xls-r-2B-te / trainer_state.json
chmanoj's picture
End of training
460a6d6
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 149.9928057553957,
"global_step": 10350,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.14,
"learning_rate": 9.000000000000001e-09,
"loss": 28.1828,
"step": 10
},
{
"epoch": 0.29,
"learning_rate": 2.4e-08,
"loss": 24.2341,
"step": 20
},
{
"epoch": 0.43,
"learning_rate": 3.9e-08,
"loss": 24.4626,
"step": 30
},
{
"epoch": 0.58,
"learning_rate": 5.250000000000001e-08,
"loss": 25.4891,
"step": 40
},
{
"epoch": 0.72,
"learning_rate": 6.75e-08,
"loss": 21.8962,
"step": 50
},
{
"epoch": 0.86,
"learning_rate": 8.25e-08,
"loss": 26.2134,
"step": 60
},
{
"epoch": 1.01,
"learning_rate": 9.75e-08,
"loss": 22.6548,
"step": 70
},
{
"epoch": 1.16,
"learning_rate": 1.125e-07,
"loss": 24.8818,
"step": 80
},
{
"epoch": 1.3,
"learning_rate": 1.2750000000000002e-07,
"loss": 23.207,
"step": 90
},
{
"epoch": 1.45,
"learning_rate": 1.41e-07,
"loss": 21.792,
"step": 100
},
{
"epoch": 1.59,
"learning_rate": 1.56e-07,
"loss": 22.6451,
"step": 110
},
{
"epoch": 1.73,
"learning_rate": 1.71e-07,
"loss": 17.8776,
"step": 120
},
{
"epoch": 1.88,
"learning_rate": 1.845e-07,
"loss": 21.7572,
"step": 130
},
{
"epoch": 2.03,
"learning_rate": 1.995e-07,
"loss": 19.17,
"step": 140
},
{
"epoch": 2.17,
"learning_rate": 2.145e-07,
"loss": 18.2221,
"step": 150
},
{
"epoch": 2.32,
"learning_rate": 2.295e-07,
"loss": 17.0341,
"step": 160
},
{
"epoch": 2.46,
"learning_rate": 2.445e-07,
"loss": 14.7646,
"step": 170
},
{
"epoch": 2.6,
"learning_rate": 2.5949999999999996e-07,
"loss": 14.9477,
"step": 180
},
{
"epoch": 2.75,
"learning_rate": 2.745e-07,
"loss": 12.1337,
"step": 190
},
{
"epoch": 2.89,
"learning_rate": 2.895e-07,
"loss": 11.5327,
"step": 200
},
{
"epoch": 3.04,
"learning_rate": 3.045e-07,
"loss": 7.1502,
"step": 210
},
{
"epoch": 3.19,
"learning_rate": 3.195e-07,
"loss": 4.7373,
"step": 220
},
{
"epoch": 3.33,
"learning_rate": 3.3450000000000004e-07,
"loss": 4.3251,
"step": 230
},
{
"epoch": 3.47,
"learning_rate": 3.4950000000000005e-07,
"loss": 4.0801,
"step": 240
},
{
"epoch": 3.62,
"learning_rate": 3.645e-07,
"loss": 3.9993,
"step": 250
},
{
"epoch": 3.76,
"learning_rate": 3.7950000000000003e-07,
"loss": 3.8435,
"step": 260
},
{
"epoch": 3.91,
"learning_rate": 3.9450000000000005e-07,
"loss": 3.7385,
"step": 270
},
{
"epoch": 4.06,
"learning_rate": 4.0950000000000006e-07,
"loss": 3.9388,
"step": 280
},
{
"epoch": 4.2,
"learning_rate": 4.2449999999999997e-07,
"loss": 3.6508,
"step": 290
},
{
"epoch": 4.35,
"learning_rate": 4.395e-07,
"loss": 3.6744,
"step": 300
},
{
"epoch": 4.49,
"learning_rate": 4.545e-07,
"loss": 3.6055,
"step": 310
},
{
"epoch": 4.63,
"learning_rate": 4.695e-07,
"loss": 3.5857,
"step": 320
},
{
"epoch": 4.78,
"learning_rate": 4.845e-07,
"loss": 3.6282,
"step": 330
},
{
"epoch": 4.92,
"learning_rate": 4.995e-07,
"loss": 3.5538,
"step": 340
},
{
"epoch": 5.07,
"learning_rate": 5.145e-07,
"loss": 3.7494,
"step": 350
},
{
"epoch": 5.22,
"learning_rate": 5.295e-07,
"loss": 3.5425,
"step": 360
},
{
"epoch": 5.36,
"learning_rate": 5.445e-07,
"loss": 3.5777,
"step": 370
},
{
"epoch": 5.5,
"learning_rate": 5.595e-07,
"loss": 3.5485,
"step": 380
},
{
"epoch": 5.65,
"learning_rate": 5.745000000000001e-07,
"loss": 3.575,
"step": 390
},
{
"epoch": 5.79,
"learning_rate": 5.895e-07,
"loss": 3.5488,
"step": 400
},
{
"epoch": 5.94,
"learning_rate": 6.045000000000001e-07,
"loss": 3.5145,
"step": 410
},
{
"epoch": 6.09,
"learning_rate": 6.195e-07,
"loss": 3.7089,
"step": 420
},
{
"epoch": 6.23,
"learning_rate": 6.345e-07,
"loss": 3.5021,
"step": 430
},
{
"epoch": 6.37,
"learning_rate": 6.495e-07,
"loss": 3.5196,
"step": 440
},
{
"epoch": 6.52,
"learning_rate": 6.645000000000001e-07,
"loss": 3.5193,
"step": 450
},
{
"epoch": 6.66,
"learning_rate": 6.795e-07,
"loss": 3.505,
"step": 460
},
{
"epoch": 6.81,
"learning_rate": 6.945000000000001e-07,
"loss": 3.5159,
"step": 470
},
{
"epoch": 6.95,
"learning_rate": 7.094999999999999e-07,
"loss": 3.4936,
"step": 480
},
{
"epoch": 7.1,
"learning_rate": 7.245e-07,
"loss": 3.6754,
"step": 490
},
{
"epoch": 7.24,
"learning_rate": 7.395e-07,
"loss": 3.5031,
"step": 500
},
{
"epoch": 7.24,
"eval_loss": 4.951496124267578,
"eval_runtime": 80.925,
"eval_samples_per_second": 13.741,
"eval_steps_per_second": 3.435,
"eval_wer": 1.0,
"step": 500
},
{
"epoch": 7.39,
"learning_rate": 7.545e-07,
"loss": 3.4977,
"step": 510
},
{
"epoch": 7.53,
"learning_rate": 7.695e-07,
"loss": 3.4927,
"step": 520
},
{
"epoch": 7.68,
"learning_rate": 7.845000000000001e-07,
"loss": 3.4699,
"step": 530
},
{
"epoch": 7.82,
"learning_rate": 7.995e-07,
"loss": 3.4756,
"step": 540
},
{
"epoch": 7.96,
"learning_rate": 8.145000000000001e-07,
"loss": 3.467,
"step": 550
},
{
"epoch": 8.12,
"learning_rate": 8.295000000000001e-07,
"loss": 3.6391,
"step": 560
},
{
"epoch": 8.26,
"learning_rate": 8.444999999999999e-07,
"loss": 3.4643,
"step": 570
},
{
"epoch": 8.4,
"learning_rate": 8.595e-07,
"loss": 3.4442,
"step": 580
},
{
"epoch": 8.55,
"learning_rate": 8.745e-07,
"loss": 3.4775,
"step": 590
},
{
"epoch": 8.69,
"learning_rate": 8.895e-07,
"loss": 3.4583,
"step": 600
},
{
"epoch": 8.83,
"learning_rate": 9.045e-07,
"loss": 3.4588,
"step": 610
},
{
"epoch": 8.98,
"learning_rate": 9.195e-07,
"loss": 3.4522,
"step": 620
},
{
"epoch": 9.13,
"learning_rate": 9.345e-07,
"loss": 3.6193,
"step": 630
},
{
"epoch": 9.27,
"learning_rate": 9.495000000000001e-07,
"loss": 3.4395,
"step": 640
},
{
"epoch": 9.42,
"learning_rate": 9.645e-07,
"loss": 3.4333,
"step": 650
},
{
"epoch": 9.56,
"learning_rate": 9.795000000000001e-07,
"loss": 3.452,
"step": 660
},
{
"epoch": 9.71,
"learning_rate": 9.945000000000002e-07,
"loss": 3.4371,
"step": 670
},
{
"epoch": 9.85,
"learning_rate": 1.0095e-06,
"loss": 3.4405,
"step": 680
},
{
"epoch": 9.99,
"learning_rate": 1.0245e-06,
"loss": 3.4329,
"step": 690
},
{
"epoch": 10.14,
"learning_rate": 1.0395e-06,
"loss": 3.6168,
"step": 700
},
{
"epoch": 10.29,
"learning_rate": 1.0545e-06,
"loss": 3.4226,
"step": 710
},
{
"epoch": 10.43,
"learning_rate": 1.0694999999999999e-06,
"loss": 3.4074,
"step": 720
},
{
"epoch": 10.58,
"learning_rate": 1.0845e-06,
"loss": 3.409,
"step": 730
},
{
"epoch": 10.72,
"learning_rate": 1.0995e-06,
"loss": 3.4156,
"step": 740
},
{
"epoch": 10.86,
"learning_rate": 1.1145e-06,
"loss": 3.4124,
"step": 750
},
{
"epoch": 11.01,
"learning_rate": 1.1295e-06,
"loss": 3.5837,
