xls-r-300m-npsc-seq2seq / trainer_state.json
Rolv-Arild's picture
End of training
a991a30
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 19.9992034411343,
"global_step": 15680,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.0000000000000003e-07,
"loss": 11.2012,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 1.3e-06,
"loss": 11.0313,
"step": 20
},
{
"epoch": 0.04,
"learning_rate": 2.3e-06,
"loss": 9.915,
"step": 30
},
{
"epoch": 0.05,
"learning_rate": 3.3e-06,
"loss": 8.2315,
"step": 40
},
{
"epoch": 0.06,
"learning_rate": 4.2999999999999995e-06,
"loss": 7.1969,
"step": 50
},
{
"epoch": 0.08,
"learning_rate": 5.3e-06,
"loss": 6.5699,
"step": 60
},
{
"epoch": 0.09,
"learning_rate": 6.300000000000001e-06,
"loss": 5.9241,
"step": 70
},
{
"epoch": 0.1,
"learning_rate": 7.2999999999999996e-06,
"loss": 5.4646,
"step": 80
},
{
"epoch": 0.11,
"learning_rate": 8.3e-06,
"loss": 5.1939,
"step": 90
},
{
"epoch": 0.13,
"learning_rate": 9.3e-06,
"loss": 5.0085,
"step": 100
},
{
"epoch": 0.14,
"learning_rate": 1.03e-05,
"loss": 5.5541,
"step": 110
},
{
"epoch": 0.15,
"learning_rate": 1.13e-05,
"loss": 4.9172,
"step": 120
},
{
"epoch": 0.17,
"learning_rate": 1.23e-05,
"loss": 4.6835,
"step": 130
},
{
"epoch": 0.18,
"learning_rate": 1.3300000000000001e-05,
"loss": 4.5962,
"step": 140
},
{
"epoch": 0.19,
"learning_rate": 1.43e-05,
"loss": 4.3211,
"step": 150
},
{
"epoch": 0.2,
"learning_rate": 1.53e-05,
"loss": 4.8684,
"step": 160
},
{
"epoch": 0.22,
"learning_rate": 1.63e-05,
"loss": 4.4981,
"step": 170
},
{
"epoch": 0.23,
"learning_rate": 1.73e-05,
"loss": 4.3235,
"step": 180
},
{
"epoch": 0.24,
"learning_rate": 1.83e-05,
"loss": 4.2043,
"step": 190
},
{
"epoch": 0.25,
"learning_rate": 1.93e-05,
"loss": 3.9079,
"step": 200
},
{
"epoch": 0.27,
"learning_rate": 2.0300000000000002e-05,
"loss": 4.4068,
"step": 210
},
{
"epoch": 0.28,
"learning_rate": 2.13e-05,
"loss": 4.0767,
"step": 220
},
{
"epoch": 0.29,
"learning_rate": 2.23e-05,
"loss": 4.0105,
"step": 230
},
{
"epoch": 0.31,
"learning_rate": 2.3300000000000004e-05,
"loss": 3.8937,
"step": 240
},
{
"epoch": 0.32,
"learning_rate": 2.43e-05,
"loss": 3.4475,
"step": 250
},
{
"epoch": 0.33,
"learning_rate": 2.5300000000000002e-05,
"loss": 4.1418,
"step": 260
},
{
"epoch": 0.34,
"learning_rate": 2.6300000000000002e-05,
"loss": 3.8227,
"step": 270
},
{
"epoch": 0.36,
"learning_rate": 2.7300000000000003e-05,
"loss": 3.693,
"step": 280
},
{
"epoch": 0.37,
"learning_rate": 2.83e-05,
"loss": 3.6436,
"step": 290
},
{
"epoch": 0.38,
"learning_rate": 2.93e-05,
"loss": 3.2461,
"step": 300
},
{
"epoch": 0.4,
"learning_rate": 3.03e-05,
"loss": 3.9707,
"step": 310
},
{
"epoch": 0.41,
"learning_rate": 3.13e-05,
"loss": 3.6917,
"step": 320
},
{
"epoch": 0.42,
"learning_rate": 3.2300000000000006e-05,
"loss": 3.5928,
"step": 330
},
{
"epoch": 0.43,
"learning_rate": 3.33e-05,
"loss": 3.5301,
"step": 340
},
{
"epoch": 0.45,
"learning_rate": 3.430000000000001e-05,
"loss": 2.9921,
"step": 350
},
{
"epoch": 0.46,
"learning_rate": 3.53e-05,
"loss": 3.8641,
"step": 360
},
{
"epoch": 0.47,
"learning_rate": 3.63e-05,
"loss": 3.5432,
"step": 370
},
{
"epoch": 0.48,
"learning_rate": 3.73e-05,
"loss": 3.487,
"step": 380
},
{
"epoch": 0.5,
"learning_rate": 3.83e-05,
"loss": 3.3875,
"step": 390
},
{
"epoch": 0.51,
"learning_rate": 3.9300000000000007e-05,
"loss": 2.888,
"step": 400
},
{
"epoch": 0.51,
"eval_loss": 3.7320261001586914,
"eval_runtime": 946.9905,
"eval_samples_per_second": 6.582,
"eval_steps_per_second": 0.824,
"eval_wer": 0.9440106155491065,
"step": 400
},
{
"epoch": 0.52,
"learning_rate": 4.0300000000000004e-05,
"loss": 3.8227,
"step": 410
},
{
"epoch": 0.54,
"learning_rate": 4.13e-05,
"loss": 3.4862,
"step": 420
},
{
"epoch": 0.55,
"learning_rate": 4.23e-05,
"loss": 3.3335,
"step": 430
},
{
"epoch": 0.56,
"learning_rate": 4.33e-05,
"loss": 3.2408,
"step": 440
},
{
"epoch": 0.57,
"learning_rate": 4.43e-05,
"loss": 2.5279,
"step": 450
},
{
"epoch": 0.59,
"learning_rate": 4.53e-05,
"loss": 3.7253,
"step": 460
},
{
"epoch": 0.6,
"learning_rate": 4.630000000000001e-05,
"loss": 3.3742,
"step": 470
},
{
"epoch": 0.61,
"learning_rate": 4.73e-05,
"loss": 3.2955,
"step": 480
},
{
"epoch": 0.62,
"learning_rate": 4.83e-05,
"loss": 3.1341,
"step": 490
},
{
"epoch": 0.64,
"learning_rate": 4.93e-05,
"loss": 2.6154,
"step": 500
},
{
"epoch": 0.65,
"learning_rate": 5.03e-05,
"loss": 3.5928,
"step": 510
},
{
"epoch": 0.66,
"learning_rate": 5.130000000000001e-05,
"loss": 3.338,
"step": 520
},
{
"epoch": 0.68,
"learning_rate": 5.2300000000000004e-05,
"loss": 3.2339,
"step": 530
},
{
"epoch": 0.69,
"learning_rate": 5.330000000000001e-05,
"loss": 3.0817,
"step": 540
},
{
"epoch": 0.7,
"learning_rate": 5.4300000000000005e-05,
"loss": 2.453,
"step": 550
},
{
"epoch": 0.71,
"learning_rate": 5.530000000000001e-05,
"loss": 3.5553,
"step": 560
},
{
"epoch": 0.73,
"learning_rate": 5.63e-05,
"loss": 3.248,
"step": 570
},
{
"epoch": 0.74,
"learning_rate": 5.73e-05,
"loss": 3.1946,
"step": 580
},
{
"epoch": 0.75,
"learning_rate": 5.83e-05,
"loss": 3.0223,
"step": 590
},
{
"epoch": 0.76,
"learning_rate": 5.93e-05,
"loss": 2.3511,
"step": 600
},
{
"epoch": 0.78,
"learning_rate": 6.03e-05,
"loss": 3.5242,
"step": 610
},
{
"epoch": 0.79,
"learning_rate": 6.13e-05,
"loss": 3.1946,
"step": 620
},
{
"epoch": 0.8,
"learning_rate": 6.23e-05,
"loss": 3.1166,
"step": 630
},
{
"epoch": 0.82,
"learning_rate": 6.330000000000001e-05,
"loss": 2.9555,
"step": 640
},
{
"epoch": 0.83,
"learning_rate": 6.43e-05,
"loss": 2.2808,
"step": 650
},
{
"epoch": 0.84,
"learning_rate": 6.53e-05,
"loss": 3.5894,
"step": 660
},
{
"epoch": 0.85,
"learning_rate": 6.630000000000001e-05,
"loss": 3.203,
"step": 670
},
{
"epoch": 0.87,
"learning_rate": 6.730000000000001e-05,
"loss": 3.0484,
"step": 680
},
{
"epoch": 0.88,
"learning_rate": 6.83e-05,
"loss": 2.8733,
"step": 690
},
{
"epoch": 0.89,
"learning_rate": 6.93e-05,
"loss": 2.2043,
"step": 700
},
{
"epoch": 0.9,
"learning_rate": 7.03e-05,
"loss": 3.4822,
"step": 710
},
{
"epoch": 0.92,
"learning_rate": 7.13e-05,
"loss": 3.131,
"step": 720
},
{
"epoch": 0.93,
"learning_rate": 7.23e-05,
"loss": 2.9721,
"step": 730
},
{
"epoch": 0.94,
"learning_rate": 7.33e-05,
"loss": 2.8593,
"step": 740
},
{
"epoch": 0.96,
"learning_rate": 7.43e-05,
"loss": 2.2461,
"step": 750
},
{
"epoch": 0.97,
"learning_rate": 7.53e-05,
"loss": 3.4536,
"step": 760
},
{
"epoch": 0.98,
"learning_rate": 7.630000000000001e-05,
"loss": 3.0393,
"step": 770
},
{
"epoch": 0.99,
"learning_rate": 7.730000000000001e-05,
"loss": 2.7416,
"step": 780
},
{
"epoch": 1.01,
"learning_rate": 7.83e-05,
"loss": 3.0346,
"step": 790
},
{
"epoch": 1.02,
"learning_rate": 7.93e-05,
"loss": 3.1636,
"step": 800
},
{
"epoch": 1.02,
"eval_loss": 2.918823480606079,
"eval_runtime": 1358.8762,
"eval_samples_per_second": 4.587,
"eval_steps_per_second": 0.574,
"eval_wer": 1.1915762944860242,
"step": 800
},
{
"epoch": 1.03,
"learning_rate": 8.030000000000001e-05,
"loss": 2.9569,
"step": 810
},
{
"epoch": 1.05,
"learning_rate": 8.13e-05,
"loss": 2.7792,
"step": 820
},
{
"epoch": 1.06,
"learning_rate": 8.23e-05,
"loss": 2.3876,
"step": 830
},
{
"epoch": 1.07,
"learning_rate": 8.33e-05,
"loss": 2.8558,
"step": 840
},
{
"epoch": 1.08,
"learning_rate": 8.43e-05,
"loss": 3.059,
"step": 850
},
{
"epoch": 1.1,
"learning_rate": 8.53e-05,
"loss": 2.9143,
"step": 860
},
{
"epoch": 1.11,
"learning_rate": 8.63e-05,
"loss": 2.7392,
"step": 870
},
{
"epoch": 1.12,
"learning_rate": 8.730000000000001e-05,
"loss": 2.2904,
"step": 880
},
{
"epoch": 1.14,
"learning_rate": 8.83e-05,
"loss": 2.7131,
"step": 890
},
{
"epoch": 1.15,
"learning_rate": 8.93e-05,
"loss": 3.0171,
"step": 900
},
{
"epoch": 1.16,
"learning_rate": 9.030000000000001e-05,
"loss": 2.8487,
"step": 910
},
{
"epoch": 1.17,
"learning_rate": 9.130000000000001e-05,
"loss": 2.675,
"step": 920
},
{
"epoch": 1.19,
"learning_rate": 9.230000000000001e-05,
"loss": 2.1836,
"step": 930
},
{
"epoch": 1.2,
"learning_rate": 9.33e-05,
"loss": 2.7424,
"step": 940
},
{
"epoch": 1.21,
"learning_rate": 9.43e-05,
"loss": 3.0442,
"step": 950
},
{
"epoch": 1.22,
"learning_rate": 9.53e-05,
"loss": 2.8557,
"step": 960
},
{
"epoch": 1.24,
"learning_rate": 9.63e-05,
"loss": 2.6003,
"step": 970
},
{
"epoch": 1.25,
"learning_rate": 9.730000000000001e-05,
"loss": 2.1095,
"step": 980
},
{
"epoch": 1.26,
"learning_rate": 9.83e-05,
"loss": 2.6137,
"step": 990
},
{
"epoch": 1.28,
"learning_rate": 9.93e-05,
"loss": 2.9493,
"step": 1000
},
{
"epoch": 1.29,
"learning_rate": 9.997956403269755e-05,
"loss": 2.8053,
"step": 1010
},
{
"epoch": 1.3,
"learning_rate": 9.991144414168938e-05,
"loss": 2.5713,
"step": 1020
},
{
"epoch": 1.31,
"learning_rate": 9.984332425068121e-05,
"loss": 2.0882,
"step": 1030
},
{
"epoch": 1.33,
"learning_rate": 9.977520435967302e-05,
"loss": 2.6747,
"step": 1040
},
{
"epoch": 1.34,
"learning_rate": 9.970708446866486e-05,
"loss": 2.9617,
"step": 1050
},
{
"epoch": 1.35,
"learning_rate": 9.963896457765668e-05,
"loss": 2.7173,
"step": 1060
},
{
"epoch": 1.36,
"learning_rate": 9.95708446866485e-05,
"loss": 2.5349,
"step": 1070
},
{
"epoch": 1.38,
"learning_rate": 9.950272479564033e-05,
"loss": 2.1039,
"step": 1080
},
{
"epoch": 1.39,
"learning_rate": 9.943460490463216e-05,
"loss": 2.5893,
"step": 1090
},
{
"epoch": 1.4,
"learning_rate": 9.936648501362397e-05,
"loss": 2.8321,
"step": 1100
},
{
"epoch": 1.42,
"learning_rate": 9.929836512261581e-05,
"loss": 2.6416,
"step": 1110
},
{
"epoch": 1.43,
"learning_rate": 9.923024523160763e-05,
"loss": 2.4261,
"step": 1120
},
{
"epoch": 1.44,
"learning_rate": 9.916212534059946e-05,
"loss": 1.9565,
"step": 1130
},
{
"epoch": 1.45,
"learning_rate": 9.909400544959128e-05,
"loss": 2.5986,
"step": 1140
},
{
"epoch": 1.47,
"learning_rate": 9.902588555858311e-05,
"loss": 2.7577,
"step": 1150
},
{
"epoch": 1.48,
"learning_rate": 9.895776566757494e-05,
"loss": 2.5457,
"step": 1160
},
{
"epoch": 1.49,
"learning_rate": 9.888964577656676e-05,
"loss": 2.2434,
"step": 1170
},
{
"epoch": 1.5,
"learning_rate": 9.882152588555859e-05,
"loss": 1.6913,
"step": 1180
},
{
"epoch": 1.52,
"learning_rate": 9.875340599455042e-05,
"loss": 2.6432,
"step": 1190
},
{
"epoch": 1.53,
"learning_rate": 9.868528610354225e-05,
"loss": 2.773,
"step": 1200
},
{
"epoch": 1.53,
"eval_loss": 2.3347249031066895,
"eval_runtime": 1367.3502,
"eval_samples_per_second": 4.558,
"eval_steps_per_second": 0.57,
"eval_wer": 1.013412631739728,
"step": 1200
},
{
"epoch": 1.54,
"learning_rate": 9.861716621253406e-05,
"loss": 2.3554,
"step": 1210
},
{
"epoch": 1.56,
"learning_rate": 9.85490463215259e-05,
"loss": 1.9262,
"step": 1220
},
{
"epoch": 1.57,
"learning_rate": 9.848092643051771e-05,
"loss": 1.4834,
"step": 1230
},
{
"epoch": 1.58,
"learning_rate": 9.841280653950954e-05,
"loss": 2.4816,
"step": 1240
},
{
"epoch": 1.59,
"learning_rate": 9.834468664850137e-05,
"loss": 2.415,
"step": 1250
},
{
"epoch": 1.61,
"learning_rate": 9.82765667574932e-05,
"loss": 1.9016,
"step": 1260
},
{
"epoch": 1.62,
"learning_rate": 9.820844686648501e-05,
"loss": 1.5463,
"step": 1270
},
{
"epoch": 1.63,
"learning_rate": 9.814032697547685e-05,
"loss": 1.2685,
"step": 1280
},
{
"epoch": 1.64,
"learning_rate": 9.807220708446867e-05,
"loss": 2.2386,
"step": 1290
},
{
"epoch": 1.66,
"learning_rate": 9.800408719346049e-05,
"loss": 1.9767,
"step": 1300
},
{
"epoch": 1.67,
"learning_rate": 9.793596730245232e-05,
"loss": 1.5275,
"step": 1310
},
{
"epoch": 1.68,
"learning_rate": 9.786784741144415e-05,
"loss": 1.2746,
"step": 1320
},
{
"epoch": 1.7,
"learning_rate": 9.779972752043597e-05,
"loss": 1.1684,
"step": 1330
},
{
"epoch": 1.71,
"learning_rate": 9.77316076294278e-05,
"loss": 2.0181,
"step": 1340
},
{
"epoch": 1.72,
"learning_rate": 9.766348773841963e-05,
"loss": 1.6714,
"step": 1350
},
{
"epoch": 1.73,
"learning_rate": 9.759536784741144e-05,
"loss": 1.2632,
"step": 1360
},
{
"epoch": 1.75,
"learning_rate": 9.752724795640328e-05,
"loss": 1.1007,
"step": 1370
},
{
"epoch": 1.76,
"learning_rate": 9.74591280653951e-05,
"loss": 1.0076,
"step": 1380
},
{
"epoch": 1.77,
"learning_rate": 9.739100817438692e-05,
"loss": 1.8759,
"step": 1390
},
{
"epoch": 1.79,
"learning_rate": 9.732288828337875e-05,
"loss": 1.4086,
"step": 1400
},
{
"epoch": 1.8,
"learning_rate": 9.725476839237058e-05,
"loss": 1.1063,
"step": 1410
},
{
"epoch": 1.81,
"learning_rate": 9.718664850136239e-05,
"loss": 1.0004,
"step": 1420
},
{
"epoch": 1.82,
"learning_rate": 9.711852861035423e-05,
"loss": 0.9749,
"step": 1430
},
{
"epoch": 1.84,
"learning_rate": 9.705040871934605e-05,
"loss": 1.5007,
"step": 1440
},
{
"epoch": 1.85,
"learning_rate": 9.698228882833788e-05,
"loss": 1.2057,
"step": 1450
},
{
"epoch": 1.86,
"learning_rate": 9.69141689373297e-05,
"loss": 1.0082,
"step": 1460
},
{
"epoch": 1.87,
"learning_rate": 9.684604904632153e-05,
"loss": 0.891,
"step": 1470
},
{
"epoch": 1.89,
"learning_rate": 9.677792915531336e-05,
"loss": 0.8808,
"step": 1480
},
{
"epoch": 1.9,
"learning_rate": 9.670980926430518e-05,
"loss": 1.4966,
"step": 1490
},
{
"epoch": 1.91,
"learning_rate": 9.664168937329701e-05,
"loss": 1.1203,
"step": 1500
},
{
"epoch": 1.93,
"learning_rate": 9.657356948228883e-05,
"loss": 0.914,
"step": 1510
},
{
"epoch": 1.94,
"learning_rate": 9.650544959128067e-05,
"loss": 0.8256,
"step": 1520
},
{
"epoch": 1.95,
"learning_rate": 9.643732970027248e-05,
"loss": 0.8674,
"step": 1530
},
{
"epoch": 1.96,
"learning_rate": 9.636920980926431e-05,
"loss": 1.2761,
"step": 1540
},
{
"epoch": 1.98,
"learning_rate": 9.630108991825613e-05,
"loss": 0.981,
"step": 1550
},
{
"epoch": 1.99,
"learning_rate": 9.623297002724796e-05,
"loss": 0.8637,
"step": 1560
},
{
"epoch": 2.0,
"learning_rate": 9.616485013623978e-05,
"loss": 1.0717,
"step": 1570
},
{
"epoch": 2.02,
"learning_rate": 9.609673024523162e-05,
"loss": 1.0224,
"step": 1580
},
{
"epoch": 2.03,
"learning_rate": 9.602861035422343e-05,
"loss": 0.76,
"step": 1590
},
{
"epoch": 2.04,
"learning_rate": 9.596049046321526e-05,
"loss": 0.7198,
"step": 1600
},
{
"epoch": 2.04,
"eval_loss": 0.6678358316421509,
"eval_runtime": 1372.0406,
"eval_samples_per_second": 4.543,
"eval_steps_per_second": 0.568,
"eval_wer": 0.4826351764166794,
"step": 1600
},
{
"epoch": 2.05,
"learning_rate": 9.589237057220708e-05,
"loss": 0.6698,
"step": 1610
},
{
"epoch": 2.07,
"learning_rate": 9.582425068119891e-05,
"loss": 0.8794,
"step": 1620
},
{
"epoch": 2.08,
"learning_rate": 9.575613079019074e-05,
"loss": 0.9707,
"step": 1630
},
{
"epoch": 2.09,
"learning_rate": 9.568801089918257e-05,
"loss": 0.7405,
"step": 1640
},
{
"epoch": 2.1,
"learning_rate": 9.56198910081744e-05,
"loss": 0.6886,
"step": 1650
},
{
"epoch": 2.12,
"learning_rate": 9.555177111716622e-05,
"loss": 0.6535,
"step": 1660
},
{
"epoch": 2.13,
"learning_rate": 9.548365122615805e-05,
"loss": 0.8923,
"step": 1670
},
{
"epoch": 2.14,
"learning_rate": 9.541553133514986e-05,
"loss": 0.8998,
"step": 1680
},
{
"epoch": 2.16,
"learning_rate": 9.53474114441417e-05,
"loss": 0.7139,
"step": 1690
},
{
"epoch": 2.17,
"learning_rate": 9.527929155313352e-05,
"loss": 0.6413,
"step": 1700
},
{
"epoch": 2.18,
"learning_rate": 9.521117166212534e-05,
"loss": 0.6463,
"step": 1710
},
{
"epoch": 2.19,
"learning_rate": 9.514305177111717e-05,
"loss": 0.7948,
"step": 1720
},
{
"epoch": 2.21,
"learning_rate": 9.5074931880109e-05,
"loss": 0.8649,
"step": 1730
},
{
"epoch": 2.22,
"learning_rate": 9.500681198910081e-05,
"loss": 0.7054,
"step": 1740
},
{
"epoch": 2.23,
"learning_rate": 9.493869209809265e-05,
"loss": 0.6181,
"step": 1750
},
{
"epoch": 2.24,
"learning_rate": 9.487057220708447e-05,
"loss": 0.5805,
"step": 1760
},
{
"epoch": 2.26,
"learning_rate": 9.48024523160763e-05,
"loss": 0.7917,
"step": 1770
},
{
"epoch": 2.27,
"learning_rate": 9.473433242506812e-05,
"loss": 0.796,
"step": 1780
},
{
"epoch": 2.28,
"learning_rate": 9.466621253405995e-05,
"loss": 0.6605,
"step": 1790
},
{
"epoch": 2.3,
"learning_rate": 9.459809264305178e-05,
"loss": 0.5949,
"step": 1800
},
{
"epoch": 2.31,
"learning_rate": 9.45299727520436e-05,
"loss": 0.603,
"step": 1810
},
{
"epoch": 2.32,
"learning_rate": 9.446185286103543e-05,
"loss": 0.7413,
"step": 1820
},
{
"epoch": 2.33,
"learning_rate": 9.439373297002725e-05,
"loss": 0.725,
"step": 1830
},
{
"epoch": 2.35,
"learning_rate": 9.432561307901909e-05,
"loss": 0.6509,
"step": 1840
},
{
"epoch": 2.36,
"learning_rate": 9.42574931880109e-05,
"loss": 0.5928,
"step": 1850
},
{
"epoch": 2.37,
"learning_rate": 9.418937329700273e-05,
"loss": 0.5688,
"step": 1860
},
{
"epoch": 2.38,
"learning_rate": 9.412125340599455e-05,
"loss": 0.7283,
"step": 1870
},
{
"epoch": 2.4,
"learning_rate": 9.405313351498638e-05,
"loss": 0.7412,
"step": 1880
},
{
"epoch": 2.41,
"learning_rate": 9.39850136239782e-05,
"loss": 0.5876,
"step": 1890
},
{
"epoch": 2.42,
"learning_rate": 9.391689373297004e-05,
"loss": 0.5753,
"step": 1900
},
{
"epoch": 2.44,
"learning_rate": 9.384877384196185e-05,
"loss": 0.5513,
"step": 1910
},
{
"epoch": 2.45,
"learning_rate": 9.378065395095368e-05,
"loss": 0.7226,
"step": 1920
},
{
"epoch": 2.46,
"learning_rate": 9.37125340599455e-05,
"loss": 0.7162,
"step": 1930
},
{
"epoch": 2.47,
"learning_rate": 9.364441416893733e-05,
"loss": 0.5622,
"step": 1940
},
{
"epoch": 2.49,
"learning_rate": 9.357629427792916e-05,
"loss": 0.5364,
"step": 1950
},
{
"epoch": 2.5,
"learning_rate": 9.350817438692099e-05,
"loss": 0.5314,
"step": 1960
},
{
"epoch": 2.51,
"learning_rate": 9.344005449591281e-05,
"loss": 0.7133,
"step": 1970
},
{
"epoch": 2.53,
"learning_rate": 9.337193460490463e-05,
"loss": 0.6619,
"step": 1980
},
{
"epoch": 2.54,
"learning_rate": 9.330381471389647e-05,
"loss": 0.5651,
"step": 1990
},
{
"epoch": 2.55,
"learning_rate": 9.323569482288828e-05,
"loss": 0.5255,
"step": 2000
},
{
"epoch": 2.55,
"eval_loss": 0.4605100154876709,
"eval_runtime": 1363.3153,
"eval_samples_per_second": 4.572,
"eval_steps_per_second": 0.572,
"eval_wer": 0.4135290972964717,
"step": 2000
},
{
"epoch": 2.56,
"learning_rate": 9.316757493188011e-05,
"loss": 0.5396,
"step": 2010
},
{
"epoch": 2.58,
"learning_rate": 9.309945504087194e-05,
"loss": 0.6388,
"step": 2020
},
{
"epoch": 2.59,
"learning_rate": 9.303133514986376e-05,
"loss": 0.6487,
"step": 2030
},
{
"epoch": 2.6,
"learning_rate": 9.296321525885558e-05,
"loss": 0.5491,
"step": 2040
},
{
"epoch": 2.61,
"learning_rate": 9.289509536784742e-05,
"loss": 0.5008,
"step": 2050
},
{
"epoch": 2.63,
"learning_rate": 9.282697547683923e-05,
"loss": 0.5197,
"step": 2060
},
{
"epoch": 2.64,
