Splend1dchan's picture
add trainer_state
cf94878
{
"best_metric": 1.0490120649337769,
"best_model_checkpoint": "/work/twskvfb446/facebook/wav2vec2-large-lv60_voidful/phoneme_byt5_SpeechMixEEDT5_w2v2-large_t5lephone-small_bs256/checkpoint-7868",
"epoch": 13.998445480790584,
"global_step": 7868,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 4.000000000000001e-06,
"loss": 4.9943,
"step": 10
},
{
"epoch": 0.04,
"learning_rate": 8.000000000000001e-06,
"loss": 4.4785,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 1.2e-05,
"loss": 4.191,
"step": 30
},
{
"epoch": 0.07,
"learning_rate": 1.6000000000000003e-05,
"loss": 3.9363,
"step": 40
},
{
"epoch": 0.09,
"learning_rate": 2e-05,
"loss": 3.6209,
"step": 50
},
{
"epoch": 0.11,
"learning_rate": 2.4e-05,
"loss": 3.3071,
"step": 60
},
{
"epoch": 0.12,
"learning_rate": 2.8e-05,
"loss": 3.0288,
"step": 70
},
{
"epoch": 0.14,
"learning_rate": 3.2000000000000005e-05,
"loss": 2.844,
"step": 80
},
{
"epoch": 0.16,
"learning_rate": 3.6e-05,
"loss": 2.7064,
"step": 90
},
{
"epoch": 0.18,
"learning_rate": 4e-05,
"loss": 2.5839,
"step": 100
},
{
"epoch": 0.2,
"learning_rate": 3.997613365155132e-05,
"loss": 2.505,
"step": 110
},
{
"epoch": 0.21,
"learning_rate": 3.995226730310263e-05,
"loss": 2.4443,
"step": 120
},
{
"epoch": 0.23,
"learning_rate": 3.992840095465394e-05,
"loss": 2.3822,
"step": 130
},
{
"epoch": 0.25,
"learning_rate": 3.9904534606205255e-05,
"loss": 2.2834,
"step": 140
},
{
"epoch": 0.27,
"learning_rate": 3.988066825775656e-05,
"loss": 2.1641,
"step": 150
},
{
"epoch": 0.28,
"learning_rate": 3.985680190930788e-05,
"loss": 2.092,
"step": 160
},
{
"epoch": 0.3,
"learning_rate": 3.983293556085919e-05,
"loss": 2.025,
"step": 170
},
{
"epoch": 0.32,
"learning_rate": 3.9809069212410506e-05,
"loss": 1.9894,
"step": 180
},
{
"epoch": 0.34,
"learning_rate": 3.978520286396182e-05,
"loss": 1.9449,
"step": 190
},
{
"epoch": 0.36,
"learning_rate": 3.976133651551313e-05,
"loss": 1.9187,
"step": 200
},
{
"epoch": 0.37,
"learning_rate": 3.973747016706444e-05,
"loss": 1.8856,
"step": 210
},
{
"epoch": 0.39,
"learning_rate": 3.971360381861575e-05,
"loss": 1.8647,
"step": 220
},
{
"epoch": 0.41,
"learning_rate": 3.9689737470167066e-05,
"loss": 1.8505,
"step": 230
},
{
"epoch": 0.43,
"learning_rate": 3.966587112171838e-05,
"loss": 1.8383,
"step": 240
},
{
"epoch": 0.44,
"learning_rate": 3.9642004773269695e-05,
"loss": 1.8252,
"step": 250
},
{
"epoch": 0.46,
"learning_rate": 3.961813842482101e-05,
"loss": 1.8093,
"step": 260
},
{
"epoch": 0.48,
"learning_rate": 3.959427207637232e-05,
"loss": 1.7959,
"step": 270
},
{
"epoch": 0.5,
"learning_rate": 3.957040572792363e-05,
"loss": 1.7828,
"step": 280
},
{
"epoch": 0.52,
"learning_rate": 3.954653937947494e-05,
"loss": 1.7763,
"step": 290
},
{
"epoch": 0.53,
"learning_rate": 3.9522673031026254e-05,
"loss": 1.7738,
"step": 300
},
{
"epoch": 0.55,
"learning_rate": 3.949880668257757e-05,
"loss": 1.7619,
"step": 310
},
{
"epoch": 0.57,
"learning_rate": 3.947494033412888e-05,
"loss": 1.7543,
"step": 320
},
{
"epoch": 0.59,
"learning_rate": 3.94510739856802e-05,
"loss": 1.7457,
"step": 330
},
{
"epoch": 0.6,
"learning_rate": 3.9427207637231506e-05,
"loss": 1.74,
"step": 340
},
{
"epoch": 0.62,
"learning_rate": 3.940334128878282e-05,
"loss": 1.7339,
"step": 350
},
{
"epoch": 0.64,
"learning_rate": 3.937947494033413e-05,
"loss": 1.7328,
"step": 360
},
{
"epoch": 0.66,
"learning_rate": 3.935560859188544e-05,
"loss": 1.7171,
"step": 370
},
{
"epoch": 0.68,
"learning_rate": 3.933174224343676e-05,
"loss": 1.713,
"step": 380
},
{
"epoch": 0.69,
"learning_rate": 3.930787589498807e-05,
"loss": 1.7056,
"step": 390
},
{
"epoch": 0.71,
"learning_rate": 3.9284009546539386e-05,
"loss": 1.704,
"step": 400
},
{
"epoch": 0.73,
"learning_rate": 3.9260143198090694e-05,
"loss": 1.6972,
"step": 410
},
{
"epoch": 0.75,
"learning_rate": 3.923627684964201e-05,
"loss": 1.695,
"step": 420
},
{
"epoch": 0.76,
"learning_rate": 3.9212410501193317e-05,
"loss": 1.6824,
"step": 430
},
{
"epoch": 0.78,
"learning_rate": 3.918854415274463e-05,
"loss": 1.6853,
"step": 440
},
{
"epoch": 0.8,
"learning_rate": 3.9164677804295946e-05,
"loss": 1.6796,
"step": 450
},
{
"epoch": 0.82,
"learning_rate": 3.914081145584726e-05,
"loss": 1.6829,
"step": 460
},
{
"epoch": 0.83,
"learning_rate": 3.9116945107398575e-05,
"loss": 1.6773,
"step": 470
},
{
"epoch": 0.85,
"learning_rate": 3.909307875894988e-05,
"loss": 1.672,
"step": 480
},
{
"epoch": 0.87,
"learning_rate": 3.90692124105012e-05,
"loss": 1.6617,
"step": 490
},
{
"epoch": 0.89,
"learning_rate": 3.9045346062052505e-05,
"loss": 1.6661,
"step": 500
},
{
"epoch": 0.91,
"learning_rate": 3.902147971360382e-05,
"loss": 1.66,
"step": 510
},
{
"epoch": 0.92,
"learning_rate": 3.8997613365155134e-05,
"loss": 1.6593,
"step": 520
},
{
"epoch": 0.94,
"learning_rate": 3.897374701670645e-05,
"loss": 1.6523,
"step": 530
},
{
"epoch": 0.96,
"learning_rate": 3.894988066825776e-05,
"loss": 1.6483,
"step": 540
},
{
"epoch": 0.98,
"learning_rate": 3.892601431980907e-05,
"loss": 1.6415,
"step": 550
},
{
"epoch": 0.99,
"learning_rate": 3.8902147971360386e-05,
"loss": 1.6417,
"step": 560
},
{
"epoch": 1.0,
"eval_cer": 4.8418803418803416,
"eval_loss": 1.442280650138855,
"eval_runtime": 4.8026,
"eval_samples_per_second": 2.082,
"eval_steps_per_second": 0.208,
"eval_wer": 1.1428571428571428,
"step": 562
},
{
"epoch": 1.01,
"learning_rate": 3.8878281622911693e-05,
"loss": 1.7786,
"step": 570
},
{
"epoch": 1.03,
"learning_rate": 3.885441527446301e-05,
"loss": 1.6225,
"step": 580
},
{
"epoch": 1.05,
"learning_rate": 3.883054892601432e-05,
"loss": 1.6146,
"step": 590
},
{
"epoch": 1.07,
"learning_rate": 3.880668257756564e-05,
"loss": 1.6069,
"step": 600
},
{
"epoch": 1.09,
"learning_rate": 3.878281622911695e-05,
"loss": 1.5974,
"step": 610
},
{
"epoch": 1.1,
"learning_rate": 3.875894988066826e-05,
"loss": 1.5803,
"step": 620
},
{
"epoch": 1.12,
"learning_rate": 3.8735083532219574e-05,
"loss": 1.5726,
"step": 630
},
{
"epoch": 1.14,
"learning_rate": 3.871121718377089e-05,
"loss": 1.564,
"step": 640
},
{
"epoch": 1.16,
"learning_rate": 3.8687350835322197e-05,
"loss": 1.5567,
"step": 650
},
{
"epoch": 1.17,
"learning_rate": 3.866348448687351e-05,
"loss": 1.556,
"step": 660
},
{
"epoch": 1.19,
"learning_rate": 3.8639618138424826e-05,
"loss": 1.5439,
"step": 670
},
{
"epoch": 1.21,
"learning_rate": 3.861575178997614e-05,
"loss": 1.5427,
"step": 680
},
{
"epoch": 1.23,
"learning_rate": 3.859188544152745e-05,
"loss": 1.5358,
"step": 690
},
{
"epoch": 1.25,
"learning_rate": 3.856801909307876e-05,
"loss": 1.5316,
"step": 700
},
{
"epoch": 1.26,
"learning_rate": 3.854415274463008e-05,
"loss": 1.5197,
"step": 710
},
{
"epoch": 1.28,
"learning_rate": 3.8520286396181385e-05,
"loss": 1.5193,
"step": 720
},
{
"epoch": 1.3,
"learning_rate": 3.84964200477327e-05,
"loss": 1.5078,
"step": 730
},
{
"epoch": 1.32,
"learning_rate": 3.8472553699284014e-05,
"loss": 1.5033,
"step": 740
},
{
"epoch": 1.33,
"learning_rate": 3.844868735083532e-05,
"loss": 1.4989,
"step": 750
},
{
"epoch": 1.35,
"learning_rate": 3.8424821002386637e-05,
"loss": 1.4983,
"step": 760
},
{
"epoch": 1.37,
"learning_rate": 3.840095465393795e-05,
"loss": 1.4935,
"step": 770
},
{
"epoch": 1.39,
"learning_rate": 3.8377088305489266e-05,
"loss": 1.4875,
"step": 780
},
{
"epoch": 1.41,
"learning_rate": 3.8353221957040573e-05,
"loss": 1.4827,
"step": 790
},
{
"epoch": 1.42,
"learning_rate": 3.832935560859189e-05,
"loss": 1.4844,
"step": 800
},
{
"epoch": 1.44,
"learning_rate": 3.83054892601432e-05,
"loss": 1.4736,
"step": 810
},
{
"epoch": 1.46,
"learning_rate": 3.828162291169451e-05,
"loss": 1.4708,
"step": 820
},
{
"epoch": 1.48,
"learning_rate": 3.8257756563245825e-05,
"loss": 1.4708,
"step": 830
},
{
"epoch": 1.49,
"learning_rate": 3.823389021479714e-05,
"loss": 1.467,
"step": 840
},
{
"epoch": 1.51,
"learning_rate": 3.8210023866348454e-05,
"loss": 1.4656,
"step": 850
},
{
"epoch": 1.53,
"learning_rate": 3.818615751789976e-05,
"loss": 1.4631,
"step": 860
},
{
"epoch": 1.55,
"learning_rate": 3.8162291169451077e-05,
"loss": 1.4568,
"step": 870
},
{
"epoch": 1.56,
"learning_rate": 3.813842482100239e-05,
"loss": 1.4552,
"step": 880
},
{
"epoch": 1.58,
"learning_rate": 3.81145584725537e-05,
"loss": 1.4493,
"step": 890
},
{
"epoch": 1.6,
"learning_rate": 3.8090692124105013e-05,
"loss": 1.4449,
"step": 900
},
{
"epoch": 1.62,
"learning_rate": 3.806682577565633e-05,
"loss": 1.447,
"step": 910
},
{
"epoch": 1.64,
"learning_rate": 3.804295942720764e-05,
"loss": 1.4419,
"step": 920
},
{
"epoch": 1.65,
"learning_rate": 3.801909307875896e-05,
"loss": 1.4418,
"step": 930
},
{
"epoch": 1.67,
"learning_rate": 3.7995226730310265e-05,
"loss": 1.4414,
"step": 940
},
{
"epoch": 1.69,
"learning_rate": 3.797136038186158e-05,
"loss": 1.4353,
"step": 950
},
{
"epoch": 1.71,
"learning_rate": 3.794749403341289e-05,
"loss": 1.429,
"step": 960
},
{
"epoch": 1.72,
"learning_rate": 3.79236276849642e-05,
"loss": 1.4349,
"step": 970
},
{
"epoch": 1.74,
"learning_rate": 3.7899761336515517e-05,
"loss": 1.4319,
"step": 980
},
{
"epoch": 1.76,
"learning_rate": 3.787589498806683e-05,
"loss": 1.4227,
"step": 990
},
{
"epoch": 1.78,
"learning_rate": 3.7852028639618146e-05,