"step": 760
},
{
"epoch": 11.16,
"learning_rate": 1.1445e-06,
"loss": 3.4058,
"step": 770
},
{
"epoch": 11.3,
"learning_rate": 1.1595e-06,
"loss": 3.4128,
"step": 780
},
{
"epoch": 11.45,
"learning_rate": 1.1745000000000001e-06,
"loss": 3.385,
"step": 790
},
{
"epoch": 11.59,
"learning_rate": 1.1895e-06,
"loss": 3.3935,
"step": 800
},
{
"epoch": 11.73,
"learning_rate": 1.2045e-06,
"loss": 3.3816,
"step": 810
},
{
"epoch": 11.88,
"learning_rate": 1.2195e-06,
"loss": 3.3713,
"step": 820
},
{
"epoch": 12.03,
"learning_rate": 1.2345e-06,
"loss": 3.551,
"step": 830
},
{
"epoch": 12.17,
"learning_rate": 1.2495e-06,
"loss": 3.3547,
"step": 840
},
{
"epoch": 12.32,
"learning_rate": 1.2645e-06,
"loss": 3.3473,
"step": 850
},
{
"epoch": 12.46,
"learning_rate": 1.2795e-06,
"loss": 3.3273,
"step": 860
},
{
"epoch": 12.6,
"learning_rate": 1.2945e-06,
"loss": 3.3225,
"step": 870
},
{
"epoch": 12.75,
"learning_rate": 1.3095000000000001e-06,
"loss": 3.3156,
"step": 880
},
{
"epoch": 12.89,
"learning_rate": 1.3245e-06,
"loss": 3.3005,
"step": 890
},
{
"epoch": 13.04,
"learning_rate": 1.3395e-06,
"loss": 3.4521,
"step": 900
},
{
"epoch": 13.19,
"learning_rate": 1.3545000000000001e-06,
"loss": 3.2617,
"step": 910
},
{
"epoch": 13.33,
"learning_rate": 1.3695000000000002e-06,
"loss": 3.2502,
"step": 920
},
{
"epoch": 13.47,
"learning_rate": 1.3845e-06,
"loss": 3.2332,
"step": 930
},
{
"epoch": 13.62,
"learning_rate": 1.3995000000000001e-06,
"loss": 3.2122,
"step": 940
},
{
"epoch": 13.76,
"learning_rate": 1.4145e-06,
"loss": 3.1723,
"step": 950
},
{
"epoch": 13.91,
"learning_rate": 1.4295e-06,
"loss": 3.1439,
"step": 960
},
{
"epoch": 14.06,
"learning_rate": 1.4444999999999999e-06,
"loss": 3.2843,
"step": 970
},
{
"epoch": 14.2,
"learning_rate": 1.4595e-06,
"loss": 3.0663,
"step": 980
},
{
"epoch": 14.35,
"learning_rate": 1.4745e-06,
"loss": 3.046,
"step": 990
},
{
"epoch": 14.49,
"learning_rate": 1.4895e-06,
"loss": 3.0108,
"step": 1000
},
{
"epoch": 14.49,
"eval_loss": 3.3814377784729004,
"eval_runtime": 80.8852,
"eval_samples_per_second": 13.748,
"eval_steps_per_second": 3.437,
"eval_wer": 1.0,
"step": 1000
},
{
"epoch": 14.63,
"learning_rate": 1.5045e-06,
"loss": 2.9622,
"step": 1010
},
{
"epoch": 14.78,
"learning_rate": 1.5194999999999998e-06,
"loss": 2.9512,
"step": 1020
},
{
"epoch": 14.92,
"learning_rate": 1.5344999999999999e-06,
"loss": 2.8982,
"step": 1030
},
{
"epoch": 15.07,
"learning_rate": 1.5495e-06,
"loss": 2.9837,
"step": 1040
},
{
"epoch": 15.22,
"learning_rate": 1.5645e-06,
"loss": 2.7965,
"step": 1050
},
{
"epoch": 15.36,
"learning_rate": 1.5794999999999999e-06,
"loss": 2.7453,
"step": 1060
},
{
"epoch": 15.5,
"learning_rate": 1.5945e-06,
"loss": 2.6937,
"step": 1070
},
{
"epoch": 15.65,
"learning_rate": 1.6095e-06,
"loss": 2.6168,
"step": 1080
},
{
"epoch": 15.79,
"learning_rate": 1.6245e-06,
"loss": 2.6216,
"step": 1090
},
{
"epoch": 15.94,
"learning_rate": 1.6395e-06,
"loss": 2.559,
"step": 1100
},
{
"epoch": 16.09,
"learning_rate": 1.6545e-06,
"loss": 2.6089,
"step": 1110
},
{
"epoch": 16.23,
"learning_rate": 1.6695e-06,
"loss": 2.4751,
"step": 1120
},
{
"epoch": 16.37,
"learning_rate": 1.6845000000000001e-06,
"loss": 2.3754,
"step": 1130
},
{
"epoch": 16.52,
"learning_rate": 1.6995e-06,
"loss": 2.3927,
"step": 1140
},
{
"epoch": 16.66,
"learning_rate": 1.7145e-06,
"loss": 2.3235,
"step": 1150
},
{
"epoch": 16.81,
"learning_rate": 1.7295000000000001e-06,
"loss": 2.2497,
"step": 1160
},
{
"epoch": 16.95,
"learning_rate": 1.7445000000000002e-06,
"loss": 2.2205,
"step": 1170
},
{
"epoch": 17.1,
"learning_rate": 1.7595e-06,
"loss": 2.2836,
"step": 1180
},
{
"epoch": 17.24,
"learning_rate": 1.7745000000000001e-06,
"loss": 2.1324,
"step": 1190
},
{
"epoch": 17.39,
"learning_rate": 1.7895000000000002e-06,
"loss": 2.0397,
"step": 1200
},
{
"epoch": 17.53,
"learning_rate": 1.8045e-06,
"loss": 2.0696,
"step": 1210
},
{
"epoch": 17.68,
"learning_rate": 1.8195000000000001e-06,
"loss": 2.0017,
"step": 1220
},
{
"epoch": 17.82,
"learning_rate": 1.8345000000000002e-06,
"loss": 1.9907,
"step": 1230
},
{
"epoch": 17.96,
"learning_rate": 1.8495000000000002e-06,
"loss": 1.9553,
"step": 1240
},
{
"epoch": 18.12,
"learning_rate": 1.8645e-06,
"loss": 2.0103,
"step": 1250
},
{
"epoch": 18.26,
"learning_rate": 1.8795e-06,
"loss": 1.9132,
"step": 1260
},
{
"epoch": 18.4,
"learning_rate": 1.8944999999999998e-06,
"loss": 1.8362,
"step": 1270
},
{
"epoch": 18.55,
"learning_rate": 1.9095e-06,
"loss": 1.8067,
"step": 1280
},
{
"epoch": 18.69,
"learning_rate": 1.9245e-06,
"loss": 1.8177,
"step": 1290
},
{
"epoch": 18.83,
"learning_rate": 1.9395e-06,
"loss": 1.7522,
"step": 1300
},
{
"epoch": 18.98,
"learning_rate": 1.9545e-06,
"loss": 1.7703,
"step": 1310
},
{
"epoch": 19.13,
"learning_rate": 1.9695e-06,
"loss": 1.7663,
"step": 1320
},
{
"epoch": 19.27,
"learning_rate": 1.9845e-06,
"loss": 1.7104,
"step": 1330
},
{
"epoch": 19.42,
"learning_rate": 1.9995e-06,
"loss": 1.6223,
"step": 1340
},
{
"epoch": 19.56,
"learning_rate": 2.0145e-06,
"loss": 1.6236,
"step": 1350
},
{
"epoch": 19.71,
"learning_rate": 2.0295000000000002e-06,
"loss": 1.6141,
"step": 1360
},
{
"epoch": 19.85,
"learning_rate": 2.0445e-06,
"loss": 1.5409,
"step": 1370
},
{
"epoch": 19.99,
"learning_rate": 2.0595e-06,
"loss": 1.6208,
"step": 1380
},
{
"epoch": 20.14,
"learning_rate": 2.0745000000000002e-06,
"loss": 1.5611,
"step": 1390
},
{
"epoch": 20.29,
"learning_rate": 2.0895e-06,
"loss": 1.5126,
"step": 1400
},
{
"epoch": 20.43,
"learning_rate": 2.1045e-06,
"loss": 1.4751,
"step": 1410
},
{
"epoch": 20.58,
"learning_rate": 2.1195e-06,
"loss": 1.4671,
"step": 1420
},
{
"epoch": 20.72,
"learning_rate": 2.1345e-06,
"loss": 1.4835,
"step": 1430
},
{
"epoch": 20.86,
"learning_rate": 2.1495000000000003e-06,
"loss": 1.4408,
"step": 1440
},
{
"epoch": 21.01,
"learning_rate": 2.1645e-06,
"loss": 1.5207,
"step": 1450
},
{
"epoch": 21.16,
"learning_rate": 2.1795e-06,
"loss": 1.3442,
"step": 1460
},
{
"epoch": 21.3,
"learning_rate": 2.1945000000000003e-06,
"loss": 1.4264,
"step": 1470
},
{
"epoch": 21.45,
"learning_rate": 2.2095e-06,
"loss": 1.3635,
"step": 1480
},
{
"epoch": 21.59,
"learning_rate": 2.2245e-06,
"loss": 1.3543,
"step": 1490
},
{
"epoch": 21.73,
"learning_rate": 2.2395000000000003e-06,
"loss": 1.392,
"step": 1500
},
{
"epoch": 21.73,
"eval_loss": 0.6381995677947998,
"eval_runtime": 80.6605,
"eval_samples_per_second": 13.786,
"eval_steps_per_second": 3.447,