"learning_rate": 9.275885558583107e-05,
"loss": 0.6516,
"step": 2070
},
{
"epoch": 2.65,
"learning_rate": 9.26907356948229e-05,
"loss": 0.6303,
"step": 2080
},
{
"epoch": 2.67,
"learning_rate": 9.262261580381471e-05,
"loss": 0.5336,
"step": 2090
},
{
"epoch": 2.68,
"learning_rate": 9.255449591280656e-05,
"loss": 0.4828,
"step": 2100
},
{
"epoch": 2.69,
"learning_rate": 9.248637602179837e-05,
"loss": 0.5037,
"step": 2110
},
{
"epoch": 2.7,
"learning_rate": 9.24182561307902e-05,
"loss": 0.6519,
"step": 2120
},
{
"epoch": 2.72,
"learning_rate": 9.235013623978202e-05,
"loss": 0.6046,
"step": 2130
},
{
"epoch": 2.73,
"learning_rate": 9.228201634877385e-05,
"loss": 0.5294,
"step": 2140
},
{
"epoch": 2.74,
"learning_rate": 9.221389645776567e-05,
"loss": 0.4708,
"step": 2150
},
{
"epoch": 2.75,
"learning_rate": 9.21457765667575e-05,
"loss": 0.5075,
"step": 2160
},
{
"epoch": 2.77,
"learning_rate": 9.207765667574932e-05,
"loss": 0.5829,
"step": 2170
},
{
"epoch": 2.78,
"learning_rate": 9.200953678474115e-05,
"loss": 0.5873,
"step": 2180
},
{
"epoch": 2.79,
"learning_rate": 9.194141689373297e-05,
"loss": 0.4926,
"step": 2190
},
{
"epoch": 2.81,
"learning_rate": 9.18732970027248e-05,
"loss": 0.4843,
"step": 2200
},
{
"epoch": 2.82,
"learning_rate": 9.180517711171663e-05,
"loss": 0.4875,
"step": 2210
},
{
"epoch": 2.83,
"learning_rate": 9.173705722070846e-05,
"loss": 0.6328,
"step": 2220
},
{
"epoch": 2.84,
"learning_rate": 9.166893732970028e-05,
"loss": 0.5735,
"step": 2230
},
{
"epoch": 2.86,
"learning_rate": 9.16008174386921e-05,
"loss": 0.4783,
"step": 2240
},
{
"epoch": 2.87,
"learning_rate": 9.153269754768394e-05,
"loss": 0.4412,
"step": 2250
},
{
"epoch": 2.88,
"learning_rate": 9.146457765667575e-05,
"loss": 0.5008,
"step": 2260
},
{
"epoch": 2.89,
"learning_rate": 9.139645776566758e-05,
"loss": 0.5338,
"step": 2270
},
{
"epoch": 2.91,
"learning_rate": 9.132833787465941e-05,
"loss": 0.5799,
"step": 2280
},
{
"epoch": 2.92,
"learning_rate": 9.126021798365123e-05,
"loss": 0.4821,
"step": 2290
},
{
"epoch": 2.93,
"learning_rate": 9.119209809264305e-05,
"loss": 0.4763,
"step": 2300
},
{
"epoch": 2.95,
"learning_rate": 9.112397820163489e-05,
"loss": 0.4834,
"step": 2310
},
{
"epoch": 2.96,
"learning_rate": 9.10558583106267e-05,
"loss": 0.5971,
"step": 2320
},
{
"epoch": 2.97,
"learning_rate": 9.098773841961853e-05,
"loss": 0.531,
"step": 2330
},
{
"epoch": 2.98,
"learning_rate": 9.091961852861036e-05,
"loss": 0.4768,
"step": 2340
},
{
"epoch": 3.0,
"learning_rate": 9.085149863760218e-05,
"loss": 0.4849,
"step": 2350
},
{
"epoch": 3.01,
"learning_rate": 9.078337874659401e-05,
"loss": 0.6024,
"step": 2360
},
{
"epoch": 3.02,
"learning_rate": 9.071525885558584e-05,
"loss": 0.4439,
"step": 2370
},
{
"epoch": 3.04,
"learning_rate": 9.064713896457767e-05,
"loss": 0.3636,
"step": 2380
},
{
"epoch": 3.05,
"learning_rate": 9.057901907356948e-05,
"loss": 0.3485,
"step": 2390
},
{
"epoch": 3.06,
"learning_rate": 9.051089918256132e-05,
"loss": 0.3961,
"step": 2400
},
{
"epoch": 3.06,
"eval_loss": 0.4266389012336731,
"eval_runtime": 1426.8916,
"eval_samples_per_second": 4.368,
"eval_steps_per_second": 0.547,
"eval_wer": 0.3954769360012219,
"step": 2400
},
{
"epoch": 3.07,
"learning_rate": 9.044277929155313e-05,
"loss": 0.5748,
"step": 2410
},
{
"epoch": 3.09,
"learning_rate": 9.037465940054496e-05,
"loss": 0.4057,
"step": 2420
},
{
"epoch": 3.1,
"learning_rate": 9.030653950953679e-05,
"loss": 0.3898,
"step": 2430
},
{
"epoch": 3.11,
"learning_rate": 9.023841961852862e-05,
"loss": 0.3239,
"step": 2440
},
{
"epoch": 3.12,
"learning_rate": 9.017029972752043e-05,
"loss": 0.3935,
"step": 2450
},
{
"epoch": 3.14,
"learning_rate": 9.010217983651227e-05,
"loss": 0.5595,
"step": 2460
},
{
"epoch": 3.15,
"learning_rate": 9.003405994550409e-05,
"loss": 0.4205,
"step": 2470
},
{
"epoch": 3.16,
"learning_rate": 8.996594005449591e-05,
"loss": 0.355,
"step": 2480
},
{
"epoch": 3.18,
"learning_rate": 8.989782016348774e-05,
"loss": 0.3479,
"step": 2490
},
{
"epoch": 3.19,
"learning_rate": 8.982970027247957e-05,
"loss": 0.4093,
"step": 2500
},
{
"epoch": 3.2,
"learning_rate": 8.97615803814714e-05,
"loss": 0.542,
"step": 2510
},
{
"epoch": 3.21,
"learning_rate": 8.969346049046322e-05,
"loss": 0.3946,
"step": 2520
},
{
"epoch": 3.23,
"learning_rate": 8.962534059945505e-05,
"loss": 0.3697,
"step": 2530
},
{
"epoch": 3.24,
"learning_rate": 8.955722070844688e-05,
"loss": 0.3516,
"step": 2540
},
{
"epoch": 3.25,
"learning_rate": 8.94891008174387e-05,
"loss": 0.3849,
"step": 2550
},
{
"epoch": 3.27,
"learning_rate": 8.942098092643052e-05,
"loss": 0.5427,
"step": 2560
},
{
"epoch": 3.28,
"learning_rate": 8.935286103542236e-05,
"loss": 0.4053,
"step": 2570
},
{
"epoch": 3.29,
"learning_rate": 8.928474114441417e-05,
"loss": 0.3635,
"step": 2580
},
{
"epoch": 3.3,
"learning_rate": 8.9216621253406e-05,
"loss": 0.3638,
"step": 2590
},
{
"epoch": 3.32,
"learning_rate": 8.914850136239783e-05,
"loss": 0.4078,
"step": 2600
},
{
"epoch": 3.33,
"learning_rate": 8.908038147138965e-05,
"loss": 0.5507,
"step": 2610
},
{
"epoch": 3.34,
"learning_rate": 8.901226158038147e-05,
"loss": 0.3924,
"step": 2620
},
{
"epoch": 3.35,
"learning_rate": 8.894414168937331e-05,
"loss": 0.3673,
"step": 2630
},
{
"epoch": 3.37,
"learning_rate": 8.887602179836512e-05,
"loss": 0.3392,
"step": 2640
},
{
"epoch": 3.38,
"learning_rate": 8.880790190735695e-05,
"loss": 0.3742,
"step": 2650
},
{
"epoch": 3.39,
"learning_rate": 8.873978201634878e-05,
"loss": 0.5251,
"step": 2660
},
{
"epoch": 3.41,
"learning_rate": 8.86716621253406e-05,
"loss": 0.3981,
"step": 2670
},
{
"epoch": 3.42,
"learning_rate": 8.860354223433243e-05,
"loss": 0.3673,
"step": 2680
},
{
"epoch": 3.43,
"learning_rate": 8.853542234332426e-05,
"loss": 0.3505,
"step": 2690
},
{
"epoch": 3.44,
"learning_rate": 8.846730245231609e-05,
"loss": 0.3614,
"step": 2700
},
{
"epoch": 3.46,
"learning_rate": 8.83991825613079e-05,
"loss": 0.5592,
"step": 2710
},
{
"epoch": 3.47,
"learning_rate": 8.833106267029974e-05,
"loss": 0.3909,
"step": 2720
},
{
"epoch": 3.48,
"learning_rate": 8.826294277929155e-05,
"loss": 0.3616,
"step": 2730
},
{
"epoch": 3.49,
"learning_rate": 8.819482288828338e-05,
"loss": 0.3378,
"step": 2740
},
{
"epoch": 3.51,
"learning_rate": 8.812670299727521e-05,
"loss": 0.4051,
"step": 2750
},
{
"epoch": 3.52,
"learning_rate": 8.805858310626704e-05,
"loss": 0.5289,
"step": 2760
},
{
"epoch": 3.53,
"learning_rate": 8.799046321525885e-05,
"loss": 0.3972,
"step": 2770
},
{
"epoch": 3.55,
"learning_rate": 8.792234332425069e-05,
"loss": 0.3723,
"step": 2780
},
{
"epoch": 3.56,
"learning_rate": 8.78542234332425e-05,
"loss": 0.3565,
"step": 2790
},
{
"epoch": 3.57,
"learning_rate": 8.778610354223433e-05,
"loss": 0.3424,
"step": 2800
},
{
"epoch": 3.57,
"eval_loss": 0.3786042034626007,
"eval_runtime": 1357.3296,
"eval_samples_per_second": 4.592,
"eval_steps_per_second": 0.575,
"eval_wer": 0.3741026424316481,
"step": 2800
},
{
"epoch": 3.58,
"learning_rate": 8.771798365122616e-05,
"loss": 0.5377,
"step": 2810
},
{
"epoch": 3.6,
"learning_rate": 8.764986376021799e-05,
"loss": 0.3955,
"step": 2820
},
{
"epoch": 3.61,
"learning_rate": 8.758174386920981e-05,
"loss": 0.3742,
"step": 2830
},
{
"epoch": 3.62,
"learning_rate": 8.751362397820164e-05,
"loss": 0.3342,
"step": 2840
},
{
"epoch": 3.63,
"learning_rate": 8.744550408719347e-05,
"loss": 0.4129,
"step": 2850
},
{
"epoch": 3.65,
"learning_rate": 8.737738419618528e-05,
"loss": 0.5368,
"step": 2860
},
{
"epoch": 3.66,
"learning_rate": 8.730926430517712e-05,
"loss": 0.3735,
"step": 2870
},
{
"epoch": 3.67,
"learning_rate": 8.724114441416894e-05,
"loss": 0.3438,
"step": 2880
},
{
"epoch": 3.69,
"learning_rate": 8.717302452316076e-05,
"loss": 0.3248,
"step": 2890
},
{
"epoch": 3.7,
"learning_rate": 8.710490463215259e-05,
"loss": 0.3426,
"step": 2900
},
{
"epoch": 3.71,
"learning_rate": 8.703678474114442e-05,
"loss": 0.4977,
"step": 2910
},
{
"epoch": 3.72,
"learning_rate": 8.696866485013623e-05,
"loss": 0.3857,
"step": 2920
},
{
"epoch": 3.74,
"learning_rate": 8.690054495912807e-05,
"loss": 0.3688,
"step": 2930
},
{
"epoch": 3.75,
"learning_rate": 8.683242506811989e-05,
"loss": 0.3203,
"step": 2940
},
{
"epoch": 3.76,
"learning_rate": 8.676430517711172e-05,
"loss": 0.3877,
"step": 2950
},
{
"epoch": 3.77,
"learning_rate": 8.669618528610354e-05,
"loss": 0.4694,
"step": 2960
},
{
"epoch": 3.79,
"learning_rate": 8.662806539509537e-05,
"loss": 0.3847,
"step": 2970
},
{
"epoch": 3.8,
"learning_rate": 8.65599455040872e-05,
"loss": 0.3363,
"step": 2980
},
{
"epoch": 3.81,
"learning_rate": 8.649182561307902e-05,
"loss": 0.3188,
"step": 2990
},
{
"epoch": 3.83,
"learning_rate": 8.642370572207085e-05,
"loss": 0.3828,
"step": 3000
},
{
"epoch": 3.84,
"learning_rate": 8.635558583106268e-05,
"loss": 0.4805,
"step": 3010
},
{
"epoch": 3.85,
"learning_rate": 8.62874659400545e-05,
"loss": 0.3729,
"step": 3020
},
{
"epoch": 3.86,
"learning_rate": 8.621934604904632e-05,
"loss": 0.3452,
"step": 3030
},
{
"epoch": 3.88,
"learning_rate": 8.615122615803816e-05,
"loss": 0.3196,
"step": 3040
},
{
"epoch": 3.89,
"learning_rate": 8.608310626702997e-05,
"loss": 0.3589,
"step": 3050
},
{
"epoch": 3.9,
"learning_rate": 8.60149863760218e-05,
"loss": 0.4738,
"step": 3060
},
{
"epoch": 3.92,
"learning_rate": 8.594686648501363e-05,
"loss": 0.3593,
"step": 3070
},
{
"epoch": 3.93,
"learning_rate": 8.587874659400546e-05,
"loss": 0.3296,
"step": 3080
},
{
"epoch": 3.94,
"learning_rate": 8.581062670299727e-05,
"loss": 0.326,
"step": 3090
},
{
"epoch": 3.95,
"learning_rate": 8.574250681198911e-05,
"loss": 0.3493,
"step": 3100
},
{
"epoch": 3.97,
"learning_rate": 8.567438692098093e-05,
"loss": 0.422,
"step": 3110
},
{
"epoch": 3.98,
"learning_rate": 8.560626702997275e-05,
"loss": 0.362,
"step": 3120
},
{
"epoch": 3.99,
"learning_rate": 8.553814713896458e-05,
"loss": 0.329,
"step": 3130
},
{
"epoch": 4.01,
"learning_rate": 8.547002724795641e-05,
"loss": 0.4632,
"step": 3140
},
{
"epoch": 4.02,
"learning_rate": 8.540190735694823e-05,
"loss": 0.3265,
"step": 3150
},
{
"epoch": 4.03,
"learning_rate": 8.533378746594006e-05,
"loss": 0.2534,
"step": 3160
},
{
"epoch": 4.04,
"learning_rate": 8.526566757493189e-05,
"loss": 0.2548,
"step": 3170
},
{
"epoch": 4.06,
"learning_rate": 8.51975476839237e-05,
"loss": 0.2539,
"step": 3180
},
{
"epoch": 4.07,
"learning_rate": 8.512942779291554e-05,
"loss": 0.4004,
"step": 3190
},
{
"epoch": 4.08,
"learning_rate": 8.506130790190736e-05,
"loss": 0.3858,
"step": 3200
},
{
"epoch": 4.08,
"eval_loss": 0.31610095500946045,
"eval_runtime": 1426.1412,
"eval_samples_per_second": 4.371,
"eval_steps_per_second": 0.547,
"eval_wer": 0.35519130899648693,
"step": 3200
},
{
"epoch": 4.09,
"learning_rate": 8.499318801089918e-05,
"loss": 0.2862,
"step": 3210
},
{
"epoch": 4.11,
"learning_rate": 8.492506811989101e-05,
"loss": 0.2404,
"step": 3220
},
{
"epoch": 4.12,
"learning_rate": 8.485694822888284e-05,
"loss": 0.2461,
"step": 3230
},
{
"epoch": 4.13,
"learning_rate": 8.478882833787465e-05,
"loss": 0.3903,
"step": 3240
},
{
"epoch": 4.15,
"learning_rate": 8.47207084468665e-05,
"loss": 0.3463,
"step": 3250
},
{
"epoch": 4.16,
"learning_rate": 8.465258855585831e-05,
"loss": 0.2812,
"step": 3260
},
{
"epoch": 4.17,
"learning_rate": 8.458446866485014e-05,
"loss": 0.2613,
"step": 3270
},
{
"epoch": 4.18,
"learning_rate": 8.451634877384196e-05,
"loss": 0.2712,
"step": 3280
},
{
"epoch": 4.2,
"learning_rate": 8.444822888283379e-05,
"loss": 0.4027,
"step": 3290
},
{
"epoch": 4.21,
"learning_rate": 8.438010899182562e-05,
"loss": 0.3261,
"step": 3300
},
{
"epoch": 4.22,
"learning_rate": 8.431198910081744e-05,
"loss": 0.2902,
"step": 3310
},
{
"epoch": 4.23,
"learning_rate": 8.424386920980927e-05,
"loss": 0.2529,
"step": 3320
},
{
"epoch": 4.25,
"learning_rate": 8.417574931880109e-05,
"loss": 0.2809,
"step": 3330
},
{
"epoch": 4.26,
"learning_rate": 8.410762942779293e-05,
"loss": 0.3984,
"step": 3340
},
{
"epoch": 4.27,
"learning_rate": 8.403950953678474e-05,
"loss": 0.3262,
"step": 3350
},
{
"epoch": 4.29,
"learning_rate": 8.397138964577657e-05,
"loss": 0.2792,
"step": 3360
},
{
"epoch": 4.3,
"learning_rate": 8.39032697547684e-05,
"loss": 0.2571,
"step": 3370
},
{
"epoch": 4.31,
"learning_rate": 8.383514986376022e-05,
"loss": 0.2702,
"step": 3380
},
{
"epoch": 4.32,
"learning_rate": 8.376702997275204e-05,
"loss": 0.4027,
"step": 3390
},
{
"epoch": 4.34,
"learning_rate": 8.369891008174388e-05,
"loss": 0.3302,
"step": 3400
},
{
"epoch": 4.35,
"learning_rate": 8.363079019073569e-05,
"loss": 0.2765,
"step": 3410
},
{
"epoch": 4.36,
"learning_rate": 8.356267029972752e-05,
"loss": 0.2596,
"step": 3420
},
{
"epoch": 4.37,
"learning_rate": 8.349455040871936e-05,
"loss": 0.2913,
"step": 3430
},
{
"epoch": 4.39,
"learning_rate": 8.342643051771117e-05,
"loss": 0.3909,
"step": 3440
},
{
"epoch": 4.4,
"learning_rate": 8.335831062670301e-05,
"loss": 0.3138,
"step": 3450
},
{
"epoch": 4.41,
"learning_rate": 8.329019073569483e-05,
"loss": 0.2773,
"step": 3460
},
{
"epoch": 4.43,
"learning_rate": 8.322207084468665e-05,
"loss": 0.2594,
"step": 3470
},
{
"epoch": 4.44,
"learning_rate": 8.315395095367848e-05,
"loss": 0.2463,
"step": 3480
},
{
"epoch": 4.45,
"learning_rate": 8.308583106267031e-05,
"loss": 0.3788,
"step": 3490
},
{
"epoch": 4.46,
"learning_rate": 8.301771117166212e-05,
"loss": 0.3399,
"step": 3500
},
{
"epoch": 4.48,
"learning_rate": 8.294959128065396e-05,
"loss": 0.2706,
"step": 3510
},
{
"epoch": 4.49,
"learning_rate": 8.288147138964578e-05,
"loss": 0.2421,
"step": 3520
},
{
"epoch": 4.5,
"learning_rate": 8.28133514986376e-05,
"loss": 0.2637,
"step": 3530
},
{
"epoch": 4.51,
"learning_rate": 8.274523160762943e-05,
"loss": 0.381,
"step": 3540
},
{
"epoch": 4.53,
"learning_rate": 8.267711171662126e-05,
"loss": 0.3187,
"step": 3550
},
{
"epoch": 4.54,
"learning_rate": 8.260899182561309e-05,
"loss": 0.2779,
"step": 3560
},
{
"epoch": 4.55,
"learning_rate": 8.254087193460491e-05,
"loss": 0.2455,
"step": 3570
},
{
"epoch": 4.57,
"learning_rate": 8.247275204359674e-05,
"loss": 0.2843,
"step": 3580
},
{
"epoch": 4.58,
"learning_rate": 8.240463215258856e-05,
"loss": 0.4318,
"step": 3590
},
{
"epoch": 4.59,
"learning_rate": 8.23365122615804e-05,
"loss": 0.3218,
"step": 3600
},
{
"epoch": 4.59,
"eval_loss": 0.3028896749019623,
"eval_runtime": 1356.3055,
"eval_samples_per_second": 4.596,
"eval_steps_per_second": 0.575,
"eval_wer": 0.35104819001069193,
"step": 3600
},
{
"epoch": 4.6,
"learning_rate": 8.226839237057221e-05,
"loss": 0.2763,
"step": 3610
},
{
"epoch": 4.62,
"learning_rate": 8.220027247956404e-05,
"loss": 0.2707,
"step": 3620
},
{
"epoch": 4.63,
"learning_rate": 8.213215258855586e-05,
"loss": 0.2201,
"step": 3630
},
{
"epoch": 4.64,
"learning_rate": 8.206403269754769e-05,
"loss": 0.3985,
"step": 3640
},
{
"epoch": 4.66,
"learning_rate": 8.19959128065395e-05,
"loss": 0.3283,
"step": 3650
},
{
"epoch": 4.67,
"learning_rate": 8.192779291553135e-05,
"loss": 0.2688,
"step": 3660
},
{
"epoch": 4.68,
"learning_rate": 8.185967302452316e-05,
"loss": 0.2559,
"step": 3670
},
{
"epoch": 4.69,
"learning_rate": 8.179155313351499e-05,
"loss": 0.2632,
"step": 3680
},
{
"epoch": 4.71,
"learning_rate": 8.172343324250681e-05,
"loss": 0.381,
"step": 3690
},
{
"epoch": 4.72,
"learning_rate": 8.165531335149864e-05,
"loss": 0.3152,
"step": 3700
},
{
"epoch": 4.73,
"learning_rate": 8.158719346049047e-05,
"loss": 0.2753,
"step": 3710
},
{
"epoch": 4.74,
"learning_rate": 8.15190735694823e-05,
"loss": 0.2612,
"step": 3720
},
{
"epoch": 4.76,
"learning_rate": 8.145095367847412e-05,
"loss": 0.2563,
"step": 3730
},
{
"epoch": 4.77,
"learning_rate": 8.138283378746594e-05,
"loss": 0.4019,
"step": 3740
},
{
"epoch": 4.78,
"learning_rate": 8.131471389645778e-05,
"loss": 0.3191,
"step": 3750
},
{
"epoch": 4.8,
"learning_rate": 8.124659400544959e-05,
"loss": 0.2542,
"step": 3760
},
{
"epoch": 4.81,
"learning_rate": 8.117847411444142e-05,
"loss": 0.2379,
"step": 3770
},
{
"epoch": 4.82,
"learning_rate": 8.111035422343325e-05,
"loss": 0.2729,
"step": 3780
},
{
"epoch": 4.83,
"learning_rate": 8.104223433242507e-05,
"loss": 0.4223,
"step": 3790
},
{
"epoch": 4.85,
"learning_rate": 8.097411444141689e-05,
"loss": 0.3398,
"step": 3800
},
{
"epoch": 4.86,
"learning_rate": 8.090599455040873e-05,
"loss": 0.275,
"step": 3810
},
{
"epoch": 4.87,
"learning_rate": 8.083787465940054e-05,
"loss": 0.2567,
"step": 3820
},
{
"epoch": 4.88,
"learning_rate": 8.076975476839237e-05,
"loss": 0.2844,
"step": 3830
},
{
"epoch": 4.9,
"learning_rate": 8.07016348773842e-05,
"loss": 0.3447,
"step": 3840
},
{
"epoch": 4.91,
"learning_rate": 8.063351498637602e-05,
"loss": 0.3174,
"step": 3850
},
{
"epoch": 4.92,
"learning_rate": 8.056539509536785e-05,
"loss": 0.2743,
"step": 3860
},
{
"epoch": 4.94,
"learning_rate": 8.049727520435968e-05,
"loss": 0.2494,
"step": 3870
},
{
"epoch": 4.95,
"learning_rate": 8.04291553133515e-05,
"loss": 0.2413,
"step": 3880
},
{
"epoch": 4.96,
"learning_rate": 8.036103542234332e-05,
"loss": 0.3719,
"step": 3890
},
{
"epoch": 4.97,
"learning_rate": 8.029291553133516e-05,
"loss": 0.3094,
"step": 3900
},
{
"epoch": 4.99,
"learning_rate": 8.022479564032698e-05,
"loss": 0.2559,
"step": 3910
},
{
"epoch": 5.0,
"learning_rate": 8.015667574931882e-05,
"loss": 0.292,
"step": 3920
},
{
"epoch": 5.01,
"learning_rate": 8.008855585831063e-05,
"loss": 0.3344,
"step": 3930
},
{
"epoch": 5.03,
"learning_rate": 8.002043596730246e-05,
"loss": 0.2311,
"step": 3940
},
{
"epoch": 5.04,
"learning_rate": 7.995231607629428e-05,
"loss": 0.1952,
"step": 3950
},
{
"epoch": 5.05,
"learning_rate": 7.988419618528611e-05,
"loss": 0.1826,
"step": 3960
},
{
"epoch": 5.06,
"learning_rate": 7.981607629427793e-05,
"loss": 0.2663,
"step": 3970
},
{
"epoch": 5.08,
"learning_rate": 7.974795640326977e-05,
"loss": 0.3244,
"step": 3980
},
{
"epoch": 5.09,
"learning_rate": 7.967983651226158e-05,
"loss": 0.2331,
"step": 3990
},
{
"epoch": 5.1,
"learning_rate": 7.961171662125341e-05,
"loss": 0.199,
"step": 4000
},
{
"epoch": 5.1,
"eval_loss": 0.29879283905029297,
"eval_runtime": 1354.5596,
"eval_samples_per_second": 4.601,
"eval_steps_per_second": 0.576,
"eval_wer": 0.34178822361386896,
"step": 4000
},
{
"epoch": 5.11,
"learning_rate": 7.954359673024523e-05,
"loss": 0.1886,
"step": 4010
},
{
"epoch": 5.13,
"learning_rate": 7.947547683923706e-05,
"loss": 0.2599,
"step": 4020
},
{
"epoch": 5.14,
"learning_rate": 7.940735694822889e-05,
"loss": 0.3075,
"step": 4030
},
{
"epoch": 5.15,
"learning_rate": 7.933923705722072e-05,
"loss": 0.2248,