"loss": 1.4191,
"step": 1000
},
{
"epoch": 1.8,
"learning_rate": 3.7828162291169453e-05,
"loss": 1.4207,
"step": 1010
},
{
"epoch": 1.81,
"learning_rate": 3.780429594272077e-05,
"loss": 1.4152,
"step": 1020
},
{
"epoch": 1.83,
"learning_rate": 3.7780429594272076e-05,
"loss": 1.4132,
"step": 1030
},
{
"epoch": 1.85,
"learning_rate": 3.775656324582339e-05,
"loss": 1.4123,
"step": 1040
},
{
"epoch": 1.87,
"learning_rate": 3.7732696897374705e-05,
"loss": 1.407,
"step": 1050
},
{
"epoch": 1.88,
"learning_rate": 3.770883054892602e-05,
"loss": 1.4065,
"step": 1060
},
{
"epoch": 1.9,
"learning_rate": 3.7684964200477334e-05,
"loss": 1.4069,
"step": 1070
},
{
"epoch": 1.92,
"learning_rate": 3.766109785202864e-05,
"loss": 1.4129,
"step": 1080
},
{
"epoch": 1.94,
"learning_rate": 3.7637231503579957e-05,
"loss": 1.4008,
"step": 1090
},
{
"epoch": 1.96,
"learning_rate": 3.7613365155131264e-05,
"loss": 1.3963,
"step": 1100
},
{
"epoch": 1.97,
"learning_rate": 3.758949880668258e-05,
"loss": 1.3955,
"step": 1110
},
{
"epoch": 1.99,
"learning_rate": 3.7565632458233893e-05,
"loss": 1.3941,
"step": 1120
},
{
"epoch": 2.0,
"eval_cer": 4.794871794871795,
"eval_loss": 1.2559969425201416,
"eval_runtime": 4.8324,
"eval_samples_per_second": 2.069,
"eval_steps_per_second": 0.207,
"eval_wer": 1.0649350649350648,
"step": 1124
},
{
"epoch": 2.01,
"learning_rate": 3.754176610978521e-05,
"loss": 1.5184,
"step": 1130
},
{
"epoch": 2.03,
"learning_rate": 3.751789976133652e-05,
"loss": 1.3922,
"step": 1140
},
{
"epoch": 2.05,
"learning_rate": 3.749403341288783e-05,
"loss": 1.3902,
"step": 1150
},
{
"epoch": 2.06,
"learning_rate": 3.7470167064439145e-05,
"loss": 1.3891,
"step": 1160
},
{
"epoch": 2.08,
"learning_rate": 3.744630071599045e-05,
"loss": 1.387,
"step": 1170
},
{
"epoch": 2.1,
"learning_rate": 3.742243436754177e-05,
"loss": 1.377,
"step": 1180
},
{
"epoch": 2.12,
"learning_rate": 3.739856801909308e-05,
"loss": 1.3807,
"step": 1190
},
{
"epoch": 2.14,
"learning_rate": 3.7374701670644397e-05,
"loss": 1.3814,
"step": 1200
},
{
"epoch": 2.15,
"learning_rate": 3.735083532219571e-05,
"loss": 1.3826,
"step": 1210
},
{
"epoch": 2.17,
"learning_rate": 3.732696897374702e-05,
"loss": 1.3762,
"step": 1220
},
{
"epoch": 2.19,
"learning_rate": 3.7303102625298333e-05,
"loss": 1.375,
"step": 1230
},
{
"epoch": 2.21,
"learning_rate": 3.727923627684964e-05,
"loss": 1.3777,
"step": 1240
},
{
"epoch": 2.22,
"learning_rate": 3.7255369928400956e-05,
"loss": 1.3676,
"step": 1250
},
{
"epoch": 2.24,
"learning_rate": 3.723150357995227e-05,
"loss": 1.3678,
"step": 1260
},
{
"epoch": 2.26,
"learning_rate": 3.7207637231503585e-05,
"loss": 1.3743,
"step": 1270
},
{
"epoch": 2.28,
"learning_rate": 3.71837708830549e-05,
"loss": 1.3717,
"step": 1280
},
{
"epoch": 2.29,
"learning_rate": 3.715990453460621e-05,
"loss": 1.3657,
"step": 1290
},
{
"epoch": 2.31,
"learning_rate": 3.713603818615752e-05,
"loss": 1.3609,
"step": 1300
},
{
"epoch": 2.33,
"learning_rate": 3.711217183770883e-05,
"loss": 1.3649,
"step": 1310
},
{
"epoch": 2.35,
"learning_rate": 3.7088305489260144e-05,
"loss": 1.3633,
"step": 1320
},
{
"epoch": 2.37,
"learning_rate": 3.706443914081146e-05,
"loss": 1.3576,
"step": 1330
},
{
"epoch": 2.38,
"learning_rate": 3.7040572792362773e-05,
"loss": 1.3635,
"step": 1340
},
{
"epoch": 2.4,
"learning_rate": 3.701670644391409e-05,
"loss": 1.3583,
"step": 1350
},
{
"epoch": 2.42,
"learning_rate": 3.6992840095465396e-05,
"loss": 1.354,
"step": 1360
},
{
"epoch": 2.44,
"learning_rate": 3.696897374701671e-05,
"loss": 1.3562,
"step": 1370
},
{
"epoch": 2.45,
"learning_rate": 3.694510739856802e-05,
"loss": 1.3581,
"step": 1380
},
{
"epoch": 2.47,
"learning_rate": 3.692124105011933e-05,
"loss": 1.3536,
"step": 1390
},
{
"epoch": 2.49,
"learning_rate": 3.689737470167065e-05,
"loss": 1.3502,
"step": 1400
},
{
"epoch": 2.51,
"learning_rate": 3.687350835322196e-05,
"loss": 1.3484,
"step": 1410
},
{
"epoch": 2.53,
"learning_rate": 3.6849642004773277e-05,
"loss": 1.3437,
"step": 1420
},
{
"epoch": 2.54,
"learning_rate": 3.6825775656324584e-05,
"loss": 1.3521,
"step": 1430
},
{
"epoch": 2.56,
"learning_rate": 3.68019093078759e-05,
"loss": 1.3466,
"step": 1440
},
{
"epoch": 2.58,
"learning_rate": 3.677804295942721e-05,
"loss": 1.3434,
"step": 1450
},
{
"epoch": 2.6,
"learning_rate": 3.675417661097852e-05,
"loss": 1.3409,
"step": 1460
},
{
"epoch": 2.61,
"learning_rate": 3.6730310262529836e-05,
"loss": 1.3444,
"step": 1470
},
{
"epoch": 2.63,
"learning_rate": 3.670644391408115e-05,
"loss": 1.3463,
"step": 1480
},
{
"epoch": 2.65,
"learning_rate": 3.6682577565632465e-05,
"loss": 1.3402,
"step": 1490
},
{
"epoch": 2.67,
"learning_rate": 3.665871121718377e-05,
"loss": 1.3381,
"step": 1500
},
{
"epoch": 2.69,
"learning_rate": 3.663484486873509e-05,
"loss": 1.3344,
"step": 1510
},
{
"epoch": 2.7,
"learning_rate": 3.6610978520286395e-05,
"loss": 1.3362,
"step": 1520
},
{
"epoch": 2.72,
"learning_rate": 3.658711217183771e-05,
"loss": 1.34,
"step": 1530
},
{
"epoch": 2.74,
"learning_rate": 3.6563245823389024e-05,
"loss": 1.3396,
"step": 1540
},
{
"epoch": 2.76,
"learning_rate": 3.653937947494034e-05,
"loss": 1.3375,
"step": 1550
},
{
"epoch": 2.77,
"learning_rate": 3.6515513126491654e-05,
"loss": 1.3297,
"step": 1560
},
{
"epoch": 2.79,
"learning_rate": 3.649164677804296e-05,
"loss": 1.3236,
"step": 1570
},
{
"epoch": 2.81,
"learning_rate": 3.6467780429594276e-05,
"loss": 1.3289,
"step": 1580
},
{
"epoch": 2.83,
"learning_rate": 3.6443914081145584e-05,
"loss": 1.3264,
"step": 1590
},
{
"epoch": 2.85,
"learning_rate": 3.64200477326969e-05,
"loss": 1.3268,
"step": 1600
},
{
"epoch": 2.86,
"learning_rate": 3.639618138424821e-05,
"loss": 1.3243,
"step": 1610
},
{
"epoch": 2.88,
"learning_rate": 3.637231503579953e-05,
"loss": 1.3308,
"step": 1620
},
{
"epoch": 2.9,
"learning_rate": 3.634844868735084e-05,
"loss": 1.3245,
"step": 1630
},
{
"epoch": 2.92,
"learning_rate": 3.632458233890215e-05,
"loss": 1.321,
"step": 1640
},
{
"epoch": 2.93,
"learning_rate": 3.6300715990453464e-05,
"loss": 1.3189,
"step": 1650
},
{
"epoch": 2.95,
"learning_rate": 3.627684964200477e-05,
"loss": 1.3241,
"step": 1660
},
{
"epoch": 2.97,
"learning_rate": 3.625298329355609e-05,
"loss": 1.3239,
"step": 1670
},
{
"epoch": 2.99,
"learning_rate": 3.62291169451074e-05,
"loss": 1.323,
"step": 1680
},
{
"epoch": 3.0,
"eval_cer": 4.784188034188034,
"eval_loss": 1.2157490253448486,
"eval_runtime": 4.5874,
"eval_samples_per_second": 2.18,
"eval_steps_per_second": 0.218,
"eval_wer": 1.0129870129870129,
"step": 1686
},
{
"epoch": 3.01,
"learning_rate": 3.6205250596658716e-05,
"loss": 1.4321,
"step": 1690
},
{
"epoch": 3.02,
"learning_rate": 3.618138424821003e-05,
"loss": 1.3181,
"step": 1700
},
{
"epoch": 3.04,
"learning_rate": 3.615751789976134e-05,
"loss": 1.3191,
"step": 1710
},
{
"epoch": 3.06,
"learning_rate": 3.613365155131265e-05,
"loss": 1.3159,
"step": 1720
},
{
"epoch": 3.08,
"learning_rate": 3.610978520286396e-05,
"loss": 1.3134,
"step": 1730
},
{
"epoch": 3.1,
"learning_rate": 3.6085918854415275e-05,
"loss": 1.3112,
"step": 1740
},
{
"epoch": 3.11,
"learning_rate": 3.606205250596659e-05,
"loss": 1.3127,
"step": 1750
},
{
"epoch": 3.13,
"learning_rate": 3.6038186157517904e-05,
"loss": 1.3097,
"step": 1760
},
{
"epoch": 3.15,
"learning_rate": 3.601431980906922e-05,
"loss": 1.3095,
"step": 1770
},
{
"epoch": 3.17,
"learning_rate": 3.599045346062053e-05,
"loss": 1.3081,
"step": 1780
},
{
"epoch": 3.18,
"learning_rate": 3.596658711217184e-05,
"loss": 1.3117,
"step": 1790
},
{
"epoch": 3.2,
"learning_rate": 3.594272076372315e-05,
"loss": 1.3107,
"step": 1800
},
{
"epoch": 3.22,
"learning_rate": 3.5918854415274464e-05,
"loss": 1.3031,
"step": 1810
},
{
"epoch": 3.24,
"learning_rate": 3.589498806682578e-05,
"loss": 1.3074,
"step": 1820
},
{
"epoch": 3.26,
"learning_rate": 3.587112171837709e-05,
"loss": 1.3048,
"step": 1830
},
{
"epoch": 3.27,
"learning_rate": 3.584725536992841e-05,
"loss": 1.3014,
"step": 1840
},
{
"epoch": 3.29,
"learning_rate": 3.5823389021479715e-05,
"loss": 1.3064,
"step": 1850
},
{
"epoch": 3.31,
"learning_rate": 3.579952267303103e-05,
"loss": 1.3072,
"step": 1860
},
{
"epoch": 3.33,
"learning_rate": 3.577565632458234e-05,
"loss": 1.301,
"step": 1870
},
{
"epoch": 3.34,
"learning_rate": 3.575178997613365e-05,
"loss": 1.2968,
"step": 1880
},
{
"epoch": 3.36,
"learning_rate": 3.572792362768497e-05,
"loss": 1.2954,
"step": 1890
},
{
"epoch": 3.38,
"learning_rate": 3.570405727923628e-05,
"loss": 1.2983,
"step": 1900
},
{
"epoch": 3.4,
"learning_rate": 3.5680190930787596e-05,
"loss": 1.2948,
"step": 1910
},
{
"epoch": 3.42,
"learning_rate": 3.5656324582338904e-05,
"loss": 1.2944,
"step": 1920
},
{
"epoch": 3.43,
"learning_rate": 3.563245823389022e-05,
"loss": 1.2964,
"step": 1930
},
{
"epoch": 3.45,
"learning_rate": 3.560859188544153e-05,
"loss": 1.2975,
"step": 1940
},
{
"epoch": 3.47,
"learning_rate": 3.558472553699284e-05,
"loss": 1.2959,
"step": 1950
},
{
"epoch": 3.49,
"learning_rate": 3.5560859188544155e-05,
"loss": 1.2894,
"step": 1960
},
{
"epoch": 3.5,
"learning_rate": 3.553699284009547e-05,
"loss": 1.2868,
"step": 1970
},
{
"epoch": 3.52,
"learning_rate": 3.5513126491646784e-05,
"loss": 1.3002,
"step": 1980
},