"eval_wer": 0.7275336322869955,
"step": 1500
},
{
"epoch": 21.88,
"learning_rate": 2.2544999999999998e-06,
"loss": 1.2893,
"step": 1510
},
{
"epoch": 22.03,
"learning_rate": 2.2695e-06,
"loss": 1.4047,
"step": 1520
},
{
"epoch": 22.17,
"learning_rate": 2.2845e-06,
"loss": 1.2601,
"step": 1530
},
{
"epoch": 22.32,
"learning_rate": 2.2994999999999998e-06,
"loss": 1.2695,
"step": 1540
},
{
"epoch": 22.46,
"learning_rate": 2.3145e-06,
"loss": 1.2669,
"step": 1550
},
{
"epoch": 22.6,
"learning_rate": 2.3295e-06,
"loss": 1.2321,
"step": 1560
},
{
"epoch": 22.75,
"learning_rate": 2.3445e-06,
"loss": 1.2905,
"step": 1570
},
{
"epoch": 22.89,
"learning_rate": 2.3595e-06,
"loss": 1.1821,
"step": 1580
},
{
"epoch": 23.04,
"learning_rate": 2.3745e-06,
"loss": 1.2748,
"step": 1590
},
{
"epoch": 23.19,
"learning_rate": 2.3895e-06,
"loss": 1.2028,
"step": 1600
},
{
"epoch": 23.33,
"learning_rate": 2.4045e-06,
"loss": 1.1953,
"step": 1610
},
{
"epoch": 23.47,
"learning_rate": 2.4195e-06,
"loss": 1.1711,
"step": 1620
},
{
"epoch": 23.62,
"learning_rate": 2.4345e-06,
"loss": 1.1731,
"step": 1630
},
{
"epoch": 23.76,
"learning_rate": 2.4495e-06,
"loss": 1.1925,
"step": 1640
},
{
"epoch": 23.91,
"learning_rate": 2.4645000000000003e-06,
"loss": 1.1525,
"step": 1650
},
{
"epoch": 24.06,
"learning_rate": 2.4795e-06,
"loss": 1.2419,
"step": 1660
},
{
"epoch": 24.2,
"learning_rate": 2.4945e-06,
"loss": 1.1294,
"step": 1670
},
{
"epoch": 24.35,
"learning_rate": 2.5095000000000003e-06,
"loss": 1.1105,
"step": 1680
},
{
"epoch": 24.49,
"learning_rate": 2.5245e-06,
"loss": 1.1414,
"step": 1690
},
{
"epoch": 24.63,
"learning_rate": 2.5395e-06,
"loss": 1.0687,
"step": 1700
},
{
"epoch": 24.78,
"learning_rate": 2.5545000000000003e-06,
"loss": 1.1196,
"step": 1710
},
{
"epoch": 24.92,
"learning_rate": 2.5695e-06,
"loss": 1.1384,
"step": 1720
},
{
"epoch": 25.07,
"learning_rate": 2.5845e-06,
"loss": 1.1583,
"step": 1730
},
{
"epoch": 25.22,
"learning_rate": 2.5995000000000003e-06,
"loss": 1.0956,
"step": 1740
},
{
"epoch": 25.36,
"learning_rate": 2.6145e-06,
"loss": 1.0292,
"step": 1750
},
{
"epoch": 25.5,
"learning_rate": 2.6295e-06,
"loss": 1.1084,
"step": 1760
},
{
"epoch": 25.65,
"learning_rate": 2.6445e-06,
"loss": 1.0195,
"step": 1770
},
{
"epoch": 25.79,
"learning_rate": 2.6595e-06,
"loss": 1.1022,
"step": 1780
},
{
"epoch": 25.94,
"learning_rate": 2.6745e-06,
"loss": 1.0944,
"step": 1790
},
{
"epoch": 26.09,
"learning_rate": 2.6895e-06,
"loss": 1.0941,
"step": 1800
},
{
"epoch": 26.23,
"learning_rate": 2.7045e-06,
"loss": 1.05,
"step": 1810
},
{
"epoch": 26.37,
"learning_rate": 2.7195e-06,
"loss": 1.0454,
"step": 1820
},
{
"epoch": 26.52,
"learning_rate": 2.7345e-06,
"loss": 1.0814,
"step": 1830
},
{
"epoch": 26.66,
"learning_rate": 2.7495e-06,
"loss": 0.9989,
"step": 1840
},
{
"epoch": 26.81,
"learning_rate": 2.7645e-06,
"loss": 0.9971,
"step": 1850
},
{
"epoch": 26.95,
"learning_rate": 2.7795000000000002e-06,
"loss": 1.0312,
"step": 1860
},
{
"epoch": 27.1,
"learning_rate": 2.7945e-06,
"loss": 1.0308,
"step": 1870
},
{
"epoch": 27.24,
"learning_rate": 2.8095e-06,
"loss": 1.0391,
"step": 1880
},
{
"epoch": 27.39,
"learning_rate": 2.8245000000000002e-06,
"loss": 0.9528,
"step": 1890
},
{
"epoch": 27.53,
"learning_rate": 2.8395e-06,
"loss": 0.9994,
"step": 1900
},
{
"epoch": 27.68,
"learning_rate": 2.8545e-06,
"loss": 0.9735,
"step": 1910
},
{
"epoch": 27.82,
"learning_rate": 2.8695000000000002e-06,
"loss": 0.9864,
"step": 1920
},
{
"epoch": 27.96,
"learning_rate": 2.8845e-06,
"loss": 0.9995,
"step": 1930
},
{
"epoch": 28.12,
"learning_rate": 2.8995000000000004e-06,
"loss": 0.9822,
"step": 1940
},
{
"epoch": 28.26,
"learning_rate": 2.9145000000000002e-06,
"loss": 1.0025,
"step": 1950
},
{
"epoch": 28.4,
"learning_rate": 2.9295e-06,
"loss": 0.9237,
"step": 1960
},
{
"epoch": 28.55,
"learning_rate": 2.9445000000000004e-06,
"loss": 0.9515,
"step": 1970
},
{
"epoch": 28.69,
"learning_rate": 2.9595e-06,
"loss": 0.9652,
"step": 1980
},
{
"epoch": 28.83,
"learning_rate": 2.9745e-06,
"loss": 0.899,
"step": 1990
},
{
"epoch": 28.98,
"learning_rate": 2.9895000000000003e-06,
"loss": 0.9757,
"step": 2000
},
{
"epoch": 28.98,
"eval_loss": 0.5245834589004517,
"eval_runtime": 80.4644,
"eval_samples_per_second": 13.82,
"eval_steps_per_second": 3.455,
"eval_wer": 0.6414349775784753,
"step": 2000
},
{
"epoch": 29.13,
"learning_rate": 2.9989221556886227e-06,
"loss": 0.9038,
"step": 2010
},
{
"epoch": 29.27,
"learning_rate": 2.9953293413173654e-06,
"loss": 0.9578,
"step": 2020
},
{
"epoch": 29.42,
"learning_rate": 2.9917365269461077e-06,
"loss": 0.9013,
"step": 2030
},
{
"epoch": 29.56,
"learning_rate": 2.9881437125748503e-06,
"loss": 0.9225,
"step": 2040
},
{
"epoch": 29.71,
"learning_rate": 2.984550898203593e-06,
"loss": 0.9209,
"step": 2050
},
{
"epoch": 29.85,
"learning_rate": 2.9809580838323352e-06,
"loss": 0.8796,
"step": 2060
},
{
"epoch": 29.99,
"learning_rate": 2.977365269461078e-06,
"loss": 0.931,
"step": 2070
},
{
"epoch": 30.14,
"learning_rate": 2.9737724550898206e-06,
"loss": 0.9217,
"step": 2080
},
{
"epoch": 30.29,
"learning_rate": 2.970179640718563e-06,
"loss": 0.9474,
"step": 2090
},
{
"epoch": 30.43,
"learning_rate": 2.9665868263473055e-06,
"loss": 0.9161,
"step": 2100
},
{
"epoch": 30.58,
"learning_rate": 2.962994011976048e-06,
"loss": 0.8708,
"step": 2110
},
{
"epoch": 30.72,
"learning_rate": 2.9594011976047904e-06,
"loss": 0.921,
"step": 2120
},
{
"epoch": 30.86,
"learning_rate": 2.955808383233533e-06,
"loss": 0.8486,
"step": 2130
},
{
"epoch": 31.01,
"learning_rate": 2.9522155688622753e-06,
"loss": 0.9301,
"step": 2140
},
{
"epoch": 31.16,
"learning_rate": 2.948622754491018e-06,
"loss": 0.8314,
"step": 2150
},
{
"epoch": 31.3,
"learning_rate": 2.9450299401197607e-06,
"loss": 0.8669,
"step": 2160
},
{
"epoch": 31.45,
"learning_rate": 2.941437125748503e-06,
"loss": 0.8842,
"step": 2170
},
{
"epoch": 31.59,
"learning_rate": 2.9378443113772456e-06,
"loss": 0.8329,
"step": 2180
},
{
"epoch": 31.73,
"learning_rate": 2.9342514970059882e-06,
"loss": 0.9133,
"step": 2190
},
{
"epoch": 31.88,
"learning_rate": 2.9306586826347305e-06,
"loss": 0.852,
"step": 2200
},
{
"epoch": 32.03,
"learning_rate": 2.927065868263473e-06,
"loss": 0.9374,
"step": 2210
},
{
"epoch": 32.17,
"learning_rate": 2.923473053892216e-06,
"loss": 0.8135,
"step": 2220
},
{
"epoch": 32.32,
"learning_rate": 2.919880239520958e-06,
"loss": 0.852,
"step": 2230
},
{