"step": 4040
},
{
"epoch": 5.17,
"learning_rate": 7.927111716621254e-05,
"loss": 0.2008,
"step": 4050
},
{
"epoch": 5.18,
"learning_rate": 7.920299727520436e-05,
"loss": 0.1997,
"step": 4060
},
{
"epoch": 5.19,
"learning_rate": 7.91348773841962e-05,
"loss": 0.2697,
"step": 4070
},
{
"epoch": 5.2,
"learning_rate": 7.906675749318801e-05,
"loss": 0.3081,
"step": 4080
},
{
"epoch": 5.22,
"learning_rate": 7.899863760217984e-05,
"loss": 0.2232,
"step": 4090
},
{
"epoch": 5.23,
"learning_rate": 7.893051771117167e-05,
"loss": 0.2146,
"step": 4100
},
{
"epoch": 5.24,
"learning_rate": 7.88623978201635e-05,
"loss": 0.1739,
"step": 4110
},
{
"epoch": 5.25,
"learning_rate": 7.879427792915531e-05,
"loss": 0.2945,
"step": 4120
},
{
"epoch": 5.27,
"learning_rate": 7.872615803814715e-05,
"loss": 0.3158,
"step": 4130
},
{
"epoch": 5.28,
"learning_rate": 7.865803814713896e-05,
"loss": 0.243,
"step": 4140
},
{
"epoch": 5.29,
"learning_rate": 7.858991825613079e-05,
"loss": 0.2088,
"step": 4150
},
{
"epoch": 5.31,
"learning_rate": 7.852179836512262e-05,
"loss": 0.1899,
"step": 4160
},
{
"epoch": 5.32,
"learning_rate": 7.845367847411444e-05,
"loss": 0.2968,
"step": 4170
},
{
"epoch": 5.33,
"learning_rate": 7.838555858310627e-05,
"loss": 0.3202,
"step": 4180
},
{
"epoch": 5.34,
"learning_rate": 7.83174386920981e-05,
"loss": 0.2636,
"step": 4190
},
{
"epoch": 5.36,
"learning_rate": 7.824931880108993e-05,
"loss": 0.2132,
"step": 4200
},
{
"epoch": 5.37,
"learning_rate": 7.818119891008174e-05,
"loss": 0.1976,
"step": 4210
},
{
"epoch": 5.38,
"learning_rate": 7.811307901907358e-05,
"loss": 0.2709,
"step": 4220
},
{
"epoch": 5.4,
"learning_rate": 7.80449591280654e-05,
"loss": 0.3094,
"step": 4230
},
{
"epoch": 5.41,
"learning_rate": 7.797683923705722e-05,
"loss": 0.2282,
"step": 4240
},
{
"epoch": 5.42,
"learning_rate": 7.790871934604905e-05,
"loss": 0.2148,
"step": 4250
},
{
"epoch": 5.43,
"learning_rate": 7.784059945504088e-05,
"loss": 0.2079,
"step": 4260
},
{
"epoch": 5.45,
"learning_rate": 7.777247956403269e-05,
"loss": 0.2455,
"step": 4270
},
{
"epoch": 5.46,
"learning_rate": 7.770435967302453e-05,
"loss": 0.3046,
"step": 4280
},
{
"epoch": 5.47,
"learning_rate": 7.763623978201635e-05,
"loss": 0.2257,
"step": 4290
},
{
"epoch": 5.48,
"learning_rate": 7.756811989100817e-05,
"loss": 0.2027,
"step": 4300
},
{
"epoch": 5.5,
"learning_rate": 7.75e-05,
"loss": 0.2004,
"step": 4310
},
{
"epoch": 5.51,
"learning_rate": 7.743188010899183e-05,
"loss": 0.2498,
"step": 4320
},
{
"epoch": 5.52,
"learning_rate": 7.736376021798365e-05,
"loss": 0.3068,
"step": 4330
},
{
"epoch": 5.54,
"learning_rate": 7.729564032697548e-05,
"loss": 0.2259,
"step": 4340
},
{
"epoch": 5.55,
"learning_rate": 7.722752043596731e-05,
"loss": 0.1961,
"step": 4350
},
{
"epoch": 5.56,
"learning_rate": 7.715940054495914e-05,
"loss": 0.2078,
"step": 4360
},
{
"epoch": 5.57,
"learning_rate": 7.709128065395096e-05,
"loss": 0.256,
"step": 4370
},
{
"epoch": 5.59,
"learning_rate": 7.702316076294278e-05,
"loss": 0.3219,
"step": 4380
},
{
"epoch": 5.6,
"learning_rate": 7.695504087193462e-05,
"loss": 0.234,
"step": 4390
},
{
"epoch": 5.61,
"learning_rate": 7.688692098092643e-05,
"loss": 0.2054,
"step": 4400
},
{
"epoch": 5.61,
"eval_loss": 0.2873386740684509,
"eval_runtime": 1362.5508,
"eval_samples_per_second": 4.575,
"eval_steps_per_second": 0.572,
"eval_wer": 0.34335382617992977,
"step": 4400
},
{
"epoch": 5.62,
"learning_rate": 7.681880108991826e-05,
"loss": 0.1902,
"step": 4410
},
{
"epoch": 5.64,
"learning_rate": 7.675068119891009e-05,
"loss": 0.2802,
"step": 4420
},
{
"epoch": 5.65,
"learning_rate": 7.668256130790191e-05,
"loss": 0.307,
"step": 4430
},
{
"epoch": 5.66,
"learning_rate": 7.661444141689373e-05,
"loss": 0.2302,
"step": 4440
},
{
"epoch": 5.68,
"learning_rate": 7.654632152588557e-05,
"loss": 0.2121,
"step": 4450
},
{
"epoch": 5.69,
"learning_rate": 7.647820163487738e-05,
"loss": 0.1782,
"step": 4460
},
{
"epoch": 5.7,
"learning_rate": 7.641008174386921e-05,
"loss": 0.2857,
"step": 4470
},
{
"epoch": 5.71,
"learning_rate": 7.634196185286104e-05,
"loss": 0.305,
"step": 4480
},
{
"epoch": 5.73,
"learning_rate": 7.627384196185286e-05,
"loss": 0.2318,
"step": 4490
},
{
"epoch": 5.74,
"learning_rate": 7.620572207084469e-05,
"loss": 0.2035,
"step": 4500
},
{
"epoch": 5.75,
"learning_rate": 7.613760217983652e-05,
"loss": 0.1914,
"step": 4510
},
{
"epoch": 5.76,
"learning_rate": 7.606948228882835e-05,
"loss": 0.2506,
"step": 4520
},
{
"epoch": 5.78,
"learning_rate": 7.600136239782016e-05,
"loss": 0.2988,
"step": 4530
},
{
"epoch": 5.79,
"learning_rate": 7.5933242506812e-05,
"loss": 0.2363,
"step": 4540
},
{
"epoch": 5.8,
"learning_rate": 7.586512261580381e-05,
"loss": 0.2106,
"step": 4550
},
{
"epoch": 5.82,
"learning_rate": 7.579700272479564e-05,
"loss": 0.1792,
"step": 4560
},
{
"epoch": 5.83,
"learning_rate": 7.572888283378747e-05,
"loss": 0.2635,
"step": 4570
},
{
"epoch": 5.84,
"learning_rate": 7.56607629427793e-05,
"loss": 0.3131,
"step": 4580
},
{
"epoch": 5.85,
"learning_rate": 7.559264305177111e-05,
"loss": 0.2375,
"step": 4590
},
{
"epoch": 5.87,
"learning_rate": 7.552452316076295e-05,
"loss": 0.1996,
"step": 4600
},
{
"epoch": 5.88,
"learning_rate": 7.545640326975477e-05,
"loss": 0.1848,
"step": 4610
},
{
"epoch": 5.89,
"learning_rate": 7.538828337874659e-05,
"loss": 0.2375,
"step": 4620
},
{
"epoch": 5.9,
"learning_rate": 7.532016348773842e-05,
"loss": 0.3027,
"step": 4630
},
{
"epoch": 5.92,
"learning_rate": 7.525204359673025e-05,
"loss": 0.2376,
"step": 4640
},
{
"epoch": 5.93,
"learning_rate": 7.518392370572207e-05,
"loss": 0.2072,
"step": 4650
},
{
"epoch": 5.94,
"learning_rate": 7.51158038147139e-05,
"loss": 0.2128,
"step": 4660
},
{
"epoch": 5.96,
"learning_rate": 7.504768392370573e-05,
"loss": 0.3047,
"step": 4670
},
{
"epoch": 5.97,
"learning_rate": 7.497956403269754e-05,
"loss": 0.2724,
"step": 4680
},
{
"epoch": 5.98,
"learning_rate": 7.491144414168938e-05,
"loss": 0.2071,
"step": 4690
},
{
"epoch": 5.99,
"learning_rate": 7.48433242506812e-05,
"loss": 0.2074,
"step": 4700
},
{
"epoch": 6.01,
"learning_rate": 7.477520435967302e-05,
"loss": 0.3043,
"step": 4710
},
{
"epoch": 6.02,
"learning_rate": 7.470708446866485e-05,
"loss": 0.2099,
"step": 4720
},
{
"epoch": 6.03,
"learning_rate": 7.463896457765668e-05,
"loss": 0.1647,
"step": 4730
},
{
"epoch": 6.05,
"learning_rate": 7.45708446866485e-05,
"loss": 0.1382,
"step": 4740
},
{
"epoch": 6.06,
"learning_rate": 7.450272479564033e-05,
"loss": 0.1726,
"step": 4750
},
{
"epoch": 6.07,
"learning_rate": 7.443460490463216e-05,
"loss": 0.2836,
"step": 4760
},
{
"epoch": 6.08,
"learning_rate": 7.436648501362398e-05,
"loss": 0.2129,
"step": 4770
},
{
"epoch": 6.1,
"learning_rate": 7.429836512261582e-05,
"loss": 0.1673,
"step": 4780
},
{
"epoch": 6.11,
"learning_rate": 7.423024523160763e-05,
"loss": 0.1534,
"step": 4790
},
{
"epoch": 6.12,
"learning_rate": 7.416212534059946e-05,
"loss": 0.1704,
"step": 4800
},
{
"epoch": 6.12,
"eval_loss": 0.31291627883911133,
"eval_runtime": 1357.9405,
"eval_samples_per_second": 4.59,
"eval_steps_per_second": 0.574,
"eval_wer": 0.34319153810905756,
"step": 4800
},
{
"epoch": 6.14,
"learning_rate": 7.409400544959128e-05,
"loss": 0.2631,
"step": 4810
},
{
"epoch": 6.15,
"learning_rate": 7.402588555858311e-05,
"loss": 0.2144,
"step": 4820
},
{
"epoch": 6.16,
"learning_rate": 7.395776566757494e-05,
"loss": 0.1788,
"step": 4830
},
{
"epoch": 6.17,
"learning_rate": 7.388964577656677e-05,
"loss": 0.148,
"step": 4840
},
{
"epoch": 6.19,
"learning_rate": 7.382152588555858e-05,
"loss": 0.2017,
"step": 4850
},
{
"epoch": 6.2,
"learning_rate": 7.375340599455042e-05,
"loss": 0.302,
"step": 4860
},
{
"epoch": 6.21,
"learning_rate": 7.368528610354223e-05,
"loss": 0.2176,
"step": 4870
},
{
"epoch": 6.22,
"learning_rate": 7.361716621253406e-05,
"loss": 0.1862,
"step": 4880
},
{
"epoch": 6.24,
"learning_rate": 7.354904632152589e-05,
"loss": 0.156,
"step": 4890
},
{
"epoch": 6.25,
"learning_rate": 7.348092643051772e-05,
"loss": 0.1761,
"step": 4900
},
{
"epoch": 6.26,
"learning_rate": 7.341280653950954e-05,
"loss": 0.2865,
"step": 4910
},
{
"epoch": 6.28,
"learning_rate": 7.334468664850137e-05,
"loss": 0.215,
"step": 4920
},
{
"epoch": 6.29,
"learning_rate": 7.32765667574932e-05,
"loss": 0.1806,
"step": 4930
},
{
"epoch": 6.3,
"learning_rate": 7.320844686648501e-05,
"loss": 0.1613,
"step": 4940
},
{
"epoch": 6.31,
"learning_rate": 7.314032697547685e-05,
"loss": 0.1898,
"step": 4950
},
{
"epoch": 6.33,
"learning_rate": 7.307220708446867e-05,
"loss": 0.2798,
"step": 4960
},
{
"epoch": 6.34,
"learning_rate": 7.30040871934605e-05,
"loss": 0.2264,
"step": 4970
},
{
"epoch": 6.35,
"learning_rate": 7.293596730245232e-05,
"loss": 0.1772,
"step": 4980
},
{
"epoch": 6.36,
"learning_rate": 7.286784741144415e-05,
"loss": 0.1632,
"step": 4990
},
{
"epoch": 6.38,
"learning_rate": 7.279972752043596e-05,
"loss": 0.1935,
"step": 5000
},
{
"epoch": 6.39,
"learning_rate": 7.27316076294278e-05,
"loss": 0.2846,
"step": 5010
},
{
"epoch": 6.4,
"learning_rate": 7.266348773841962e-05,
"loss": 0.2205,
"step": 5020
},
{
"epoch": 6.42,
"learning_rate": 7.259536784741144e-05,
"loss": 0.1869,
"step": 5030
},
{
"epoch": 6.43,
"learning_rate": 7.252724795640327e-05,
"loss": 0.1546,
"step": 5040
},
{
"epoch": 6.44,
"learning_rate": 7.24591280653951e-05,
"loss": 0.1831,
"step": 5050
},
{
"epoch": 6.45,
"learning_rate": 7.239100817438693e-05,
"loss": 0.2718,
"step": 5060
},
{
"epoch": 6.47,
"learning_rate": 7.232288828337875e-05,
"loss": 0.2188,
"step": 5070
},
{
"epoch": 6.48,
"learning_rate": 7.225476839237058e-05,
"loss": 0.1769,
"step": 5080
},
{
"epoch": 6.49,
"learning_rate": 7.21866485013624e-05,
"loss": 0.1684,
"step": 5090
},
{
"epoch": 6.5,
"learning_rate": 7.211852861035424e-05,
"loss": 0.1683,
"step": 5100
},
{
"epoch": 6.52,
"learning_rate": 7.205040871934605e-05,
"loss": 0.3011,
"step": 5110
},
{
"epoch": 6.53,
"learning_rate": 7.198228882833788e-05,
"loss": 0.2114,
"step": 5120
},
{
"epoch": 6.54,
"learning_rate": 7.19141689373297e-05,
"loss": 0.1741,
"step": 5130
},
{
"epoch": 6.56,
"learning_rate": 7.184604904632153e-05,
"loss": 0.1467,
"step": 5140
},
{
"epoch": 6.57,
"learning_rate": 7.177792915531335e-05,
"loss": 0.1705,
"step": 5150
},
{
"epoch": 6.58,
"learning_rate": 7.170980926430519e-05,
"loss": 0.296,
"step": 5160
},
{
"epoch": 6.59,
"learning_rate": 7.1641689373297e-05,
"loss": 0.2137,
"step": 5170
},
{
"epoch": 6.61,
"learning_rate": 7.157356948228883e-05,
"loss": 0.1733,
"step": 5180
},
{
"epoch": 6.62,
"learning_rate": 7.150544959128065e-05,
"loss": 0.1585,
"step": 5190
},
{
"epoch": 6.63,
"learning_rate": 7.143732970027248e-05,
"loss": 0.1805,
"step": 5200
},
{
"epoch": 6.63,
"eval_loss": 0.296330988407135,
"eval_runtime": 1357.6829,
"eval_samples_per_second": 4.591,
"eval_steps_per_second": 0.575,
"eval_wer": 0.3412727203299221,
"step": 5200
},
{
"epoch": 6.64,
"learning_rate": 7.136920980926431e-05,
"loss": 0.2849,
"step": 5210
},
{
"epoch": 6.66,
"learning_rate": 7.130108991825614e-05,
"loss": 0.2099,
"step": 5220
},
{
"epoch": 6.67,
"learning_rate": 7.123297002724796e-05,
"loss": 0.1777,
"step": 5230
},
{
"epoch": 6.68,
"learning_rate": 7.116485013623978e-05,
"loss": 0.1496,
"step": 5240
},
{
"epoch": 6.7,
"learning_rate": 7.109673024523162e-05,
"loss": 0.1906,
"step": 5250
},
{
"epoch": 6.71,
"learning_rate": 7.102861035422343e-05,
"loss": 0.2884,
"step": 5260
},
{
"epoch": 6.72,
"learning_rate": 7.096049046321526e-05,
"loss": 0.2049,
"step": 5270
},
{
"epoch": 6.73,
"learning_rate": 7.089237057220709e-05,
"loss": 0.1815,
"step": 5280
},
{
"epoch": 6.75,
"learning_rate": 7.082425068119891e-05,
"loss": 0.156,
"step": 5290
},
{
"epoch": 6.76,
"learning_rate": 7.075613079019074e-05,
"loss": 0.1857,
"step": 5300
},
{
"epoch": 6.77,
"learning_rate": 7.068801089918257e-05,
"loss": 0.277,
"step": 5310
},
{
"epoch": 6.79,
"learning_rate": 7.061989100817438e-05,
"loss": 0.2278,
"step": 5320
},
{
"epoch": 6.8,
"learning_rate": 7.055177111716622e-05,
"loss": 0.1667,
"step": 5330
},
{
"epoch": 6.81,
"learning_rate": 7.048365122615804e-05,
"loss": 0.1474,
"step": 5340
},
{
"epoch": 6.82,
"learning_rate": 7.041553133514986e-05,
"loss": 0.164,
"step": 5350
},
{
"epoch": 6.84,
"learning_rate": 7.034741144414169e-05,
"loss": 0.2885,
"step": 5360
},
{
"epoch": 6.85,
"learning_rate": 7.027929155313352e-05,
"loss": 0.2117,
"step": 5370
},
{
"epoch": 6.86,
"learning_rate": 7.021117166212535e-05,
"loss": 0.1828,
"step": 5380
},
{
"epoch": 6.87,
"learning_rate": 7.014305177111717e-05,
"loss": 0.1589,
"step": 5390
},
{
"epoch": 6.89,
"learning_rate": 7.0074931880109e-05,
"loss": 0.1649,
"step": 5400
},
{
"epoch": 6.9,
"learning_rate": 7.000681198910082e-05,
"loss": 0.282,
"step": 5410
},
{
"epoch": 6.91,
"learning_rate": 6.993869209809266e-05,
"loss": 0.211,
"step": 5420
},
{
"epoch": 6.93,
"learning_rate": 6.987057220708447e-05,
"loss": 0.1849,
"step": 5430
},
{
"epoch": 6.94,
"learning_rate": 6.98024523160763e-05,
"loss": 0.1615,
"step": 5440
},
{
"epoch": 6.95,
"learning_rate": 6.973433242506812e-05,
"loss": 0.1879,
"step": 5450
},
{
"epoch": 6.96,
"learning_rate": 6.966621253405995e-05,
"loss": 0.2815,
"step": 5460
},
{
"epoch": 6.98,
"learning_rate": 6.959809264305177e-05,
"loss": 0.2039,
"step": 5470
},
{
"epoch": 6.99,
"learning_rate": 6.95299727520436e-05,
"loss": 0.1596,
"step": 5480
},
{
"epoch": 7.0,
"learning_rate": 6.946185286103542e-05,
"loss": 0.2583,
"step": 5490
},
{
"epoch": 7.02,
"learning_rate": 6.939373297002725e-05,
"loss": 0.1986,
"step": 5500
},
{
"epoch": 7.03,
"learning_rate": 6.932561307901907e-05,
"loss": 0.1591,
"step": 5510
},
{
"epoch": 7.04,
"learning_rate": 6.92574931880109e-05,
"loss": 0.1369,
"step": 5520
},
{
"epoch": 7.05,
"learning_rate": 6.918937329700273e-05,
"loss": 0.123,
"step": 5530
},
{
"epoch": 7.07,
"learning_rate": 6.912125340599456e-05,
"loss": 0.2453,
"step": 5540
},
{
"epoch": 7.08,
"learning_rate": 6.905313351498638e-05,
"loss": 0.1955,
"step": 5550
},
{
"epoch": 7.09,
"learning_rate": 6.89850136239782e-05,
"loss": 0.1656,
"step": 5560
},
{
"epoch": 7.1,
"learning_rate": 6.891689373297004e-05,
"loss": 0.1371,
"step": 5570
},
{
"epoch": 7.12,
"learning_rate": 6.884877384196185e-05,
"loss": 0.1227,
"step": 5580
},
{
"epoch": 7.13,
"learning_rate": 6.878065395095368e-05,
"loss": 0.2073,
"step": 5590
},
{
"epoch": 7.14,
"learning_rate": 6.871253405994551e-05,
"loss": 0.2091,
"step": 5600
},
{
"epoch": 7.14,
"eval_loss": 0.27548345923423767,
"eval_runtime": 1422.2095,
"eval_samples_per_second": 4.383,
"eval_steps_per_second": 0.548,
"eval_wer": 0.3328528333587903,
"step": 5600
},
{
"epoch": 7.16,
"learning_rate": 6.864441416893733e-05,
"loss": 0.1649,
"step": 5610
},
{
"epoch": 7.17,
"learning_rate": 6.857629427792915e-05,
"loss": 0.1315,
"step": 5620
},
{
"epoch": 7.18,
"learning_rate": 6.850817438692099e-05,
"loss": 0.1456,
"step": 5630
},
{
"epoch": 7.19,
"learning_rate": 6.84400544959128e-05,
"loss": 0.1979,
"step": 5640
},
{
"epoch": 7.21,
"learning_rate": 6.837193460490463e-05,
"loss": 0.2183,
"step": 5650
},
{
"epoch": 7.22,
"learning_rate": 6.830381471389646e-05,
"loss": 0.1624,
"step": 5660
},
{
"epoch": 7.23,
"learning_rate": 6.823569482288828e-05,
"loss": 0.1402,
"step": 5670
},
{
"epoch": 7.24,
"learning_rate": 6.816757493188011e-05,
"loss": 0.1308,
"step": 5680
},
{
"epoch": 7.26,
"learning_rate": 6.809945504087194e-05,
"loss": 0.2221,
"step": 5690
},
{
"epoch": 7.27,
"learning_rate": 6.803133514986377e-05,
"loss": 0.2085,
"step": 5700
},
{
"epoch": 7.28,
"learning_rate": 6.796321525885558e-05,
"loss": 0.1621,
"step": 5710
},
{
"epoch": 7.3,
"learning_rate": 6.789509536784742e-05,
"loss": 0.1396,
"step": 5720
},
{
"epoch": 7.31,
"learning_rate": 6.782697547683924e-05,
"loss": 0.134,
"step": 5730
},
{
"epoch": 7.32,
"learning_rate": 6.775885558583108e-05,
"loss": 0.2114,
"step": 5740
},
{
"epoch": 7.33,
"learning_rate": 6.769073569482289e-05,
"loss": 0.2041,
"step": 5750
},
{
"epoch": 7.35,
"learning_rate": 6.762261580381472e-05,
"loss": 0.1704,
"step": 5760
},
{
"epoch": 7.36,
"learning_rate": 6.755449591280654e-05,
"loss": 0.1387,
"step": 5770
},
{
"epoch": 7.37,
"learning_rate": 6.748637602179837e-05,
"loss": 0.1347,
"step": 5780
},
{
"epoch": 7.38,
"learning_rate": 6.741825613079019e-05,
"loss": 0.2059,
"step": 5790
},
{
"epoch": 7.4,
"learning_rate": 6.735013623978203e-05,
"loss": 0.2099,
"step": 5800
},
{
"epoch": 7.41,
"learning_rate": 6.728201634877384e-05,
"loss": 0.1647,
"step": 5810
},
{
"epoch": 7.42,
"learning_rate": 6.721389645776567e-05,
"loss": 0.1292,
"step": 5820
},
{
"epoch": 7.44,
"learning_rate": 6.71457765667575e-05,
"loss": 0.1363,
"step": 5830
},
{
"epoch": 7.45,
"learning_rate": 6.707765667574932e-05,
"loss": 0.243,
"step": 5840
},
{
"epoch": 7.46,
"learning_rate": 6.700953678474115e-05,
"loss": 0.2221,
"step": 5850
},
{
"epoch": 7.47,
"learning_rate": 6.694141689373298e-05,
"loss": 0.1653,
"step": 5860
},
{
"epoch": 7.49,
"learning_rate": 6.68732970027248e-05,
"loss": 0.1463,
"step": 5870
},
{
"epoch": 7.5,
"learning_rate": 6.680517711171662e-05,
"loss": 0.128,
"step": 5880
},
{
"epoch": 7.51,
"learning_rate": 6.673705722070846e-05,
"loss": 0.228,
"step": 5890
},
{
"epoch": 7.53,
"learning_rate": 6.666893732970027e-05,
"loss": 0.2107,
"step": 5900
},
{
"epoch": 7.54,
"learning_rate": 6.66008174386921e-05,
"loss": 0.1571,
"step": 5910
},
{
"epoch": 7.55,
"learning_rate": 6.653269754768393e-05,
"loss": 0.1351,
"step": 5920
},
{
"epoch": 7.56,
"learning_rate": 6.646457765667575e-05,
"loss": 0.1337,
"step": 5930
},
{
"epoch": 7.58,
"learning_rate": 6.639645776566757e-05,
"loss": 0.2023,
"step": 5940
},
{
"epoch": 7.59,
"learning_rate": 6.632833787465941e-05,
"loss": 0.2083,
"step": 5950
},
{
"epoch": 7.6,
"learning_rate": 6.626021798365122e-05,
"loss": 0.1521,
"step": 5960
},
{
"epoch": 7.61,
"learning_rate": 6.619209809264305e-05,
"loss": 0.1219,
"step": 5970
},
{
"epoch": 7.63,
"learning_rate": 6.612397820163488e-05,
"loss": 0.1414,
"step": 5980
},
{
"epoch": 7.64,
"learning_rate": 6.60558583106267e-05,
"loss": 0.2388,
"step": 5990
},
{
"epoch": 7.65,
"learning_rate": 6.598773841961853e-05,
"loss": 0.1971,
"step": 6000
},
{
"epoch": 7.65,
"eval_loss": 0.2706403434276581,
"eval_runtime": 1357.2506,
"eval_samples_per_second": 4.592,
"eval_steps_per_second": 0.575,
"eval_wer": 0.33087673743699403,
"step": 6000
},
{
"epoch": 7.67,
"learning_rate": 6.591961852861036e-05,
"loss": 0.1618,
"step": 6010
},
{
"epoch": 7.68,
"learning_rate": 6.585149863760219e-05,
