05,
"loss": 1.4768,
"step": 11760
},
{
"epoch": 9.97,
"learning_rate": 4.502624894157494e-05,
"loss": 1.4836,
"step": 11770
},
{
"epoch": 9.97,
"learning_rate": 4.5022438611346316e-05,
"loss": 1.479,
"step": 11780
},
{
"epoch": 9.98,
"learning_rate": 4.50182049110923e-05,
"loss": 1.5087,
"step": 11790
},
{
"epoch": 9.99,
"learning_rate": 4.501397121083827e-05,
"loss": 1.4942,
"step": 11800
},
{
"epoch": 10.0,
"learning_rate": 4.5009737510584256e-05,
"loss": 1.5324,
"step": 11810
},
{
"epoch": 10.0,
"eval_cer": 0.24601168524427133,
"eval_loss": 1.2874042987823486,
"eval_mer": 0.66684788517995,
"eval_runtime": 28.2038,
"eval_samples_per_second": 77.224,
"eval_steps_per_second": 4.857,
"step": 11810
},
{
"epoch": 10.01,
"learning_rate": 4.5005503810330227e-05,
"loss": 1.5366,
"step": 11820
},
{
"epoch": 10.02,
"learning_rate": 4.500127011007621e-05,
"loss": 1.5396,
"step": 11830
},
{
"epoch": 10.03,
"learning_rate": 4.499703640982219e-05,
"loss": 1.541,
"step": 11840
},
{
"epoch": 10.03,
"learning_rate": 4.4992802709568166e-05,
"loss": 1.5522,
"step": 11850
},
{
"epoch": 10.04,
"learning_rate": 4.4988569009314144e-05,
"loss": 1.7348,
"step": 11860
},
{
"epoch": 10.05,
"learning_rate": 4.498433530906012e-05,
"loss": 1.5764,
"step": 11870
},
{
"epoch": 10.06,
"learning_rate": 4.49801016088061e-05,
"loss": 1.5444,
"step": 11880
},
{
"epoch": 10.07,
"learning_rate": 4.4975867908552077e-05,
"loss": 1.6274,
"step": 11890
},
{
"epoch": 10.08,
"learning_rate": 4.4971634208298054e-05,
"loss": 1.7053,
"step": 11900
},
{
"epoch": 10.08,
"learning_rate": 4.496740050804403e-05,
"loss": 1.7295,
"step": 11910
},
{
"epoch": 10.09,
"learning_rate": 4.496316680779001e-05,
"loss": 1.6501,
"step": 11920
},
{
"epoch": 10.1,
"learning_rate": 4.495893310753599e-05,
"loss": 1.7354,
"step": 11930
},
{
"epoch": 10.11,
"learning_rate": 4.4954699407281965e-05,
"loss": 1.761,
"step": 11940
},
{
"epoch": 10.12,
"learning_rate": 4.495046570702794e-05,
"loss": 1.9456,
"step": 11950
},
{
"epoch": 10.13,
"learning_rate": 4.494623200677393e-05,
"loss": 2.3037,
"step": 11960
},
{
"epoch": 10.14,
"learning_rate": 4.49419983065199e-05,
"loss": 1.8925,
"step": 11970
},
{
"epoch": 10.14,
"learning_rate": 4.493776460626588e-05,
"loss": 1.9024,
"step": 11980
},
{
"epoch": 10.15,
"learning_rate": 4.493353090601185e-05,
"loss": 1.8458,
"step": 11990
},
{
"epoch": 10.16,
"learning_rate": 4.492929720575784e-05,
"loss": 1.8047,
"step": 12000
},
{
"epoch": 10.17,
"learning_rate": 4.4925063505503815e-05,
"loss": 2.0453,
"step": 12010
},
{
"epoch": 10.18,
"learning_rate": 4.492082980524979e-05,
"loss": 1.7405,
"step": 12020
},
{
"epoch": 10.19,
"learning_rate": 4.491659610499577e-05,
"loss": 1.8958,
"step": 12030
},
{
"epoch": 10.19,
"learning_rate": 4.491236240474175e-05,
"loss": 2.0114,
"step": 12040
},
{
"epoch": 10.2,
"learning_rate": 4.4908128704487725e-05,
"loss": 1.9778,
"step": 12050
},
{
"epoch": 10.21,
"learning_rate": 4.49038950042337e-05,