"epoch": 32.46,
"learning_rate": 2.9162874251497007e-06,
"loss": 0.8536,
"step": 2240
},
{
"epoch": 32.6,
"learning_rate": 2.9126946107784434e-06,
"loss": 0.8405,
"step": 2250
},
{
"epoch": 32.75,
"learning_rate": 2.9091017964071857e-06,
"loss": 0.8529,
"step": 2260
},
{
"epoch": 32.89,
"learning_rate": 2.9055089820359283e-06,
"loss": 0.8294,
"step": 2270
},
{
"epoch": 33.04,
"learning_rate": 2.9019161676646706e-06,
"loss": 0.9009,
"step": 2280
},
{
"epoch": 33.19,
"learning_rate": 2.8983233532934132e-06,
"loss": 0.8367,
"step": 2290
},
{
"epoch": 33.33,
"learning_rate": 2.894730538922156e-06,
"loss": 0.8131,
"step": 2300
},
{
"epoch": 33.47,
"learning_rate": 2.891137724550898e-06,
"loss": 0.859,
"step": 2310
},
{
"epoch": 33.62,
"learning_rate": 2.887544910179641e-06,
"loss": 0.8162,
"step": 2320
},
{
"epoch": 33.76,
"learning_rate": 2.8839520958083835e-06,
"loss": 0.8391,
"step": 2330
},
{
"epoch": 33.91,
"learning_rate": 2.8803592814371257e-06,
"loss": 0.7834,
"step": 2340
},
{
"epoch": 34.06,
"learning_rate": 2.8767664670658684e-06,
"loss": 0.854,
"step": 2350
},
{
"epoch": 34.2,
"learning_rate": 2.873173652694611e-06,
"loss": 0.8008,
"step": 2360
},
{
"epoch": 34.35,
"learning_rate": 2.8695808383233533e-06,
"loss": 0.7912,
"step": 2370
},
{
"epoch": 34.49,
"learning_rate": 2.865988023952096e-06,
"loss": 0.8211,
"step": 2380
},
{
"epoch": 34.63,
"learning_rate": 2.8623952095808386e-06,
"loss": 0.7528,
"step": 2390
},
{
"epoch": 34.78,
"learning_rate": 2.858802395209581e-06,
"loss": 0.8423,
"step": 2400
},
{
"epoch": 34.92,
"learning_rate": 2.8552095808383236e-06,
"loss": 0.7877,
"step": 2410
},
{
"epoch": 35.07,
"learning_rate": 2.851616766467066e-06,
"loss": 0.8529,
"step": 2420
},
{
"epoch": 35.22,
"learning_rate": 2.8480239520958085e-06,
"loss": 0.7859,
"step": 2430
},
{
"epoch": 35.36,
"learning_rate": 2.844431137724551e-06,
"loss": 0.7691,
"step": 2440
},
{
"epoch": 35.5,
"learning_rate": 2.8408383233532934e-06,
"loss": 0.8178,
"step": 2450
},
{
"epoch": 35.65,
"learning_rate": 2.837245508982036e-06,
"loss": 0.7588,
"step": 2460
},
{
"epoch": 35.79,
"learning_rate": 2.8336526946107787e-06,
"loss": 0.7684,
"step": 2470
},
{
"epoch": 35.94,
"learning_rate": 2.830059880239521e-06,
"loss": 0.7786,
"step": 2480
},
{
"epoch": 36.09,
"learning_rate": 2.8264670658682632e-06,
"loss": 0.7964,
"step": 2490
},
{
"epoch": 36.23,
"learning_rate": 2.8228742514970063e-06,
"loss": 0.8414,
"step": 2500
},
{
"epoch": 36.23,
"eval_loss": 0.4910103678703308,
"eval_runtime": 80.7809,
"eval_samples_per_second": 13.766,
"eval_steps_per_second": 3.441,
"eval_wer": 0.6130941704035875,
"step": 2500
},
{
"epoch": 36.37,
"learning_rate": 2.8192814371257486e-06,
"loss": 0.7242,
"step": 2510
},
{
"epoch": 36.52,
"learning_rate": 2.8156886227544912e-06,
"loss": 0.7916,
"step": 2520
},
{
"epoch": 36.66,
"learning_rate": 2.812095808383234e-06,
"loss": 0.7427,
"step": 2530
},
{
"epoch": 36.81,
"learning_rate": 2.808502994011976e-06,
"loss": 0.7362,
"step": 2540
},
{
"epoch": 36.95,
"learning_rate": 2.804910179640719e-06,
"loss": 0.7939,
"step": 2550
},
{
"epoch": 37.1,
"learning_rate": 2.801317365269461e-06,
"loss": 0.7787,
"step": 2560
},
{
"epoch": 37.24,
"learning_rate": 2.7977245508982037e-06,
"loss": 0.8297,
"step": 2570
},
{
"epoch": 37.39,
"learning_rate": 2.7941317365269464e-06,
"loss": 0.7301,
"step": 2580
},
{
"epoch": 37.53,
"learning_rate": 2.7905389221556886e-06,
"loss": 0.7588,
"step": 2590
},
{
"epoch": 37.68,
"learning_rate": 2.7869461077844313e-06,
"loss": 0.7481,
"step": 2600
},
{
"epoch": 37.82,
"learning_rate": 2.783353293413174e-06,
"loss": 0.7445,
"step": 2610
},
{
"epoch": 37.96,
"learning_rate": 2.7797604790419162e-06,
"loss": 0.7681,
"step": 2620
},
{
"epoch": 38.12,
"learning_rate": 2.7761676646706585e-06,
"loss": 0.759,
"step": 2630
},
{
"epoch": 38.26,
"learning_rate": 2.7725748502994016e-06,
"loss": 0.787,
"step": 2640
},
{
"epoch": 38.4,
"learning_rate": 2.768982035928144e-06,
"loss": 0.7404,
"step": 2650
},
{
"epoch": 38.55,
"learning_rate": 2.765389221556886e-06,
"loss": 0.7421,
"step": 2660
},
{
"epoch": 38.69,
"learning_rate": 2.7617964071856287e-06,
"loss": 0.7413,
"step": 2670
},
{
"epoch": 38.83,
"learning_rate": 2.7582035928143714e-06,
"loss": 0.7258,
"step": 2680
},
{
"epoch": 38.98,
"learning_rate": 2.7546107784431136e-06,
"loss": 0.7655,
"step": 2690
},
{
"epoch": 39.13,
"learning_rate": 2.7510179640718563e-06,
"loss": 0.7107,
"step": 2700
},
{
"epoch": 39.27,
"learning_rate": 2.747425149700599e-06,
"loss": 0.7439,
"step": 2710
},
{
"epoch": 39.42,
"learning_rate": 2.743832335329341e-06,
"loss": 0.7167,
"step": 2720
},
{
"epoch": 39.56,
"learning_rate": 2.740239520958084e-06,
"loss": 0.7142,
"step": 2730
},
{
"epoch": 39.71,
"learning_rate": 2.736646706586826e-06,
"loss": 0.7401,
"step": 2740
},
{
"epoch": 39.85,
"learning_rate": 2.733053892215569e-06,
"loss": 0.7247,
"step": 2750
},
{
"epoch": 39.99,
"learning_rate": 2.7294610778443115e-06,
"loss": 0.7702,
"step": 2760
},
{
"epoch": 40.14,
"learning_rate": 2.7258682634730537e-06,
"loss": 0.715,
"step": 2770
},
{
"epoch": 40.29,
"learning_rate": 2.722275449101797e-06,
"loss": 0.7349,
"step": 2780
},
{
"epoch": 40.43,
"learning_rate": 2.718682634730539e-06,
"loss": 0.7213,
"step": 2790
},
{
"epoch": 40.58,
"learning_rate": 2.7150898203592813e-06,
"loss": 0.7136,
"step": 2800
},
{
"epoch": 40.72,
"learning_rate": 2.711497005988024e-06,
"loss": 0.7592,
"step": 2810
},
{
"epoch": 40.86,
"learning_rate": 2.7079041916167666e-06,
"loss": 0.6764,
"step": 2820
},
{
"epoch": 41.01,
"learning_rate": 2.704311377245509e-06,
"loss": 0.7919,
"step": 2830
},
{
"epoch": 41.16,
"learning_rate": 2.7007185628742515e-06,
"loss": 0.662,
"step": 2840
},
{
"epoch": 41.3,
"learning_rate": 2.697125748502994e-06,
"loss": 0.6946,
"step": 2850
},
{
"epoch": 41.45,
"learning_rate": 2.6935329341317364e-06,
"loss": 0.7335,
"step": 2860
},
{
"epoch": 41.59,
"learning_rate": 2.689940119760479e-06,
"loss": 0.6908,
"step": 2870
},
{
"epoch": 41.73,
"learning_rate": 2.6863473053892214e-06,
"loss": 0.7628,
"step": 2880
},
{
"epoch": 41.88,
"learning_rate": 2.682754491017964e-06,
"loss": 0.679,
"step": 2890
},
{
"epoch": 42.03,
"learning_rate": 2.6791616766467067e-06,
"loss": 0.7441,
"step": 2900
},
{
"epoch": 42.17,
"learning_rate": 2.675568862275449e-06,
"loss": 0.6892,
"step": 2910
},
{
"epoch": 42.32,
"learning_rate": 2.6719760479041916e-06,
"loss": 0.7205,
"step": 2920
},
{