"loss": 0.1335,
"step": 6020
},
{
"epoch": 7.69,
"learning_rate": 6.5783378746594e-05,
"loss": 0.1397,
"step": 6030
},
{
"epoch": 7.7,
"learning_rate": 6.571525885558584e-05,
"loss": 0.2312,
"step": 6040
},
{
"epoch": 7.72,
"learning_rate": 6.564713896457766e-05,
"loss": 0.2257,
"step": 6050
},
{
"epoch": 7.73,
"learning_rate": 6.557901907356948e-05,
"loss": 0.1647,
"step": 6060
},
{
"epoch": 7.74,
"learning_rate": 6.551089918256131e-05,
"loss": 0.1399,
"step": 6070
},
{
"epoch": 7.75,
"learning_rate": 6.544277929155314e-05,
"loss": 0.1251,
"step": 6080
},
{
"epoch": 7.77,
"learning_rate": 6.537465940054495e-05,
"loss": 0.2068,
"step": 6090
},
{
"epoch": 7.78,
"learning_rate": 6.530653950953679e-05,
"loss": 0.1983,
"step": 6100
},
{
"epoch": 7.79,
"learning_rate": 6.523841961852862e-05,
"loss": 0.1569,
"step": 6110
},
{
"epoch": 7.81,
"learning_rate": 6.517029972752043e-05,
"loss": 0.1373,
"step": 6120
},
{
"epoch": 7.82,
"learning_rate": 6.510217983651227e-05,
"loss": 0.126,
"step": 6130
},
{
"epoch": 7.83,
"learning_rate": 6.503405994550409e-05,
"loss": 0.1909,
"step": 6140
},
{
"epoch": 7.84,
"learning_rate": 6.496594005449591e-05,
"loss": 0.2118,
"step": 6150
},
{
"epoch": 7.86,
"learning_rate": 6.489782016348774e-05,
"loss": 0.1755,
"step": 6160
},
{
"epoch": 7.87,
"learning_rate": 6.482970027247957e-05,
"loss": 0.1371,
"step": 6170
},
{
"epoch": 7.88,
"learning_rate": 6.476158038147138e-05,
"loss": 0.1302,
"step": 6180
},
{
"epoch": 7.89,
"learning_rate": 6.469346049046322e-05,
"loss": 0.2641,
"step": 6190
},
{
"epoch": 7.91,
"learning_rate": 6.462534059945504e-05,
"loss": 0.2094,
"step": 6200
},
{
"epoch": 7.92,
"learning_rate": 6.455722070844688e-05,
"loss": 0.1446,
"step": 6210
},
{
"epoch": 7.93,
"learning_rate": 6.448910081743869e-05,
"loss": 0.1363,
"step": 6220
},
{
"epoch": 7.95,
"learning_rate": 6.442098092643052e-05,
"loss": 0.1349,
"step": 6230
},
{
"epoch": 7.96,
"learning_rate": 6.435286103542235e-05,
"loss": 0.2243,
"step": 6240
},
{
"epoch": 7.97,
"learning_rate": 6.428474114441417e-05,
"loss": 0.2035,
"step": 6250
},
{
"epoch": 7.98,
"learning_rate": 6.4216621253406e-05,
"loss": 0.1409,
"step": 6260
},
{
"epoch": 8.0,
"learning_rate": 6.414850136239783e-05,
"loss": 0.1474,
"step": 6270
},
{
"epoch": 8.01,
"learning_rate": 6.408038147138966e-05,
"loss": 0.2361,
"step": 6280
},
{
"epoch": 8.02,
"learning_rate": 6.401226158038147e-05,
"loss": 0.1616,
"step": 6290
},
{
"epoch": 8.04,
"learning_rate": 6.394414168937331e-05,
"loss": 0.1255,
"step": 6300
},
{
"epoch": 8.05,
"learning_rate": 6.387602179836512e-05,
"loss": 0.1027,
"step": 6310
},
{
"epoch": 8.06,
"learning_rate": 6.380790190735695e-05,
"loss": 0.1261,
"step": 6320
},
{
"epoch": 8.07,
"learning_rate": 6.373978201634878e-05,
"loss": 0.2168,
"step": 6330
},
{
"epoch": 8.09,
"learning_rate": 6.36716621253406e-05,
"loss": 0.1548,
"step": 6340
},
{
"epoch": 8.1,
"learning_rate": 6.360354223433242e-05,
"loss": 0.1257,
"step": 6350
},
{
"epoch": 8.11,
"learning_rate": 6.353542234332426e-05,
"loss": 0.1043,
"step": 6360
},
{
"epoch": 8.12,
"learning_rate": 6.346730245231608e-05,
"loss": 0.1502,
"step": 6370
},
{
"epoch": 8.14,
"learning_rate": 6.33991825613079e-05,
"loss": 0.2203,
"step": 6380
},
{
"epoch": 8.15,
"learning_rate": 6.333106267029973e-05,
"loss": 0.1471,
"step": 6390
},
{
"epoch": 8.16,
"learning_rate": 6.326294277929156e-05,
"loss": 0.1237,
"step": 6400
},
{
"epoch": 8.16,
"eval_loss": 0.28225475549697876,
"eval_runtime": 1353.8722,
"eval_samples_per_second": 4.604,
"eval_steps_per_second": 0.576,
"eval_wer": 0.3269627310218421,
"step": 6400
},
{
"epoch": 8.18,
"learning_rate": 6.319482288828338e-05,
"loss": 0.0982,
"step": 6410
},
{
"epoch": 8.19,
"learning_rate": 6.312670299727521e-05,
"loss": 0.1334,
"step": 6420
},
{
"epoch": 8.2,
"learning_rate": 6.305858310626704e-05,
"loss": 0.2213,
"step": 6430
},
{
"epoch": 8.21,
"learning_rate": 6.299046321525885e-05,
"loss": 0.1592,
"step": 6440
},
{
"epoch": 8.23,
"learning_rate": 6.29223433242507e-05,
"loss": 0.1368,
"step": 6450
},
{
"epoch": 8.24,
"learning_rate": 6.285422343324251e-05,
"loss": 0.1116,
"step": 6460
},
{
"epoch": 8.25,
"learning_rate": 6.278610354223433e-05,
"loss": 0.1464,
"step": 6470
},
{
"epoch": 8.27,
"learning_rate": 6.271798365122616e-05,
"loss": 0.2219,
"step": 6480
},
{
"epoch": 8.28,
"learning_rate": 6.264986376021799e-05,
"loss": 0.1525,
"step": 6490
},
{
"epoch": 8.29,
"learning_rate": 6.25817438692098e-05,
"loss": 0.1158,
"step": 6500
},
{
"epoch": 8.3,
"learning_rate": 6.251362397820164e-05,
"loss": 0.1072,
"step": 6510
},
{
"epoch": 8.32,
"learning_rate": 6.244550408719346e-05,
"loss": 0.1393,
"step": 6520
},
{
"epoch": 8.33,
"learning_rate": 6.237738419618529e-05,
"loss": 0.2376,
"step": 6530
},
{
"epoch": 8.34,
"learning_rate": 6.230926430517711e-05,
"loss": 0.1616,
"step": 6540
},
{
"epoch": 8.35,
"learning_rate": 6.224114441416894e-05,
"loss": 0.1236,
"step": 6550
},
{
"epoch": 8.37,
"learning_rate": 6.217302452316077e-05,
"loss": 0.0995,
"step": 6560
},
{
"epoch": 8.38,
"learning_rate": 6.21049046321526e-05,
"loss": 0.1579,
"step": 6570
},
{
"epoch": 8.39,
"learning_rate": 6.203678474114442e-05,
"loss": 0.2111,
"step": 6580
},
{
"epoch": 8.41,
"learning_rate": 6.196866485013624e-05,
"loss": 0.1493,
"step": 6590
},
{
"epoch": 8.42,
"learning_rate": 6.190054495912808e-05,
"loss": 0.126,
"step": 6600
},
{
"epoch": 8.43,
"learning_rate": 6.183242506811989e-05,
"loss": 0.1424,
"step": 6610
},
{
"epoch": 8.44,
"learning_rate": 6.176430517711172e-05,
"loss": 0.1535,
"step": 6620
},
{
"epoch": 8.46,
"learning_rate": 6.169618528610354e-05,
"loss": 0.2261,
"step": 6630
},
{
"epoch": 8.47,
"learning_rate": 6.162806539509537e-05,
"loss": 0.1551,
"step": 6640
},
{
"epoch": 8.48,
"learning_rate": 6.15599455040872e-05,
"loss": 0.1835,
"step": 6650
},
{
"epoch": 8.49,
"learning_rate": 6.149182561307903e-05,
"loss": 0.1058,
"step": 6660
},
{
"epoch": 8.51,
"learning_rate": 6.142370572207084e-05,
"loss": 0.1685,
"step": 6670
},
{
"epoch": 8.52,
"learning_rate": 6.135558583106268e-05,
"loss": 0.2288,
"step": 6680
},
{
"epoch": 8.53,
"learning_rate": 6.12874659400545e-05,
"loss": 0.1564,
"step": 6690
},
{
"epoch": 8.55,
"learning_rate": 6.121934604904632e-05,
"loss": 0.1162,
"step": 6700
},
{
"epoch": 8.56,
"learning_rate": 6.115122615803815e-05,
"loss": 0.1167,
"step": 6710
},
{
"epoch": 8.57,
"learning_rate": 6.108310626702998e-05,
"loss": 0.145,
"step": 6720
},
{
"epoch": 8.58,
"learning_rate": 6.10149863760218e-05,
"loss": 0.231,
"step": 6730
},
{
"epoch": 8.6,
"learning_rate": 6.094686648501363e-05,
"loss": 0.1575,
"step": 6740
},
{
"epoch": 8.61,
"learning_rate": 6.087874659400545e-05,
"loss": 0.1249,
"step": 6750
},
{
"epoch": 8.62,
"learning_rate": 6.081062670299727e-05,
"loss": 0.1176,
"step": 6760
},
{
"epoch": 8.63,
"learning_rate": 6.074250681198911e-05,
"loss": 0.1393,
"step": 6770
},
{
"epoch": 8.65,
"learning_rate": 6.067438692098093e-05,
"loss": 0.2434,
"step": 6780
},
{
"epoch": 8.66,
"learning_rate": 6.0606267029972755e-05,
"loss": 0.1537,
"step": 6790
},
{
"epoch": 8.67,
"learning_rate": 6.053814713896458e-05,
"loss": 0.123,
"step": 6800
},
{
"epoch": 8.67,
"eval_loss": 0.27536651492118835,
"eval_runtime": 1356.6363,
"eval_samples_per_second": 4.594,
"eval_steps_per_second": 0.575,
"eval_wer": 0.3246143271727509,
"step": 6800
},
{
"epoch": 8.69,
"learning_rate": 6.047002724795641e-05,
"loss": 0.1009,
"step": 6810
},
{
"epoch": 8.7,
"learning_rate": 6.040190735694823e-05,
"loss": 0.1458,
"step": 6820
},
{
"epoch": 8.71,
"learning_rate": 6.0333787465940064e-05,
"loss": 0.2283,
"step": 6830
},
{
"epoch": 8.72,
"learning_rate": 6.0265667574931884e-05,
"loss": 0.1539,
"step": 6840
},
{
"epoch": 8.74,
"learning_rate": 6.0197547683923705e-05,
"loss": 0.1197,
"step": 6850
},
{
"epoch": 8.75,
"learning_rate": 6.012942779291554e-05,
"loss": 0.1078,
"step": 6860
},
{
"epoch": 8.76,
"learning_rate": 6.006130790190736e-05,
"loss": 0.1415,
"step": 6870
},
{
"epoch": 8.77,
"learning_rate": 5.999318801089918e-05,
"loss": 0.226,
"step": 6880
},
{
"epoch": 8.79,
"learning_rate": 5.9925068119891014e-05,
"loss": 0.146,
"step": 6890
},
{
"epoch": 8.8,
"learning_rate": 5.9856948228882835e-05,
"loss": 0.1146,
"step": 6900
},
{
"epoch": 8.81,
"learning_rate": 5.9788828337874655e-05,
"loss": 0.1093,
"step": 6910
},
{
"epoch": 8.83,
"learning_rate": 5.972070844686649e-05,
"loss": 0.1492,
"step": 6920
},
{
"epoch": 8.84,
"learning_rate": 5.965258855585831e-05,
"loss": 0.2254,
"step": 6930
},
{
"epoch": 8.85,
"learning_rate": 5.958446866485014e-05,
"loss": 0.1561,
"step": 6940
},
{
"epoch": 8.86,
"learning_rate": 5.9516348773841965e-05,
"loss": 0.1232,
"step": 6950
},
{
"epoch": 8.88,
"learning_rate": 5.944822888283379e-05,
"loss": 0.1036,
"step": 6960
},
{
"epoch": 8.89,
"learning_rate": 5.938010899182561e-05,
"loss": 0.1354,
"step": 6970
},
{
"epoch": 8.9,
"learning_rate": 5.9311989100817447e-05,
"loss": 0.2194,
"step": 6980
},
{
"epoch": 8.92,
"learning_rate": 5.924386920980927e-05,
"loss": 0.1511,
"step": 6990
},
{
"epoch": 8.93,
"learning_rate": 5.917574931880109e-05,
"loss": 0.1192,
"step": 7000
},
{
"epoch": 8.94,
"learning_rate": 5.910762942779292e-05,
"loss": 0.1051,
"step": 7010
},
{
"epoch": 8.95,
"learning_rate": 5.903950953678474e-05,
"loss": 0.1295,
"step": 7020
},
{
"epoch": 8.97,
"learning_rate": 5.897138964577656e-05,
"loss": 0.2105,
"step": 7030
},
{
"epoch": 8.98,
"learning_rate": 5.89032697547684e-05,
"loss": 0.1368,
"step": 7040
},
{
"epoch": 8.99,
"learning_rate": 5.883514986376022e-05,
"loss": 0.1107,
"step": 7050
},
{
"epoch": 9.01,
"learning_rate": 5.876702997275204e-05,
"loss": 0.2054,
"step": 7060
},
{
"epoch": 9.02,
"learning_rate": 5.869891008174387e-05,
"loss": 0.141,
"step": 7070
},
{
"epoch": 9.03,
"learning_rate": 5.863079019073569e-05,
"loss": 0.1179,
"step": 7080
},
{
"epoch": 9.04,
"learning_rate": 5.856267029972752e-05,
"loss": 0.0953,
"step": 7090
},
{
"epoch": 9.06,
"learning_rate": 5.8494550408719354e-05,
"loss": 0.114,
"step": 7100
},
{
"epoch": 9.07,
"learning_rate": 5.8426430517711175e-05,
"loss": 0.1698,
"step": 7110
},
{
"epoch": 9.08,
"learning_rate": 5.835831062670301e-05,
"loss": 0.1508,
"step": 7120
},
{
"epoch": 9.09,
"learning_rate": 5.829019073569483e-05,
"loss": 0.1086,
"step": 7130
},
{
"epoch": 9.11,
"learning_rate": 5.822207084468665e-05,
"loss": 0.1121,
"step": 7140
},
{
"epoch": 9.12,
"learning_rate": 5.8153950953678484e-05,
"loss": 0.1243,
"step": 7150
},
{
"epoch": 9.13,
"learning_rate": 5.8085831062670304e-05,
"loss": 0.1963,
"step": 7160
},
{
"epoch": 9.15,
"learning_rate": 5.8017711171662125e-05,
"loss": 0.1519,
"step": 7170
},
{
"epoch": 9.16,
"learning_rate": 5.794959128065396e-05,
"loss": 0.1153,
"step": 7180
},
{
"epoch": 9.17,
"learning_rate": 5.788147138964578e-05,
"loss": 0.1292,
"step": 7190
},
{
"epoch": 9.18,
"learning_rate": 5.78133514986376e-05,
"loss": 0.103,
"step": 7200
},
{
"epoch": 9.18,
"eval_loss": 0.2917274832725525,
"eval_runtime": 1363.2635,
"eval_samples_per_second": 4.572,
"eval_steps_per_second": 0.572,
"eval_wer": 0.3271536581640446,
"step": 7200
},
{
"epoch": 9.2,
"learning_rate": 5.7745231607629434e-05,
"loss": 0.2107,
"step": 7210
},
{
"epoch": 9.21,
"learning_rate": 5.7677111716621255e-05,
"loss": 0.1528,
"step": 7220
},
{
"epoch": 9.22,
"learning_rate": 5.760899182561308e-05,
"loss": 0.1125,
"step": 7230
},
{
"epoch": 9.23,
"learning_rate": 5.754087193460491e-05,
"loss": 0.0891,
"step": 7240
},
{
"epoch": 9.25,
"learning_rate": 5.747275204359674e-05,
"loss": 0.1036,
"step": 7250
},
{
"epoch": 9.26,
"learning_rate": 5.740463215258856e-05,
"loss": 0.1822,
"step": 7260
},
{
"epoch": 9.27,
"learning_rate": 5.733651226158039e-05,
"loss": 0.1684,
"step": 7270
},
{
"epoch": 9.29,
"learning_rate": 5.726839237057221e-05,
"loss": 0.1207,
"step": 7280
},
{
"epoch": 9.3,
"learning_rate": 5.720027247956403e-05,
"loss": 0.0943,
"step": 7290
},
{
"epoch": 9.31,
"learning_rate": 5.7132152588555867e-05,
"loss": 0.1049,
"step": 7300
},
{
"epoch": 9.32,
"learning_rate": 5.706403269754769e-05,
"loss": 0.179,
"step": 7310
},
{
"epoch": 9.34,
"learning_rate": 5.699591280653951e-05,
"loss": 0.1544,
"step": 7320
},
{
"epoch": 9.35,
"learning_rate": 5.692779291553134e-05,
"loss": 0.1168,
"step": 7330
},
{
"epoch": 9.36,
"learning_rate": 5.685967302452316e-05,
"loss": 0.091,
"step": 7340
},
{
"epoch": 9.37,
"learning_rate": 5.679155313351498e-05,
"loss": 0.0994,
"step": 7350
},
{
"epoch": 9.39,
"learning_rate": 5.672343324250682e-05,
"loss": 0.1756,
"step": 7360
},
{
"epoch": 9.4,
"learning_rate": 5.665531335149864e-05,
"loss": 0.1602,
"step": 7370
},
{
"epoch": 9.41,
"learning_rate": 5.6587193460490465e-05,
"loss": 0.1205,
"step": 7380
},
{
"epoch": 9.43,
"learning_rate": 5.651907356948229e-05,
"loss": 0.0882,
"step": 7390
},
{
"epoch": 9.44,
"learning_rate": 5.645095367847412e-05,
"loss": 0.1134,
"step": 7400
},
{
"epoch": 9.45,
"learning_rate": 5.638283378746594e-05,
"loss": 0.1802,
"step": 7410
},
{
"epoch": 9.46,
"learning_rate": 5.6314713896457774e-05,
"loss": 0.1509,
"step": 7420
},
{
"epoch": 9.48,
"learning_rate": 5.6246594005449595e-05,
"loss": 0.1117,
"step": 7430
},
{
"epoch": 9.49,
"learning_rate": 5.6178474114441415e-05,
"loss": 0.1064,
"step": 7440
},
{
"epoch": 9.5,
"learning_rate": 5.611035422343325e-05,
"loss": 0.0984,
"step": 7450
},
{
"epoch": 9.51,
"learning_rate": 5.604223433242507e-05,
"loss": 0.1623,
"step": 7460
},
{
"epoch": 9.53,
"learning_rate": 5.597411444141689e-05,
"loss": 0.1516,
"step": 7470
},
{
"epoch": 9.54,
"learning_rate": 5.5905994550408724e-05,
"loss": 0.1151,
"step": 7480
},
{
"epoch": 9.55,
"learning_rate": 5.5837874659400545e-05,
"loss": 0.1012,
"step": 7490
},
{
"epoch": 9.57,
"learning_rate": 5.5769754768392365e-05,
"loss": 0.0917,
"step": 7500
},
{
"epoch": 9.58,
"learning_rate": 5.57016348773842e-05,
"loss": 0.2109,
"step": 7510
},
{
"epoch": 9.59,
"learning_rate": 5.563351498637602e-05,
"loss": 0.1456,
"step": 7520
},
{
"epoch": 9.6,
"learning_rate": 5.556539509536785e-05,
"loss": 0.1244,
"step": 7530
},
{
"epoch": 9.62,
"learning_rate": 5.5497275204359675e-05,
"loss": 0.0984,
"step": 7540
},
{
"epoch": 9.63,
"learning_rate": 5.54291553133515e-05,
"loss": 0.0984,
"step": 7550
},
{
"epoch": 9.64,
"learning_rate": 5.536103542234332e-05,
"loss": 0.1904,
"step": 7560
},
{
"epoch": 9.66,
"learning_rate": 5.529291553133516e-05,
"loss": 0.1523,
"step": 7570
},
{
"epoch": 9.67,
"learning_rate": 5.522479564032698e-05,
"loss": 0.1162,
"step": 7580
},
{
"epoch": 9.68,
"learning_rate": 5.515667574931881e-05,
"loss": 0.095,
"step": 7590
},
{
"epoch": 9.69,
"learning_rate": 5.508855585831063e-05,
"loss": 0.1143,
"step": 7600
},
{
"epoch": 9.69,
"eval_loss": 0.28853729367256165,
"eval_runtime": 1355.1162,
"eval_samples_per_second": 4.6,
"eval_steps_per_second": 0.576,
"eval_wer": 0.33048533679547887,
"step": 7600
},
{
"epoch": 9.71,
"learning_rate": 5.502043596730245e-05,
"loss": 0.1862,
"step": 7610
},
{
"epoch": 9.72,
"learning_rate": 5.4952316076294287e-05,
"loss": 0.164,
"step": 7620
},
{
"epoch": 9.73,
"learning_rate": 5.488419618528611e-05,
"loss": 0.1247,
"step": 7630
},
{
"epoch": 9.74,
"learning_rate": 5.481607629427793e-05,
"loss": 0.0893,
"step": 7640
},
{
"epoch": 9.76,
"learning_rate": 5.474795640326976e-05,
"loss": 0.1028,
"step": 7650
},
{
"epoch": 9.77,
"learning_rate": 5.467983651226158e-05,
"loss": 0.2013,
"step": 7660
},
{
"epoch": 9.78,
"learning_rate": 5.46117166212534e-05,
"loss": 0.1607,
"step": 7670
},
{
"epoch": 9.8,
"learning_rate": 5.454359673024524e-05,
"loss": 0.1262,
"step": 7680
},
{
"epoch": 9.81,
"learning_rate": 5.447547683923706e-05,
"loss": 0.103,
"step": 7690
},
{
"epoch": 9.82,
"learning_rate": 5.4407356948228885e-05,
"loss": 0.098,
"step": 7700
},
{
"epoch": 9.83,
"learning_rate": 5.433923705722072e-05,
"loss": 0.2012,
"step": 7710
},
{
"epoch": 9.85,
"learning_rate": 5.427111716621254e-05,
"loss": 0.1486,
"step": 7720
},
{
"epoch": 9.86,
"learning_rate": 5.420299727520436e-05,
"loss": 0.11,
"step": 7730
},
{
"epoch": 9.87,
"learning_rate": 5.4134877384196194e-05,
"loss": 0.0958,
"step": 7740
},
{
"epoch": 9.88,
"learning_rate": 5.4066757493188015e-05,
"loss": 0.1152,
"step": 7750
},
{
"epoch": 9.9,
"learning_rate": 5.3998637602179835e-05,
"loss": 0.1947,
"step": 7760
},
{
"epoch": 9.91,
"learning_rate": 5.393051771117167e-05,
"loss": 0.1528,
"step": 7770
},
{
"epoch": 9.92,
"learning_rate": 5.386239782016349e-05,
"loss": 0.1232,
"step": 7780
},
{
"epoch": 9.94,
"learning_rate": 5.379427792915531e-05,
"loss": 0.0953,
"step": 7790
},
{
"epoch": 9.95,
"learning_rate": 5.3726158038147144e-05,
"loss": 0.1072,
"step": 7800
},
{
"epoch": 9.96,
"learning_rate": 5.3658038147138965e-05,
"loss": 0.1809,
"step": 7810
},
{
"epoch": 9.97,
"learning_rate": 5.358991825613079e-05,
"loss": 0.1366,
"step": 7820
},
{
"epoch": 9.99,
"learning_rate": 5.352179836512262e-05,
"loss": 0.1063,
"step": 7830
},
{
"epoch": 10.0,
"learning_rate": 5.345367847411445e-05,
"loss": 0.1365,
"step": 7840
},
{
"epoch": 10.01,
"learning_rate": 5.338555858310627e-05,
"loss": 0.1612,
"step": 7850
},
{
"epoch": 10.03,
"learning_rate": 5.33174386920981e-05,
"loss": 0.1143,
"step": 7860
},
{
"epoch": 10.04,
"learning_rate": 5.324931880108992e-05,
"loss": 0.0963,
"step": 7870
},
{
"epoch": 10.05,
"learning_rate": 5.318119891008174e-05,
"loss": 0.085,
"step": 7880
},
{
"epoch": 10.06,
"learning_rate": 5.311307901907358e-05,
"loss": 0.1373,
"step": 7890
},
{
"epoch": 10.08,
"learning_rate": 5.30449591280654e-05,
"loss": 0.168,
"step": 7900
},
{
"epoch": 10.09,
"learning_rate": 5.297683923705722e-05,
"loss": 0.1151,
"step": 7910
},
{
"epoch": 10.1,
"learning_rate": 5.290871934604905e-05,
"loss": 0.0903,
"step": 7920
},
{
"epoch": 10.11,
"learning_rate": 5.284059945504087e-05,
"loss": 0.0753,
"step": 7930
},
{
"epoch": 10.13,
"learning_rate": 5.277247956403269e-05,
"loss": 0.1385,
"step": 7940
},
{
"epoch": 10.14,
"learning_rate": 5.270435967302453e-05,
"loss": 0.1558,
"step": 7950
},
{
"epoch": 10.15,
"learning_rate": 5.263623978201635e-05,
"loss": 0.1149,
"step": 7960
},
{
"epoch": 10.17,
"learning_rate": 5.2568119891008175e-05,
"loss": 0.0961,
"step": 7970
},
{
"epoch": 10.18,
"learning_rate": 5.25e-05,
"loss": 0.086,
"step": 7980
},
{
"epoch": 10.19,
"learning_rate": 5.243188010899183e-05,
"loss": 0.1316,
"step": 7990
},
{
"epoch": 10.2,
"learning_rate": 5.236376021798365e-05,
"loss": 0.156,
"step": 8000
},
{
"epoch": 10.2,
"eval_loss": 0.281043142080307,
"eval_runtime": 1358.69,