{
"epoch": 3.54,
"learning_rate": 3.548926014319809e-05,
"loss": 1.2964,
"step": 1990
},
{
"epoch": 3.56,
"learning_rate": 3.546539379474941e-05,
"loss": 1.2967,
"step": 2000
},
{
"epoch": 3.58,
"learning_rate": 3.544152744630072e-05,
"loss": 1.2947,
"step": 2010
},
{
"epoch": 3.59,
"learning_rate": 3.541766109785203e-05,
"loss": 1.2908,
"step": 2020
},
{
"epoch": 3.61,
"learning_rate": 3.5393794749403344e-05,
"loss": 1.2893,
"step": 2030
},
{
"epoch": 3.63,
"learning_rate": 3.536992840095466e-05,
"loss": 1.2872,
"step": 2040
},
{
"epoch": 3.65,
"learning_rate": 3.5346062052505966e-05,
"loss": 1.286,
"step": 2050
},
{
"epoch": 3.66,
"learning_rate": 3.532219570405728e-05,
"loss": 1.2875,
"step": 2060
},
{
"epoch": 3.68,
"learning_rate": 3.5298329355608595e-05,
"loss": 1.2898,
"step": 2070
},
{
"epoch": 3.7,
"learning_rate": 3.527446300715991e-05,
"loss": 1.2854,
"step": 2080
},
{
"epoch": 3.72,
"learning_rate": 3.525059665871122e-05,
"loss": 1.288,
"step": 2090
},
{
"epoch": 3.74,
"learning_rate": 3.522673031026253e-05,
"loss": 1.2852,
"step": 2100
},
{
"epoch": 3.75,
"learning_rate": 3.520286396181385e-05,
"loss": 1.2856,
"step": 2110
},
{
"epoch": 3.77,
"learning_rate": 3.5178997613365155e-05,
"loss": 1.281,
"step": 2120
},
{
"epoch": 3.79,
"learning_rate": 3.515513126491647e-05,
"loss": 1.2879,
"step": 2130
},
{
"epoch": 3.81,
"learning_rate": 3.5131264916467784e-05,
"loss": 1.2849,
"step": 2140
},
{
"epoch": 3.82,
"learning_rate": 3.51073985680191e-05,
"loss": 1.2858,
"step": 2150
},
{
"epoch": 3.84,
"learning_rate": 3.5083532219570406e-05,
"loss": 1.2851,
"step": 2160
},
{
"epoch": 3.86,
"learning_rate": 3.505966587112172e-05,
"loss": 1.2801,
"step": 2170
},
{
"epoch": 3.88,
"learning_rate": 3.5035799522673035e-05,
"loss": 1.289,
"step": 2180
},
{
"epoch": 3.9,
"learning_rate": 3.501193317422434e-05,
"loss": 1.2839,
"step": 2190
},
{
"epoch": 3.91,
"learning_rate": 3.498806682577566e-05,
"loss": 1.2742,
"step": 2200
},
{
"epoch": 3.93,
"learning_rate": 3.496420047732697e-05,
"loss": 1.2762,
"step": 2210
},
{
"epoch": 3.95,
"learning_rate": 3.494033412887829e-05,
"loss": 1.2748,
"step": 2220
},
{
"epoch": 3.97,
"learning_rate": 3.49164677804296e-05,
"loss": 1.277,
"step": 2230
},
{
"epoch": 3.98,
"learning_rate": 3.489260143198091e-05,
"loss": 1.275,
"step": 2240
},
{
"epoch": 4.0,
"eval_cer": 4.788461538461538,
"eval_loss": 1.1931421756744385,
"eval_runtime": 4.8209,
"eval_samples_per_second": 2.074,
"eval_steps_per_second": 0.207,
"eval_wer": 0.974025974025974,
"step": 2248
},
{
"epoch": 4.0,
"learning_rate": 3.4868735083532224e-05,
"loss": 1.3891,
"step": 2250
},
{
"epoch": 4.02,
"learning_rate": 3.484486873508353e-05,
"loss": 1.2736,
"step": 2260
},
{
"epoch": 4.04,
"learning_rate": 3.4821002386634846e-05,
"loss": 1.2706,
"step": 2270
},
{
"epoch": 4.06,
"learning_rate": 3.479713603818616e-05,
"loss": 1.2725,
"step": 2280
},
{
"epoch": 4.07,
"learning_rate": 3.4773269689737475e-05,
"loss": 1.2807,
"step": 2290
},
{
"epoch": 4.09,
"learning_rate": 3.474940334128879e-05,
"loss": 1.2753,
"step": 2300
},
{
"epoch": 4.11,
"learning_rate": 3.47255369928401e-05,
"loss": 1.2717,
"step": 2310
},
{
"epoch": 4.13,
"learning_rate": 3.470167064439141e-05,
"loss": 1.2686,
"step": 2320
},
{
"epoch": 4.15,
"learning_rate": 3.467780429594272e-05,
"loss": 1.2662,
"step": 2330
},
{
"epoch": 4.16,
"learning_rate": 3.4653937947494035e-05,
"loss": 1.2721,
"step": 2340
},
{
"epoch": 4.18,
"learning_rate": 3.463007159904535e-05,
"loss": 1.2682,
"step": 2350
},
{
"epoch": 4.2,
"learning_rate": 3.4606205250596664e-05,
"loss": 1.2746,
"step": 2360
},
{
"epoch": 4.22,
"learning_rate": 3.458233890214798e-05,
"loss": 1.2697,
"step": 2370
},
{
"epoch": 4.23,
"learning_rate": 3.4558472553699286e-05,
"loss": 1.2666,
"step": 2380
},
{
"epoch": 4.25,
"learning_rate": 3.45346062052506e-05,
"loss": 1.266,
"step": 2390
},
{
"epoch": 4.27,
"learning_rate": 3.451073985680191e-05,
"loss": 1.2704,
"step": 2400
},
{
"epoch": 4.29,
"learning_rate": 3.448687350835322e-05,
"loss": 1.2611,
"step": 2410
},
{
"epoch": 4.31,
"learning_rate": 3.446300715990454e-05,
"loss": 1.2663,
"step": 2420
},
{
"epoch": 4.32,
"learning_rate": 3.443914081145585e-05,
"loss": 1.2622,
"step": 2430
},
{
"epoch": 4.34,
"learning_rate": 3.441527446300717e-05,
"loss": 1.2643,
"step": 2440
},
{
"epoch": 4.36,
"learning_rate": 3.4391408114558475e-05,
"loss": 1.2654,
"step": 2450
},
{
"epoch": 4.38,
"learning_rate": 3.436754176610979e-05,
"loss": 1.265,
"step": 2460
},
{
"epoch": 4.39,
"learning_rate": 3.43436754176611e-05,
"loss": 1.2564,
"step": 2470
},
{
"epoch": 4.41,
"learning_rate": 3.431980906921241e-05,
"loss": 1.2609,
"step": 2480
},
{
"epoch": 4.43,
"learning_rate": 3.4295942720763726e-05,
"loss": 1.2589,
"step": 2490
},
{
"epoch": 4.45,
"learning_rate": 3.427207637231504e-05,
"loss": 1.2597,
"step": 2500
},
{
"epoch": 4.47,
"learning_rate": 3.4248210023866355e-05,
"loss": 1.2553,
"step": 2510
},
{
"epoch": 4.48,
"learning_rate": 3.422434367541766e-05,
"loss": 1.2581,
"step": 2520
},
{
"epoch": 4.5,
"learning_rate": 3.420047732696898e-05,
"loss": 1.2588,
"step": 2530
},
{
"epoch": 4.52,
"learning_rate": 3.4176610978520285e-05,
"loss": 1.262,
"step": 2540
},
{
"epoch": 4.54,
"learning_rate": 3.41527446300716e-05,
"loss": 1.2535,
"step": 2550
},
{
"epoch": 4.55,
"learning_rate": 3.4128878281622915e-05,
"loss": 1.2547,
"step": 2560
},
{
"epoch": 4.57,
"learning_rate": 3.410501193317423e-05,
"loss": 1.2494,
"step": 2570
},
{
"epoch": 4.59,
"learning_rate": 3.4081145584725544e-05,
"loss": 1.2523,
"step": 2580
},
{
"epoch": 4.61,
"learning_rate": 3.405727923627685e-05,
"loss": 1.2553,
"step": 2590
},
{
"epoch": 4.63,
"learning_rate": 3.4033412887828166e-05,
"loss": 1.2548,
"step": 2600
},
{
"epoch": 4.64,
"learning_rate": 3.4009546539379474e-05,
"loss": 1.2525,
"step": 2610
},
{
"epoch": 4.66,
"learning_rate": 3.398568019093079e-05,
"loss": 1.2562,
"step": 2620
},
{
"epoch": 4.68,
"learning_rate": 3.39618138424821e-05,
"loss": 1.2571,
"step": 2630
},
{
"epoch": 4.7,
"learning_rate": 3.393794749403342e-05,
"loss": 1.2558,
"step": 2640
},
{
"epoch": 4.71,
"learning_rate": 3.391408114558473e-05,
"loss": 1.2509,
"step": 2650
},
{
"epoch": 4.73,
"learning_rate": 3.389021479713604e-05,
"loss": 1.2489,
"step": 2660
},
{
"epoch": 4.75,
"learning_rate": 3.3866348448687355e-05,
"loss": 1.2475,
"step": 2670
},
{
"epoch": 4.77,
"learning_rate": 3.384248210023866e-05,
"loss": 1.2468,
"step": 2680
},
{
"epoch": 4.79,
"learning_rate": 3.381861575178998e-05,
"loss": 1.2529,
"step": 2690
},
{
"epoch": 4.8,
"learning_rate": 3.379474940334129e-05,
"loss": 1.2524,
"step": 2700
},
{
"epoch": 4.82,
"learning_rate": 3.3770883054892606e-05,
"loss": 1.2504,
"step": 2710
},
{
"epoch": 4.84,
"learning_rate": 3.374701670644392e-05,
"loss": 1.2499,
"step": 2720
},
{
"epoch": 4.86,
"learning_rate": 3.372315035799523e-05,
"loss": 1.2493,
"step": 2730
},
{
"epoch": 4.87,
"learning_rate": 3.369928400954654e-05,
"loss": 1.2517,
"step": 2740
},
{
"epoch": 4.89,
"learning_rate": 3.367541766109785e-05,
"loss": 1.2502,
"step": 2750
},
{
"epoch": 4.91,
"learning_rate": 3.3651551312649165e-05,
"loss": 1.2455,
"step": 2760
},
{
"epoch": 4.93,
"learning_rate": 3.362768496420048e-05,
"loss": 1.2444,
"step": 2770
},
{
"epoch": 4.95,
"learning_rate": 3.3603818615751795e-05,
"loss": 1.2407,
"step": 2780
},
{
"epoch": 4.96,
"learning_rate": 3.357995226730311e-05,
"loss": 1.247,
"step": 2790
},
{
"epoch": 4.98,
"learning_rate": 3.355608591885442e-05,
"loss": 1.2476,
"step": 2800
},
{
"epoch": 5.0,
"learning_rate": 3.353221957040573e-05,
"loss": 1.2373,
"step": 2810
},
{
"epoch": 5.0,
"eval_cer": 4.77991452991453,
"eval_loss": 1.2068004608154297,
"eval_runtime": 4.7887,
"eval_samples_per_second": 2.088,
"eval_steps_per_second": 0.209,
"eval_wer": 1.0649350649350648,
"step": 2810
},
{
"epoch": 5.02,
"learning_rate": 3.350835322195704e-05,
"loss": 1.3503,
"step": 2820
},
{
"epoch": 5.04,
"learning_rate": 3.3484486873508354e-05,
"loss": 1.2483,
"step": 2830
},
{
"epoch": 5.05,
"learning_rate": 3.346062052505967e-05,
"loss": 1.2401,
"step": 2840
},
{
"epoch": 5.07,
"learning_rate": 3.343675417661098e-05,
"loss": 1.2351,
"step": 2850
},
{
"epoch": 5.09,
"learning_rate": 3.34128878281623e-05,
"loss": 1.2424,
"step": 2860
},
{
"epoch": 5.11,
"learning_rate": 3.3389021479713605e-05,
"loss": 1.2392,
"step": 2870
},
{
"epoch": 5.12,
"learning_rate": 3.336515513126492e-05,
"loss": 1.2397,
"step": 2880
},
{
"epoch": 5.14,
"learning_rate": 3.334128878281623e-05,
"loss": 1.2388,
"step": 2890
},
{
"epoch": 5.16,
"learning_rate": 3.331742243436754e-05,
"loss": 1.2345,
"step": 2900
},
{
"epoch": 5.18,
"learning_rate": 3.329355608591886e-05,
"loss": 1.2432,
"step": 2910
},
{
"epoch": 5.2,
"learning_rate": 3.326968973747017e-05,
"loss": 1.2374,
"step": 2920
},
{
"epoch": 5.21,
"learning_rate": 3.3245823389021486e-05,
"loss": 1.2345,
"step": 2930
},
{
"epoch": 5.23,
"learning_rate": 3.3221957040572794e-05,
"loss": 1.2349,
"step": 2940
},
{
"epoch": 5.25,
"learning_rate": 3.319809069212411e-05,
"loss": 1.2321,
"step": 2950
},
{
"epoch": 5.27,
"learning_rate": 3.3174224343675416e-05,
"loss": 1.2305,
"step": 2960
},
{
"epoch": 5.28,
"learning_rate": 3.315035799522673e-05,