"epoch": 42.46,
"learning_rate": 2.6683832335329343e-06,
"loss": 0.7189,
"step": 2930
},
{
"epoch": 42.6,
"learning_rate": 2.6647904191616765e-06,
"loss": 0.6834,
"step": 2940
},
{
"epoch": 42.75,
"learning_rate": 2.661197604790419e-06,
"loss": 0.6674,
"step": 2950
},
{
"epoch": 42.89,
"learning_rate": 2.657604790419162e-06,
"loss": 0.6581,
"step": 2960
},
{
"epoch": 43.04,
"learning_rate": 2.654011976047904e-06,
"loss": 0.7261,
"step": 2970
},
{
"epoch": 43.19,
"learning_rate": 2.6504191616766468e-06,
"loss": 0.6802,
"step": 2980
},
{
"epoch": 43.33,
"learning_rate": 2.6468263473053894e-06,
"loss": 0.665,
"step": 2990
},
{
"epoch": 43.47,
"learning_rate": 2.6432335329341317e-06,
"loss": 0.6932,
"step": 3000
},
{
"epoch": 43.47,
"eval_loss": 0.4982631802558899,
"eval_runtime": 80.6907,
"eval_samples_per_second": 13.781,
"eval_steps_per_second": 3.445,
"eval_wer": 0.5912107623318386,
"step": 3000
},
{
"epoch": 43.62,
"learning_rate": 2.6396407185628744e-06,
"loss": 0.6916,
"step": 3010
},
{
"epoch": 43.76,
"learning_rate": 2.6360479041916166e-06,
"loss": 0.7148,
"step": 3020
},
{
"epoch": 43.91,
"learning_rate": 2.6324550898203593e-06,
"loss": 0.6593,
"step": 3030
},
{
"epoch": 44.06,
"learning_rate": 2.628862275449102e-06,
"loss": 0.7295,
"step": 3040
},
{
"epoch": 44.2,
"learning_rate": 2.625269461077844e-06,
"loss": 0.6767,
"step": 3050
},
{
"epoch": 44.35,
"learning_rate": 2.621676646706587e-06,
"loss": 0.6541,
"step": 3060
},
{
"epoch": 44.49,
"learning_rate": 2.6180838323353295e-06,
"loss": 0.7219,
"step": 3070
},
{
"epoch": 44.63,
"learning_rate": 2.6144910179640718e-06,
"loss": 0.6099,
"step": 3080
},
{
"epoch": 44.78,
"learning_rate": 2.6108982035928144e-06,
"loss": 0.701,
"step": 3090
},
{
"epoch": 44.92,
"learning_rate": 2.607305389221557e-06,
"loss": 0.6684,
"step": 3100
},
{
"epoch": 45.07,
"learning_rate": 2.6037125748502994e-06,
"loss": 0.6927,
"step": 3110
},
{
"epoch": 45.22,
"learning_rate": 2.600119760479042e-06,
"loss": 0.6896,
"step": 3120
},
{
"epoch": 45.36,
"learning_rate": 2.5965269461077847e-06,
"loss": 0.6419,
"step": 3130
},
{
"epoch": 45.5,
"learning_rate": 2.592934131736527e-06,
"loss": 0.7004,
"step": 3140
},
{
"epoch": 45.65,
"learning_rate": 2.5893413173652696e-06,
"loss": 0.6232,
"step": 3150
},
{
"epoch": 45.79,
"learning_rate": 2.585748502994012e-06,
"loss": 0.6833,
"step": 3160
},
{
"epoch": 45.94,
"learning_rate": 2.5821556886227545e-06,
"loss": 0.6662,
"step": 3170
},
{
"epoch": 46.09,
"learning_rate": 2.578562874251497e-06,
"loss": 0.6761,
"step": 3180
},
{
"epoch": 46.23,
"learning_rate": 2.5749700598802394e-06,
"loss": 0.6844,
"step": 3190
},
{
"epoch": 46.37,
"learning_rate": 2.571377245508982e-06,
"loss": 0.6176,
"step": 3200
},
{
"epoch": 46.52,
"learning_rate": 2.5677844311377248e-06,
"loss": 0.6937,
"step": 3210
},
{
"epoch": 46.66,
"learning_rate": 2.564191616766467e-06,
"loss": 0.6268,
"step": 3220
},
{
"epoch": 46.81,
"learning_rate": 2.5605988023952097e-06,
"loss": 0.6689,
"step": 3230
},
{
"epoch": 46.95,
"learning_rate": 2.5570059880239523e-06,
"loss": 0.6746,
"step": 3240
},
{
"epoch": 47.1,
"learning_rate": 2.5534131736526946e-06,
"loss": 0.6692,
"step": 3250
},
{
"epoch": 47.24,
"learning_rate": 2.5498203592814373e-06,
"loss": 0.6989,
"step": 3260
},
{
"epoch": 47.39,
"learning_rate": 2.54622754491018e-06,
"loss": 0.5954,
"step": 3270
},
{
"epoch": 47.53,
"learning_rate": 2.542634730538922e-06,
"loss": 0.6302,
"step": 3280
},
{
"epoch": 47.68,
"learning_rate": 2.539041916167665e-06,
"loss": 0.6569,
"step": 3290
},
{
"epoch": 47.82,
"learning_rate": 2.535449101796407e-06,
"loss": 0.6566,
"step": 3300
},
{
"epoch": 47.96,
"learning_rate": 2.5318562874251498e-06,
"loss": 0.6705,
"step": 3310
},
{
"epoch": 48.12,
"learning_rate": 2.5282634730538924e-06,
"loss": 0.6465,
"step": 3320
},
{
"epoch": 48.26,
"learning_rate": 2.5246706586826347e-06,
"loss": 0.6533,
"step": 3330
},
{
"epoch": 48.4,
"learning_rate": 2.5210778443113773e-06,
"loss": 0.6155,
"step": 3340
},
{
"epoch": 48.55,
"learning_rate": 2.51748502994012e-06,
"loss": 0.6535,
"step": 3350
},
{
"epoch": 48.69,
"learning_rate": 2.5138922155688623e-06,
"loss": 0.638,
"step": 3360
},
{
"epoch": 48.83,
"learning_rate": 2.5102994011976045e-06,
"loss": 0.625,
"step": 3370
},
{
"epoch": 48.98,
"learning_rate": 2.5067065868263476e-06,
"loss": 0.681,
"step": 3380
},
{
"epoch": 49.13,
"learning_rate": 2.50311377245509e-06,
"loss": 0.6304,
"step": 3390
},
{
"epoch": 49.27,
"learning_rate": 2.4995209580838325e-06,
"loss": 0.6438,
"step": 3400
},
{
"epoch": 49.42,
"learning_rate": 2.495928143712575e-06,
"loss": 0.622,
"step": 3410
},
{
"epoch": 49.56,
"learning_rate": 2.4923353293413174e-06,
"loss": 0.6357,
"step": 3420
},
{
"epoch": 49.71,
"learning_rate": 2.48874251497006e-06,
"loss": 0.6538,
"step": 3430
},
{
"epoch": 49.85,
"learning_rate": 2.4851497005988023e-06,
"loss": 0.6229,
"step": 3440
},
{
"epoch": 49.99,
"learning_rate": 2.481556886227545e-06,
"loss": 0.6526,
"step": 3450
},
{
"epoch": 50.14,
"learning_rate": 2.4779640718562877e-06,
"loss": 0.629,
"step": 3460
},
{
"epoch": 50.29,
"learning_rate": 2.47437125748503e-06,
"loss": 0.6474,
"step": 3470
},
{
"epoch": 50.43,
"learning_rate": 2.4707784431137726e-06,
"loss": 0.6075,
"step": 3480
},
{
"epoch": 50.58,
"learning_rate": 2.4671856287425153e-06,
"loss": 0.641,
"step": 3490
},
{
"epoch": 50.72,
"learning_rate": 2.4635928143712575e-06,
"loss": 0.6191,
"step": 3500
},
{
"epoch": 50.72,
"eval_loss": 0.46035492420196533,
"eval_runtime": 80.9575,
"eval_samples_per_second": 13.736,
"eval_steps_per_second": 3.434,
"eval_wer": 0.5732735426008969,
"step": 3500
},
{
"epoch": 50.86,
"learning_rate": 2.4599999999999997e-06,
"loss": 0.6304,
"step": 3510
},
{
"epoch": 51.01,
"learning_rate": 2.456407185628743e-06,
"loss": 0.6854,
"step": 3520
},
{
"epoch": 51.16,
"learning_rate": 2.452814371257485e-06,
"loss": 0.5921,
"step": 3530
},
{
"epoch": 51.3,
"learning_rate": 2.4492215568862273e-06,
"loss": 0.6037,
"step": 3540
},
{
"epoch": 51.45,
"learning_rate": 2.4456287425149704e-06,
"loss": 0.6371,
"step": 3550
},
{
"epoch": 51.59,
"learning_rate": 2.4420359281437127e-06,
"loss": 0.6078,
"step": 3560
},
{
"epoch": 51.73,
"learning_rate": 2.438443113772455e-06,
"loss": 0.6275,
"step": 3570
},
{
"epoch": 51.88,
"learning_rate": 2.4348502994011976e-06,
"loss": 0.5948,
"step": 3580
},
{
"epoch": 52.03,
"learning_rate": 2.4312574850299402e-06,
"loss": 0.6681,
"step": 3590