"loss": 2.3987,
"step": 12060
},
{
"epoch": 10.22,
"learning_rate": 4.489966130397968e-05,
"loss": 1.948,
"step": 12070
},
{
"epoch": 10.23,
"learning_rate": 4.489542760372566e-05,
"loss": 1.8516,
"step": 12080
},
{
"epoch": 10.24,
"learning_rate": 4.4891193903471636e-05,
"loss": 2.0011,
"step": 12090
},
{
"epoch": 10.25,
"learning_rate": 4.488696020321761e-05,
"loss": 2.0376,
"step": 12100
},
{
"epoch": 10.25,
"learning_rate": 4.4883149872988996e-05,
"loss": 2.3684,
"step": 12110
},
{
"epoch": 10.26,
"learning_rate": 4.4878916172734974e-05,
"loss": 1.8877,
"step": 12120
},
{
"epoch": 10.27,
"learning_rate": 4.487468247248095e-05,
"loss": 2.0353,
"step": 12130
},
{
"epoch": 10.28,
"learning_rate": 4.487044877222693e-05,
"loss": 2.001,
"step": 12140
},
{
"epoch": 10.29,
"learning_rate": 4.4866215071972906e-05,
"loss": 2.2172,
"step": 12150
},
{
"epoch": 10.3,
"learning_rate": 4.4861981371718884e-05,
"loss": 2.6229,
"step": 12160
},
{
"epoch": 10.3,
"learning_rate": 4.485774767146486e-05,
"loss": 1.9102,
"step": 12170
},
{
"epoch": 10.31,
"learning_rate": 4.485351397121084e-05,
"loss": 1.9907,
"step": 12180
},
{
"epoch": 10.32,
"learning_rate": 4.484928027095682e-05,
"loss": 2.0939,
"step": 12190
},
{
"epoch": 10.33,
"learning_rate": 4.4845046570702794e-05,
"loss": 2.1526,
"step": 12200
},
{
"epoch": 10.34,
"learning_rate": 4.484081287044877e-05,
"loss": 2.4638,
"step": 12210
},
{
"epoch": 10.35,
"learning_rate": 4.4836579170194757e-05,
"loss": 2.013,
"step": 12220
},
{
"epoch": 10.36,
"learning_rate": 4.483234546994073e-05,
"loss": 1.9353,
"step": 12230
},
{
"epoch": 10.36,
"learning_rate": 4.482811176968671e-05,
"loss": 2.0008,
"step": 12240
},
{
"epoch": 10.37,
"learning_rate": 4.482387806943268e-05,
"loss": 2.1421,
"step": 12250
},
{
"epoch": 10.38,
"learning_rate": 4.481964436917867e-05,
"loss": 2.3906,
"step": 12260
},
{
"epoch": 10.39,
"learning_rate": 4.481541066892464e-05,
"loss": 1.9363,
"step": 12270
},
{
"epoch": 10.4,
"learning_rate": 4.481117696867062e-05,
"loss": 1.9357,
"step": 12280
},
{
"epoch": 10.41,
"learning_rate": 4.480694326841659e-05,
"loss": 2.0629,
"step": 12290
},
{
"epoch": 10.41,
"learning_rate": 4.480270956816258e-05,
"loss": 2.108,
"step": 12300
},
{
"epoch": 10.42,
"learning_rate": 4.4798475867908555e-05,
"loss": 2.5918,
"step": 12310
},
{
"epoch": 10.43,
"learning_rate": 4.479424216765453e-05,
"loss": 1.9791,
"step": 12320
},
{
"epoch": 10.44,
"learning_rate": 4.479000846740051e-05,
"loss": 1.9266,
"step": 12330
},
{
"epoch": 10.45,
"learning_rate": 4.478577476714649e-05,
"loss": 2.0217,
"step": 12340
},
{
"epoch": 10.46,
"learning_rate": 4.4781541066892466e-05,
"loss": 2.2233,
"step": 12350
},
{
"epoch": 10.47,
"learning_rate": 4.477730736663844e-05,
"loss": 2.4653,
"step": 12360
},
{
"epoch": 10.47,
"learning_rate": 4.477307366638442e-05,
"loss": 1.985,
"step": 12370
},
{
"epoch": 10.48,
"learning_rate": 4.47688399661304e-05,
"loss": 1.9862,
"step": 12380
},
{