"loss": 1.2371,
"step": 2970
},
{
"epoch": 5.3,
"learning_rate": 3.3126491646778045e-05,
"loss": 1.238,
"step": 2980
},
{
"epoch": 5.32,
"learning_rate": 3.310262529832936e-05,
"loss": 1.236,
"step": 2990
},
{
"epoch": 5.34,
"learning_rate": 3.3078758949880675e-05,
"loss": 1.2323,
"step": 3000
},
{
"epoch": 5.36,
"learning_rate": 3.305489260143198e-05,
"loss": 1.2266,
"step": 3010
},
{
"epoch": 5.37,
"learning_rate": 3.30310262529833e-05,
"loss": 1.2322,
"step": 3020
},
{
"epoch": 5.39,
"learning_rate": 3.3007159904534605e-05,
"loss": 1.2321,
"step": 3030
},
{
"epoch": 5.41,
"learning_rate": 3.298329355608592e-05,
"loss": 1.227,
"step": 3040
},
{
"epoch": 5.43,
"learning_rate": 3.2959427207637234e-05,
"loss": 1.2316,
"step": 3050
},
{
"epoch": 5.44,
"learning_rate": 3.293556085918855e-05,
"loss": 1.2317,
"step": 3060
},
{
"epoch": 5.46,
"learning_rate": 3.291169451073986e-05,
"loss": 1.23,
"step": 3070
},
{
"epoch": 5.48,
"learning_rate": 3.288782816229117e-05,
"loss": 1.2321,
"step": 3080
},
{
"epoch": 5.5,
"learning_rate": 3.2863961813842485e-05,
"loss": 1.2326,
"step": 3090
},
{
"epoch": 5.52,
"learning_rate": 3.284009546539379e-05,
"loss": 1.2276,
"step": 3100
},
{
"epoch": 5.53,
"learning_rate": 3.281622911694511e-05,
"loss": 1.2284,
"step": 3110
},
{
"epoch": 5.55,
"learning_rate": 3.279236276849642e-05,
"loss": 1.2298,
"step": 3120
},
{
"epoch": 5.57,
"learning_rate": 3.276849642004774e-05,
"loss": 1.2305,
"step": 3130
},
{
"epoch": 5.59,
"learning_rate": 3.274463007159905e-05,
"loss": 1.2284,
"step": 3140
},
{
"epoch": 5.6,
"learning_rate": 3.272076372315036e-05,
"loss": 1.2315,
"step": 3150
},
{
"epoch": 5.62,
"learning_rate": 3.2696897374701674e-05,
"loss": 1.2266,
"step": 3160
},
{
"epoch": 5.64,
"learning_rate": 3.267303102625299e-05,
"loss": 1.2248,
"step": 3170
},
{
"epoch": 5.66,
"learning_rate": 3.2649164677804296e-05,
"loss": 1.2237,
"step": 3180
},
{
"epoch": 5.68,
"learning_rate": 3.262529832935561e-05,
"loss": 1.2272,
"step": 3190
},
{
"epoch": 5.69,
"learning_rate": 3.2601431980906925e-05,
"loss": 1.223,
"step": 3200
},
{
"epoch": 5.71,
"learning_rate": 3.257756563245824e-05,
"loss": 1.2256,
"step": 3210
},
{
"epoch": 5.73,
"learning_rate": 3.255369928400955e-05,
"loss": 1.2199,
"step": 3220
},
{
"epoch": 5.75,
"learning_rate": 3.252983293556086e-05,
"loss": 1.2244,
"step": 3230
},
{
"epoch": 5.76,
"learning_rate": 3.250596658711218e-05,
"loss": 1.2191,
"step": 3240
},
{
"epoch": 5.78,
"learning_rate": 3.2482100238663485e-05,
"loss": 1.2297,
"step": 3250
},
{
"epoch": 5.8,
"learning_rate": 3.24582338902148e-05,
"loss": 1.2235,
"step": 3260
},
{
"epoch": 5.82,
"learning_rate": 3.2434367541766114e-05,
"loss": 1.2217,
"step": 3270
},
{
"epoch": 5.83,
"learning_rate": 3.241050119331743e-05,
"loss": 1.22,
"step": 3280
},
{
"epoch": 5.85,
"learning_rate": 3.2386634844868736e-05,
"loss": 1.2273,
"step": 3290
},
{
"epoch": 5.87,
"learning_rate": 3.236276849642005e-05,
"loss": 1.2171,
"step": 3300
},
{
"epoch": 5.89,
"learning_rate": 3.2338902147971365e-05,
"loss": 1.2251,
"step": 3310
},
{
"epoch": 5.91,
"learning_rate": 3.231503579952267e-05,
"loss": 1.2199,
"step": 3320
},
{
"epoch": 5.92,
"learning_rate": 3.229116945107399e-05,
"loss": 1.2178,
"step": 3330
},
{
"epoch": 5.94,
"learning_rate": 3.22673031026253e-05,
"loss": 1.2198,
"step": 3340
},
{
"epoch": 5.96,
"learning_rate": 3.224343675417661e-05,
"loss": 1.2186,
"step": 3350
},
{
"epoch": 5.98,
"learning_rate": 3.2219570405727925e-05,
"loss": 1.2116,
"step": 3360
},
{
"epoch": 5.99,
"learning_rate": 3.219570405727924e-05,
"loss": 1.2191,
"step": 3370
},
{
"epoch": 6.0,
"eval_cer": 4.788461538461538,
"eval_loss": 1.1857426166534424,
"eval_runtime": 4.8064,
"eval_samples_per_second": 2.081,
"eval_steps_per_second": 0.208,
"eval_wer": 1.025974025974026,
"step": 3372
},
{
"epoch": 6.01,
"learning_rate": 3.2171837708830554e-05,
"loss": 1.3261,
"step": 3380
},
{
"epoch": 6.03,
"learning_rate": 3.214797136038186e-05,
"loss": 1.2206,
"step": 3390
},
{
"epoch": 6.05,
"learning_rate": 3.2124105011933176e-05,
"loss": 1.2121,
"step": 3400
},
{
"epoch": 6.07,
"learning_rate": 3.210023866348449e-05,
"loss": 1.2154,
"step": 3410
},
{
"epoch": 6.09,
"learning_rate": 3.20763723150358e-05,
"loss": 1.2153,
"step": 3420
},
{
"epoch": 6.1,
"learning_rate": 3.205250596658711e-05,
"loss": 1.2137,
"step": 3430
},
{
"epoch": 6.12,
"learning_rate": 3.202863961813843e-05,
"loss": 1.2168,
"step": 3440
},
{
"epoch": 6.14,
"learning_rate": 3.200477326968974e-05,
"loss": 1.2152,
"step": 3450
},
{
"epoch": 6.16,
"learning_rate": 3.198090692124105e-05,
"loss": 1.2102,
"step": 3460
},
{
"epoch": 6.17,
"learning_rate": 3.1957040572792365e-05,
"loss": 1.2101,
"step": 3470
},
{
"epoch": 6.19,
"learning_rate": 3.193317422434368e-05,
"loss": 1.2107,
"step": 3480
},
{
"epoch": 6.21,
"learning_rate": 3.190930787589499e-05,
"loss": 1.2146,
"step": 3490
},
{
"epoch": 6.23,
"learning_rate": 3.18854415274463e-05,
"loss": 1.2136,
"step": 3500
},
{
"epoch": 6.25,
"learning_rate": 3.1861575178997616e-05,
"loss": 1.2109,
"step": 3510
},
{
"epoch": 6.26,
"learning_rate": 3.183770883054893e-05,
"loss": 1.1998,
"step": 3520
},
{
"epoch": 6.28,
"learning_rate": 3.1813842482100245e-05,
"loss": 1.2116,
"step": 3530
},
{
"epoch": 6.3,
"learning_rate": 3.178997613365155e-05,
"loss": 1.2004,
"step": 3540
},
{
"epoch": 6.32,
"learning_rate": 3.176610978520287e-05,
"loss": 1.2144,
"step": 3550
},
{
"epoch": 6.33,
"learning_rate": 3.1742243436754176e-05,
"loss": 1.2048,
"step": 3560
},
{
"epoch": 6.35,
"learning_rate": 3.171837708830549e-05,
"loss": 1.1998,
"step": 3570
},
{
"epoch": 6.37,
"learning_rate": 3.1694510739856805e-05,
"loss": 1.2075,
"step": 3580
},
{
"epoch": 6.39,
"learning_rate": 3.167064439140812e-05,
"loss": 1.2091,
"step": 3590
},
{
"epoch": 6.41,
"learning_rate": 3.1646778042959434e-05,
"loss": 1.2061,
"step": 3600
},
{
"epoch": 6.42,
"learning_rate": 3.162291169451074e-05,
"loss": 1.2081,
"step": 3610
},
{
"epoch": 6.44,
"learning_rate": 3.1599045346062056e-05,
"loss": 1.2037,
"step": 3620
},
{
"epoch": 6.46,
"learning_rate": 3.1575178997613364e-05,
"loss": 1.2091,
"step": 3630
},
{
"epoch": 6.48,
"learning_rate": 3.155131264916468e-05,
"loss": 1.2054,
"step": 3640
},
{
"epoch": 6.49,
"learning_rate": 3.152744630071599e-05,
"loss": 1.2096,
"step": 3650
},
{
"epoch": 6.51,
"learning_rate": 3.150357995226731e-05,
"loss": 1.2104,
"step": 3660
},
{
"epoch": 6.53,
"learning_rate": 3.147971360381862e-05,
"loss": 1.2022,
"step": 3670
},
{
"epoch": 6.55,
"learning_rate": 3.145584725536993e-05,
"loss": 1.2037,
"step": 3680
},
{
"epoch": 6.56,
"learning_rate": 3.1431980906921245e-05,
"loss": 1.2099,
"step": 3690
},
{
"epoch": 6.58,
"learning_rate": 3.140811455847255e-05,
"loss": 1.2037,
"step": 3700
},
{
"epoch": 6.6,
"learning_rate": 3.138424821002387e-05,
"loss": 1.2025,
"step": 3710
},
{
"epoch": 6.62,
"learning_rate": 3.136038186157518e-05,
"loss": 1.2029,
"step": 3720
},
{
"epoch": 6.64,
"learning_rate": 3.1336515513126496e-05,
"loss": 1.2072,
"step": 3730
},
{
"epoch": 6.65,
"learning_rate": 3.131264916467781e-05,
"loss": 1.2022,
"step": 3740
},
{
"epoch": 6.67,
"learning_rate": 3.128878281622912e-05,
"loss": 1.2047,
"step": 3750
},
{
"epoch": 6.69,
"learning_rate": 3.126491646778043e-05,
"loss": 1.2019,
"step": 3760
},
{
"epoch": 6.71,
"learning_rate": 3.124105011933174e-05,
"loss": 1.2083,
"step": 3770
},
{
"epoch": 6.72,
"learning_rate": 3.1217183770883056e-05,
"loss": 1.2015,
"step": 3780
},
{
"epoch": 6.74,
"learning_rate": 3.119331742243437e-05,
"loss": 1.2069,
"step": 3790
},
{
"epoch": 6.76,
"learning_rate": 3.1169451073985685e-05,
"loss": 1.2006,
"step": 3800
},
{
"epoch": 6.78,
"learning_rate": 3.1145584725537e-05,
"loss": 1.1969,
"step": 3810
},
{
"epoch": 6.8,
"learning_rate": 3.112171837708831e-05,
"loss": 1.1971,
"step": 3820
},
{
"epoch": 6.81,
"learning_rate": 3.109785202863962e-05,
"loss": 1.1997,
"step": 3830
},
{
"epoch": 6.83,
"learning_rate": 3.107398568019093e-05,
"loss": 1.2082,
"step": 3840
},
{
"epoch": 6.85,
"learning_rate": 3.1050119331742244e-05,
"loss": 1.1952,
"step": 3850
},
{
"epoch": 6.87,
"learning_rate": 3.102625298329356e-05,
"loss": 1.1896,
"step": 3860
},
{
"epoch": 6.88,
"learning_rate": 3.100238663484487e-05,
"loss": 1.1958,
"step": 3870
},
{
"epoch": 6.9,
"learning_rate": 3.097852028639619e-05,
"loss": 1.1938,
"step": 3880
},
{
"epoch": 6.92,
"learning_rate": 3.0954653937947496e-05,
"loss": 1.1991,
"step": 3890
},
{
"epoch": 6.94,
"learning_rate": 3.093078758949881e-05,
"loss": 1.1969,
"step": 3900
},
{
"epoch": 6.96,
"learning_rate": 3.090692124105012e-05,
"loss": 1.1958,
"step": 3910
},
{
"epoch": 6.97,
"learning_rate": 3.088305489260143e-05,
"loss": 1.1876,
"step": 3920
},
{
"epoch": 6.99,
"learning_rate": 3.085918854415275e-05,
"loss": 1.1942,
"step": 3930
},
{
"epoch": 7.0,
"eval_cer": 4.7905982905982905,
"eval_loss": 1.1601146459579468,
"eval_runtime": 4.8741,
"eval_samples_per_second": 2.052,
"eval_steps_per_second": 0.205,
"eval_wer": 0.987012987012987,
"step": 3934
},
{
"epoch": 7.01,
"learning_rate": 3.083532219570406e-05,
"loss": 1.3012,
"step": 3940
},
{
"epoch": 7.03,
"learning_rate": 3.0811455847255376e-05,
"loss": 1.1941,
"step": 3950