},
{
"epoch": 52.17,
"learning_rate": 2.4276646706586825e-06,
"loss": 0.5871,
"step": 3600
},
{
"epoch": 52.32,
"learning_rate": 2.424071856287425e-06,
"loss": 0.5858,
"step": 3610
},
{
"epoch": 52.46,
"learning_rate": 2.420479041916168e-06,
"loss": 0.6237,
"step": 3620
},
{
"epoch": 52.6,
"learning_rate": 2.4168862275449105e-06,
"loss": 0.5776,
"step": 3630
},
{
"epoch": 52.75,
"learning_rate": 2.4132934131736527e-06,
"loss": 0.6677,
"step": 3640
},
{
"epoch": 52.89,
"learning_rate": 2.409700598802395e-06,
"loss": 0.6013,
"step": 3650
},
{
"epoch": 53.04,
"learning_rate": 2.406107784431138e-06,
"loss": 0.6523,
"step": 3660
},
{
"epoch": 53.19,
"learning_rate": 2.4025149700598803e-06,
"loss": 0.5959,
"step": 3670
},
{
"epoch": 53.33,
"learning_rate": 2.3989221556886226e-06,
"loss": 0.6164,
"step": 3680
},
{
"epoch": 53.47,
"learning_rate": 2.3953293413173657e-06,
"loss": 0.6211,
"step": 3690
},
{
"epoch": 53.62,
"learning_rate": 2.391736526946108e-06,
"loss": 0.5589,
"step": 3700
},
{
"epoch": 53.76,
"learning_rate": 2.38814371257485e-06,
"loss": 0.6315,
"step": 3710
},
{
"epoch": 53.91,
"learning_rate": 2.384550898203593e-06,
"loss": 0.6044,
"step": 3720
},
{
"epoch": 54.06,
"learning_rate": 2.3809580838323355e-06,
"loss": 0.6244,
"step": 3730
},
{
"epoch": 54.2,
"learning_rate": 2.3773652694610777e-06,
"loss": 0.61,
"step": 3740
},
{
"epoch": 54.35,
"learning_rate": 2.3737724550898204e-06,
"loss": 0.5776,
"step": 3750
},
{
"epoch": 54.49,
"learning_rate": 2.370179640718563e-06,
"loss": 0.6291,
"step": 3760
},
{
"epoch": 54.63,
"learning_rate": 2.3665868263473053e-06,
"loss": 0.589,
"step": 3770
},
{
"epoch": 54.78,
"learning_rate": 2.362994011976048e-06,
"loss": 0.6022,
"step": 3780
},
{
"epoch": 54.92,
"learning_rate": 2.3594011976047902e-06,
"loss": 0.6051,
"step": 3790
},
{
"epoch": 55.07,
"learning_rate": 2.355808383233533e-06,
"loss": 0.6141,
"step": 3800
},
{
"epoch": 55.22,
"learning_rate": 2.3522155688622756e-06,
"loss": 0.6146,
"step": 3810
},
{
"epoch": 55.36,
"learning_rate": 2.348622754491018e-06,
"loss": 0.5678,
"step": 3820
},
{
"epoch": 55.5,
"learning_rate": 2.3450299401197605e-06,
"loss": 0.6029,
"step": 3830
},
{
"epoch": 55.65,
"learning_rate": 2.341437125748503e-06,
"loss": 0.5621,
"step": 3840
},
{
"epoch": 55.79,
"learning_rate": 2.3378443113772454e-06,
"loss": 0.6003,
"step": 3850
},
{
"epoch": 55.94,
"learning_rate": 2.334251497005988e-06,
"loss": 0.5909,
"step": 3860
},
{
"epoch": 56.09,
"learning_rate": 2.3306586826347307e-06,
"loss": 0.6047,
"step": 3870
},
{
"epoch": 56.23,
"learning_rate": 2.327065868263473e-06,
"loss": 0.5976,
"step": 3880
},
{
"epoch": 56.37,
"learning_rate": 2.3234730538922156e-06,
"loss": 0.554,
"step": 3890
},
{
"epoch": 56.52,
"learning_rate": 2.3198802395209583e-06,
"loss": 0.6121,
"step": 3900
},
{
"epoch": 56.66,
"learning_rate": 2.3162874251497006e-06,
"loss": 0.571,
"step": 3910
},
{
"epoch": 56.81,
"learning_rate": 2.3126946107784432e-06,
"loss": 0.5974,
"step": 3920
},
{
"epoch": 56.95,
"learning_rate": 2.3091017964071855e-06,
"loss": 0.6142,
"step": 3930
},
{
"epoch": 57.1,
"learning_rate": 2.305508982035928e-06,
"loss": 0.594,
"step": 3940
},
{
"epoch": 57.24,
"learning_rate": 2.301916167664671e-06,
"loss": 0.6407,
"step": 3950
},
{
"epoch": 57.39,
"learning_rate": 2.298323353293413e-06,
"loss": 0.5777,
"step": 3960
},
{
"epoch": 57.53,
"learning_rate": 2.2947305389221557e-06,
"loss": 0.6078,
"step": 3970
},
{
"epoch": 57.68,
"learning_rate": 2.2911377245508984e-06,
"loss": 0.5771,
"step": 3980
},
{
"epoch": 57.82,
"learning_rate": 2.2875449101796406e-06,
"loss": 0.5865,
"step": 3990
},
{
"epoch": 57.96,
"learning_rate": 2.2839520958083833e-06,
"loss": 0.6102,
"step": 4000
},
{
"epoch": 57.96,
"eval_loss": 0.4474981129169464,
"eval_runtime": 80.9709,
"eval_samples_per_second": 13.733,
"eval_steps_per_second": 3.433,
"eval_wer": 0.5743497757847533,
"step": 4000
},
{
"epoch": 58.12,
"learning_rate": 2.280359281437126e-06,
"loss": 0.5916,
"step": 4010
},
{
"epoch": 58.26,
"learning_rate": 2.2767664670658682e-06,
"loss": 0.5633,
"step": 4020
},
{
"epoch": 58.4,
"learning_rate": 2.273173652694611e-06,
"loss": 0.5488,
"step": 4030
},
{
"epoch": 58.55,
"learning_rate": 2.2695808383233536e-06,
"loss": 0.5994,
"step": 4040
},
{
"epoch": 58.69,
"learning_rate": 2.265988023952096e-06,
"loss": 0.5759,
"step": 4050
},
{
"epoch": 58.83,
"learning_rate": 2.2623952095808385e-06,
"loss": 0.5603,
"step": 4060
},
{
"epoch": 58.98,
"learning_rate": 2.2588023952095807e-06,
"loss": 0.5872,
"step": 4070
},
{
"epoch": 59.13,
"learning_rate": 2.2552095808383234e-06,
"loss": 0.5874,
"step": 4080
},
{
"epoch": 59.27,
"learning_rate": 2.251616766467066e-06,
"loss": 0.5918,
"step": 4090
},
{
"epoch": 59.42,
"learning_rate": 2.2480239520958083e-06,
"loss": 0.562,
"step": 4100
},
{
"epoch": 59.56,
"learning_rate": 2.244431137724551e-06,
"loss": 0.5606,
"step": 4110
},
{
"epoch": 59.71,
"learning_rate": 2.2408383233532936e-06,
"loss": 0.5868,
"step": 4120
},
{
"epoch": 59.85,
"learning_rate": 2.237245508982036e-06,
"loss": 0.5859,
"step": 4130
},
{
"epoch": 59.99,
"learning_rate": 2.2336526946107785e-06,
"loss": 0.5842,
"step": 4140
},
{
"epoch": 60.14,
"learning_rate": 2.230059880239521e-06,
"loss": 0.5699,
"step": 4150
},
{
"epoch": 60.29,
"learning_rate": 2.2264670658682635e-06,
"loss": 0.5861,
"step": 4160
},
{
"epoch": 60.43,
"learning_rate": 2.222874251497006e-06,
"loss": 0.5621,
"step": 4170
},
{
"epoch": 60.58,
"learning_rate": 2.219281437125749e-06,
"loss": 0.5708,
"step": 4180
},
{
"epoch": 60.72,
"learning_rate": 2.215688622754491e-06,
"loss": 0.5743,
"step": 4190
},
{
"epoch": 60.86,
"learning_rate": 2.2120958083832337e-06,
"loss": 0.5367,
"step": 4200
},
{
"epoch": 61.01,
"learning_rate": 2.208502994011976e-06,
"loss": 0.6458,
"step": 4210
},
{
"epoch": 61.16,
"learning_rate": 2.2049101796407186e-06,
"loss": 0.5474,
"step": 4220
},
{
"epoch": 61.3,
"learning_rate": 2.2013173652694613e-06,
"loss": 0.5463,
"step": 4230
},
{
"epoch": 61.45,
"learning_rate": 2.1977245508982035e-06,
"loss": 0.5747,
"step": 4240
},
{
"epoch": 61.59,
"learning_rate": 2.194131736526946e-06,
"loss": 0.5529,
"step": 4250
},
{
"epoch": 61.73,
"learning_rate": 2.190538922155689e-06,
"loss": 0.5905,
"step": 4260
},
{
"epoch": 61.88,
"learning_rate": 2.186946107784431e-06,
"loss": 0.5203,
"step": 4270
},
{
"epoch": 62.03,
"learning_rate": 2.1833532934131738e-06,
"loss": 0.5807,
"step": 4280