"eval_samples_per_second": 4.588,
"eval_steps_per_second": 0.574,
"eval_wer": 0.3288147243012067,
"step": 8000
},
{
"epoch": 10.22,
"learning_rate": 5.2295640326975484e-05,
"loss": 0.1093,
"step": 8010
},
{
"epoch": 10.23,
"learning_rate": 5.2227520435967305e-05,
"loss": 0.098,
"step": 8020
},
{
"epoch": 10.24,
"learning_rate": 5.215940054495914e-05,
"loss": 0.071,
"step": 8030
},
{
"epoch": 10.25,
"learning_rate": 5.209128065395096e-05,
"loss": 0.1449,
"step": 8040
},
{
"epoch": 10.27,
"learning_rate": 5.202316076294278e-05,
"loss": 0.1516,
"step": 8050
},
{
"epoch": 10.28,
"learning_rate": 5.1955040871934614e-05,
"loss": 0.1174,
"step": 8060
},
{
"epoch": 10.29,
"learning_rate": 5.1886920980926434e-05,
"loss": 0.0923,
"step": 8070
},
{
"epoch": 10.31,
"learning_rate": 5.1818801089918255e-05,
"loss": 0.0845,
"step": 8080
},
{
"epoch": 10.32,
"learning_rate": 5.175068119891009e-05,
"loss": 0.1245,
"step": 8090
},
{
"epoch": 10.33,
"learning_rate": 5.168256130790191e-05,
"loss": 0.1581,
"step": 8100
},
{
"epoch": 10.34,
"learning_rate": 5.161444141689373e-05,
"loss": 0.1186,
"step": 8110
},
{
"epoch": 10.36,
"learning_rate": 5.1546321525885564e-05,
"loss": 0.0911,
"step": 8120
},
{
"epoch": 10.37,
"learning_rate": 5.1478201634877385e-05,
"loss": 0.0854,
"step": 8130
},
{
"epoch": 10.38,
"learning_rate": 5.141008174386921e-05,
"loss": 0.1656,
"step": 8140
},
{
"epoch": 10.4,
"learning_rate": 5.134196185286104e-05,
"loss": 0.1625,
"step": 8150
},
{
"epoch": 10.41,
"learning_rate": 5.127384196185287e-05,
"loss": 0.1175,
"step": 8160
},
{
"epoch": 10.42,
"learning_rate": 5.120572207084469e-05,
"loss": 0.0917,
"step": 8170
},
{
"epoch": 10.43,
"learning_rate": 5.113760217983652e-05,
"loss": 0.0878,
"step": 8180
},
{
"epoch": 10.45,
"learning_rate": 5.106948228882834e-05,
"loss": 0.1276,
"step": 8190
},
{
"epoch": 10.46,
"learning_rate": 5.100136239782016e-05,
"loss": 0.1524,
"step": 8200
},
{
"epoch": 10.47,
"learning_rate": 5.0933242506811997e-05,
"loss": 0.1095,
"step": 8210
},
{
"epoch": 10.48,
"learning_rate": 5.086512261580382e-05,
"loss": 0.0993,
"step": 8220
},
{
"epoch": 10.5,
"learning_rate": 5.079700272479564e-05,
"loss": 0.0883,
"step": 8230
},
{
"epoch": 10.51,
"learning_rate": 5.072888283378747e-05,
"loss": 0.1267,
"step": 8240
},
{
"epoch": 10.52,
"learning_rate": 5.066076294277929e-05,
"loss": 0.1568,
"step": 8250
},
{
"epoch": 10.54,
"learning_rate": 5.059264305177111e-05,
"loss": 0.1185,
"step": 8260
},
{
"epoch": 10.55,
"learning_rate": 5.052452316076295e-05,
"loss": 0.0867,
"step": 8270
},
{
"epoch": 10.56,
"learning_rate": 5.045640326975477e-05,
"loss": 0.0829,
"step": 8280
},
{
"epoch": 10.57,
"learning_rate": 5.0388283378746595e-05,
"loss": 0.1458,
"step": 8290
},
{
"epoch": 10.59,
"learning_rate": 5.032016348773842e-05,
"loss": 0.1596,
"step": 8300
},
{
"epoch": 10.6,
"learning_rate": 5.025204359673025e-05,
"loss": 0.1166,
"step": 8310
},
{
"epoch": 10.61,
"learning_rate": 5.018392370572207e-05,
"loss": 0.092,
"step": 8320
},
{
"epoch": 10.62,
"learning_rate": 5.0115803814713904e-05,
"loss": 0.0755,
"step": 8330
},
{
"epoch": 10.64,
"learning_rate": 5.0047683923705725e-05,
"loss": 0.1201,
"step": 8340
},
{
"epoch": 10.65,
"learning_rate": 4.997956403269755e-05,
"loss": 0.1542,
"step": 8350
},
{
"epoch": 10.66,
"learning_rate": 4.991144414168937e-05,
"loss": 0.11,
"step": 8360
},
{
"epoch": 10.68,
"learning_rate": 4.98433242506812e-05,
"loss": 0.0914,
"step": 8370
},
{
"epoch": 10.69,
"learning_rate": 4.977520435967303e-05,
"loss": 0.0864,
"step": 8380
},
{
"epoch": 10.7,
"learning_rate": 4.9707084468664854e-05,
"loss": 0.1088,
"step": 8390
},
{
"epoch": 10.71,
"learning_rate": 4.9638964577656675e-05,
"loss": 0.167,
"step": 8400
},
{
"epoch": 10.71,
"eval_loss": 0.2688770592212677,
"eval_runtime": 1426.2094,
"eval_samples_per_second": 4.37,
"eval_steps_per_second": 0.547,
"eval_wer": 0.3232014663204521,
"step": 8400
},
{
"epoch": 10.73,
"learning_rate": 4.95708446866485e-05,
"loss": 0.1123,
"step": 8410
},
{
"epoch": 10.74,
"learning_rate": 4.950272479564033e-05,
"loss": 0.0967,
"step": 8420
},
{
"epoch": 10.75,
"learning_rate": 4.943460490463216e-05,
"loss": 0.0923,
"step": 8430
},
{
"epoch": 10.76,
"learning_rate": 4.9366485013623984e-05,
"loss": 0.1188,
"step": 8440
},
{
"epoch": 10.78,
"learning_rate": 4.929836512261581e-05,
"loss": 0.162,
"step": 8450
},
{
"epoch": 10.79,
"learning_rate": 4.923024523160763e-05,
"loss": 0.1141,
"step": 8460
},
{
"epoch": 10.8,
"learning_rate": 4.916212534059946e-05,
"loss": 0.0887,
"step": 8470
},
{
"epoch": 10.82,
"learning_rate": 4.909400544959129e-05,
"loss": 0.0782,
"step": 8480
},
{
"epoch": 10.83,
"learning_rate": 4.902588555858311e-05,
"loss": 0.1172,
"step": 8490
},
{
"epoch": 10.84,
"learning_rate": 4.8957765667574935e-05,
"loss": 0.1609,
"step": 8500
},
{
"epoch": 10.85,
"learning_rate": 4.888964577656676e-05,
"loss": 0.1158,
"step": 8510
},
{
"epoch": 10.87,
"learning_rate": 4.882152588555858e-05,
"loss": 0.0916,
"step": 8520
},
{
"epoch": 10.88,
"learning_rate": 4.875340599455041e-05,
"loss": 0.0741,
"step": 8530
},
{
"epoch": 10.89,
"learning_rate": 4.868528610354224e-05,
"loss": 0.1279,
"step": 8540
},
{
"epoch": 10.9,
"learning_rate": 4.861716621253406e-05,
"loss": 0.1497,
"step": 8550
},
{
"epoch": 10.92,
"learning_rate": 4.8549046321525885e-05,
"loss": 0.1137,
"step": 8560
},
{
"epoch": 10.93,
"learning_rate": 4.848092643051771e-05,
"loss": 0.0854,
"step": 8570
},
{
"epoch": 10.94,
"learning_rate": 4.841280653950954e-05,
"loss": 0.0817,
"step": 8580
},
{
"epoch": 10.96,
"learning_rate": 4.834468664850137e-05,
"loss": 0.1257,
"step": 8590
},
{
"epoch": 10.97,
"learning_rate": 4.8276566757493194e-05,
"loss": 0.1407,
"step": 8600
},
{
"epoch": 10.98,
"learning_rate": 4.820844686648502e-05,
"loss": 0.094,
"step": 8610
},
{
"epoch": 10.99,
"learning_rate": 4.814032697547684e-05,
"loss": 0.086,
"step": 8620
},
{
"epoch": 11.01,
"learning_rate": 4.807220708446867e-05,
"loss": 0.1743,
"step": 8630
},
{
"epoch": 11.02,
"learning_rate": 4.80040871934605e-05,
"loss": 0.1049,
"step": 8640
},
{
"epoch": 11.03,
"learning_rate": 4.793596730245232e-05,
"loss": 0.0851,
"step": 8650
},
{
"epoch": 11.05,
"learning_rate": 4.7867847411444145e-05,
"loss": 0.0711,
"step": 8660
},
{
"epoch": 11.06,
"learning_rate": 4.779972752043597e-05,
"loss": 0.0907,
"step": 8670
},
{
"epoch": 11.07,
"learning_rate": 4.773160762942779e-05,
"loss": 0.1484,
"step": 8680
},
{
"epoch": 11.08,
"learning_rate": 4.766348773841962e-05,
"loss": 0.1114,
"step": 8690
},
{
"epoch": 11.1,
"learning_rate": 4.759536784741145e-05,
"loss": 0.0901,
"step": 8700
},
{
"epoch": 11.11,
"learning_rate": 4.752724795640327e-05,
"loss": 0.0707,
"step": 8710
},
{
"epoch": 11.12,
"learning_rate": 4.7459128065395095e-05,
"loss": 0.0971,
"step": 8720
},
{
"epoch": 11.14,
"learning_rate": 4.739100817438692e-05,
"loss": 0.1691,
"step": 8730
},
{
"epoch": 11.15,
"learning_rate": 4.732288828337875e-05,
"loss": 0.1155,
"step": 8740
},
{
"epoch": 11.16,
"learning_rate": 4.725476839237058e-05,
"loss": 0.0854,
"step": 8750
},
{
"epoch": 11.17,
"learning_rate": 4.7186648501362404e-05,
"loss": 0.0676,
"step": 8760
},
{
"epoch": 11.19,
"learning_rate": 4.7118528610354225e-05,
"loss": 0.0866,
"step": 8770
},
{
"epoch": 11.2,
"learning_rate": 4.705040871934605e-05,
"loss": 0.1466,
"step": 8780
},
{
"epoch": 11.21,
"learning_rate": 4.698228882833788e-05,
"loss": 0.1183,
"step": 8790
},
{
"epoch": 11.22,
"learning_rate": 4.69141689373297e-05,
"loss": 0.0815,
"step": 8800
},
{
"epoch": 11.22,
"eval_loss": 0.28985264897346497,
"eval_runtime": 1359.0731,
"eval_samples_per_second": 4.586,
"eval_steps_per_second": 0.574,
"eval_wer": 0.32356422789063694,
"step": 8800
},
{
"epoch": 11.24,
"learning_rate": 4.684604904632153e-05,
"loss": 0.0679,
"step": 8810
},
{
"epoch": 11.25,
"learning_rate": 4.6777929155313355e-05,
"loss": 0.0858,
"step": 8820
},
{
"epoch": 11.26,
"learning_rate": 4.670980926430518e-05,
"loss": 0.14,
"step": 8830
},
{
"epoch": 11.28,
"learning_rate": 4.6641689373297e-05,
"loss": 0.1108,
"step": 8840
},
{
"epoch": 11.29,
"learning_rate": 4.657356948228883e-05,
"loss": 0.0885,
"step": 8850
},
{
"epoch": 11.3,
"learning_rate": 4.650544959128066e-05,
"loss": 0.0676,
"step": 8860
},
{
"epoch": 11.31,
"learning_rate": 4.643732970027248e-05,
"loss": 0.0892,
"step": 8870
},
{
"epoch": 11.33,
"learning_rate": 4.6369209809264305e-05,
"loss": 0.1677,
"step": 8880
},
{
"epoch": 11.34,
"learning_rate": 4.630108991825613e-05,
"loss": 0.1307,
"step": 8890
},
{
"epoch": 11.35,
"learning_rate": 4.623297002724796e-05,
"loss": 0.1091,
"step": 8900
},
{
"epoch": 11.36,
"learning_rate": 4.616485013623979e-05,
"loss": 0.0746,
"step": 8910
},
{
"epoch": 11.38,
"learning_rate": 4.6096730245231614e-05,
"loss": 0.0818,
"step": 8920
},
{
"epoch": 11.39,
"learning_rate": 4.6028610354223435e-05,
"loss": 0.147,
"step": 8930
},
{
"epoch": 11.4,
"learning_rate": 4.596049046321526e-05,
"loss": 0.1256,
"step": 8940
},
{
"epoch": 11.42,
"learning_rate": 4.589237057220709e-05,
"loss": 0.0888,
"step": 8950
},
{
"epoch": 11.43,
"learning_rate": 4.582425068119891e-05,
"loss": 0.071,
"step": 8960
},
{
"epoch": 11.44,
"learning_rate": 4.575613079019074e-05,
"loss": 0.0862,
"step": 8970
},
{
"epoch": 11.45,
"learning_rate": 4.5688010899182565e-05,
"loss": 0.1401,
"step": 8980
},
{
"epoch": 11.47,
"learning_rate": 4.5619891008174385e-05,
"loss": 0.1185,
"step": 8990
},
{
"epoch": 11.48,
"learning_rate": 4.555177111716621e-05,
"loss": 0.0934,
"step": 9000
},
{
"epoch": 11.49,
"learning_rate": 4.548365122615804e-05,
"loss": 0.0785,
"step": 9010
},
{
"epoch": 11.5,
"learning_rate": 4.541553133514986e-05,
"loss": 0.0778,
"step": 9020
},
{
"epoch": 11.52,
"learning_rate": 4.534741144414169e-05,
"loss": 0.1465,
"step": 9030
},
{
"epoch": 11.53,
"learning_rate": 4.5279291553133515e-05,
"loss": 0.1161,
"step": 9040
},
{
"epoch": 11.54,
"learning_rate": 4.521117166212534e-05,
"loss": 0.0896,
"step": 9050
},
{
"epoch": 11.56,
"learning_rate": 4.514305177111717e-05,
"loss": 0.0755,
"step": 9060
},
{
"epoch": 11.57,
"learning_rate": 4.5074931880109e-05,
"loss": 0.0906,
"step": 9070
},
{
"epoch": 11.58,
"learning_rate": 4.5006811989100824e-05,
"loss": 0.1451,
"step": 9080
},
{
"epoch": 11.59,
"learning_rate": 4.4938692098092645e-05,
"loss": 0.1147,
"step": 9090
},
{
"epoch": 11.61,
"learning_rate": 4.487057220708447e-05,
"loss": 0.0807,
"step": 9100
},
{
"epoch": 11.62,
"learning_rate": 4.48024523160763e-05,
"loss": 0.0787,
"step": 9110
},
{
"epoch": 11.63,
"learning_rate": 4.473433242506812e-05,
"loss": 0.0931,
"step": 9120
},
{
"epoch": 11.64,
"learning_rate": 4.466621253405995e-05,
"loss": 0.1479,
"step": 9130
},
{
"epoch": 11.66,
"learning_rate": 4.4598092643051775e-05,
"loss": 0.1033,
"step": 9140
},
{
"epoch": 11.67,
"learning_rate": 4.4529972752043595e-05,
"loss": 0.0866,
"step": 9150
},
{
"epoch": 11.68,
"learning_rate": 4.446185286103542e-05,
"loss": 0.0759,
"step": 9160
},
{
"epoch": 11.7,
"learning_rate": 4.439373297002725e-05,
"loss": 0.1024,
"step": 9170
},
{
"epoch": 11.71,
"learning_rate": 4.432561307901908e-05,
"loss": 0.1796,
"step": 9180
},
{
"epoch": 11.72,
"learning_rate": 4.4257493188010904e-05,
"loss": 0.1174,
"step": 9190
},
{
"epoch": 11.73,
"learning_rate": 4.418937329700273e-05,
"loss": 0.0844,
"step": 9200
},
{
"epoch": 11.73,
"eval_loss": 0.2798416316509247,
"eval_runtime": 1354.4726,
"eval_samples_per_second": 4.602,
"eval_steps_per_second": 0.576,
"eval_wer": 0.3224854895371926,
"step": 9200
},
{
"epoch": 11.75,
"learning_rate": 4.412125340599455e-05,
"loss": 0.0712,
"step": 9210
},
{
"epoch": 11.76,
"learning_rate": 4.405313351498638e-05,
"loss": 0.0797,
"step": 9220
},
{
"epoch": 11.77,
"learning_rate": 4.398501362397821e-05,
"loss": 0.1652,
"step": 9230
},
{
"epoch": 11.79,
"learning_rate": 4.391689373297003e-05,
"loss": 0.1123,
"step": 9240
},
{
"epoch": 11.8,
"learning_rate": 4.3848773841961855e-05,
"loss": 0.0905,
"step": 9250
},
{
"epoch": 11.81,
"learning_rate": 4.378065395095368e-05,
"loss": 0.0705,
"step": 9260
},
{
"epoch": 11.82,
"learning_rate": 4.37125340599455e-05,
"loss": 0.083,
"step": 9270
},
{
"epoch": 11.84,
"learning_rate": 4.364441416893733e-05,
"loss": 0.159,
"step": 9280
},
{
"epoch": 11.85,
"learning_rate": 4.357629427792916e-05,
"loss": 0.114,
"step": 9290
},
{
"epoch": 11.86,
"learning_rate": 4.3508174386920985e-05,
"loss": 0.0832,
"step": 9300
},
{
"epoch": 11.87,
"learning_rate": 4.3440054495912805e-05,
"loss": 0.0679,
"step": 9310
},
{
"epoch": 11.89,
"learning_rate": 4.337193460490463e-05,
"loss": 0.0839,
"step": 9320
},
{
"epoch": 11.9,
"learning_rate": 4.330381471389646e-05,
"loss": 0.1515,
"step": 9330
},
{
"epoch": 11.91,
"learning_rate": 4.323569482288829e-05,
"loss": 0.1111,
"step": 9340
},
{
"epoch": 11.93,
"learning_rate": 4.3167574931880114e-05,
"loss": 0.0872,
"step": 9350
},
{
"epoch": 11.94,
"learning_rate": 4.309945504087194e-05,
"loss": 0.0759,
"step": 9360
},
{
"epoch": 11.95,
"learning_rate": 4.303133514986376e-05,
"loss": 0.087,
"step": 9370
},
{
"epoch": 11.96,
"learning_rate": 4.296321525885559e-05,
"loss": 0.1595,
"step": 9380
},
{
"epoch": 11.98,
"learning_rate": 4.289509536784742e-05,
"loss": 0.0986,
"step": 9390
},
{
"epoch": 11.99,
"learning_rate": 4.282697547683924e-05,
"loss": 0.0768,
"step": 9400
},
{
"epoch": 12.0,
"learning_rate": 4.2758855585831065e-05,
"loss": 0.1419,
"step": 9410
},
{
"epoch": 12.02,
"learning_rate": 4.269073569482289e-05,
"loss": 0.1068,
"step": 9420
},
{
"epoch": 12.03,
"learning_rate": 4.262261580381471e-05,
"loss": 0.0839,
"step": 9430
},
{
"epoch": 12.04,
"learning_rate": 4.255449591280654e-05,
"loss": 0.0717,
"step": 9440
},
{
"epoch": 12.05,
"learning_rate": 4.248637602179837e-05,
"loss": 0.0673,
"step": 9450
},
{
"epoch": 12.07,
"learning_rate": 4.241825613079019e-05,
"loss": 0.1271,
"step": 9460
},
{
"epoch": 12.08,
"learning_rate": 4.2350136239782015e-05,
"loss": 0.116,
"step": 9470
},
{
"epoch": 12.09,
"learning_rate": 4.228201634877384e-05,
"loss": 0.0861,
"step": 9480
},
{
"epoch": 12.1,
"learning_rate": 4.221389645776567e-05,
"loss": 0.0669,
"step": 9490
},
{
"epoch": 12.12,
"learning_rate": 4.21457765667575e-05,
"loss": 0.0641,
"step": 9500
},
{
"epoch": 12.13,
"learning_rate": 4.2077656675749324e-05,
"loss": 0.1306,
"step": 9510
},
{
"epoch": 12.14,
"learning_rate": 4.200953678474115e-05,
"loss": 0.1095,
"step": 9520
},
{
"epoch": 12.16,
"learning_rate": 4.194141689373297e-05,
"loss": 0.0887,
"step": 9530
},
{
"epoch": 12.17,
"learning_rate": 4.18732970027248e-05,
"loss": 0.0654,
"step": 9540
},
{
"epoch": 12.18,
"learning_rate": 4.180517711171663e-05,
"loss": 0.078,
"step": 9550
},
{
"epoch": 12.19,
"learning_rate": 4.173705722070845e-05,
"loss": 0.1252,
"step": 9560
},
{
"epoch": 12.21,
"learning_rate": 4.1668937329700275e-05,
"loss": 0.1113,
"step": 9570
},
{
"epoch": 12.22,
"learning_rate": 4.16008174386921e-05,
"loss": 0.0859,
"step": 9580
},
{
"epoch": 12.23,
"learning_rate": 4.153269754768392e-05,
"loss": 0.0796,
"step": 9590
},
{
"epoch": 12.24,
"learning_rate": 4.146457765667575e-05,
"loss": 0.0775,
"step": 9600
},
{
"epoch": 12.24,
"eval_loss": 0.2894273102283478,
"eval_runtime": 1358.164,
"eval_samples_per_second": 4.589,
"eval_steps_per_second": 0.574,
"eval_wer": 0.3224186650374217,
"step": 9600
},
{
"epoch": 12.26,
"learning_rate": 4.139645776566758e-05,
"loss": 0.1213,
"step": 9610
},
{
"epoch": 12.27,
"learning_rate": 4.13283378746594e-05,
"loss": 0.1165,
"step": 9620
},
{
"epoch": 12.28,
"learning_rate": 4.1260217983651225e-05,
"loss": 0.0892,
"step": 9630
},
{
"epoch": 12.3,
"learning_rate": 4.119209809264305e-05,
"loss": 0.069,
"step": 9640
},
{
"epoch": 12.31,
"learning_rate": 4.112397820163488e-05,
"loss": 0.0692,
"step": 9650
},
{
"epoch": 12.32,
"learning_rate": 4.105585831062671e-05,
"loss": 0.121,
"step": 9660
},
{
"epoch": 12.33,
"learning_rate": 4.0987738419618534e-05,
"loss": 0.1186,
"step": 9670
},
{
"epoch": 12.35,
"learning_rate": 4.0919618528610355e-05,
"loss": 0.082,
"step": 9680
},
{
"epoch": 12.36,
"learning_rate": 4.085149863760218e-05,
"loss": 0.0721,
"step": 9690
},
{
"epoch": 12.37,
"learning_rate": 4.078337874659401e-05,
"loss": 0.0555,
"step": 9700
},
{
"epoch": 12.38,
"learning_rate": 4.071525885558583e-05,
"loss": 0.1086,
"step": 9710
},
{
"epoch": 12.4,
"learning_rate": 4.064713896457766e-05,
"loss": 0.1175,
"step": 9720
},
{
"epoch": 12.41,
"learning_rate": 4.0579019073569485e-05,
"loss": 0.0856,
"step": 9730
},
{
"epoch": 12.42,
"learning_rate": 4.0510899182561305e-05,
"loss": 0.0777,
"step": 9740
},
{
"epoch": 12.44,
"learning_rate": 4.044277929155313e-05,
"loss": 0.0705,
"step": 9750
},
{
"epoch": 12.45,
"learning_rate": 4.037465940054496e-05,
"loss": 0.1141,
"step": 9760
},
{
"epoch": 12.46,
"learning_rate": 4.030653950953679e-05,
"loss": 0.1149,
"step": 9770
},
{
"epoch": 12.47,
"learning_rate": 4.0238419618528614e-05,
"loss": 0.0973,
"step": 9780
},
{
"epoch": 12.49,
"learning_rate": 4.017029972752044e-05,
"loss": 0.071,
"step": 9790
},
{
"epoch": 12.5,
"learning_rate": 4.010217983651227e-05,
"loss": 0.0653,
"step": 9800
},
{
"epoch": 12.51,
"learning_rate": 4.003405994550409e-05,
"loss": 0.1146,
"step": 9810
},
{
"epoch": 12.53,
"learning_rate": 3.996594005449592e-05,
"loss": 0.1125,
"step": 9820
},
{
"epoch": 12.54,
"learning_rate": 3.9897820163487744e-05,
"loss": 0.0917,
"step": 9830
},
{
"epoch": 12.55,
"learning_rate": 3.9829700272479565e-05,
"loss": 0.0703,
"step": 9840
},
{
"epoch": 12.56,
"learning_rate": 3.976158038147139e-05,
"loss": 0.0643,
"step": 9850
},
{
"epoch": 12.58,
"learning_rate": 3.969346049046322e-05,
"loss": 0.1285,
"step": 9860
},
{
"epoch": 12.59,
"learning_rate": 3.962534059945504e-05,
"loss": 0.1159,
"step": 9870
},
{
"epoch": 12.6,
"learning_rate": 3.955722070844687e-05,
"loss": 0.0922,
"step": 9880
},
{
"epoch": 12.61,
"learning_rate": 3.9489100817438695e-05,
"loss": 0.0686,
"step": 9890
},
{
"epoch": 12.63,
"learning_rate": 3.9420980926430515e-05,
"loss": 0.0609,
"step": 9900
},
{
"epoch": 12.64,
"learning_rate": 3.935286103542234e-05,
"loss": 0.126,
"step": 9910
},
{
"epoch": 12.65,
"learning_rate": 3.928474114441417e-05,
"loss": 0.1126,
"step": 9920
},
{
"epoch": 12.67,
"learning_rate": 3.9216621253406e-05,
"loss": 0.0827,
"step": 9930
},
{
"epoch": 12.68,
"learning_rate": 3.9148501362397824e-05,
"loss": 0.0679,
"step": 9940
},
{
"epoch": 12.69,
"learning_rate": 3.908038147138965e-05,
"loss": 0.0673,
"step": 9950
},
{
"epoch": 12.7,
"learning_rate": 3.901226158038147e-05,