},
{
"epoch": 7.05,
"learning_rate": 3.0787589498806684e-05,
"loss": 1.1899,
"step": 3960
},
{
"epoch": 7.06,
"learning_rate": 3.0763723150358e-05,
"loss": 1.1886,
"step": 3970
},
{
"epoch": 7.08,
"learning_rate": 3.0739856801909307e-05,
"loss": 1.1903,
"step": 3980
},
{
"epoch": 7.1,
"learning_rate": 3.071599045346062e-05,
"loss": 1.1918,
"step": 3990
},
{
"epoch": 7.12,
"learning_rate": 3.0692124105011936e-05,
"loss": 1.1976,
"step": 4000
},
{
"epoch": 7.14,
"learning_rate": 3.066825775656325e-05,
"loss": 1.2001,
"step": 4010
},
{
"epoch": 7.15,
"learning_rate": 3.0644391408114565e-05,
"loss": 1.1896,
"step": 4020
},
{
"epoch": 7.17,
"learning_rate": 3.062052505966587e-05,
"loss": 1.1937,
"step": 4030
},
{
"epoch": 7.19,
"learning_rate": 3.059665871121719e-05,
"loss": 1.1903,
"step": 4040
},
{
"epoch": 7.21,
"learning_rate": 3.0572792362768495e-05,
"loss": 1.1891,
"step": 4050
},
{
"epoch": 7.22,
"learning_rate": 3.054892601431981e-05,
"loss": 1.1805,
"step": 4060
},
{
"epoch": 7.24,
"learning_rate": 3.0525059665871124e-05,
"loss": 1.1919,
"step": 4070
},
{
"epoch": 7.26,
"learning_rate": 3.050119331742244e-05,
"loss": 1.1904,
"step": 4080
},
{
"epoch": 7.28,
"learning_rate": 3.047732696897375e-05,
"loss": 1.1927,
"step": 4090
},
{
"epoch": 7.29,
"learning_rate": 3.0453460620525064e-05,
"loss": 1.1898,
"step": 4100
},
{
"epoch": 7.31,
"learning_rate": 3.0429594272076372e-05,
"loss": 1.1928,
"step": 4110
},
{
"epoch": 7.33,
"learning_rate": 3.0405727923627687e-05,
"loss": 1.1831,
"step": 4120
},
{
"epoch": 7.35,
"learning_rate": 3.0381861575178998e-05,
"loss": 1.1833,
"step": 4130
},
{
"epoch": 7.37,
"learning_rate": 3.0357995226730313e-05,
"loss": 1.1863,
"step": 4140
},
{
"epoch": 7.38,
"learning_rate": 3.0334128878281627e-05,
"loss": 1.1807,
"step": 4150
},
{
"epoch": 7.4,
"learning_rate": 3.031026252983294e-05,
"loss": 1.1854,
"step": 4160
},
{
"epoch": 7.42,
"learning_rate": 3.0286396181384253e-05,
"loss": 1.1835,
"step": 4170
},
{
"epoch": 7.44,
"learning_rate": 3.0262529832935564e-05,
"loss": 1.1852,
"step": 4180
},
{
"epoch": 7.45,
"learning_rate": 3.0238663484486875e-05,
"loss": 1.186,
"step": 4190
},
{
"epoch": 7.47,
"learning_rate": 3.0214797136038187e-05,
"loss": 1.1841,
"step": 4200
},
{
"epoch": 7.49,
"learning_rate": 3.01909307875895e-05,
"loss": 1.1806,
"step": 4210
},
{
"epoch": 7.51,
"learning_rate": 3.0167064439140816e-05,
"loss": 1.1876,
"step": 4220
},
{
"epoch": 7.53,
"learning_rate": 3.0143198090692127e-05,
"loss": 1.1802,
"step": 4230
},
{
"epoch": 7.54,
"learning_rate": 3.011933174224344e-05,
"loss": 1.1848,
"step": 4240
},
{
"epoch": 7.56,
"learning_rate": 3.0095465393794753e-05,
"loss": 1.1814,
"step": 4250
},
{
"epoch": 7.58,
"learning_rate": 3.0071599045346064e-05,
"loss": 1.1838,
"step": 4260
},
{
"epoch": 7.6,
"learning_rate": 3.0047732696897375e-05,
"loss": 1.1799,
"step": 4270
},
{
"epoch": 7.61,
"learning_rate": 3.002386634844869e-05,
"loss": 1.1791,
"step": 4280
},
{
"epoch": 7.63,
"learning_rate": 3.0000000000000004e-05,
"loss": 1.18,
"step": 4290
},
{
"epoch": 7.65,
"learning_rate": 2.9976133651551315e-05,
"loss": 1.1862,
"step": 4300
},
{
"epoch": 7.67,
"learning_rate": 2.995226730310263e-05,
"loss": 1.1711,
"step": 4310
},
{
"epoch": 7.69,
"learning_rate": 2.992840095465394e-05,
"loss": 1.1776,
"step": 4320
},
{
"epoch": 7.7,
"learning_rate": 2.9904534606205252e-05,
"loss": 1.18,
"step": 4330
},
{
"epoch": 7.72,
"learning_rate": 2.9880668257756563e-05,
"loss": 1.1819,
"step": 4340
},
{
"epoch": 7.74,
"learning_rate": 2.9856801909307878e-05,
"loss": 1.1766,
"step": 4350
},
{
"epoch": 7.76,
"learning_rate": 2.983293556085919e-05,
"loss": 1.1756,
"step": 4360
},
{
"epoch": 7.77,
"learning_rate": 2.9809069212410504e-05,
"loss": 1.1812,
"step": 4370
},
{
"epoch": 7.79,
"learning_rate": 2.978520286396182e-05,
"loss": 1.1764,
"step": 4380
},
{
"epoch": 7.81,
"learning_rate": 2.976133651551313e-05,
"loss": 1.1816,
"step": 4390
},
{
"epoch": 7.83,
"learning_rate": 2.973747016706444e-05,
"loss": 1.1737,
"step": 4400
},
{
"epoch": 7.85,
"learning_rate": 2.9713603818615752e-05,
"loss": 1.1788,
"step": 4410
},
{
"epoch": 7.86,
"learning_rate": 2.9689737470167067e-05,
"loss": 1.1813,
"step": 4420
},
{
"epoch": 7.88,
"learning_rate": 2.9665871121718378e-05,
"loss": 1.1766,
"step": 4430
},
{
"epoch": 7.9,
"learning_rate": 2.9642004773269692e-05,
"loss": 1.1721,
"step": 4440
},
{
"epoch": 7.92,
"learning_rate": 2.9618138424821007e-05,
"loss": 1.1788,
"step": 4450
},
{
"epoch": 7.93,
"learning_rate": 2.9594272076372318e-05,
"loss": 1.1747,
"step": 4460
},
{
"epoch": 7.95,
"learning_rate": 2.9570405727923633e-05,
"loss": 1.1741,
"step": 4470
},
{
"epoch": 7.97,
"learning_rate": 2.954653937947494e-05,
"loss": 1.1713,
"step": 4480
},
{
"epoch": 7.99,
"learning_rate": 2.9522673031026255e-05,
"loss": 1.1716,
"step": 4490
},
{
"epoch": 8.0,
"eval_cer": 4.777777777777778,
"eval_loss": 1.1047004461288452,
"eval_runtime": 4.6551,
"eval_samples_per_second": 2.148,
"eval_steps_per_second": 0.215,
"eval_wer": 0.948051948051948,
"step": 4496
},
{
"epoch": 8.01,
"learning_rate": 2.9498806682577566e-05,
"loss": 1.276,
"step": 4500
},
{
"epoch": 8.02,
"learning_rate": 2.947494033412888e-05,
"loss": 1.1693,
"step": 4510
},
{
"epoch": 8.04,
"learning_rate": 2.9451073985680195e-05,
"loss": 1.1742,
"step": 4520
},
{
"epoch": 8.06,
"learning_rate": 2.9427207637231507e-05,
"loss": 1.1741,
"step": 4530
},
{
"epoch": 8.08,
"learning_rate": 2.940334128878282e-05,
"loss": 1.1713,
"step": 4540
},
{
"epoch": 8.1,
"learning_rate": 2.937947494033413e-05,
"loss": 1.1656,
"step": 4550
},
{
"epoch": 8.11,
"learning_rate": 2.9355608591885443e-05,
"loss": 1.1666,
"step": 4560
},
{
"epoch": 8.13,
"learning_rate": 2.9331742243436755e-05,
"loss": 1.169,
"step": 4570
},
{
"epoch": 8.15,
"learning_rate": 2.930787589498807e-05,
"loss": 1.1749,
"step": 4580
},
{
"epoch": 8.17,
"learning_rate": 2.9284009546539384e-05,
"loss": 1.1651,
"step": 4590
},
{
"epoch": 8.18,
"learning_rate": 2.9260143198090695e-05,
"loss": 1.1711,
"step": 4600
},
{
"epoch": 8.2,
"learning_rate": 2.923627684964201e-05,
"loss": 1.1713,
"step": 4610
},
{
"epoch": 8.22,
"learning_rate": 2.9212410501193317e-05,
"loss": 1.1656,
"step": 4620
},
{
"epoch": 8.24,
"learning_rate": 2.9188544152744632e-05,
"loss": 1.1618,
"step": 4630
},
{
"epoch": 8.26,
"learning_rate": 2.9164677804295943e-05,
"loss": 1.1703,
"step": 4640
},
{
"epoch": 8.27,
"learning_rate": 2.9140811455847258e-05,
"loss": 1.1734,
"step": 4650
},
{
"epoch": 8.29,
"learning_rate": 2.9116945107398572e-05,
"loss": 1.1687,
"step": 4660
},
{
"epoch": 8.31,
"learning_rate": 2.9093078758949883e-05,
"loss": 1.1704,
"step": 4670
},
{
"epoch": 8.33,
"learning_rate": 2.9069212410501198e-05,
"loss": 1.1651,
"step": 4680
},
{
"epoch": 8.34,
"learning_rate": 2.9045346062052506e-05,
"loss": 1.1627,
"step": 4690
},
{
"epoch": 8.36,
"learning_rate": 2.902147971360382e-05,
"loss": 1.1639,
"step": 4700
},
{
"epoch": 8.38,
"learning_rate": 2.899761336515513e-05,
"loss": 1.1606,
"step": 4710
},
{
"epoch": 8.4,
"learning_rate": 2.8973747016706446e-05,
"loss": 1.1643,
"step": 4720
},
{
"epoch": 8.42,
"learning_rate": 2.894988066825776e-05,
"loss": 1.1694,
"step": 4730
},
{
"epoch": 8.43,
"learning_rate": 2.8926014319809072e-05,
"loss": 1.1623,
"step": 4740
},
{
"epoch": 8.45,
"learning_rate": 2.8902147971360387e-05,
"loss": 1.163,
"step": 4750
},
{
"epoch": 8.47,
"learning_rate": 2.8878281622911694e-05,
"loss": 1.1623,
"step": 4760
},
{
"epoch": 8.49,
"learning_rate": 2.885441527446301e-05,
"loss": 1.1631,
"step": 4770
},
{
"epoch": 8.5,
"learning_rate": 2.883054892601432e-05,
"loss": 1.1607,
"step": 4780
},
{
"epoch": 8.52,
"learning_rate": 2.8806682577565635e-05,
"loss": 1.158,
"step": 4790
},
{
"epoch": 8.54,
"learning_rate": 2.878281622911695e-05,
"loss": 1.1641,
"step": 4800
},
{
"epoch": 8.56,
"learning_rate": 2.875894988066826e-05,
"loss": 1.1615,
"step": 4810
},
{
"epoch": 8.58,
"learning_rate": 2.8735083532219575e-05,
"loss": 1.1618,
"step": 4820
},
{
"epoch": 8.59,
"learning_rate": 2.8711217183770886e-05,
"loss": 1.1626,
"step": 4830
},
{
"epoch": 8.61,
"learning_rate": 2.8687350835322197e-05,
"loss": 1.1596,
"step": 4840
},
{
"epoch": 8.63,
"learning_rate": 2.866348448687351e-05,
"loss": 1.1568,
"step": 4850
},
{
"epoch": 8.65,
"learning_rate": 2.8639618138424823e-05,
"loss": 1.1555,
"step": 4860
},
{
"epoch": 8.66,
"learning_rate": 2.8615751789976138e-05,
"loss": 1.1623,
"step": 4870
},
{
"epoch": 8.68,
"learning_rate": 2.859188544152745e-05,
"loss": 1.1525,
"step": 4880
},
{
"epoch": 8.7,
"learning_rate": 2.8568019093078763e-05,
"loss": 1.1599,
"step": 4890
},
{
"epoch": 8.72,
"learning_rate": 2.8544152744630075e-05,
"loss": 1.1656,
"step": 4900
},
{
"epoch": 8.74,
"learning_rate": 2.8520286396181386e-05,
"loss": 1.1528,
"step": 4910
},
{
"epoch": 8.75,
"learning_rate": 2.8496420047732697e-05,
"loss": 1.1603,
"step": 4920
},
{
"epoch": 8.77,
"learning_rate": 2.847255369928401e-05,
"loss": 1.1589,
"step": 4930
},
{
"epoch": 8.79,
"learning_rate": 2.8448687350835323e-05,
"loss": 1.1599,
"step": 4940
},
{
"epoch": 8.81,
"learning_rate": 2.8424821002386637e-05,
"loss": 1.1562,
"step": 4950
},
{
"epoch": 8.82,
"learning_rate": 2.8400954653937952e-05,