},
{
"epoch": 62.17,
"learning_rate": 2.1797604790419165e-06,
"loss": 0.5689,
"step": 4290
},
{
"epoch": 62.32,
"learning_rate": 2.1761676646706587e-06,
"loss": 0.5615,
"step": 4300
},
{
"epoch": 62.46,
"learning_rate": 2.1725748502994014e-06,
"loss": 0.5708,
"step": 4310
},
{
"epoch": 62.6,
"learning_rate": 2.168982035928144e-06,
"loss": 0.5625,
"step": 4320
},
{
"epoch": 62.75,
"learning_rate": 2.1653892215568863e-06,
"loss": 0.5837,
"step": 4330
},
{
"epoch": 62.89,
"learning_rate": 2.161796407185629e-06,
"loss": 0.5387,
"step": 4340
},
{
"epoch": 63.04,
"learning_rate": 2.158203592814371e-06,
"loss": 0.6188,
"step": 4350
},
{
"epoch": 63.19,
"learning_rate": 2.154610778443114e-06,
"loss": 0.5374,
"step": 4360
},
{
"epoch": 63.33,
"learning_rate": 2.1510179640718565e-06,
"loss": 0.5437,
"step": 4370
},
{
"epoch": 63.47,
"learning_rate": 2.1474251497005988e-06,
"loss": 0.5877,
"step": 4380
},
{
"epoch": 63.62,
"learning_rate": 2.1438323353293414e-06,
"loss": 0.5315,
"step": 4390
},
{
"epoch": 63.76,
"learning_rate": 2.140239520958084e-06,
"loss": 0.5548,
"step": 4400
},
{
"epoch": 63.91,
"learning_rate": 2.1366467065868264e-06,
"loss": 0.5295,
"step": 4410
},
{
"epoch": 64.06,
"learning_rate": 2.1330538922155686e-06,
"loss": 0.5922,
"step": 4420
},
{
"epoch": 64.2,
"learning_rate": 2.1294610778443117e-06,
"loss": 0.5589,
"step": 4430
},
{
"epoch": 64.35,
"learning_rate": 2.125868263473054e-06,
"loss": 0.5451,
"step": 4440
},
{
"epoch": 64.49,
"learning_rate": 2.122275449101796e-06,
"loss": 0.6053,
"step": 4450
},
{
"epoch": 64.63,
"learning_rate": 2.1186826347305393e-06,
"loss": 0.5375,
"step": 4460
},
{
"epoch": 64.78,
"learning_rate": 2.1150898203592815e-06,
"loss": 0.5256,
"step": 4470
},
{
"epoch": 64.92,
"learning_rate": 2.1114970059880238e-06,
"loss": 0.5475,
"step": 4480
},
{
"epoch": 65.07,
"learning_rate": 2.1079041916167664e-06,
"loss": 0.5621,
"step": 4490
},
{
"epoch": 65.22,
"learning_rate": 2.104311377245509e-06,
"loss": 0.5607,
"step": 4500
},
{
"epoch": 65.22,
"eval_loss": 0.42857515811920166,
"eval_runtime": 80.8819,
"eval_samples_per_second": 13.748,
"eval_steps_per_second": 3.437,
"eval_wer": 0.5589237668161435,
"step": 4500
},
{
"epoch": 65.36,
"learning_rate": 2.1007185628742518e-06,
"loss": 0.5288,
"step": 4510
},
{
"epoch": 65.5,
"learning_rate": 2.097125748502994e-06,
"loss": 0.5573,
"step": 4520
},
{
"epoch": 65.65,
"learning_rate": 2.0935329341317367e-06,
"loss": 0.5314,
"step": 4530
},
{
"epoch": 65.79,
"learning_rate": 2.0899401197604794e-06,
"loss": 0.5628,
"step": 4540
},
{
"epoch": 65.94,
"learning_rate": 2.0863473053892216e-06,
"loss": 0.5499,
"step": 4550
},
{
"epoch": 66.09,
"learning_rate": 2.082754491017964e-06,
"loss": 0.5716,
"step": 4560
},
{
"epoch": 66.23,
"learning_rate": 2.079161676646707e-06,
"loss": 0.5617,
"step": 4570
},
{
"epoch": 66.37,
"learning_rate": 2.075568862275449e-06,
"loss": 0.5167,
"step": 4580
},
{
"epoch": 66.52,
"learning_rate": 2.0719760479041914e-06,
"loss": 0.5668,
"step": 4590
},
{
"epoch": 66.66,
"learning_rate": 2.0683832335329345e-06,
"loss": 0.5533,
"step": 4600
},
{
"epoch": 66.81,
"learning_rate": 2.0647904191616768e-06,
"loss": 0.5373,
"step": 4610
},
{
"epoch": 66.95,
"learning_rate": 2.061197604790419e-06,
"loss": 0.5612,
"step": 4620
},
{
"epoch": 67.1,
"learning_rate": 2.0576047904191617e-06,
"loss": 0.5395,
"step": 4630
},
{
"epoch": 67.24,
"learning_rate": 2.0540119760479043e-06,
"loss": 0.5766,
"step": 4640
},
{
"epoch": 67.39,
"learning_rate": 2.0504191616766466e-06,
"loss": 0.5166,
"step": 4650
},
{
"epoch": 67.53,
"learning_rate": 2.0468263473053893e-06,
"loss": 0.5726,
"step": 4660
},
{
"epoch": 67.68,
"learning_rate": 2.043233532934132e-06,
"loss": 0.5259,
"step": 4670
},
{
"epoch": 67.82,
"learning_rate": 2.039640718562874e-06,
"loss": 0.5266,
"step": 4680
},
{
"epoch": 67.96,
"learning_rate": 2.036047904191617e-06,
"loss": 0.5713,
"step": 4690
},
{
"epoch": 68.12,
"learning_rate": 2.032455089820359e-06,
"loss": 0.5634,
"step": 4700
},
{
"epoch": 68.26,
"learning_rate": 2.0288622754491018e-06,
"loss": 0.5545,
"step": 4710
},
{
"epoch": 68.4,
"learning_rate": 2.0252694610778444e-06,
"loss": 0.5101,
"step": 4720
},
{
"epoch": 68.55,
"learning_rate": 2.0216766467065867e-06,
"loss": 0.5611,
"step": 4730
},
{
"epoch": 68.69,
"learning_rate": 2.0180838323353293e-06,
"loss": 0.5337,
"step": 4740
},
{
"epoch": 68.83,
"learning_rate": 2.014491017964072e-06,
"loss": 0.5167,
"step": 4750
},
{
"epoch": 68.98,
"learning_rate": 2.0108982035928143e-06,
"loss": 0.5796,
"step": 4760
},
{
"epoch": 69.13,
"learning_rate": 2.007305389221557e-06,
"loss": 0.5183,
"step": 4770
},
{
"epoch": 69.27,
"learning_rate": 2.0037125748502996e-06,
"loss": 0.5395,
"step": 4780
},
{
"epoch": 69.42,
"learning_rate": 2.000119760479042e-06,
"loss": 0.5188,
"step": 4790
},
{
"epoch": 69.56,
"learning_rate": 1.9965269461077845e-06,
"loss": 0.5366,
"step": 4800
},
{
"epoch": 69.71,
"learning_rate": 1.9929341317365267e-06,
"loss": 0.5493,
"step": 4810
},
{
"epoch": 69.85,
"learning_rate": 1.9893413173652694e-06,
"loss": 0.524,
"step": 4820
},
{
"epoch": 69.99,
"learning_rate": 1.985748502994012e-06,
"loss": 0.5601,
"step": 4830
},
{
"epoch": 70.14,
"learning_rate": 1.9821556886227543e-06,
"loss": 0.5134,
"step": 4840
},
{
"epoch": 70.29,
"learning_rate": 1.978562874251497e-06,
"loss": 0.5296,
"step": 4850
},
{
"epoch": 70.43,
"learning_rate": 1.9749700598802397e-06,
"loss": 0.5411,
"step": 4860
},
{
"epoch": 70.58,
"learning_rate": 1.971377245508982e-06,
"loss": 0.5255,
"step": 4870
},
{
"epoch": 70.72,
"learning_rate": 1.9677844311377246e-06,
"loss": 0.5417,
"step": 4880
},
{
"epoch": 70.86,
"learning_rate": 1.9641916167664672e-06,
"loss": 0.5164,
"step": 4890
},
{
"epoch": 71.01,
"learning_rate": 1.9605988023952095e-06,
"loss": 0.5867,
"step": 4900
},
{
"epoch": 71.16,
"learning_rate": 1.957005988023952e-06,
"loss": 0.5307,
"step": 4910
},
{
"epoch": 71.3,
"learning_rate": 1.953413173652695e-06,
"loss": 0.5333,
"step": 4920
},
{
"epoch": 71.45,
"learning_rate": 1.949820359281437e-06,
"loss": 0.5522,
"step": 4930
},
{
"epoch": 71.59,
"learning_rate": 1.9462275449101797e-06,
"loss": 0.5099,
"step": 4940
},
{
"epoch": 71.73,
"learning_rate": 1.942634730538922e-06,
"loss": 0.5319,
"step": 4950
},
{
"epoch": 71.88,
"learning_rate": 1.9390419161676647e-06,
"loss": 0.4936,
"step": 4960
},
{
"epoch": 72.03,
"learning_rate": 1.9354491017964073e-06,
"loss": 0.5462,