"loss": 0.1171,
"step": 9960
},
{
"epoch": 12.72,
"learning_rate": 3.89441416893733e-05,
"loss": 0.1119,
"step": 9970
},
{
"epoch": 12.73,
"learning_rate": 3.887602179836513e-05,
"loss": 0.0927,
"step": 9980
},
{
"epoch": 12.74,
"learning_rate": 3.8807901907356954e-05,
"loss": 0.0805,
"step": 9990
},
{
"epoch": 12.75,
"learning_rate": 3.8739782016348775e-05,
"loss": 0.0677,
"step": 10000
},
{
"epoch": 12.75,
"eval_loss": 0.28383344411849976,
"eval_runtime": 1361.9523,
"eval_samples_per_second": 4.577,
"eval_steps_per_second": 0.573,
"eval_wer": 0.32036619825874446,
"step": 10000
},
{
"epoch": 12.77,
"learning_rate": 3.86716621253406e-05,
"loss": 0.1114,
"step": 10010
},
{
"epoch": 12.78,
"learning_rate": 3.860354223433243e-05,
"loss": 0.1127,
"step": 10020
},
{
"epoch": 12.79,
"learning_rate": 3.853542234332425e-05,
"loss": 0.08,
"step": 10030
},
{
"epoch": 12.81,
"learning_rate": 3.846730245231608e-05,
"loss": 0.0703,
"step": 10040
},
{
"epoch": 12.82,
"learning_rate": 3.8399182561307905e-05,
"loss": 0.0738,
"step": 10050
},
{
"epoch": 12.83,
"learning_rate": 3.8331062670299725e-05,
"loss": 0.115,
"step": 10060
},
{
"epoch": 12.84,
"learning_rate": 3.826294277929155e-05,
"loss": 0.1193,
"step": 10070
},
{
"epoch": 12.86,
"learning_rate": 3.819482288828338e-05,
"loss": 0.0817,
"step": 10080
},
{
"epoch": 12.87,
"learning_rate": 3.812670299727521e-05,
"loss": 0.061,
"step": 10090
},
{
"epoch": 12.88,
"learning_rate": 3.8058583106267034e-05,
"loss": 0.0629,
"step": 10100
},
{
"epoch": 12.89,
"learning_rate": 3.799046321525886e-05,
"loss": 0.113,
"step": 10110
},
{
"epoch": 12.91,
"learning_rate": 3.792234332425068e-05,
"loss": 0.1108,
"step": 10120
},
{
"epoch": 12.92,
"learning_rate": 3.785422343324251e-05,
"loss": 0.0959,
"step": 10130
},
{
"epoch": 12.93,
"learning_rate": 3.778610354223434e-05,
"loss": 0.0624,
"step": 10140
},
{
"epoch": 12.95,
"learning_rate": 3.771798365122616e-05,
"loss": 0.0729,
"step": 10150
},
{
"epoch": 12.96,
"learning_rate": 3.7649863760217985e-05,
"loss": 0.118,
"step": 10160
},
{
"epoch": 12.97,
"learning_rate": 3.758174386920981e-05,
"loss": 0.1069,
"step": 10170
},
{
"epoch": 12.98,
"learning_rate": 3.751362397820163e-05,
"loss": 0.0815,
"step": 10180
},
{
"epoch": 13.0,
"learning_rate": 3.744550408719346e-05,
"loss": 0.0708,
"step": 10190
},
{
"epoch": 13.01,
"learning_rate": 3.737738419618529e-05,
"loss": 0.1372,
"step": 10200
},
{
"epoch": 13.02,
"learning_rate": 3.7309264305177115e-05,
"loss": 0.0833,
"step": 10210
},
{
"epoch": 13.04,
"learning_rate": 3.7241144414168935e-05,
"loss": 0.0669,
"step": 10220
},
{
"epoch": 13.05,
"learning_rate": 3.717302452316076e-05,
"loss": 0.0548,
"step": 10230
},
{
"epoch": 13.06,
"learning_rate": 3.710490463215259e-05,
"loss": 0.0677,
"step": 10240
},
{
"epoch": 13.07,
"learning_rate": 3.703678474114442e-05,
"loss": 0.1136,
"step": 10250
},
{
"epoch": 13.09,
"learning_rate": 3.6968664850136244e-05,
"loss": 0.087,
"step": 10260
},
{
"epoch": 13.1,
"learning_rate": 3.690054495912807e-05,
"loss": 0.0735,
"step": 10270
},
{
"epoch": 13.11,
"learning_rate": 3.683242506811989e-05,
"loss": 0.0486,
"step": 10280
},
{
"epoch": 13.12,
"learning_rate": 3.676430517711172e-05,
"loss": 0.0615,
"step": 10290
},
{
"epoch": 13.14,
"learning_rate": 3.669618528610355e-05,
"loss": 0.1125,
"step": 10300
},
{
"epoch": 13.15,
"learning_rate": 3.662806539509537e-05,
"loss": 0.0852,
"step": 10310
},
{
"epoch": 13.16,
"learning_rate": 3.6559945504087195e-05,
"loss": 0.0758,
"step": 10320
},
{
"epoch": 13.18,
"learning_rate": 3.649182561307902e-05,
"loss": 0.0538,
"step": 10330
},
{
"epoch": 13.19,
"learning_rate": 3.642370572207084e-05,
"loss": 0.0628,
"step": 10340
},
{
"epoch": 13.2,
"learning_rate": 3.635558583106267e-05,
"loss": 0.1344,
"step": 10350
},
{
"epoch": 13.21,
"learning_rate": 3.62874659400545e-05,
"loss": 0.0847,
"step": 10360
},
{
"epoch": 13.23,
"learning_rate": 3.621934604904632e-05,
"loss": 0.0728,
"step": 10370
},
{
"epoch": 13.24,
"learning_rate": 3.6151226158038145e-05,
"loss": 0.058,
"step": 10380
},
{
"epoch": 13.25,
"learning_rate": 3.608310626702997e-05,
"loss": 0.066,
"step": 10390
},
{
"epoch": 13.27,
"learning_rate": 3.60149863760218e-05,
"loss": 0.1383,
"step": 10400
},
{
"epoch": 13.27,
"eval_loss": 0.2958827614784241,
"eval_runtime": 1362.145,
"eval_samples_per_second": 4.576,
"eval_steps_per_second": 0.573,
"eval_wer": 0.3211299068275546,
"step": 10400
},
{
"epoch": 13.28,
"learning_rate": 3.594686648501363e-05,
"loss": 0.0874,
"step": 10410
},
{
"epoch": 13.29,
"learning_rate": 3.5878746594005454e-05,
"loss": 0.0677,
"step": 10420
},
{
"epoch": 13.3,
"learning_rate": 3.5810626702997275e-05,
"loss": 0.0481,
"step": 10430
},
{
"epoch": 13.32,
"learning_rate": 3.57425068119891e-05,
"loss": 0.0682,
"step": 10440
},
{
"epoch": 13.33,
"learning_rate": 3.567438692098093e-05,
"loss": 0.1395,
"step": 10450
},
{
"epoch": 13.34,
"learning_rate": 3.560626702997276e-05,
"loss": 0.0879,
"step": 10460
},
{
"epoch": 13.35,
"learning_rate": 3.553814713896458e-05,
"loss": 0.0668,
"step": 10470
},
{
"epoch": 13.37,
"learning_rate": 3.5470027247956405e-05,
"loss": 0.0565,
"step": 10480
},
{
"epoch": 13.38,
"learning_rate": 3.540190735694823e-05,
"loss": 0.076,
"step": 10490
},
{
"epoch": 13.39,
"learning_rate": 3.533378746594005e-05,
"loss": 0.1275,
"step": 10500
},
{
"epoch": 13.41,
"learning_rate": 3.526566757493188e-05,
"loss": 0.0807,
"step": 10510
},
{
"epoch": 13.42,
"learning_rate": 3.519754768392371e-05,
"loss": 0.0743,
"step": 10520
},
{
"epoch": 13.43,
"learning_rate": 3.5129427792915535e-05,
"loss": 0.0607,
"step": 10530
},
{
"epoch": 13.44,
"learning_rate": 3.506130790190736e-05,
"loss": 0.0709,
"step": 10540
},
{
"epoch": 13.46,
"learning_rate": 3.499318801089919e-05,
"loss": 0.1291,
"step": 10550
},
{
"epoch": 13.47,
"learning_rate": 3.492506811989101e-05,
"loss": 0.0856,
"step": 10560
},
{
"epoch": 13.48,
"learning_rate": 3.485694822888284e-05,
"loss": 0.068,
"step": 10570
},
{
"epoch": 13.49,
"learning_rate": 3.4788828337874664e-05,
"loss": 0.0611,
"step": 10580
},
{
"epoch": 13.51,
"learning_rate": 3.4720708446866485e-05,
"loss": 0.0812,
"step": 10590
},
{
"epoch": 13.52,
"learning_rate": 3.465258855585831e-05,
"loss": 0.1167,
"step": 10600
},
{
"epoch": 13.53,
"learning_rate": 3.458446866485014e-05,
"loss": 0.0918,
"step": 10610
},
{
"epoch": 13.55,
"learning_rate": 3.451634877384196e-05,
"loss": 0.0617,
"step": 10620
},
{
"epoch": 13.56,
"learning_rate": 3.444822888283379e-05,
"loss": 0.0641,
"step": 10630
},
{
"epoch": 13.57,
"learning_rate": 3.4380108991825615e-05,
"loss": 0.0783,
"step": 10640
},
{
"epoch": 13.58,
"learning_rate": 3.4311989100817435e-05,
"loss": 0.1344,
"step": 10650
},
{
"epoch": 13.6,
"learning_rate": 3.424386920980926e-05,
"loss": 0.0866,
"step": 10660
},
{
"epoch": 13.61,
"learning_rate": 3.417574931880109e-05,
"loss": 0.0709,
"step": 10670
},
{
"epoch": 13.62,
"learning_rate": 3.410762942779292e-05,
"loss": 0.0547,
"step": 10680
},
{
"epoch": 13.63,
"learning_rate": 3.4039509536784744e-05,
"loss": 0.0879,
"step": 10690
},
{
"epoch": 13.65,
"learning_rate": 3.397138964577657e-05,
"loss": 0.124,
"step": 10700
},
{
"epoch": 13.66,
"learning_rate": 3.39032697547684e-05,
"loss": 0.0929,
"step": 10710
},
{
"epoch": 13.67,
"learning_rate": 3.383514986376022e-05,
"loss": 0.0844,
"step": 10720
},
{
"epoch": 13.69,
"learning_rate": 3.376702997275205e-05,
"loss": 0.0602,
"step": 10730
},
{
"epoch": 13.7,
"learning_rate": 3.3698910081743874e-05,
"loss": 0.0663,
"step": 10740
},
{
"epoch": 13.71,
"learning_rate": 3.3630790190735695e-05,
"loss": 0.1273,
"step": 10750
},
{
"epoch": 13.72,
"learning_rate": 3.356267029972752e-05,
"loss": 0.0865,
"step": 10760
},
{
"epoch": 13.74,
"learning_rate": 3.349455040871935e-05,
"loss": 0.0716,
"step": 10770
},
{
"epoch": 13.75,
"learning_rate": 3.342643051771117e-05,
"loss": 0.0654,
"step": 10780
},
{
"epoch": 13.76,
"learning_rate": 3.3358310626703e-05,
"loss": 0.0767,
"step": 10790
},
{
"epoch": 13.77,
"learning_rate": 3.3290190735694825e-05,
"loss": 0.1233,
"step": 10800
},
{
"epoch": 13.77,
"eval_loss": 0.29217278957366943,
"eval_runtime": 1363.9978,
"eval_samples_per_second": 4.57,
"eval_steps_per_second": 0.572,
"eval_wer": 0.3212540094699863,
"step": 10800
},
{
"epoch": 13.79,
"learning_rate": 3.3222070844686645e-05,
"loss": 0.0851,
"step": 10810
},
{
"epoch": 13.8,
"learning_rate": 3.315395095367847e-05,
"loss": 0.0695,
"step": 10820
},
{
"epoch": 13.81,
"learning_rate": 3.30858310626703e-05,
"loss": 0.0589,
"step": 10830
},
{
"epoch": 13.83,
"learning_rate": 3.301771117166213e-05,
"loss": 0.0692,
"step": 10840
},
{
"epoch": 13.84,
"learning_rate": 3.2949591280653954e-05,
"loss": 0.1202,
"step": 10850
},
{
"epoch": 13.85,
"learning_rate": 3.288147138964578e-05,
"loss": 0.0794,
"step": 10860
},
{
"epoch": 13.86,
"learning_rate": 3.28133514986376e-05,
"loss": 0.0652,
"step": 10870
},
{
"epoch": 13.88,
"learning_rate": 3.274523160762943e-05,
"loss": 0.055,
"step": 10880
},
{
"epoch": 13.89,
"learning_rate": 3.267711171662126e-05,
"loss": 0.0576,
"step": 10890
},
{
"epoch": 13.9,
"learning_rate": 3.2608991825613084e-05,
"loss": 0.1079,
"step": 10900
},
{
"epoch": 13.92,
"learning_rate": 3.2540871934604905e-05,
"loss": 0.0861,
"step": 10910
},
{
"epoch": 13.93,
"learning_rate": 3.247275204359673e-05,
"loss": 0.0703,
"step": 10920
},
{
"epoch": 13.94,
"learning_rate": 3.240463215258856e-05,
"loss": 0.052,
"step": 10930
},
{
"epoch": 13.95,
"learning_rate": 3.233651226158038e-05,
"loss": 0.0741,
"step": 10940
},
{
"epoch": 13.97,
"learning_rate": 3.226839237057221e-05,
"loss": 0.1237,
"step": 10950
},
{
"epoch": 13.98,
"learning_rate": 3.2200272479564035e-05,
"loss": 0.0753,
"step": 10960
},
{
"epoch": 13.99,
"learning_rate": 3.2132152588555855e-05,
"loss": 0.0648,
"step": 10970
},
{
"epoch": 14.01,
"learning_rate": 3.206403269754768e-05,
"loss": 0.1136,
"step": 10980
},
{
"epoch": 14.02,
"learning_rate": 3.199591280653951e-05,
"loss": 0.0857,
"step": 10990
},
{
"epoch": 14.03,
"learning_rate": 3.192779291553134e-05,
"loss": 0.0645,
"step": 11000
},
{
"epoch": 14.04,
"learning_rate": 3.1859673024523164e-05,
"loss": 0.0549,
"step": 11010
},
{
"epoch": 14.06,
"learning_rate": 3.179155313351499e-05,
"loss": 0.0598,
"step": 11020
},
{
"epoch": 14.07,
"learning_rate": 3.172343324250681e-05,
"loss": 0.1101,
"step": 11030
},
{
"epoch": 14.08,
"learning_rate": 3.165531335149864e-05,
"loss": 0.0832,
"step": 11040
},
{
"epoch": 14.09,
"learning_rate": 3.158719346049047e-05,
"loss": 0.066,
"step": 11050
},
{
"epoch": 14.11,
"learning_rate": 3.151907356948229e-05,
"loss": 0.0602,
"step": 11060
},
{
"epoch": 14.12,
"learning_rate": 3.1450953678474115e-05,
"loss": 0.0611,
"step": 11070
},
{
"epoch": 14.13,
"learning_rate": 3.138283378746594e-05,
"loss": 0.0953,
"step": 11080
},
{
"epoch": 14.15,
"learning_rate": 3.131471389645776e-05,
"loss": 0.0902,
"step": 11090
},
{
"epoch": 14.16,
"learning_rate": 3.124659400544959e-05,
"loss": 0.0605,
"step": 11100
},
{
"epoch": 14.17,
"learning_rate": 3.117847411444142e-05,
"loss": 0.0536,
"step": 11110
},
{
"epoch": 14.18,
"learning_rate": 3.1110354223433245e-05,
"loss": 0.0545,
"step": 11120
},
{
"epoch": 14.2,
"learning_rate": 3.104223433242507e-05,
"loss": 0.0958,
"step": 11130
},
{
"epoch": 14.21,
"learning_rate": 3.09741144414169e-05,
"loss": 0.0857,
"step": 11140
},
{
"epoch": 14.22,
"learning_rate": 3.0905994550408727e-05,
"loss": 0.0664,
"step": 11150
},
{
"epoch": 14.23,
"learning_rate": 3.083787465940055e-05,
"loss": 0.0511,
"step": 11160
},
{
"epoch": 14.25,
"learning_rate": 3.0769754768392374e-05,
"loss": 0.0616,
"step": 11170
},
{
"epoch": 14.26,
"learning_rate": 3.07016348773842e-05,
"loss": 0.1027,
"step": 11180
},
{
"epoch": 14.27,
"learning_rate": 3.063351498637602e-05,
"loss": 0.0851,
"step": 11190
},
{
"epoch": 14.29,
"learning_rate": 3.056539509536785e-05,
"loss": 0.0688,
"step": 11200
},
{
"epoch": 14.29,
"eval_loss": 0.2903039753437042,
"eval_runtime": 1371.1539,
"eval_samples_per_second": 4.546,
"eval_steps_per_second": 0.569,
"eval_wer": 0.32086260882847106,
"step": 11200
},
{
"epoch": 14.3,
"learning_rate": 3.0497275204359677e-05,
"loss": 0.0593,
"step": 11210
},
{
"epoch": 14.31,
"learning_rate": 3.0429155313351497e-05,
"loss": 0.0553,
"step": 11220
},
{
"epoch": 14.32,
"learning_rate": 3.0361035422343325e-05,
"loss": 0.0969,
"step": 11230
},
{
"epoch": 14.34,
"learning_rate": 3.0292915531335152e-05,
"loss": 0.0858,
"step": 11240
},
{
"epoch": 14.35,
"learning_rate": 3.0224795640326976e-05,
"loss": 0.0708,
"step": 11250
},
{
"epoch": 14.36,
"learning_rate": 3.0156675749318803e-05,
"loss": 0.0539,
"step": 11260
},
{
"epoch": 14.37,
"learning_rate": 3.008855585831063e-05,
"loss": 0.0664,
"step": 11270
},
{
"epoch": 14.39,
"learning_rate": 3.002043596730245e-05,
"loss": 0.0867,
"step": 11280
},
{
"epoch": 14.4,
"learning_rate": 2.995231607629428e-05,
"loss": 0.0821,
"step": 11290
},
{
"epoch": 14.41,
"learning_rate": 2.9884196185286106e-05,
"loss": 0.069,
"step": 11300
},
{
"epoch": 14.43,
"learning_rate": 2.981607629427793e-05,
"loss": 0.0555,
"step": 11310
},
{
"epoch": 14.44,
"learning_rate": 2.9747956403269757e-05,
"loss": 0.0501,
"step": 11320
},
{
"epoch": 14.45,
"learning_rate": 2.9679836512261584e-05,
"loss": 0.0969,
"step": 11330
},
{
"epoch": 14.46,
"learning_rate": 2.9611716621253405e-05,
"loss": 0.0821,
"step": 11340
},
{
"epoch": 14.48,
"learning_rate": 2.9543596730245232e-05,
"loss": 0.0683,
"step": 11350
},
{
"epoch": 14.49,
"learning_rate": 2.947547683923706e-05,
"loss": 0.0515,
"step": 11360
},
{
"epoch": 14.5,
"learning_rate": 2.9407356948228887e-05,
"loss": 0.0518,
"step": 11370
},
{
"epoch": 14.51,
"learning_rate": 2.9339237057220707e-05,
"loss": 0.0908,
"step": 11380
},
{
"epoch": 14.53,
"learning_rate": 2.9271117166212535e-05,
"loss": 0.0841,
"step": 11390
},
{
"epoch": 14.54,
"learning_rate": 2.9202997275204362e-05,
"loss": 0.0628,
"step": 11400
},
{
"epoch": 14.55,
"learning_rate": 2.9134877384196186e-05,
"loss": 0.0517,
"step": 11410
},
{
"epoch": 14.57,
"learning_rate": 2.9066757493188013e-05,
"loss": 0.055,
"step": 11420
},
{
"epoch": 14.58,
"learning_rate": 2.899863760217984e-05,
"loss": 0.0904,
"step": 11430
},
{
"epoch": 14.59,
"learning_rate": 2.893051771117166e-05,
"loss": 0.0848,
"step": 11440
},
{
"epoch": 14.6,
"learning_rate": 2.886239782016349e-05,
"loss": 0.0694,
"step": 11450
},
{
"epoch": 14.62,
"learning_rate": 2.8794277929155316e-05,
"loss": 0.0515,
"step": 11460
},
{
"epoch": 14.63,
"learning_rate": 2.872615803814714e-05,
"loss": 0.0622,
"step": 11470
},
{
"epoch": 14.64,
"learning_rate": 2.8658038147138967e-05,
"loss": 0.1064,
"step": 11480
},
{
"epoch": 14.66,
"learning_rate": 2.8589918256130794e-05,
"loss": 0.0824,
"step": 11490
},
{
"epoch": 14.67,
"learning_rate": 2.8521798365122615e-05,
"loss": 0.0665,
"step": 11500
},
{
"epoch": 14.68,
"learning_rate": 2.8453678474114442e-05,
"loss": 0.0541,
"step": 11510
},
{
"epoch": 14.69,
"learning_rate": 2.838555858310627e-05,
"loss": 0.0654,
"step": 11520
},
{
"epoch": 14.71,
"learning_rate": 2.831743869209809e-05,
"loss": 0.1008,
"step": 11530
},
{
"epoch": 14.72,
"learning_rate": 2.8249318801089917e-05,
"loss": 0.0819,
"step": 11540
},
{
"epoch": 14.73,
"learning_rate": 2.8181198910081745e-05,
"loss": 0.0713,
"step": 11550
},
{
"epoch": 14.74,
"learning_rate": 2.811307901907357e-05,
"loss": 0.057,
"step": 11560
},
{
"epoch": 14.76,
"learning_rate": 2.8044959128065396e-05,
"loss": 0.0629,
"step": 11570
},
{
"epoch": 14.77,
"learning_rate": 2.7976839237057223e-05,
"loss": 0.1028,
"step": 11580
},
{
"epoch": 14.78,
"learning_rate": 2.790871934604905e-05,
"loss": 0.0902,
"step": 11590
},
{
"epoch": 14.8,
"learning_rate": 2.784059945504087e-05,
"loss": 0.0655,
"step": 11600
},
{
"epoch": 14.8,
"eval_loss": 0.2868470549583435,
"eval_runtime": 1425.5906,
"eval_samples_per_second": 4.372,
"eval_steps_per_second": 0.547,
"eval_wer": 0.3182087215518558,
"step": 11600
},
{
"epoch": 14.81,
"learning_rate": 2.77724795640327e-05,
"loss": 0.0506,
"step": 11610
},
{
"epoch": 14.82,
"learning_rate": 2.7704359673024526e-05,
"loss": 0.0544,
"step": 11620
},
{
"epoch": 14.83,
"learning_rate": 2.763623978201635e-05,
"loss": 0.0956,
"step": 11630
},
{
"epoch": 14.85,
"learning_rate": 2.7568119891008177e-05,
"loss": 0.0852,
"step": 11640
},
{
"epoch": 14.86,
"learning_rate": 2.7500000000000004e-05,
"loss": 0.0614,
"step": 11650
},
{
"epoch": 14.87,
"learning_rate": 2.7431880108991825e-05,
"loss": 0.0484,
"step": 11660
},
{
"epoch": 14.88,
"learning_rate": 2.7363760217983652e-05,
"loss": 0.0523,
"step": 11670
},
{
"epoch": 14.9,
"learning_rate": 2.729564032697548e-05,
"loss": 0.1058,
"step": 11680
},
{
"epoch": 14.91,
"learning_rate": 2.7227520435967303e-05,
"loss": 0.0824,
"step": 11690
},
{
"epoch": 14.92,
"learning_rate": 2.715940054495913e-05,
"loss": 0.0649,
"step": 11700
},
{
"epoch": 14.94,
"learning_rate": 2.7091280653950958e-05,
"loss": 0.054,
"step": 11710
},
{
"epoch": 14.95,
"learning_rate": 2.702316076294278e-05,
"loss": 0.0536,
"step": 11720
},
{
"epoch": 14.96,
"learning_rate": 2.6955040871934606e-05,
"loss": 0.1042,
"step": 11730
},
{
"epoch": 14.97,
"learning_rate": 2.6886920980926433e-05,
"loss": 0.0756,
"step": 11740
},
{
"epoch": 14.99,
"learning_rate": 2.6818801089918254e-05,
"loss": 0.0554,
"step": 11750
},
{
"epoch": 15.0,
"learning_rate": 2.675068119891008e-05,
"loss": 0.0682,
"step": 11760
},
{
"epoch": 15.01,
"learning_rate": 2.668256130790191e-05,
"loss": 0.093,
"step": 11770
},
{
"epoch": 15.03,
"learning_rate": 2.6614441416893732e-05,
"loss": 0.0687,
"step": 11780
},
{
"epoch": 15.04,
"learning_rate": 2.654632152588556e-05,
"loss": 0.0511,
"step": 11790
},
{
"epoch": 15.05,
"learning_rate": 2.6478201634877387e-05,
"loss": 0.0451,
"step": 11800
},
{
"epoch": 15.06,
"learning_rate": 2.6410081743869208e-05,
"loss": 0.0571,
"step": 11810
},
{
"epoch": 15.08,
"learning_rate": 2.6341961852861035e-05,
"loss": 0.0889,
"step": 11820
},
{
"epoch": 15.09,
"learning_rate": 2.6273841961852862e-05,
"loss": 0.0667,
"step": 11830
},
{
"epoch": 15.1,
"learning_rate": 2.620572207084469e-05,
"loss": 0.0586,
"step": 11840
},
{
"epoch": 15.11,
"learning_rate": 2.6137602179836513e-05,
"loss": 0.0461,
"step": 11850
},
{
"epoch": 15.13,
"learning_rate": 2.606948228882834e-05,
"loss": 0.0781,
"step": 11860
},
{
"epoch": 15.14,
"learning_rate": 2.6001362397820168e-05,
"loss": 0.089,
"step": 11870
},
{
"epoch": 15.15,
"learning_rate": 2.593324250681199e-05,