"loss": 1.1593,
"step": 4960
},
{
"epoch": 8.84,
"learning_rate": 2.8377088305489263e-05,
"loss": 1.1559,
"step": 4970
},
{
"epoch": 8.86,
"learning_rate": 2.8353221957040574e-05,
"loss": 1.1578,
"step": 4980
},
{
"epoch": 8.88,
"learning_rate": 2.8329355608591886e-05,
"loss": 1.162,
"step": 4990
},
{
"epoch": 8.9,
"learning_rate": 2.83054892601432e-05,
"loss": 1.1617,
"step": 5000
},
{
"epoch": 8.91,
"learning_rate": 2.828162291169451e-05,
"loss": 1.1576,
"step": 5010
},
{
"epoch": 8.93,
"learning_rate": 2.8257756563245826e-05,
"loss": 1.1609,
"step": 5020
},
{
"epoch": 8.95,
"learning_rate": 2.823389021479714e-05,
"loss": 1.1584,
"step": 5030
},
{
"epoch": 8.97,
"learning_rate": 2.821002386634845e-05,
"loss": 1.1601,
"step": 5040
},
{
"epoch": 8.98,
"learning_rate": 2.8186157517899763e-05,
"loss": 1.1555,
"step": 5050
},
{
"epoch": 9.0,
"eval_cer": 4.7585470085470085,
"eval_loss": 1.0889137983322144,
"eval_runtime": 4.8301,
"eval_samples_per_second": 2.07,
"eval_steps_per_second": 0.207,
"eval_wer": 0.948051948051948,
"step": 5058
},
{
"epoch": 9.0,
"learning_rate": 2.8162291169451074e-05,
"loss": 1.2598,
"step": 5060
},
{
"epoch": 9.02,
"learning_rate": 2.813842482100239e-05,
"loss": 1.1559,
"step": 5070
},
{
"epoch": 9.04,
"learning_rate": 2.81145584725537e-05,
"loss": 1.1544,
"step": 5080
},
{
"epoch": 9.06,
"learning_rate": 2.8090692124105014e-05,
"loss": 1.1514,
"step": 5090
},
{
"epoch": 9.07,
"learning_rate": 2.806682577565633e-05,
"loss": 1.1548,
"step": 5100
},
{
"epoch": 9.09,
"learning_rate": 2.804295942720764e-05,
"loss": 1.1531,
"step": 5110
},
{
"epoch": 9.11,
"learning_rate": 2.8019093078758955e-05,
"loss": 1.149,
"step": 5120
},
{
"epoch": 9.13,
"learning_rate": 2.7995226730310262e-05,
"loss": 1.1512,
"step": 5130
},
{
"epoch": 9.15,
"learning_rate": 2.7971360381861577e-05,
"loss": 1.152,
"step": 5140
},
{
"epoch": 9.16,
"learning_rate": 2.7947494033412888e-05,
"loss": 1.1497,
"step": 5150
},
{
"epoch": 9.18,
"learning_rate": 2.7923627684964203e-05,
"loss": 1.151,
"step": 5160
},
{
"epoch": 9.2,
"learning_rate": 2.7899761336515517e-05,
"loss": 1.1571,
"step": 5170
},
{
"epoch": 9.22,
"learning_rate": 2.787589498806683e-05,
"loss": 1.1528,
"step": 5180
},
{
"epoch": 9.23,
"learning_rate": 2.7852028639618143e-05,
"loss": 1.1472,
"step": 5190
},
{
"epoch": 9.25,
"learning_rate": 2.782816229116945e-05,
"loss": 1.1502,
"step": 5200
},
{
"epoch": 9.27,
"learning_rate": 2.7804295942720766e-05,
"loss": 1.1487,
"step": 5210
},
{
"epoch": 9.29,
"learning_rate": 2.7780429594272077e-05,
"loss": 1.1448,
"step": 5220
},
{
"epoch": 9.31,
"learning_rate": 2.775656324582339e-05,
"loss": 1.1531,
"step": 5230
},
{
"epoch": 9.32,
"learning_rate": 2.7732696897374706e-05,
"loss": 1.151,
"step": 5240
},
{
"epoch": 9.34,
"learning_rate": 2.7708830548926017e-05,
"loss": 1.1447,
"step": 5250
},
{
"epoch": 9.36,
"learning_rate": 2.768496420047733e-05,
"loss": 1.1501,
"step": 5260
},
{
"epoch": 9.38,
"learning_rate": 2.766109785202864e-05,
"loss": 1.1473,
"step": 5270
},
{
"epoch": 9.39,
"learning_rate": 2.7637231503579954e-05,
"loss": 1.1486,
"step": 5280
},
{
"epoch": 9.41,
"learning_rate": 2.7613365155131265e-05,
"loss": 1.1547,
"step": 5290
},
{
"epoch": 9.43,
"learning_rate": 2.758949880668258e-05,
"loss": 1.1492,
"step": 5300
},
{
"epoch": 9.45,
"learning_rate": 2.7565632458233894e-05,
"loss": 1.1489,
"step": 5310
},
{
"epoch": 9.47,
"learning_rate": 2.7541766109785206e-05,
"loss": 1.1494,
"step": 5320
},
{
"epoch": 9.48,
"learning_rate": 2.751789976133652e-05,
"loss": 1.1445,
"step": 5330
},
{
"epoch": 9.5,
"learning_rate": 2.7494033412887828e-05,
"loss": 1.145,
"step": 5340
},
{
"epoch": 9.52,
"learning_rate": 2.7470167064439143e-05,
"loss": 1.147,
"step": 5350
},
{
"epoch": 9.54,
"learning_rate": 2.7446300715990454e-05,
"loss": 1.1495,
"step": 5360
},
{
"epoch": 9.55,
"learning_rate": 2.7422434367541768e-05,
"loss": 1.1551,
"step": 5370
},
{
"epoch": 9.57,
"learning_rate": 2.7398568019093083e-05,
"loss": 1.1484,
"step": 5380
},
{
"epoch": 9.59,
"learning_rate": 2.7374701670644394e-05,
"loss": 1.1546,
"step": 5390
},
{
"epoch": 9.61,
"learning_rate": 2.735083532219571e-05,
"loss": 1.1524,
"step": 5400
},
{
"epoch": 9.63,
"learning_rate": 2.7326968973747016e-05,
"loss": 1.1502,
"step": 5410
},
{
"epoch": 9.64,
"learning_rate": 2.730310262529833e-05,
"loss": 1.1464,
"step": 5420
},
{
"epoch": 9.66,
"learning_rate": 2.7279236276849642e-05,
"loss": 1.1454,
"step": 5430
},
{
"epoch": 9.68,
"learning_rate": 2.7255369928400957e-05,
"loss": 1.1416,
"step": 5440
},
{
"epoch": 9.7,
"learning_rate": 2.723150357995227e-05,
"loss": 1.1496,
"step": 5450
},
{
"epoch": 9.71,
"learning_rate": 2.7207637231503583e-05,
"loss": 1.1446,
"step": 5460
},
{
"epoch": 9.73,
"learning_rate": 2.7183770883054897e-05,
"loss": 1.144,
"step": 5470
},
{
"epoch": 9.75,
"learning_rate": 2.7159904534606208e-05,
"loss": 1.1436,
"step": 5480
},
{
"epoch": 9.77,
"learning_rate": 2.713603818615752e-05,
"loss": 1.1424,
"step": 5490
},
{
"epoch": 9.79,
"learning_rate": 2.711217183770883e-05,
"loss": 1.1466,
"step": 5500
},
{
"epoch": 9.8,
"learning_rate": 2.7088305489260145e-05,
"loss": 1.139,
"step": 5510
},
{
"epoch": 9.82,
"learning_rate": 2.706443914081146e-05,
"loss": 1.142,
"step": 5520
},
{
"epoch": 9.84,
"learning_rate": 2.704057279236277e-05,
"loss": 1.1448,
"step": 5530
},
{
"epoch": 9.86,
"learning_rate": 2.7016706443914086e-05,
"loss": 1.1374,
"step": 5540
},
{
"epoch": 9.87,
"learning_rate": 2.6992840095465397e-05,
"loss": 1.1469,
"step": 5550
},
{
"epoch": 9.89,
"learning_rate": 2.6968973747016708e-05,
"loss": 1.1398,
"step": 5560
},
{
"epoch": 9.91,
"learning_rate": 2.694510739856802e-05,
"loss": 1.142,
"step": 5570
},
{
"epoch": 9.93,
"learning_rate": 2.6921241050119334e-05,
"loss": 1.1463,
"step": 5580
},
{
"epoch": 9.95,
"learning_rate": 2.6897374701670645e-05,
"loss": 1.133,
"step": 5590
},
{
"epoch": 9.96,
"learning_rate": 2.687350835322196e-05,
"loss": 1.14,
"step": 5600
},
{
"epoch": 9.98,
"learning_rate": 2.6849642004773274e-05,
"loss": 1.1465,
"step": 5610
},
{
"epoch": 10.0,
"learning_rate": 2.6825775656324585e-05,
"loss": 1.1389,
"step": 5620
},
{
"epoch": 10.0,
"eval_cer": 4.756410256410256,
"eval_loss": 1.0790975093841553,
"eval_runtime": 4.8315,
"eval_samples_per_second": 2.07,
"eval_steps_per_second": 0.207,
"eval_wer": 0.961038961038961,
"step": 5620
},
{
"epoch": 10.02,
"learning_rate": 2.6801909307875896e-05,
"loss": 1.2382,
"step": 5630
},
{
"epoch": 10.04,
"learning_rate": 2.6778042959427208e-05,
"loss": 1.1377,
"step": 5640
},
{
"epoch": 10.05,
"learning_rate": 2.6754176610978522e-05,
"loss": 1.1432,
"step": 5650
},
{
"epoch": 10.07,
"learning_rate": 2.6730310262529833e-05,
"loss": 1.1396,
"step": 5660
},
{
"epoch": 10.09,
"learning_rate": 2.6706443914081148e-05,
"loss": 1.1391,
"step": 5670
},
{
"epoch": 10.11,
"learning_rate": 2.6682577565632463e-05,
"loss": 1.1435,
"step": 5680
},
{
"epoch": 10.12,
"learning_rate": 2.6658711217183774e-05,
"loss": 1.1434,
"step": 5690
},
{
"epoch": 10.14,
"learning_rate": 2.6634844868735085e-05,
"loss": 1.1376,
"step": 5700
},
{
"epoch": 10.16,
"learning_rate": 2.6610978520286396e-05,
"loss": 1.1327,
"step": 5710
},
{
"epoch": 10.18,
"learning_rate": 2.658711217183771e-05,
"loss": 1.1399,
"step": 5720
},
{
"epoch": 10.2,
"learning_rate": 2.6563245823389022e-05,
"loss": 1.1327,
"step": 5730
},
{
"epoch": 10.21,
"learning_rate": 2.6539379474940336e-05,
"loss": 1.1419,
"step": 5740
},
{
"epoch": 10.23,
"learning_rate": 2.651551312649165e-05,
"loss": 1.1399,
"step": 5750
},
{
"epoch": 10.25,
"learning_rate": 2.6491646778042962e-05,
"loss": 1.1369,
"step": 5760
},
{
"epoch": 10.27,
"learning_rate": 2.6467780429594277e-05,
"loss": 1.1332,
"step": 5770
},
{
"epoch": 10.28,
"learning_rate": 2.6443914081145585e-05,
"loss": 1.1368,
"step": 5780
},
{
"epoch": 10.3,
"learning_rate": 2.64200477326969e-05,
"loss": 1.1356,
"step": 5790
},
{
"epoch": 10.32,
"learning_rate": 2.639618138424821e-05,
"loss": 1.1404,
"step": 5800
},
{
"epoch": 10.34,
"learning_rate": 2.6372315035799525e-05,
"loss": 1.1387,
"step": 5810
},
{
"epoch": 10.36,
"learning_rate": 2.634844868735084e-05,
"loss": 1.1379,
"step": 5820
},
{
"epoch": 10.37,
"learning_rate": 2.632458233890215e-05,
"loss": 1.1359,
"step": 5830
},
{
"epoch": 10.39,
"learning_rate": 2.6300715990453465e-05,
"loss": 1.1399,
"step": 5840
},
{
"epoch": 10.41,
"learning_rate": 2.6276849642004773e-05,
"loss": 1.1411,
"step": 5850
},
{
"epoch": 10.43,
"learning_rate": 2.6252983293556088e-05,
"loss": 1.1372,
"step": 5860
},
{
"epoch": 10.44,
"learning_rate": 2.62291169451074e-05,
"loss": 1.1387,
"step": 5870
},
{
"epoch": 10.46,
"learning_rate": 2.6205250596658713e-05,
"loss": 1.1331,
"step": 5880
},
{
"epoch": 10.48,
"learning_rate": 2.6181384248210028e-05,
"loss": 1.1337,
"step": 5890
},
{
"epoch": 10.5,
"learning_rate": 2.615751789976134e-05,
"loss": 1.1312,
"step": 5900
},
{
"epoch": 10.52,
"learning_rate": 2.6133651551312654e-05,
"loss": 1.1315,
"step": 5910
},
{
"epoch": 10.53,
"learning_rate": 2.610978520286396e-05,
"loss": 1.128,
"step": 5920
},
{
"epoch": 10.55,
"learning_rate": 2.6085918854415276e-05,
"loss": 1.136,
"step": 5930
},
{
"epoch": 10.57,