"epoch": 10.49,
"learning_rate": 4.476460626587638e-05,
"loss": 2.0553,
"step": 12390
},
{
"epoch": 10.5,
"learning_rate": 4.4760372565622354e-05,
"loss": 2.1504,
"step": 12400
},
{
"epoch": 10.51,
"learning_rate": 4.475613886536834e-05,
"loss": 2.4646,
"step": 12410
},
{
"epoch": 10.52,
"learning_rate": 4.475190516511431e-05,
"loss": 1.9892,
"step": 12420
},
{
"epoch": 10.52,
"learning_rate": 4.474767146486029e-05,
"loss": 1.9598,
"step": 12430
},
{
"epoch": 10.53,
"learning_rate": 4.4743437764606264e-05,
"loss": 1.9647,
"step": 12440
},
{
"epoch": 10.54,
"learning_rate": 4.473920406435225e-05,
"loss": 2.1751,
"step": 12450
},
{
"epoch": 10.55,
"learning_rate": 4.473497036409822e-05,
"loss": 2.6336,
"step": 12460
},
{
"epoch": 10.56,
"learning_rate": 4.4730736663844204e-05,
"loss": 1.9717,
"step": 12470
},
{
"epoch": 10.57,
"learning_rate": 4.4726502963590175e-05,
"loss": 1.9569,
"step": 12480
},
{
"epoch": 10.58,
"learning_rate": 4.472226926333616e-05,
"loss": 2.0809,
"step": 12490
},
{
"epoch": 10.58,
"learning_rate": 4.4718035563082137e-05,
"loss": 2.2418,
"step": 12500
},
{
"epoch": 10.59,
"learning_rate": 4.4713801862828114e-05,
"loss": 2.5448,
"step": 12510
},
{
"epoch": 10.6,
"learning_rate": 4.470956816257409e-05,
"loss": 1.9617,
"step": 12520
},
{
"epoch": 10.61,
"learning_rate": 4.470533446232007e-05,
"loss": 1.9676,
"step": 12530
},
{
"epoch": 10.62,
"learning_rate": 4.4701100762066054e-05,
"loss": 2.0919,
"step": 12540
},
{
"epoch": 10.63,
"learning_rate": 4.4696867061812025e-05,
"loss": 2.2593,
"step": 12550
},
{
"epoch": 10.64,
"learning_rate": 4.469263336155801e-05,
"loss": 2.4808,
"step": 12560
},
{
"epoch": 10.64,
"learning_rate": 4.468839966130398e-05,
"loss": 2.0293,
"step": 12570
},
{
"epoch": 10.65,
"learning_rate": 4.4684165961049964e-05,
"loss": 1.9555,
"step": 12580
},
{
"epoch": 10.66,
"learning_rate": 4.4679932260795935e-05,
"loss": 2.0828,
"step": 12590
},
{
"epoch": 10.67,
"learning_rate": 4.467569856054192e-05,
"loss": 2.2231,
"step": 12600
},
{
"epoch": 10.68,
"learning_rate": 4.467146486028789e-05,
"loss": 2.6046,
"step": 12610
},
{
"epoch": 10.69,
"learning_rate": 4.4667231160033875e-05,
"loss": 1.9986,
"step": 12620
},
{
"epoch": 10.69,
"learning_rate": 4.4662997459779846e-05,
"loss": 1.9526,
"step": 12630
},
{
"epoch": 10.7,
"learning_rate": 4.465876375952583e-05,
"loss": 2.1019,
"step": 12640
},
{
"epoch": 10.71,
"learning_rate": 4.46545300592718e-05,
"loss": 2.1972,
"step": 12650
},
{
"epoch": 10.72,
"learning_rate": 4.4650296359017785e-05,
"loss": 2.6334,
"step": 12660
},
{
"epoch": 10.73,
"learning_rate": 4.464606265876376e-05,
"loss": 1.9876,
"step": 12670
},
{
"epoch": 10.74,
"learning_rate": 4.464182895850974e-05,
"loss": 1.9677,
"step": 12680
},
{
"epoch": 10.75,
"learning_rate": 4.463759525825572e-05,
"loss": 2.0477,
"step": 12690
},
{
"epoch": 10.75,
"learning_rate": 4.4633361558001696e-05,
"loss": 2.188,
"step": 12700
},
{
"epoch": 10.76,