"loss": 0.0619,
"step": 11880
},
{
"epoch": 15.17,
"learning_rate": 2.5865122615803816e-05,
"loss": 0.0508,
"step": 11890
},
{
"epoch": 15.18,
"learning_rate": 2.5797002724795643e-05,
"loss": 0.0482,
"step": 11900
},
{
"epoch": 15.19,
"learning_rate": 2.5728882833787464e-05,
"loss": 0.079,
"step": 11910
},
{
"epoch": 15.2,
"learning_rate": 2.5660762942779295e-05,
"loss": 0.0918,
"step": 11920
},
{
"epoch": 15.22,
"learning_rate": 2.5592643051771122e-05,
"loss": 0.0626,
"step": 11930
},
{
"epoch": 15.23,
"learning_rate": 2.5524523160762942e-05,
"loss": 0.0569,
"step": 11940
},
{
"epoch": 15.24,
"learning_rate": 2.545640326975477e-05,
"loss": 0.0482,
"step": 11950
},
{
"epoch": 15.25,
"learning_rate": 2.5388283378746597e-05,
"loss": 0.0682,
"step": 11960
},
{
"epoch": 15.27,
"learning_rate": 2.5320163487738418e-05,
"loss": 0.0965,
"step": 11970
},
{
"epoch": 15.28,
"learning_rate": 2.5252043596730245e-05,
"loss": 0.0629,
"step": 11980
},
{
"epoch": 15.29,
"learning_rate": 2.5183923705722072e-05,
"loss": 0.0557,
"step": 11990
},
{
"epoch": 15.31,
"learning_rate": 2.5115803814713896e-05,
"loss": 0.0449,
"step": 12000
},
{
"epoch": 15.31,
"eval_loss": 0.29593759775161743,
"eval_runtime": 1364.8447,
"eval_samples_per_second": 4.567,
"eval_steps_per_second": 0.571,
"eval_wer": 0.31715862226974184,
"step": 12000
},
{
"epoch": 15.32,
"learning_rate": 2.5047683923705723e-05,
"loss": 0.0664,
"step": 12010
},
{
"epoch": 15.33,
"learning_rate": 2.497956403269755e-05,
"loss": 0.088,
"step": 12020
},
{
"epoch": 15.34,
"learning_rate": 2.4911444141689375e-05,
"loss": 0.0641,
"step": 12030
},
{
"epoch": 15.36,
"learning_rate": 2.48433242506812e-05,
"loss": 0.0518,
"step": 12040
},
{
"epoch": 15.37,
"learning_rate": 2.4775204359673026e-05,
"loss": 0.0465,
"step": 12050
},
{
"epoch": 15.38,
"learning_rate": 2.470708446866485e-05,
"loss": 0.0814,
"step": 12060
},
{
"epoch": 15.4,
"learning_rate": 2.4638964577656677e-05,
"loss": 0.098,
"step": 12070
},
{
"epoch": 15.41,
"learning_rate": 2.4570844686648504e-05,
"loss": 0.0637,
"step": 12080
},
{
"epoch": 15.42,
"learning_rate": 2.450272479564033e-05,
"loss": 0.0536,
"step": 12090
},
{
"epoch": 15.43,
"learning_rate": 2.4434604904632156e-05,
"loss": 0.046,
"step": 12100
},
{
"epoch": 15.45,
"learning_rate": 2.436648501362398e-05,
"loss": 0.0679,
"step": 12110
},
{
"epoch": 15.46,
"learning_rate": 2.4298365122615804e-05,
"loss": 0.0886,
"step": 12120
},
{
"epoch": 15.47,
"learning_rate": 2.423024523160763e-05,
"loss": 0.061,
"step": 12130
},
{
"epoch": 15.48,
"learning_rate": 2.4162125340599455e-05,
"loss": 0.0512,
"step": 12140
},
{
"epoch": 15.5,
"learning_rate": 2.4094005449591282e-05,
"loss": 0.0457,
"step": 12150
},
{
"epoch": 15.51,
"learning_rate": 2.402588555858311e-05,
"loss": 0.0706,
"step": 12160
},
{
"epoch": 15.52,
"learning_rate": 2.3957765667574933e-05,
"loss": 0.0955,
"step": 12170
},
{
"epoch": 15.54,
"learning_rate": 2.3889645776566757e-05,
"loss": 0.0637,
"step": 12180
},
{
"epoch": 15.55,
"learning_rate": 2.3821525885558585e-05,
"loss": 0.05,
"step": 12190
},
{
"epoch": 15.56,
"learning_rate": 2.375340599455041e-05,
"loss": 0.0466,
"step": 12200
},
{
"epoch": 15.57,
"learning_rate": 2.3685286103542233e-05,
"loss": 0.0742,
"step": 12210
},
{
"epoch": 15.59,
"learning_rate": 2.361716621253406e-05,
"loss": 0.0933,
"step": 12220
},
{
"epoch": 15.6,
"learning_rate": 2.3549046321525887e-05,
"loss": 0.0702,
"step": 12230
},
{
"epoch": 15.61,
"learning_rate": 2.3480926430517714e-05,
"loss": 0.0646,
"step": 12240
},
{
"epoch": 15.62,
"learning_rate": 2.341280653950954e-05,
"loss": 0.0508,
"step": 12250
},
{
"epoch": 15.64,
"learning_rate": 2.3344686648501362e-05,
"loss": 0.065,
"step": 12260
},
{
"epoch": 15.65,
"learning_rate": 2.327656675749319e-05,
"loss": 0.0909,
"step": 12270
},
{
"epoch": 15.66,
"learning_rate": 2.3208446866485014e-05,
"loss": 0.0707,
"step": 12280
},
{
"epoch": 15.68,
"learning_rate": 2.314032697547684e-05,
"loss": 0.0536,
"step": 12290
},
{
"epoch": 15.69,
"learning_rate": 2.3072207084468668e-05,
"loss": 0.0485,
"step": 12300
},
{
"epoch": 15.7,
"learning_rate": 2.3004087193460492e-05,
"loss": 0.0644,
"step": 12310
},
{
"epoch": 15.71,
"learning_rate": 2.2935967302452316e-05,
"loss": 0.0896,
"step": 12320
},
{
"epoch": 15.73,
"learning_rate": 2.2867847411444143e-05,
"loss": 0.0724,
"step": 12330
},
{
"epoch": 15.74,
"learning_rate": 2.2799727520435967e-05,
"loss": 0.0558,
"step": 12340
},
{
"epoch": 15.75,
"learning_rate": 2.2731607629427795e-05,
"loss": 0.0443,
"step": 12350
},
{
"epoch": 15.76,
"learning_rate": 2.266348773841962e-05,
"loss": 0.0647,
"step": 12360
},
{
"epoch": 15.78,
"learning_rate": 2.2595367847411446e-05,
"loss": 0.0854,
"step": 12370
},
{
"epoch": 15.79,
"learning_rate": 2.2527247956403273e-05,
"loss": 0.07,
"step": 12380
},
{
"epoch": 15.8,
"learning_rate": 2.2459128065395097e-05,
"loss": 0.0533,
"step": 12390
},
{
"epoch": 15.82,
"learning_rate": 2.239100817438692e-05,
"loss": 0.0421,
"step": 12400
},
{
"epoch": 15.82,
"eval_loss": 0.29661157727241516,
"eval_runtime": 1360.7234,
"eval_samples_per_second": 4.581,
"eval_steps_per_second": 0.573,
"eval_wer": 0.31797006262410266,
"step": 12400
},
{
"epoch": 15.83,
"learning_rate": 2.232288828337875e-05,
"loss": 0.0668,
"step": 12410
},
{
"epoch": 15.84,
"learning_rate": 2.2254768392370572e-05,
"loss": 0.0895,
"step": 12420
},
{
"epoch": 15.85,
"learning_rate": 2.2186648501362396e-05,
"loss": 0.0717,
"step": 12430
},
{
"epoch": 15.87,
"learning_rate": 2.2118528610354224e-05,
"loss": 0.0525,
"step": 12440
},
{
"epoch": 15.88,
"learning_rate": 2.205040871934605e-05,
"loss": 0.0492,
"step": 12450
},
{
"epoch": 15.89,
"learning_rate": 2.1982288828337878e-05,
"loss": 0.0814,
"step": 12460
},
{
"epoch": 15.9,
"learning_rate": 2.1914168937329702e-05,
"loss": 0.0914,
"step": 12470
},
{
"epoch": 15.92,
"learning_rate": 2.1846049046321526e-05,
"loss": 0.0625,
"step": 12480
},
{
"epoch": 15.93,
"learning_rate": 2.1777929155313353e-05,
"loss": 0.0505,
"step": 12490
},
{
"epoch": 15.94,
"learning_rate": 2.1709809264305177e-05,
"loss": 0.0409,
"step": 12500
},
{
"epoch": 15.96,
"learning_rate": 2.1641689373297e-05,
"loss": 0.0759,
"step": 12510
},
{
"epoch": 15.97,
"learning_rate": 2.157356948228883e-05,
"loss": 0.0873,
"step": 12520
},
{
"epoch": 15.98,
"learning_rate": 2.1505449591280656e-05,
"loss": 0.0574,
"step": 12530
},
{
"epoch": 15.99,
"learning_rate": 2.143732970027248e-05,
"loss": 0.0468,
"step": 12540
},
{
"epoch": 16.01,
"learning_rate": 2.1369209809264307e-05,
"loss": 0.0902,
"step": 12550
},
{
"epoch": 16.02,
"learning_rate": 2.130108991825613e-05,
"loss": 0.0667,
"step": 12560
},
{
"epoch": 16.03,
"learning_rate": 2.123297002724796e-05,
"loss": 0.0488,
"step": 12570
},
{
"epoch": 16.05,
"learning_rate": 2.1164850136239782e-05,
"loss": 0.0498,
"step": 12580
},
{
"epoch": 16.06,
"learning_rate": 2.1096730245231606e-05,
"loss": 0.0525,
"step": 12590
},
{
"epoch": 16.07,
"learning_rate": 2.1028610354223437e-05,
"loss": 0.0809,
"step": 12600
},
{
"epoch": 16.08,
"learning_rate": 2.096049046321526e-05,
"loss": 0.0687,
"step": 12610
},
{
"epoch": 16.1,
"learning_rate": 2.0892370572207085e-05,
"loss": 0.0535,
"step": 12620
},
{
"epoch": 16.11,
"learning_rate": 2.0824250681198912e-05,
"loss": 0.0404,
"step": 12630
},
{
"epoch": 16.12,
"learning_rate": 2.0756130790190736e-05,
"loss": 0.0473,
"step": 12640
},
{
"epoch": 16.14,
"learning_rate": 2.068801089918256e-05,
"loss": 0.0861,
"step": 12650
},
{
"epoch": 16.15,
"learning_rate": 2.0619891008174387e-05,
"loss": 0.0626,
"step": 12660
},
{
"epoch": 16.16,
"learning_rate": 2.0551771117166215e-05,
"loss": 0.0545,
"step": 12670
},
{
"epoch": 16.17,
"learning_rate": 2.0483651226158042e-05,
"loss": 0.0408,
"step": 12680
},
{
"epoch": 16.19,
"learning_rate": 2.0415531335149866e-05,
"loss": 0.0507,
"step": 12690
},
{
"epoch": 16.2,
"learning_rate": 2.034741144414169e-05,
"loss": 0.0905,
"step": 12700
},
{
"epoch": 16.21,
"learning_rate": 2.0279291553133517e-05,
"loss": 0.0666,
"step": 12710
},
{
"epoch": 16.22,
"learning_rate": 2.021117166212534e-05,
"loss": 0.0541,
"step": 12720
},
{
"epoch": 16.24,
"learning_rate": 2.0143051771117165e-05,
"loss": 0.0542,
"step": 12730
},
{
"epoch": 16.25,
"learning_rate": 2.0074931880108992e-05,
"loss": 0.053,
"step": 12740
},
{
"epoch": 16.26,
"learning_rate": 2.000681198910082e-05,
"loss": 0.0833,
"step": 12750
},
{
"epoch": 16.28,
"learning_rate": 1.9938692098092644e-05,
"loss": 0.0669,
"step": 12760
},
{
"epoch": 16.29,
"learning_rate": 1.987057220708447e-05,
"loss": 0.0485,
"step": 12770
},
{
"epoch": 16.3,
"learning_rate": 1.9802452316076295e-05,
"loss": 0.0424,
"step": 12780
},
{
"epoch": 16.31,
"learning_rate": 1.9734332425068122e-05,
"loss": 0.0491,
"step": 12790
},
{
"epoch": 16.33,
"learning_rate": 1.9666212534059946e-05,
"loss": 0.0858,
"step": 12800
},
{
"epoch": 16.33,
"eval_loss": 0.29408928751945496,
"eval_runtime": 1425.5229,
"eval_samples_per_second": 4.372,
"eval_steps_per_second": 0.547,
"eval_wer": 0.3163853673438216,
"step": 12800
},
{
"epoch": 16.34,
"learning_rate": 1.959809264305177e-05,
"loss": 0.0637,
"step": 12810
},
{
"epoch": 16.35,
"learning_rate": 1.9529972752043597e-05,
"loss": 0.0578,
"step": 12820
},
{
"epoch": 16.36,
"learning_rate": 1.9461852861035425e-05,
"loss": 0.0447,
"step": 12830
},
{
"epoch": 16.38,
"learning_rate": 1.939373297002725e-05,
"loss": 0.0543,
"step": 12840
},
{
"epoch": 16.39,
"learning_rate": 1.9325613079019076e-05,
"loss": 0.0896,
"step": 12850
},
{
"epoch": 16.4,
"learning_rate": 1.92574931880109e-05,
"loss": 0.0639,
"step": 12860
},
{
"epoch": 16.42,
"learning_rate": 1.9189373297002724e-05,
"loss": 0.0601,
"step": 12870
},
{
"epoch": 16.43,
"learning_rate": 1.912125340599455e-05,
"loss": 0.044,
"step": 12880
},
{
"epoch": 16.44,
"learning_rate": 1.9053133514986375e-05,
"loss": 0.0538,
"step": 12890
},
{
"epoch": 16.45,
"learning_rate": 1.8985013623978202e-05,
"loss": 0.0752,
"step": 12900
},
{
"epoch": 16.47,
"learning_rate": 1.891689373297003e-05,
"loss": 0.0674,
"step": 12910
},
{
"epoch": 16.48,
"learning_rate": 1.8848773841961853e-05,
"loss": 0.0531,
"step": 12920
},
{
"epoch": 16.49,
"learning_rate": 1.878065395095368e-05,
"loss": 0.0451,
"step": 12930
},
{
"epoch": 16.5,
"learning_rate": 1.8712534059945505e-05,
"loss": 0.045,
"step": 12940
},
{
"epoch": 16.52,
"learning_rate": 1.864441416893733e-05,
"loss": 0.092,
"step": 12950
},
{
"epoch": 16.53,
"learning_rate": 1.8576294277929156e-05,
"loss": 0.0682,
"step": 12960
},
{
"epoch": 16.54,
"learning_rate": 1.8508174386920983e-05,
"loss": 0.0575,
"step": 12970
},
{
"epoch": 16.56,
"learning_rate": 1.8440054495912807e-05,
"loss": 0.0483,
"step": 12980
},
{
"epoch": 16.57,
"learning_rate": 1.8371934604904635e-05,
"loss": 0.0499,
"step": 12990
},
{
"epoch": 16.58,
"learning_rate": 1.830381471389646e-05,
"loss": 0.0786,
"step": 13000
},
{
"epoch": 16.59,
"learning_rate": 1.8235694822888282e-05,
"loss": 0.0647,
"step": 13010
},
{
"epoch": 16.61,
"learning_rate": 1.816757493188011e-05,
"loss": 0.0483,
"step": 13020
},
{
"epoch": 16.62,
"learning_rate": 1.8099455040871934e-05,
"loss": 0.0445,
"step": 13030
},
{
"epoch": 16.63,
"learning_rate": 1.803133514986376e-05,
"loss": 0.0482,
"step": 13040
},
{
"epoch": 16.64,
"learning_rate": 1.7963215258855588e-05,
"loss": 0.0789,
"step": 13050
},
{
"epoch": 16.66,
"learning_rate": 1.7895095367847412e-05,
"loss": 0.0685,
"step": 13060
},
{
"epoch": 16.67,
"learning_rate": 1.782697547683924e-05,
"loss": 0.053,
"step": 13070
},
{
"epoch": 16.68,
"learning_rate": 1.7758855585831063e-05,
"loss": 0.0422,
"step": 13080
},
{
"epoch": 16.7,
"learning_rate": 1.7690735694822887e-05,
"loss": 0.0411,
"step": 13090
},
{
"epoch": 16.71,
"learning_rate": 1.7622615803814715e-05,
"loss": 0.0724,
"step": 13100
},
{
"epoch": 16.72,
"learning_rate": 1.755449591280654e-05,
"loss": 0.0682,
"step": 13110
},
{
"epoch": 16.73,
"learning_rate": 1.7486376021798366e-05,
"loss": 0.0536,
"step": 13120
},
{
"epoch": 16.75,
"learning_rate": 1.7418256130790193e-05,
"loss": 0.0414,
"step": 13130
},
{
"epoch": 16.76,
"learning_rate": 1.7350136239782017e-05,
"loss": 0.0471,
"step": 13140
},
{
"epoch": 16.77,
"learning_rate": 1.7282016348773845e-05,
"loss": 0.0891,
"step": 13150
},
{
"epoch": 16.79,
"learning_rate": 1.721389645776567e-05,
"loss": 0.0565,
"step": 13160
},
{
"epoch": 16.8,
"learning_rate": 1.7145776566757492e-05,
"loss": 0.0498,
"step": 13170
},
{
"epoch": 16.81,
"learning_rate": 1.707765667574932e-05,
"loss": 0.0436,
"step": 13180
},
{
"epoch": 16.82,
"learning_rate": 1.7009536784741144e-05,
"loss": 0.0496,
"step": 13190
},
{
"epoch": 16.84,
"learning_rate": 1.694141689373297e-05,
"loss": 0.0859,
"step": 13200
},
{
"epoch": 16.84,
"eval_loss": 0.29795339703559875,
"eval_runtime": 1359.0826,
"eval_samples_per_second": 4.586,
"eval_steps_per_second": 0.574,
"eval_wer": 0.31654765541469376,
"step": 13200
},
{
"epoch": 16.85,
"learning_rate": 1.6873297002724798e-05,
"loss": 0.0628,
"step": 13210
},
{
"epoch": 16.86,
"learning_rate": 1.6805177111716622e-05,
"loss": 0.0508,
"step": 13220
},
{
"epoch": 16.87,
"learning_rate": 1.6737057220708446e-05,
"loss": 0.0443,
"step": 13230
},
{
"epoch": 16.89,
"learning_rate": 1.6668937329700273e-05,
"loss": 0.0518,
"step": 13240
},
{
"epoch": 16.9,
"learning_rate": 1.6600817438692097e-05,
"loss": 0.1028,
"step": 13250
},
{
"epoch": 16.91,
"learning_rate": 1.6532697547683925e-05,
"loss": 0.0613,
"step": 13260
},
{
"epoch": 16.93,
"learning_rate": 1.6464577656675752e-05,
"loss": 0.0516,
"step": 13270
},
{
"epoch": 16.94,
"learning_rate": 1.6396457765667576e-05,
"loss": 0.0388,
"step": 13280
},
{
"epoch": 16.95,
"learning_rate": 1.6328337874659403e-05,
"loss": 0.0495,
"step": 13290
},
{
"epoch": 16.96,
"learning_rate": 1.6260217983651227e-05,
"loss": 0.092,
"step": 13300
},
{
"epoch": 16.98,
"learning_rate": 1.619209809264305e-05,
"loss": 0.0585,
"step": 13310
},
{
"epoch": 16.99,
"learning_rate": 1.612397820163488e-05,
"loss": 0.0389,
"step": 13320
},
{
"epoch": 17.0,
"learning_rate": 1.6055858310626702e-05,
"loss": 0.0595,
"step": 13330
},
{
"epoch": 17.02,
"learning_rate": 1.598773841961853e-05,
"loss": 0.0681,
"step": 13340
},
{
"epoch": 17.03,
"learning_rate": 1.5919618528610357e-05,
"loss": 0.0566,
"step": 13350
},
{
"epoch": 17.04,
"learning_rate": 1.585149863760218e-05,
"loss": 0.0452,
"step": 13360
},
{
"epoch": 17.05,
"learning_rate": 1.5783378746594008e-05,
"loss": 0.0345,
"step": 13370
},
{
"epoch": 17.07,
"learning_rate": 1.5715258855585832e-05,
"loss": 0.0683,
"step": 13380
},
{
"epoch": 17.08,
"learning_rate": 1.5647138964577656e-05,
"loss": 0.0654,
"step": 13390
},
{
"epoch": 17.09,
"learning_rate": 1.5579019073569483e-05,
"loss": 0.0554,
"step": 13400
},
{
"epoch": 17.1,
"learning_rate": 1.5510899182561307e-05,
"loss": 0.0459,
"step": 13410
},
{
"epoch": 17.12,
"learning_rate": 1.5442779291553135e-05,
"loss": 0.0431,
"step": 13420
},
{
"epoch": 17.13,
"learning_rate": 1.5374659400544962e-05,
"loss": 0.0706,
"step": 13430
},
{
"epoch": 17.14,
"learning_rate": 1.5306539509536786e-05,
"loss": 0.0669,
"step": 13440
},
{
"epoch": 17.16,
"learning_rate": 1.523841961852861e-05,
"loss": 0.0511,
"step": 13450
},
{
"epoch": 17.17,
"learning_rate": 1.5170299727520437e-05,
"loss": 0.0458,
"step": 13460
},
{
"epoch": 17.18,
"learning_rate": 1.5102179836512261e-05,
"loss": 0.0412,
"step": 13470
},
{
"epoch": 17.19,
"learning_rate": 1.5034059945504088e-05,
"loss": 0.0718,
"step": 13480
},
{
"epoch": 17.21,
"learning_rate": 1.4965940054495914e-05,
"loss": 0.0681,
"step": 13490
},
{
"epoch": 17.22,
"learning_rate": 1.4897820163487738e-05,
"loss": 0.0563,
"step": 13500
},
{
"epoch": 17.23,
"learning_rate": 1.4829700272479565e-05,
"loss": 0.0419,
"step": 13510
},
{
"epoch": 17.24,
"learning_rate": 1.4761580381471391e-05,
"loss": 0.037,
"step": 13520
},
{
"epoch": 17.26,
"learning_rate": 1.4693460490463215e-05,
"loss": 0.0771,
"step": 13530
},
{
"epoch": 17.27,
"learning_rate": 1.4625340599455042e-05,
"loss": 0.066,
"step": 13540
},
{
"epoch": 17.28,
"learning_rate": 1.4557220708446868e-05,
"loss": 0.054,
"step": 13550
},
{
"epoch": 17.3,
"learning_rate": 1.4489100817438692e-05,
"loss": 0.0411,
"step": 13560
},
{
"epoch": 17.31,
"learning_rate": 1.4420980926430519e-05,
"loss": 0.0395,
"step": 13570
},
{
"epoch": 17.32,
"learning_rate": 1.4352861035422343e-05,
"loss": 0.0583,
"step": 13580
},
{
"epoch": 17.33,
"learning_rate": 1.428474114441417e-05,
"loss": 0.0715,
"step": 13590
},
{
"epoch": 17.35,
"learning_rate": 1.4216621253405996e-05,
"loss": 0.0561,
"step": 13600
},
{
"epoch": 17.35,
"eval_loss": 0.2964591383934021,
"eval_runtime": 1356.6797,
"eval_samples_per_second": 4.594,
"eval_steps_per_second": 0.575,
"eval_wer": 0.316490377272033,
"step": 13600
},
{
"epoch": 17.36,
"learning_rate": 1.414850136239782e-05,
"loss": 0.0422,
"step": 13610
},
{
"epoch": 17.37,
"learning_rate": 1.4080381471389647e-05,
"loss": 0.0408,
"step": 13620
},
{
"epoch": 17.38,
"learning_rate": 1.4012261580381473e-05,
"loss": 0.0636,
"step": 13630
},
{
"epoch": 17.4,
"learning_rate": 1.3944141689373297e-05,
"loss": 0.071,
"step": 13640
},
{
"epoch": 17.41,
"learning_rate": 1.3876021798365124e-05,
"loss": 0.0537,
"step": 13650
},
{
"epoch": 17.42,
"learning_rate": 1.3807901907356948e-05,
"loss": 0.0467,
"step": 13660
},
{
"epoch": 17.44,
"learning_rate": 1.3739782016348774e-05,
"loss": 0.0389,
"step": 13670
},
{
"epoch": 17.45,
"learning_rate": 1.3671662125340601e-05,
"loss": 0.0678,
"step": 13680
},
{
"epoch": 17.46,
"learning_rate": 1.3603542234332425e-05,
"loss": 0.0687,
"step": 13690
},
{
"epoch": 17.47,
"learning_rate": 1.353542234332425e-05,
"loss": 0.0534,
"step": 13700
},
{
"epoch": 17.49,
"learning_rate": 1.3467302452316078e-05,
"loss": 0.0449,
"step": 13710
},
{
"epoch": 17.5,
"learning_rate": 1.3399182561307902e-05,
"loss": 0.0421,
"step": 13720
},
{
"epoch": 17.51,
"learning_rate": 1.3331062670299729e-05,
"loss": 0.0783,
"step": 13730
},
{
"epoch": 17.53,
"learning_rate": 1.3262942779291555e-05,
"loss": 0.0682,
"step": 13740
},
{
"epoch": 17.54,
"learning_rate": 1.3194822888283379e-05,
"loss": 0.0581,
"step": 13750
},
{
"epoch": 17.55,
"learning_rate": 1.3126702997275206e-05,
"loss": 0.0446,
"step": 13760
},
{
"epoch": 17.56,
"learning_rate": 1.305858310626703e-05,
"loss": 0.0407,
"step": 13770
},
{
"epoch": 17.58,
"learning_rate": 1.2990463215258855e-05,
"loss": 0.0706,
"step": 13780
},
{
"epoch": 17.59,
"learning_rate": 1.2922343324250683e-05,
"loss": 0.069,