"learning_rate": 2.6062052505966587e-05,
"loss": 1.1284,
"step": 5940
},
{
"epoch": 10.59,
"learning_rate": 2.6038186157517902e-05,
"loss": 1.1289,
"step": 5950
},
{
"epoch": 10.6,
"learning_rate": 2.6014319809069216e-05,
"loss": 1.1307,
"step": 5960
},
{
"epoch": 10.62,
"learning_rate": 2.5990453460620528e-05,
"loss": 1.1337,
"step": 5970
},
{
"epoch": 10.64,
"learning_rate": 2.5966587112171842e-05,
"loss": 1.131,
"step": 5980
},
{
"epoch": 10.66,
"learning_rate": 2.594272076372315e-05,
"loss": 1.1269,
"step": 5990
},
{
"epoch": 10.68,
"learning_rate": 2.5918854415274465e-05,
"loss": 1.1294,
"step": 6000
},
{
"epoch": 10.69,
"learning_rate": 2.5894988066825776e-05,
"loss": 1.1323,
"step": 6010
},
{
"epoch": 10.71,
"learning_rate": 2.587112171837709e-05,
"loss": 1.1295,
"step": 6020
},
{
"epoch": 10.73,
"learning_rate": 2.5847255369928405e-05,
"loss": 1.128,
"step": 6030
},
{
"epoch": 10.75,
"learning_rate": 2.5823389021479716e-05,
"loss": 1.1304,
"step": 6040
},
{
"epoch": 10.76,
"learning_rate": 2.579952267303103e-05,
"loss": 1.1334,
"step": 6050
},
{
"epoch": 10.78,
"learning_rate": 2.577565632458234e-05,
"loss": 1.1298,
"step": 6060
},
{
"epoch": 10.8,
"learning_rate": 2.5751789976133653e-05,
"loss": 1.1292,
"step": 6070
},
{
"epoch": 10.82,
"learning_rate": 2.5727923627684964e-05,
"loss": 1.1316,
"step": 6080
},
{
"epoch": 10.83,
"learning_rate": 2.570405727923628e-05,
"loss": 1.1309,
"step": 6090
},
{
"epoch": 10.85,
"learning_rate": 2.5680190930787593e-05,
"loss": 1.1287,
"step": 6100
},
{
"epoch": 10.87,
"learning_rate": 2.5656324582338905e-05,
"loss": 1.1264,
"step": 6110
},
{
"epoch": 10.89,
"learning_rate": 2.563245823389022e-05,
"loss": 1.1258,
"step": 6120
},
{
"epoch": 10.91,
"learning_rate": 2.560859188544153e-05,
"loss": 1.1285,
"step": 6130
},
{
"epoch": 10.92,
"learning_rate": 2.558472553699284e-05,
"loss": 1.1229,
"step": 6140
},
{
"epoch": 10.94,
"learning_rate": 2.5560859188544153e-05,
"loss": 1.1322,
"step": 6150
},
{
"epoch": 10.96,
"learning_rate": 2.5536992840095467e-05,
"loss": 1.1299,
"step": 6160
},
{
"epoch": 10.98,
"learning_rate": 2.5513126491646782e-05,
"loss": 1.1307,
"step": 6170
},
{
"epoch": 10.99,
"learning_rate": 2.5489260143198093e-05,
"loss": 1.1308,
"step": 6180
},
{
"epoch": 11.0,
"eval_cer": 4.297008547008547,
"eval_loss": 1.0878677368164062,
"eval_runtime": 4.8141,
"eval_samples_per_second": 2.077,
"eval_steps_per_second": 0.208,
"eval_wer": 0.974025974025974,
"step": 6182
},
{
"epoch": 11.01,
"learning_rate": 2.5465393794749408e-05,
"loss": 1.2266,
"step": 6190
},
{
"epoch": 11.03,
"learning_rate": 2.544152744630072e-05,
"loss": 1.1325,
"step": 6200
},
{
"epoch": 11.05,
"learning_rate": 2.541766109785203e-05,
"loss": 1.1277,
"step": 6210
},
{
"epoch": 11.07,
"learning_rate": 2.539379474940334e-05,
"loss": 1.1332,
"step": 6220
},
{
"epoch": 11.09,
"learning_rate": 2.5369928400954656e-05,
"loss": 1.1235,
"step": 6230
},
{
"epoch": 11.1,
"learning_rate": 2.5346062052505967e-05,
"loss": 1.127,
"step": 6240
},
{
"epoch": 11.12,
"learning_rate": 2.532219570405728e-05,
"loss": 1.1187,
"step": 6250
},
{
"epoch": 11.14,
"learning_rate": 2.5298329355608596e-05,
"loss": 1.1235,
"step": 6260
},
{
"epoch": 11.16,
"learning_rate": 2.5274463007159907e-05,
"loss": 1.1295,
"step": 6270
},
{
"epoch": 11.17,
"learning_rate": 2.525059665871122e-05,
"loss": 1.1258,
"step": 6280
},
{
"epoch": 11.19,
"learning_rate": 2.522673031026253e-05,
"loss": 1.1173,
"step": 6290
},
{
"epoch": 11.21,
"learning_rate": 2.5202863961813844e-05,
"loss": 1.1275,
"step": 6300
},
{
"epoch": 11.23,
"learning_rate": 2.5178997613365155e-05,
"loss": 1.1241,
"step": 6310
},
{
"epoch": 11.25,
"learning_rate": 2.515513126491647e-05,
"loss": 1.1286,
"step": 6320
},
{
"epoch": 11.26,
"learning_rate": 2.5131264916467785e-05,
"loss": 1.1231,
"step": 6330
},
{
"epoch": 11.28,
"learning_rate": 2.5107398568019096e-05,
"loss": 1.123,
"step": 6340
},
{
"epoch": 11.3,
"learning_rate": 2.5083532219570407e-05,
"loss": 1.1236,
"step": 6350
},
{
"epoch": 11.32,
"learning_rate": 2.5059665871121718e-05,
"loss": 1.1244,
"step": 6360
},
{
"epoch": 11.33,
"learning_rate": 2.5035799522673033e-05,
"loss": 1.1254,
"step": 6370
},
{
"epoch": 11.35,
"learning_rate": 2.5011933174224344e-05,
"loss": 1.1251,
"step": 6380
},
{
"epoch": 11.37,
"learning_rate": 2.498806682577566e-05,
"loss": 1.1219,
"step": 6390
},
{
"epoch": 11.39,
"learning_rate": 2.4964200477326973e-05,
"loss": 1.12,
"step": 6400
},
{
"epoch": 11.41,
"learning_rate": 2.4940334128878284e-05,
"loss": 1.1189,
"step": 6410
},
{
"epoch": 11.42,
"learning_rate": 2.49164677804296e-05,
"loss": 1.1226,
"step": 6420
},
{
"epoch": 11.44,
"learning_rate": 2.4892601431980907e-05,
"loss": 1.118,
"step": 6430
},
{
"epoch": 11.46,
"learning_rate": 2.486873508353222e-05,
"loss": 1.125,
"step": 6440
},
{
"epoch": 11.48,
"learning_rate": 2.4844868735083532e-05,
"loss": 1.1208,
"step": 6450
},
{
"epoch": 11.49,
"learning_rate": 2.4821002386634847e-05,
"loss": 1.1205,
"step": 6460
},
{
"epoch": 11.51,
"learning_rate": 2.479713603818616e-05,
"loss": 1.1263,
"step": 6470
},
{
"epoch": 11.53,
"learning_rate": 2.4773269689737473e-05,
"loss": 1.1177,
"step": 6480
},
{
"epoch": 11.55,
"learning_rate": 2.4749403341288787e-05,
"loss": 1.127,
"step": 6490
},
{
"epoch": 11.56,
"learning_rate": 2.4725536992840095e-05,
"loss": 1.1226,
"step": 6500
},
{
"epoch": 11.58,
"learning_rate": 2.470167064439141e-05,
"loss": 1.1202,
"step": 6510
},
{
"epoch": 11.6,
"learning_rate": 2.467780429594272e-05,
"loss": 1.122,
"step": 6520
},
{
"epoch": 11.62,
"learning_rate": 2.4653937947494035e-05,
"loss": 1.1176,
"step": 6530
},
{
"epoch": 11.64,
"learning_rate": 2.463007159904535e-05,
"loss": 1.1171,
"step": 6540
},
{
"epoch": 11.65,
"learning_rate": 2.460620525059666e-05,
"loss": 1.1223,
"step": 6550
},
{
"epoch": 11.67,
"learning_rate": 2.4582338902147976e-05,
"loss": 1.1158,
"step": 6560
},
{
"epoch": 11.69,
"learning_rate": 2.4558472553699284e-05,
"loss": 1.1166,
"step": 6570
},
{
"epoch": 11.71,
"learning_rate": 2.4534606205250598e-05,
"loss": 1.1195,
"step": 6580
},
{
"epoch": 11.72,
"learning_rate": 2.451073985680191e-05,
"loss": 1.1172,
"step": 6590
},
{
"epoch": 11.74,
"learning_rate": 2.4486873508353224e-05,
"loss": 1.1248,
"step": 6600
},
{
"epoch": 11.76,
"learning_rate": 2.446300715990454e-05,
"loss": 1.1177,
"step": 6610
},
{
"epoch": 11.78,
"learning_rate": 2.443914081145585e-05,
"loss": 1.1217,
"step": 6620
},
{
"epoch": 11.8,
"learning_rate": 2.4415274463007164e-05,
"loss": 1.1222,
"step": 6630
},
{
"epoch": 11.81,
"learning_rate": 2.4391408114558472e-05,
"loss": 1.1153,
"step": 6640
},
{
"epoch": 11.83,
"learning_rate": 2.4367541766109787e-05,
"loss": 1.12,
"step": 6650
},
{
"epoch": 11.85,
"learning_rate": 2.4343675417661098e-05,
"loss": 1.1192,
"step": 6660
},
{
"epoch": 11.87,
"learning_rate": 2.4319809069212412e-05,
"loss": 1.116,
"step": 6670
},
{
"epoch": 11.88,
"learning_rate": 2.4295942720763727e-05,
"loss": 1.1176,
"step": 6680
},
{
"epoch": 11.9,
"learning_rate": 2.4272076372315038e-05,
"loss": 1.1188,
"step": 6690
},
{
"epoch": 11.92,
"learning_rate": 2.4248210023866353e-05,
"loss": 1.1184,
"step": 6700
},
{
"epoch": 11.94,
"learning_rate": 2.4224343675417664e-05,
"loss": 1.1178,
"step": 6710
},
{
"epoch": 11.96,
"learning_rate": 2.4200477326968975e-05,
"loss": 1.1159,
"step": 6720
},
{
"epoch": 11.97,
"learning_rate": 2.4176610978520286e-05,
"loss": 1.1158,
"step": 6730
},
{
"epoch": 11.99,
"learning_rate": 2.41527446300716e-05,
"loss": 1.1174,
"step": 6740
},
{
"epoch": 12.0,
"eval_cer": 4.730769230769231,
"eval_loss": 1.0691099166870117,
"eval_runtime": 4.8405,
"eval_samples_per_second": 2.066,
"eval_steps_per_second": 0.207,
"eval_wer": 0.961038961038961,
"step": 6744
},
{
"epoch": 12.01,
"learning_rate": 2.4128878281622915e-05,
"loss": 1.2111,
"step": 6750
},
{
"epoch": 12.03,
"learning_rate": 2.4105011933174227e-05,
"loss": 1.1112,
"step": 6760
},
{
"epoch": 12.05,
"learning_rate": 2.408114558472554e-05,
"loss": 1.1195,
"step": 6770
},
{
"epoch": 12.06,
"learning_rate": 2.4057279236276852e-05,
"loss": 1.1161,
"step": 6780
},
{
"epoch": 12.08,
"learning_rate": 2.4033412887828164e-05,
"loss": 1.1169,
"step": 6790
},
{
"epoch": 12.1,
"learning_rate": 2.4009546539379475e-05,
"loss": 1.1146,
"step": 6800
},
{
"epoch": 12.12,
"learning_rate": 2.398568019093079e-05,
"loss": 1.1129,
"step": 6810
},
{
"epoch": 12.14,
"learning_rate": 2.3961813842482104e-05,
"loss": 1.1198,
"step": 6820
},
{
"epoch": 12.15,
"learning_rate": 2.3937947494033415e-05,
"loss": 1.1119,
"step": 6830
},
{
"epoch": 12.17,
"learning_rate": 2.391408114558473e-05,
"loss": 1.1186,
"step": 6840
},
{
"epoch": 12.19,
"learning_rate": 2.389021479713604e-05,
"loss": 1.1116,
"step": 6850
},
{
"epoch": 12.21,
"learning_rate": 2.3866348448687352e-05,
"loss": 1.1108,
"step": 6860
},
{
"epoch": 12.22,
"learning_rate": 2.3842482100238663e-05,
"loss": 1.1109,
"step": 6870
},
{
"epoch": 12.24,
"learning_rate": 2.3818615751789978e-05,
"loss": 1.1153,
"step": 6880
},
{
"epoch": 12.26,
"learning_rate": 2.379474940334129e-05,
"loss": 1.1103,
"step": 6890
},
{
"epoch": 12.28,
"learning_rate": 2.3770883054892604e-05,
"loss": 1.1201,
"step": 6900
},
{
"epoch": 12.29,