"learning_rate": 4.462955122777308e-05,
"loss": 2.6296,
"step": 12710
},
{
"epoch": 10.77,
"learning_rate": 4.4625317527519056e-05,
"loss": 1.9847,
"step": 12720
},
{
"epoch": 10.78,
"learning_rate": 4.4621083827265034e-05,
"loss": 1.97,
"step": 12730
},
{
"epoch": 10.79,
"learning_rate": 4.461685012701101e-05,
"loss": 1.9851,
"step": 12740
},
{
"epoch": 10.8,
"learning_rate": 4.461261642675699e-05,
"loss": 2.2719,
"step": 12750
},
{
"epoch": 10.8,
"learning_rate": 4.4608382726502966e-05,
"loss": 2.4926,
"step": 12760
},
{
"epoch": 10.81,
"learning_rate": 4.4604149026248944e-05,
"loss": 1.9853,
"step": 12770
},
{
"epoch": 10.82,
"learning_rate": 4.459991532599492e-05,
"loss": 2.0494,
"step": 12780
},
{
"epoch": 10.83,
"learning_rate": 4.45956816257409e-05,
"loss": 2.0386,
"step": 12790
},
{
"epoch": 10.84,
"learning_rate": 4.459144792548688e-05,
"loss": 2.2637,
"step": 12800
},
{
"epoch": 10.85,
"learning_rate": 4.4587214225232854e-05,
"loss": 2.7218,
"step": 12810
},
{
"epoch": 10.86,
"learning_rate": 4.458298052497883e-05,
"loss": 1.9578,
"step": 12820
},
{
"epoch": 10.86,
"learning_rate": 4.457874682472481e-05,
"loss": 1.9637,
"step": 12830
},
{
"epoch": 10.87,
"learning_rate": 4.457451312447079e-05,
"loss": 2.055,
"step": 12840
},
{
"epoch": 10.88,
"learning_rate": 4.4570279424216765e-05,
"loss": 2.2645,
"step": 12850
},
{
"epoch": 10.89,
"learning_rate": 4.456604572396275e-05,
"loss": 2.7069,
"step": 12860
},
{
"epoch": 10.9,
"learning_rate": 4.456181202370872e-05,
"loss": 1.9068,
"step": 12870
},
{
"epoch": 10.91,
"learning_rate": 4.4557578323454705e-05,
"loss": 1.9396,
"step": 12880
},
{
"epoch": 10.91,
"learning_rate": 4.4553344623200675e-05,
"loss": 2.1007,
"step": 12890
},
{
"epoch": 10.92,
"learning_rate": 4.454911092294666e-05,
"loss": 2.2855,
"step": 12900
},
{
"epoch": 10.93,
"learning_rate": 4.454487722269264e-05,
"loss": 2.5381,
"step": 12910
},
{
"epoch": 10.94,
"learning_rate": 4.4540643522438615e-05,
"loss": 2.0182,
"step": 12920
},
{
"epoch": 10.95,
"learning_rate": 4.453640982218459e-05,
"loss": 2.0102,
"step": 12930
},
{
"epoch": 10.96,
"learning_rate": 4.453217612193057e-05,
"loss": 1.9872,
"step": 12940
},
{
"epoch": 10.97,
"learning_rate": 4.452794242167655e-05,
"loss": 2.0364,
"step": 12950
},
{
"epoch": 10.97,
"learning_rate": 4.4523708721422526e-05,
"loss": 2.5555,
"step": 12960
},
{
"epoch": 10.98,
"learning_rate": 4.45194750211685e-05,
"loss": 2.0951,
"step": 12970
},
{
"epoch": 10.99,
"learning_rate": 4.451524132091448e-05,
"loss": 2.1552,
"step": 12980
},
{
"epoch": 11.0,
"learning_rate": 4.451100762066046e-05,
"loss": 2.4549,
"step": 12990
},
{
"epoch": 11.0,
"eval_cer": 0.3198386718048592,
"eval_loss": 1.721533179283142,
"eval_mer": 0.6885397412199631,
"eval_runtime": 27.2327,
"eval_samples_per_second": 79.977,
"eval_steps_per_second": 5.031,
"step": 12991
}
],
"max_steps": 118100,
"num_train_epochs": 100,
"total_flos": 1.4197154668742943e+19,
"trial_name": null,
"trial_params": null
}