"step": 13790
},
{
"epoch": 17.6,
"learning_rate": 1.2854223433242507e-05,
"loss": 0.0551,
"step": 13800
},
{
"epoch": 17.61,
"learning_rate": 1.2786103542234332e-05,
"loss": 0.0436,
"step": 13810
},
{
"epoch": 17.63,
"learning_rate": 1.271798365122616e-05,
"loss": 0.0416,
"step": 13820
},
{
"epoch": 17.64,
"learning_rate": 1.2649863760217984e-05,
"loss": 0.0661,
"step": 13830
},
{
"epoch": 17.65,
"learning_rate": 1.2581743869209811e-05,
"loss": 0.0648,
"step": 13840
},
{
"epoch": 17.67,
"learning_rate": 1.2513623978201636e-05,
"loss": 0.055,
"step": 13850
},
{
"epoch": 17.68,
"learning_rate": 1.2445504087193462e-05,
"loss": 0.0475,
"step": 13860
},
{
"epoch": 17.69,
"learning_rate": 1.2377384196185286e-05,
"loss": 0.0417,
"step": 13870
},
{
"epoch": 17.7,
"learning_rate": 1.2309264305177112e-05,
"loss": 0.0553,
"step": 13880
},
{
"epoch": 17.72,
"learning_rate": 1.2241144414168939e-05,
"loss": 0.0669,
"step": 13890
},
{
"epoch": 17.73,
"learning_rate": 1.2173024523160763e-05,
"loss": 0.0533,
"step": 13900
},
{
"epoch": 17.74,
"learning_rate": 1.2104904632152589e-05,
"loss": 0.0457,
"step": 13910
},
{
"epoch": 17.75,
"learning_rate": 1.2036784741144414e-05,
"loss": 0.0401,
"step": 13920
},
{
"epoch": 17.77,
"learning_rate": 1.1968664850136241e-05,
"loss": 0.0702,
"step": 13930
},
{
"epoch": 17.78,
"learning_rate": 1.1900544959128065e-05,
"loss": 0.0667,
"step": 13940
},
{
"epoch": 17.79,
"learning_rate": 1.1832425068119891e-05,
"loss": 0.0543,
"step": 13950
},
{
"epoch": 17.81,
"learning_rate": 1.1764305177111717e-05,
"loss": 0.0413,
"step": 13960
},
{
"epoch": 17.82,
"learning_rate": 1.1696185286103544e-05,
"loss": 0.0411,
"step": 13970
},
{
"epoch": 17.83,
"learning_rate": 1.1628065395095368e-05,
"loss": 0.0578,
"step": 13980
},
{
"epoch": 17.84,
"learning_rate": 1.1559945504087194e-05,
"loss": 0.0687,
"step": 13990
},
{
"epoch": 17.86,
"learning_rate": 1.149182561307902e-05,
"loss": 0.0506,
"step": 14000
},
{
"epoch": 17.86,
"eval_loss": 0.29349198937416077,
"eval_runtime": 1425.365,
"eval_samples_per_second": 4.373,
"eval_steps_per_second": 0.547,
"eval_wer": 0.3147720329922102,
"step": 14000
},
{
"epoch": 17.87,
"learning_rate": 1.1423705722070845e-05,
"loss": 0.046,
"step": 14010
},
{
"epoch": 17.88,
"learning_rate": 1.135558583106267e-05,
"loss": 0.0379,
"step": 14020
},
{
"epoch": 17.89,
"learning_rate": 1.1287465940054496e-05,
"loss": 0.063,
"step": 14030
},
{
"epoch": 17.91,
"learning_rate": 1.1219346049046323e-05,
"loss": 0.064,
"step": 14040
},
{
"epoch": 17.92,
"learning_rate": 1.1151226158038147e-05,
"loss": 0.0539,
"step": 14050
},
{
"epoch": 17.93,
"learning_rate": 1.1083106267029973e-05,
"loss": 0.0415,
"step": 14060
},
{
"epoch": 17.95,
"learning_rate": 1.1014986376021799e-05,
"loss": 0.0387,
"step": 14070
},
{
"epoch": 17.96,
"learning_rate": 1.0946866485013626e-05,
"loss": 0.0647,
"step": 14080
},
{
"epoch": 17.97,
"learning_rate": 1.0885558583106268e-05,
"loss": 0.0569,
"step": 14090
},
{
"epoch": 17.98,
"learning_rate": 1.0817438692098093e-05,
"loss": 0.0442,
"step": 14100
},
{
"epoch": 18.0,
"learning_rate": 1.0749318801089919e-05,
"loss": 0.0373,
"step": 14110
},
{
"epoch": 18.01,
"learning_rate": 1.0681198910081745e-05,
"loss": 0.076,
"step": 14120
},
{
"epoch": 18.02,
"learning_rate": 1.061307901907357e-05,
"loss": 0.0564,
"step": 14130
},
{
"epoch": 18.04,
"learning_rate": 1.0544959128065396e-05,
"loss": 0.0406,
"step": 14140
},
{
"epoch": 18.05,
"learning_rate": 1.0476839237057221e-05,
"loss": 0.034,
"step": 14150
},
{
"epoch": 18.06,
"learning_rate": 1.0408719346049047e-05,
"loss": 0.0487,
"step": 14160
},
{
"epoch": 18.07,
"learning_rate": 1.0340599455040873e-05,
"loss": 0.0756,
"step": 14170
},
{
"epoch": 18.09,
"learning_rate": 1.0272479564032698e-05,
"loss": 0.0532,
"step": 14180
},
{
"epoch": 18.1,
"learning_rate": 1.0204359673024522e-05,
"loss": 0.0439,
"step": 14190
},
{
"epoch": 18.11,
"learning_rate": 1.013623978201635e-05,
"loss": 0.0335,
"step": 14200
},
{
"epoch": 18.12,
"learning_rate": 1.0068119891008175e-05,
"loss": 0.0385,
"step": 14210
},
{
"epoch": 18.14,
"learning_rate": 1e-05,
"loss": 0.073,
"step": 14220
},
{
"epoch": 18.15,
"learning_rate": 9.931880108991826e-06,
"loss": 0.0474,
"step": 14230
},
{
"epoch": 18.16,
"learning_rate": 9.863760217983652e-06,
"loss": 0.0493,
"step": 14240
},
{
"epoch": 18.18,
"learning_rate": 9.795640326975478e-06,
"loss": 0.0343,
"step": 14250
},
{
"epoch": 18.19,
"learning_rate": 9.727520435967302e-06,
"loss": 0.0383,
"step": 14260
},
{
"epoch": 18.2,
"learning_rate": 9.659400544959129e-06,
"loss": 0.0704,
"step": 14270
},
{
"epoch": 18.21,
"learning_rate": 9.591280653950955e-06,
"loss": 0.0555,
"step": 14280
},
{
"epoch": 18.23,
"learning_rate": 9.52316076294278e-06,
"loss": 0.0432,
"step": 14290
},
{
"epoch": 18.24,
"learning_rate": 9.455040871934604e-06,
"loss": 0.034,
"step": 14300
},
{
"epoch": 18.25,
"learning_rate": 9.386920980926431e-06,
"loss": 0.0527,
"step": 14310
},
{
"epoch": 18.27,
"learning_rate": 9.318801089918257e-06,
"loss": 0.0846,
"step": 14320
},
{
"epoch": 18.28,
"learning_rate": 9.250681198910083e-06,
"loss": 0.0562,
"step": 14330
},
{
"epoch": 18.29,
"learning_rate": 9.182561307901907e-06,
"loss": 0.0465,
"step": 14340
},
{
"epoch": 18.3,
"learning_rate": 9.114441416893734e-06,
"loss": 0.0361,
"step": 14350
},
{
"epoch": 18.32,
"learning_rate": 9.04632152588556e-06,
"loss": 0.038,
"step": 14360
},
{
"epoch": 18.33,
"learning_rate": 8.978201634877383e-06,
"loss": 0.0687,
"step": 14370
},
{
"epoch": 18.34,
"learning_rate": 8.91008174386921e-06,
"loss": 0.0545,
"step": 14380
},
{
"epoch": 18.35,
"learning_rate": 8.841961852861036e-06,
"loss": 0.0424,
"step": 14390
},
{
"epoch": 18.37,
"learning_rate": 8.773841961852862e-06,
"loss": 0.0312,
"step": 14400
},
{
"epoch": 18.37,
"eval_loss": 0.2964278757572174,
"eval_runtime": 1359.3539,
"eval_samples_per_second": 4.585,
"eval_steps_per_second": 0.574,
"eval_wer": 0.3153925462043684,
"step": 14400
},
{
"epoch": 18.38,
"learning_rate": 8.705722070844686e-06,
"loss": 0.036,
"step": 14410
},
{
"epoch": 18.39,
"learning_rate": 8.637602179836513e-06,
"loss": 0.0695,
"step": 14420
},
{
"epoch": 18.41,
"learning_rate": 8.569482288828339e-06,
"loss": 0.0559,
"step": 14430
},
{
"epoch": 18.42,
"learning_rate": 8.501362397820165e-06,
"loss": 0.0442,
"step": 14440
},
{
"epoch": 18.43,
"learning_rate": 8.433242506811988e-06,
"loss": 0.0388,
"step": 14450
},
{
"epoch": 18.44,
"learning_rate": 8.365122615803816e-06,
"loss": 0.039,
"step": 14460
},
{
"epoch": 18.46,
"learning_rate": 8.297002724795641e-06,
"loss": 0.0635,
"step": 14470
},
{
"epoch": 18.47,
"learning_rate": 8.228882833787465e-06,
"loss": 0.0522,
"step": 14480
},
{
"epoch": 18.48,
"learning_rate": 8.160762942779291e-06,
"loss": 0.0475,
"step": 14490
},
{
"epoch": 18.49,
"learning_rate": 8.092643051771118e-06,
"loss": 0.0347,
"step": 14500
},
{
"epoch": 18.51,
"learning_rate": 8.024523160762944e-06,
"loss": 0.0385,
"step": 14510
},
{
"epoch": 18.52,
"learning_rate": 7.956403269754768e-06,
"loss": 0.0648,
"step": 14520
},
{
"epoch": 18.53,
"learning_rate": 7.888283378746593e-06,
"loss": 0.0499,
"step": 14530
},
{
"epoch": 18.55,
"learning_rate": 7.82016348773842e-06,
"loss": 0.0446,
"step": 14540
},
{
"epoch": 18.56,
"learning_rate": 7.752043596730246e-06,
"loss": 0.0395,
"step": 14550
},
{
"epoch": 18.57,
"learning_rate": 7.68392370572207e-06,
"loss": 0.0346,
"step": 14560
},
{
"epoch": 18.58,
"learning_rate": 7.615803814713897e-06,
"loss": 0.077,
"step": 14570
},
{
"epoch": 18.6,
"learning_rate": 7.547683923705723e-06,
"loss": 0.0541,
"step": 14580
},
{
"epoch": 18.61,
"learning_rate": 7.479564032697547e-06,
"loss": 0.0488,
"step": 14590
},
{
"epoch": 18.62,
"learning_rate": 7.411444141689374e-06,
"loss": 0.0365,
"step": 14600
},
{
"epoch": 18.63,
"learning_rate": 7.343324250681199e-06,
"loss": 0.049,
"step": 14610
},
{
"epoch": 18.65,
"learning_rate": 7.275204359673026e-06,
"loss": 0.0664,
"step": 14620
},
{
"epoch": 18.66,
"learning_rate": 7.20708446866485e-06,
"loss": 0.0492,
"step": 14630
},
{
"epoch": 18.67,
"learning_rate": 7.138964577656676e-06,
"loss": 0.0475,
"step": 14640
},
{
"epoch": 18.69,
"learning_rate": 7.070844686648502e-06,
"loss": 0.0396,
"step": 14650
},
{
"epoch": 18.7,
"learning_rate": 7.0027247956403266e-06,
"loss": 0.0372,
"step": 14660
},
{
"epoch": 18.71,
"learning_rate": 6.934604904632153e-06,
"loss": 0.0682,
"step": 14670
},
{
"epoch": 18.72,
"learning_rate": 6.866485013623979e-06,
"loss": 0.0486,
"step": 14680
},
{
"epoch": 18.74,
"learning_rate": 6.798365122615804e-06,
"loss": 0.0407,
"step": 14690
},
{
"epoch": 18.75,
"learning_rate": 6.730245231607629e-06,
"loss": 0.0349,
"step": 14700
},
{
"epoch": 18.76,
"learning_rate": 6.6621253405994555e-06,
"loss": 0.0409,
"step": 14710
},
{
"epoch": 18.77,
"learning_rate": 6.594005449591281e-06,
"loss": 0.0711,
"step": 14720
},
{
"epoch": 18.79,
"learning_rate": 6.525885558583107e-06,
"loss": 0.0534,
"step": 14730
},
{
"epoch": 18.8,
"learning_rate": 6.4577656675749316e-06,
"loss": 0.0403,
"step": 14740
},
{
"epoch": 18.81,
"learning_rate": 6.389645776566758e-06,
"loss": 0.0348,
"step": 14750
},
{
"epoch": 18.83,
"learning_rate": 6.321525885558584e-06,
"loss": 0.0467,
"step": 14760
},
{
"epoch": 18.84,
"learning_rate": 6.2534059945504084e-06,
"loss": 0.0744,
"step": 14770
},
{
"epoch": 18.85,
"learning_rate": 6.185286103542234e-06,
"loss": 0.051,
"step": 14780
},
{
"epoch": 18.86,
"learning_rate": 6.1171662125340605e-06,
"loss": 0.0448,
"step": 14790
},
{
"epoch": 18.88,
"learning_rate": 6.049046321525885e-06,
"loss": 0.0403,
"step": 14800
},
{
"epoch": 18.88,
"eval_loss": 0.29674461483955383,
"eval_runtime": 1357.9469,
"eval_samples_per_second": 4.59,
"eval_steps_per_second": 0.574,
"eval_wer": 0.3159939667023064,
"step": 14800
},
{
"epoch": 18.89,
"learning_rate": 5.980926430517712e-06,
"loss": 0.039,
"step": 14810
},
{
"epoch": 18.9,
"learning_rate": 5.9128065395095365e-06,
"loss": 0.0735,
"step": 14820
},
{
"epoch": 18.92,
"learning_rate": 5.844686648501363e-06,
"loss": 0.0514,
"step": 14830
},
{
"epoch": 18.93,
"learning_rate": 5.776566757493189e-06,
"loss": 0.0386,
"step": 14840
},
{
"epoch": 18.94,
"learning_rate": 5.708446866485014e-06,
"loss": 0.0378,
"step": 14850
},
{
"epoch": 18.95,
"learning_rate": 5.64032697547684e-06,
"loss": 0.0385,
"step": 14860
},
{
"epoch": 18.97,
"learning_rate": 5.572207084468665e-06,
"loss": 0.0654,
"step": 14870
},
{
"epoch": 18.98,
"learning_rate": 5.504087193460491e-06,
"loss": 0.05,
"step": 14880
},
{
"epoch": 18.99,
"learning_rate": 5.435967302452316e-06,
"loss": 0.0317,
"step": 14890
},
{
"epoch": 19.01,
"learning_rate": 5.367847411444142e-06,
"loss": 0.0565,
"step": 14900
},
{
"epoch": 19.02,
"learning_rate": 5.299727520435967e-06,
"loss": 0.053,
"step": 14910
},
{
"epoch": 19.03,
"learning_rate": 5.231607629427794e-06,
"loss": 0.0399,
"step": 14920
},
{
"epoch": 19.04,
"learning_rate": 5.163487738419618e-06,
"loss": 0.033,
"step": 14930
},
{
"epoch": 19.06,
"learning_rate": 5.095367847411445e-06,
"loss": 0.0298,
"step": 14940
},
{
"epoch": 19.07,
"learning_rate": 5.02724795640327e-06,
"loss": 0.0633,
"step": 14950
},
{
"epoch": 19.08,
"learning_rate": 4.959128065395095e-06,
"loss": 0.0576,
"step": 14960
},
{
"epoch": 19.09,
"learning_rate": 4.891008174386921e-06,
"loss": 0.0463,
"step": 14970
},
{
"epoch": 19.11,
"learning_rate": 4.8228882833787465e-06,
"loss": 0.0389,
"step": 14980
},
{
"epoch": 19.12,
"learning_rate": 4.754768392370572e-06,
"loss": 0.039,
"step": 14990
},
{
"epoch": 19.13,
"learning_rate": 4.686648501362398e-06,
"loss": 0.0538,
"step": 15000
},
{
"epoch": 19.15,
"learning_rate": 4.618528610354224e-06,
"loss": 0.0542,
"step": 15010
},
{
"epoch": 19.16,
"learning_rate": 4.550408719346049e-06,
"loss": 0.0464,
"step": 15020
},
{
"epoch": 19.17,
"learning_rate": 4.4822888283378755e-06,
"loss": 0.032,
"step": 15030
},
{
"epoch": 19.18,
"learning_rate": 4.4141689373297e-06,
"loss": 0.0372,
"step": 15040
},
{
"epoch": 19.2,
"learning_rate": 4.346049046321527e-06,
"loss": 0.0591,
"step": 15050
},
{
"epoch": 19.21,
"learning_rate": 4.2779291553133515e-06,
"loss": 0.0531,
"step": 15060
},
{
"epoch": 19.22,
"learning_rate": 4.209809264305177e-06,
"loss": 0.045,
"step": 15070
},
{
"epoch": 19.23,
"learning_rate": 4.141689373297003e-06,
"loss": 0.0371,
"step": 15080
},
{
"epoch": 19.25,
"learning_rate": 4.073569482288828e-06,
"loss": 0.0329,
"step": 15090
},
{
"epoch": 19.26,
"learning_rate": 4.005449591280654e-06,
"loss": 0.0565,
"step": 15100
},
{
"epoch": 19.27,
"learning_rate": 3.93732970027248e-06,
"loss": 0.0526,
"step": 15110
},
{
"epoch": 19.29,
"learning_rate": 3.869209809264305e-06,
"loss": 0.045,
"step": 15120
},
{
"epoch": 19.3,
"learning_rate": 3.801089918256131e-06,
"loss": 0.0344,
"step": 15130
},
{
"epoch": 19.31,
"learning_rate": 3.732970027247957e-06,
"loss": 0.0337,
"step": 15140
},
{
"epoch": 19.32,
"learning_rate": 3.664850136239782e-06,
"loss": 0.0644,
"step": 15150
},
{
"epoch": 19.34,
"learning_rate": 3.5967302452316073e-06,
"loss": 0.0561,
"step": 15160
},
{
"epoch": 19.35,
"learning_rate": 3.5286103542234334e-06,
"loss": 0.0415,
"step": 15170
},
{
"epoch": 19.36,
"learning_rate": 3.4604904632152586e-06,
"loss": 0.0395,
"step": 15180
},
{
"epoch": 19.37,
"learning_rate": 3.3923705722070846e-06,
"loss": 0.0317,
"step": 15190
},
{
"epoch": 19.39,
"learning_rate": 3.3242506811989102e-06,
"loss": 0.0924,
"step": 15200
},
{
"epoch": 19.39,
"eval_loss": 0.29548323154449463,
"eval_runtime": 1361.5115,
"eval_samples_per_second": 4.578,
"eval_steps_per_second": 0.573,
"eval_wer": 0.31472430120665956,
"step": 15200
},
{
"epoch": 19.4,
"learning_rate": 3.256130790190736e-06,
"loss": 0.0548,
"step": 15210
},
{
"epoch": 19.41,
"learning_rate": 3.1880108991825615e-06,
"loss": 0.0441,
"step": 15220
},
{
"epoch": 19.43,
"learning_rate": 3.119891008174387e-06,
"loss": 0.037,
"step": 15230
},
{
"epoch": 19.44,
"learning_rate": 3.0517711171662127e-06,
"loss": 0.0346,
"step": 15240
},
{
"epoch": 19.45,
"learning_rate": 2.9836512261580384e-06,
"loss": 0.0635,
"step": 15250
},
{
"epoch": 19.46,
"learning_rate": 2.915531335149864e-06,
"loss": 0.0529,
"step": 15260
},
{
"epoch": 19.48,
"learning_rate": 2.8474114441416896e-06,
"loss": 0.0452,
"step": 15270
},
{
"epoch": 19.49,
"learning_rate": 2.7792915531335152e-06,
"loss": 0.0334,
"step": 15280
},
{
"epoch": 19.5,
"learning_rate": 2.711171662125341e-06,
"loss": 0.0355,
"step": 15290
},
{
"epoch": 19.51,
"learning_rate": 2.6430517711171665e-06,
"loss": 0.0614,
"step": 15300
},
{
"epoch": 19.53,
"learning_rate": 2.5749318801089917e-06,
"loss": 0.0512,
"step": 15310
},
{
"epoch": 19.54,
"learning_rate": 2.5068119891008173e-06,
"loss": 0.0467,
"step": 15320
},
{
"epoch": 19.55,
"learning_rate": 2.438692098092643e-06,
"loss": 0.0321,
"step": 15330
},
{
"epoch": 19.57,
"learning_rate": 2.3705722070844686e-06,
"loss": 0.0336,
"step": 15340
},
{
"epoch": 19.58,
"learning_rate": 2.302452316076294e-06,
"loss": 0.0552,
"step": 15350
},
{
"epoch": 19.59,
"learning_rate": 2.2343324250681202e-06,
"loss": 0.0549,
"step": 15360
},
{
"epoch": 19.6,
"learning_rate": 2.166212534059946e-06,
"loss": 0.0471,
"step": 15370
},
{
"epoch": 19.62,
"learning_rate": 2.0980926430517715e-06,
"loss": 0.0332,
"step": 15380
},
{
"epoch": 19.63,
"learning_rate": 2.029972752043597e-06,
"loss": 0.036,
"step": 15390
},
{
"epoch": 19.64,
"learning_rate": 1.9618528610354227e-06,
"loss": 0.0588,
"step": 15400
},
{
"epoch": 19.66,
"learning_rate": 1.893732970027248e-06,
"loss": 0.0578,
"step": 15410
},
{
"epoch": 19.67,
"learning_rate": 1.8256130790190735e-06,
"loss": 0.043,
"step": 15420
},
{
"epoch": 19.68,
"learning_rate": 1.7574931880108992e-06,
"loss": 0.0444,
"step": 15430
},
{
"epoch": 19.69,
"learning_rate": 1.6893732970027248e-06,
"loss": 0.0358,
"step": 15440
},
{
"epoch": 19.71,
"learning_rate": 1.6212534059945504e-06,
"loss": 0.0561,
"step": 15450
},
{
"epoch": 19.72,
"learning_rate": 1.553133514986376e-06,
"loss": 0.0544,
"step": 15460
},
{
"epoch": 19.73,
"learning_rate": 1.4850136239782019e-06,
"loss": 0.0402,
"step": 15470
},
{
"epoch": 19.74,
"learning_rate": 1.4168937329700273e-06,
"loss": 0.0356,
"step": 15480
},
{
"epoch": 19.76,
"learning_rate": 1.348773841961853e-06,
"loss": 0.0353,
"step": 15490
},
{
"epoch": 19.77,
"learning_rate": 1.2806539509536785e-06,
"loss": 0.0638,
"step": 15500
},
{
"epoch": 19.78,
"learning_rate": 1.2125340599455042e-06,
"loss": 0.0524,
"step": 15510
},
{
"epoch": 19.8,
"learning_rate": 1.1444141689373298e-06,
"loss": 0.0416,
"step": 15520
},
{
"epoch": 19.81,
"learning_rate": 1.0762942779291554e-06,
"loss": 0.037,
"step": 15530
},
{
"epoch": 19.82,
"learning_rate": 1.008174386920981e-06,
"loss": 0.0317,
"step": 15540
},
{
"epoch": 19.83,
"learning_rate": 9.400544959128065e-07,
"loss": 0.0599,
"step": 15550
},
{
"epoch": 19.85,
"learning_rate": 8.719346049046323e-07,
"loss": 0.0561,
"step": 15560
},
{
"epoch": 19.86,
"learning_rate": 8.038147138964579e-07,
"loss": 0.0411,
"step": 15570
},
{
"epoch": 19.87,
"learning_rate": 7.356948228882834e-07,
"loss": 0.0349,
"step": 15580
},
{
"epoch": 19.88,
"learning_rate": 6.67574931880109e-07,
"loss": 0.0375,
"step": 15590
},
{
"epoch": 19.9,
"learning_rate": 5.994550408719347e-07,
"loss": 0.0585,
"step": 15600
},
{
"epoch": 19.9,
"eval_loss": 0.2964596450328827,
"eval_runtime": 1427.8876,
"eval_samples_per_second": 4.365,
"eval_steps_per_second": 0.546,
"eval_wer": 0.3143997250649152,
"step": 15600
},
{
"epoch": 19.91,
"learning_rate": 5.313351498637603e-07,
"loss": 0.0515,
"step": 15610
},
{
"epoch": 19.92,
"learning_rate": 4.6321525885558585e-07,
"loss": 0.0488,
"step": 15620
},
{
"epoch": 19.94,
"learning_rate": 3.950953678474115e-07,
"loss": 0.0363,
"step": 15630
},
{
"epoch": 19.95,
"learning_rate": 3.269754768392371e-07,
"loss": 0.0416,
"step": 15640
},
{
"epoch": 19.96,
"learning_rate": 2.5885558583106267e-07,
"loss": 0.0606,
"step": 15650
},
{
"epoch": 19.97,
"learning_rate": 1.907356948228883e-07,
"loss": 0.0508,
"step": 15660
},
{
"epoch": 19.99,
"learning_rate": 1.2261580381471392e-07,
"loss": 0.0384,
"step": 15670
},
{
"epoch": 20.0,
"learning_rate": 5.449591280653952e-08,
"loss": 0.0452,
"step": 15680
},
{
"epoch": 20.0,
"step": 15680,
"total_flos": 0.0,
"train_loss": 0.45149919644986486,
"train_runtime": 150700.5368,
"train_samples_per_second": 6.664,
"train_steps_per_second": 0.104
}
],
"max_steps": 15680,
"num_train_epochs": 20,
"total_flos": 0.0,
"trial_name": null,
"trial_params": null
}