"learning_rate": 2.3747016706443918e-05,
"loss": 1.1123,
"step": 6910
},
{
"epoch": 12.31,
"learning_rate": 2.372315035799523e-05,
"loss": 1.1139,
"step": 6920
},
{
"epoch": 12.33,
"learning_rate": 2.369928400954654e-05,
"loss": 1.1124,
"step": 6930
},
{
"epoch": 12.35,
"learning_rate": 2.3675417661097852e-05,
"loss": 1.1153,
"step": 6940
},
{
"epoch": 12.37,
"learning_rate": 2.3651551312649166e-05,
"loss": 1.1096,
"step": 6950
},
{
"epoch": 12.38,
"learning_rate": 2.3627684964200477e-05,
"loss": 1.1086,
"step": 6960
},
{
"epoch": 12.4,
"learning_rate": 2.3603818615751792e-05,
"loss": 1.1099,
"step": 6970
},
{
"epoch": 12.42,
"learning_rate": 2.3579952267303107e-05,
"loss": 1.1112,
"step": 6980
},
{
"epoch": 12.44,
"learning_rate": 2.3556085918854418e-05,
"loss": 1.1135,
"step": 6990
},
{
"epoch": 12.45,
"learning_rate": 2.353221957040573e-05,
"loss": 1.113,
"step": 7000
},
{
"epoch": 12.47,
"learning_rate": 2.350835322195704e-05,
"loss": 1.1106,
"step": 7010
},
{
"epoch": 12.49,
"learning_rate": 2.3484486873508355e-05,
"loss": 1.1115,
"step": 7020
},
{
"epoch": 12.51,
"learning_rate": 2.3460620525059666e-05,
"loss": 1.1145,
"step": 7030
},
{
"epoch": 12.53,
"learning_rate": 2.343675417661098e-05,
"loss": 1.1142,
"step": 7040
},
{
"epoch": 12.54,
"learning_rate": 2.3412887828162295e-05,
"loss": 1.1122,
"step": 7050
},
{
"epoch": 12.56,
"learning_rate": 2.3389021479713606e-05,
"loss": 1.1132,
"step": 7060
},
{
"epoch": 12.58,
"learning_rate": 2.336515513126492e-05,
"loss": 1.1123,
"step": 7070
},
{
"epoch": 12.6,
"learning_rate": 2.334128878281623e-05,
"loss": 1.1019,
"step": 7080
},
{
"epoch": 12.61,
"learning_rate": 2.3317422434367543e-05,
"loss": 1.1118,
"step": 7090
},
{
"epoch": 12.63,
"learning_rate": 2.3293556085918854e-05,
"loss": 1.1148,
"step": 7100
},
{
"epoch": 12.65,
"learning_rate": 2.326968973747017e-05,
"loss": 1.1112,
"step": 7110
},
{
"epoch": 12.67,
"learning_rate": 2.3245823389021484e-05,
"loss": 1.1084,
"step": 7120
},
{
"epoch": 12.69,
"learning_rate": 2.3221957040572795e-05,
"loss": 1.1093,
"step": 7130
},
{
"epoch": 12.7,
"learning_rate": 2.319809069212411e-05,
"loss": 1.1055,
"step": 7140
},
{
"epoch": 12.72,
"learning_rate": 2.3174224343675417e-05,
"loss": 1.1042,
"step": 7150
},
{
"epoch": 12.74,
"learning_rate": 2.3150357995226732e-05,
"loss": 1.1112,
"step": 7160
},
{
"epoch": 12.76,
"learning_rate": 2.3126491646778043e-05,
"loss": 1.1113,
"step": 7170
},
{
"epoch": 12.77,
"learning_rate": 2.3102625298329358e-05,
"loss": 1.1092,
"step": 7180
},
{
"epoch": 12.79,
"learning_rate": 2.3078758949880672e-05,
"loss": 1.1074,
"step": 7190
},
{
"epoch": 12.81,
"learning_rate": 2.3054892601431983e-05,
"loss": 1.1063,
"step": 7200
},
{
"epoch": 12.83,
"learning_rate": 2.3031026252983298e-05,
"loss": 1.1079,
"step": 7210
},
{
"epoch": 12.85,
"learning_rate": 2.3007159904534606e-05,
"loss": 1.1041,
"step": 7220
},
{
"epoch": 12.86,
"learning_rate": 2.298329355608592e-05,
"loss": 1.1093,
"step": 7230
},
{
"epoch": 12.88,
"learning_rate": 2.295942720763723e-05,
"loss": 1.1055,
"step": 7240
},
{
"epoch": 12.9,
"learning_rate": 2.2935560859188546e-05,
"loss": 1.1035,
"step": 7250
},
{
"epoch": 12.92,
"learning_rate": 2.291169451073986e-05,
"loss": 1.104,
"step": 7260
},
{
"epoch": 12.93,
"learning_rate": 2.2887828162291172e-05,
"loss": 1.1079,
"step": 7270
},
{
"epoch": 12.95,
"learning_rate": 2.2863961813842486e-05,
"loss": 1.1052,
"step": 7280
},
{
"epoch": 12.97,
"learning_rate": 2.2840095465393794e-05,
"loss": 1.1054,
"step": 7290
},
{
"epoch": 12.99,
"learning_rate": 2.281622911694511e-05,
"loss": 1.1079,
"step": 7300
},
{
"epoch": 13.0,
"eval_cer": 4.756410256410256,
"eval_loss": 1.052440881729126,
"eval_runtime": 4.8878,
"eval_samples_per_second": 2.046,
"eval_steps_per_second": 0.205,
"eval_wer": 0.935064935064935,
"step": 7306
},
{
"epoch": 13.01,
"learning_rate": 2.279236276849642e-05,
"loss": 1.2048,
"step": 7310
},
{
"epoch": 13.02,
"learning_rate": 2.2768496420047734e-05,
"loss": 1.1063,
"step": 7320
},
{
"epoch": 13.04,
"learning_rate": 2.274463007159905e-05,
"loss": 1.1031,
"step": 7330
},
{
"epoch": 13.06,
"learning_rate": 2.272076372315036e-05,
"loss": 1.1081,
"step": 7340
},
{
"epoch": 13.08,
"learning_rate": 2.2696897374701675e-05,
"loss": 1.1055,
"step": 7350
},
{
"epoch": 13.1,
"learning_rate": 2.2673031026252986e-05,
"loss": 1.1054,
"step": 7360
},
{
"epoch": 13.11,
"learning_rate": 2.2649164677804297e-05,
"loss": 1.1096,
"step": 7370
},
{
"epoch": 13.13,
"learning_rate": 2.262529832935561e-05,
"loss": 1.1055,
"step": 7380
},
{
"epoch": 13.15,
"learning_rate": 2.2601431980906923e-05,
"loss": 1.1085,
"step": 7390
},
{
"epoch": 13.17,
"learning_rate": 2.2577565632458238e-05,
"loss": 1.1027,
"step": 7400
},
{
"epoch": 13.18,
"learning_rate": 2.255369928400955e-05,
"loss": 1.0999,
"step": 7410
},
{
"epoch": 13.2,
"learning_rate": 2.2529832935560863e-05,
"loss": 1.1009,
"step": 7420
},
{
"epoch": 13.22,
"learning_rate": 2.2505966587112174e-05,
"loss": 1.1078,
"step": 7430
},
{
"epoch": 13.24,
"learning_rate": 2.2482100238663486e-05,
"loss": 1.1022,
"step": 7440
},
{
"epoch": 13.26,
"learning_rate": 2.2458233890214797e-05,
"loss": 1.1037,
"step": 7450
},
{
"epoch": 13.27,
"learning_rate": 2.243436754176611e-05,
"loss": 1.1061,
"step": 7460
},
{
"epoch": 13.29,
"learning_rate": 2.2410501193317426e-05,
"loss": 1.1,
"step": 7470
},
{
"epoch": 13.31,
"learning_rate": 2.2386634844868737e-05,
"loss": 1.1011,
"step": 7480
},
{
"epoch": 13.33,
"learning_rate": 2.2362768496420052e-05,
"loss": 1.1024,
"step": 7490
},
{
"epoch": 13.34,
"learning_rate": 2.2338902147971363e-05,
"loss": 1.1005,
"step": 7500
},
{
"epoch": 13.36,
"learning_rate": 2.2315035799522674e-05,
"loss": 1.1009,
"step": 7510
},
{
"epoch": 13.38,
"learning_rate": 2.2291169451073985e-05,
"loss": 1.0998,
"step": 7520
},
{
"epoch": 13.4,
"learning_rate": 2.22673031026253e-05,
"loss": 1.0988,
"step": 7530
},
{
"epoch": 13.42,
"learning_rate": 2.224343675417661e-05,
"loss": 1.1003,
"step": 7540
},
{
"epoch": 13.43,
"learning_rate": 2.2219570405727926e-05,
"loss": 1.1013,
"step": 7550
},
{
"epoch": 13.45,
"learning_rate": 2.219570405727924e-05,
"loss": 1.101,
"step": 7560
},
{
"epoch": 13.47,
"learning_rate": 2.217183770883055e-05,
"loss": 1.1077,
"step": 7570
},
{
"epoch": 13.49,
"learning_rate": 2.2147971360381863e-05,
"loss": 1.0992,
"step": 7580
},
{
"epoch": 13.5,
"learning_rate": 2.2124105011933174e-05,
"loss": 1.1058,
"step": 7590
},
{
"epoch": 13.52,
"learning_rate": 2.210023866348449e-05,
"loss": 1.1066,
"step": 7600
},
{
"epoch": 13.54,
"learning_rate": 2.20763723150358e-05,
"loss": 1.1023,
"step": 7610
},
{
"epoch": 13.56,
"learning_rate": 2.2052505966587114e-05,
"loss": 1.0965,
"step": 7620
},
{
"epoch": 13.58,
"learning_rate": 2.202863961813843e-05,
"loss": 1.1001,
"step": 7630
},
{
"epoch": 13.59,
"learning_rate": 2.200477326968974e-05,
"loss": 1.0979,
"step": 7640
},
{
"epoch": 13.61,
"learning_rate": 2.198090692124105e-05,
"loss": 1.1039,
"step": 7650
},
{
"epoch": 13.63,
"learning_rate": 2.1957040572792362e-05,
"loss": 1.096,
"step": 7660
},
{
"epoch": 13.65,
"learning_rate": 2.1933174224343677e-05,
"loss": 1.1098,
"step": 7670
},
{
"epoch": 13.66,
"learning_rate": 2.1909307875894988e-05,
"loss": 1.1036,
"step": 7680
},
{
"epoch": 13.68,
"learning_rate": 2.1885441527446303e-05,
"loss": 1.1009,
"step": 7690
},
{
"epoch": 13.7,
"learning_rate": 2.1861575178997617e-05,
"loss": 1.0988,
"step": 7700
},
{
"epoch": 13.72,
"learning_rate": 2.183770883054893e-05,
"loss": 1.0952,
"step": 7710
},
{
"epoch": 13.74,
"learning_rate": 2.1813842482100243e-05,
"loss": 1.104,
"step": 7720
},
{
"epoch": 13.75,
"learning_rate": 2.178997613365155e-05,
"loss": 1.1001,
"step": 7730
},
{
"epoch": 13.77,
"learning_rate": 2.1766109785202865e-05,
"loss": 1.1,
"step": 7740
},
{
"epoch": 13.79,
"learning_rate": 2.1742243436754177e-05,
"loss": 1.0953,
"step": 7750
},
{
"epoch": 13.81,
"learning_rate": 2.171837708830549e-05,
"loss": 1.0947,
"step": 7760
},
{
"epoch": 13.82,
"learning_rate": 2.1694510739856806e-05,
"loss": 1.1016,
"step": 7770
},
{
"epoch": 13.84,
"learning_rate": 2.1670644391408117e-05,
"loss": 1.0941,
"step": 7780
},
{
"epoch": 13.86,
"learning_rate": 2.164677804295943e-05,
"loss": 1.0954,
"step": 7790
},
{
"epoch": 13.88,
"learning_rate": 2.162291169451074e-05,
"loss": 1.1,
"step": 7800
},
{
"epoch": 13.9,
"learning_rate": 2.1599045346062054e-05,
"loss": 1.0974,
"step": 7810
},
{
"epoch": 13.91,
"learning_rate": 2.1575178997613365e-05,
"loss": 1.0969,
"step": 7820
},
{
"epoch": 13.93,
"learning_rate": 2.155131264916468e-05,
"loss": 1.0981,
"step": 7830
},
{
"epoch": 13.95,
"learning_rate": 2.1527446300715994e-05,
"loss": 1.0993,
"step": 7840
},
{
"epoch": 13.97,
"learning_rate": 2.1503579952267305e-05,
"loss": 1.0901,
"step": 7850
},
{
"epoch": 13.98,
"learning_rate": 2.147971360381862e-05,
"loss": 1.1019,
"step": 7860
},
{
"epoch": 14.0,
"eval_cer": 4.754273504273504,
"eval_loss": 1.0490120649337769,
"eval_runtime": 4.8398,
"eval_samples_per_second": 2.066,
"eval_steps_per_second": 0.207,
"eval_wer": 0.961038961038961,
"step": 7868
}
],
"max_steps": 16860,
"num_train_epochs": 30,
"total_flos": 0.0,
"trial_name": null,
"trial_params": null
}