"step": 4970
},
{
"epoch": 72.17,
"learning_rate": 1.9318562874251496e-06,
"loss": 0.515,
"step": 4980
},
{
"epoch": 72.32,
"learning_rate": 1.9282634730538922e-06,
"loss": 0.4875,
"step": 4990
},
{
"epoch": 72.46,
"learning_rate": 1.924670658682635e-06,
"loss": 0.5493,
"step": 5000
},
{
"epoch": 72.46,
"eval_loss": 0.43060895800590515,
"eval_runtime": 80.8776,
"eval_samples_per_second": 13.749,
"eval_steps_per_second": 3.437,
"eval_wer": 0.5601793721973094,
"step": 5000
},
{
"epoch": 72.6,
"learning_rate": 1.921077844311377e-06,
"loss": 0.4906,
"step": 5010
},
{
"epoch": 72.75,
"learning_rate": 1.91748502994012e-06,
"loss": 0.5431,
"step": 5020
},
{
"epoch": 72.89,
"learning_rate": 1.9138922155688625e-06,
"loss": 0.5189,
"step": 5030
},
{
"epoch": 73.04,
"learning_rate": 1.9102994011976047e-06,
"loss": 0.569,
"step": 5040
},
{
"epoch": 73.19,
"learning_rate": 1.9067065868263472e-06,
"loss": 0.5318,
"step": 5050
},
{
"epoch": 73.33,
"learning_rate": 1.90311377245509e-06,
"loss": 0.4812,
"step": 5060
},
{
"epoch": 73.47,
"learning_rate": 1.8995209580838323e-06,
"loss": 0.5309,
"step": 5070
},
{
"epoch": 73.62,
"learning_rate": 1.8959281437125748e-06,
"loss": 0.5103,
"step": 5080
},
{
"epoch": 73.76,
"learning_rate": 1.8923353293413172e-06,
"loss": 0.553,
"step": 5090
},
{
"epoch": 73.91,
"learning_rate": 1.8887425149700601e-06,
"loss": 0.5167,
"step": 5100
},
{
"epoch": 74.06,
"learning_rate": 1.8851497005988024e-06,
"loss": 0.5641,
"step": 5110
},
{
"epoch": 74.2,
"learning_rate": 1.8815568862275448e-06,
"loss": 0.5204,
"step": 5120
},
{
"epoch": 74.35,
"learning_rate": 1.8779640718562877e-06,
"loss": 0.4981,
"step": 5130
},
{
"epoch": 74.49,
"learning_rate": 1.87437125748503e-06,
"loss": 0.5504,
"step": 5140
},
{
"epoch": 74.63,
"learning_rate": 1.8707784431137724e-06,
"loss": 0.5157,
"step": 5150
},
{
"epoch": 74.78,
"learning_rate": 1.8671856287425149e-06,
"loss": 0.503,
"step": 5160
},
{
"epoch": 74.92,
"learning_rate": 1.8635928143712575e-06,
"loss": 0.5407,
"step": 5170
},
{
"epoch": 75.07,
"learning_rate": 1.86e-06,
"loss": 0.5292,
"step": 5180
},
{
"epoch": 75.22,
"learning_rate": 1.8564071856287424e-06,
"loss": 0.5388,
"step": 5190
},
{
"epoch": 75.36,
"learning_rate": 1.852814371257485e-06,
"loss": 0.4909,
"step": 5200
},
{
"epoch": 75.5,
"learning_rate": 1.8492215568862276e-06,
"loss": 0.5203,
"step": 5210
},
{
"epoch": 75.65,
"learning_rate": 1.84562874251497e-06,
"loss": 0.4935,
"step": 5220
},
{
"epoch": 75.79,
"learning_rate": 1.8420359281437125e-06,
"loss": 0.5351,
"step": 5230
},
{
"epoch": 75.94,
"learning_rate": 1.8384431137724551e-06,
"loss": 0.5045,
"step": 5240
},
{
"epoch": 76.09,
"learning_rate": 1.8348502994011976e-06,
"loss": 0.5346,
"step": 5250
},
{
"epoch": 76.23,
"learning_rate": 1.83125748502994e-06,
"loss": 0.5302,
"step": 5260
},
{
"epoch": 76.37,
"learning_rate": 1.8276646706586827e-06,
"loss": 0.5012,
"step": 5270
},
{
"epoch": 76.52,
"learning_rate": 1.8240718562874252e-06,
"loss": 0.5178,
"step": 5280
},
{
"epoch": 76.66,
"learning_rate": 1.8204790419161676e-06,
"loss": 0.5078,
"step": 5290
},
{
"epoch": 76.81,
"learning_rate": 1.81688622754491e-06,
"loss": 0.4985,
"step": 5300
},
{
"epoch": 76.95,
"learning_rate": 1.8132934131736528e-06,
"loss": 0.5211,
"step": 5310
},
{
"epoch": 77.1,
"learning_rate": 1.8097005988023952e-06,
"loss": 0.5114,
"step": 5320
},
{
"epoch": 77.24,
"learning_rate": 1.8061077844311377e-06,
"loss": 0.5322,
"step": 5330
},
{
"epoch": 77.39,
"learning_rate": 1.8025149700598803e-06,
"loss": 0.4981,
"step": 5340
},
{
"epoch": 77.53,
"learning_rate": 1.7989221556886228e-06,
"loss": 0.5092,
"step": 5350
},
{
"epoch": 77.68,
"learning_rate": 1.7953293413173653e-06,
"loss": 0.4872,
"step": 5360
},
{
"epoch": 77.82,
"learning_rate": 1.7917365269461077e-06,
"loss": 0.4878,
"step": 5370
},
{
"epoch": 77.96,
"learning_rate": 1.7881437125748504e-06,
"loss": 0.5299,
"step": 5380
},
{
"epoch": 78.12,
"learning_rate": 1.7845508982035928e-06,
"loss": 0.5339,
"step": 5390
},
{
"epoch": 78.26,
"learning_rate": 1.7809580838323353e-06,
"loss": 0.5057,
"step": 5400
},
{
"epoch": 78.4,
"learning_rate": 1.777365269461078e-06,
"loss": 0.4833,
"step": 5410
},
{
"epoch": 78.55,
"learning_rate": 1.7737724550898204e-06,
"loss": 0.5002,
"step": 5420
},
{
"epoch": 78.69,
"learning_rate": 1.7701796407185629e-06,
"loss": 0.5115,
"step": 5430
},
{
"epoch": 78.83,
"learning_rate": 1.7665868263473053e-06,
"loss": 0.4899,
"step": 5440
},
{
"epoch": 78.98,
"learning_rate": 1.762994011976048e-06,
"loss": 0.4816,
"step": 5450
},
{
"epoch": 79.13,
"learning_rate": 1.7594011976047905e-06,
"loss": 0.5268,
"step": 5460
},
{
"epoch": 79.27,
"learning_rate": 1.755808383233533e-06,
"loss": 0.5136,
"step": 5470
},
{
"epoch": 79.42,
"learning_rate": 1.7522155688622756e-06,
"loss": 0.5003,
"step": 5480
},
{
"epoch": 79.56,
"learning_rate": 1.748622754491018e-06,
"loss": 0.4993,
"step": 5490
},
{
"epoch": 79.71,
"learning_rate": 1.7450299401197605e-06,
"loss": 0.5123,
"step": 5500
},
{
"epoch": 79.71,
"eval_loss": 0.4094592332839966,
"eval_runtime": 81.1928,
"eval_samples_per_second": 13.696,
"eval_steps_per_second": 3.424,
"eval_wer": 0.5321973094170404,
"step": 5500
},
{
"epoch": 79.85,
"learning_rate": 1.741437125748503e-06,
"loss": 0.4828,
"step": 5510
},
{
"epoch": 79.99,
"learning_rate": 1.7378443113772456e-06,
"loss": 0.5291,
"step": 5520
},
{
"epoch": 80.14,
"learning_rate": 1.734251497005988e-06,
"loss": 0.5148,
"step": 5530
},
{
"epoch": 80.29,
"learning_rate": 1.7306586826347305e-06,
"loss": 0.5161,
"step": 5540
},
{
"epoch": 80.43,
"learning_rate": 1.7270658682634732e-06,
"loss": 0.4873,
"step": 5550
},
{
"epoch": 80.58,
"learning_rate": 1.7234730538922157e-06,
"loss": 0.4778,
"step": 5560
},
{
"epoch": 80.72,
"learning_rate": 1.7198802395209581e-06,
"loss": 0.511,
"step": 5570
},
{
"epoch": 80.86,
"learning_rate": 1.7162874251497006e-06,
"loss": 0.492,
"step": 5580
},
{
"epoch": 81.01,
"learning_rate": 1.7126946107784432e-06,
"loss": 0.5407,
"step": 5590
},
{
"epoch": 81.16,
"learning_rate": 1.7091017964071857e-06,
"loss": 0.4582,
"step": 5600
},
{
"epoch": 81.3,
"learning_rate": 1.7055089820359282e-06,
"loss": 0.4732,
"step": 5610
},
{
"epoch": 81.45,
"learning_rate": 1.7019161676646708e-06,
"loss": 0.4921,
"step": 5620
},
{
"epoch": 81.59,
"learning_rate": 1.6983233532934133e-06,
"loss": 0.4952,
"step": 5630
},
{
"epoch": 81.73,
"learning_rate": 1.6947305389221557e-06,
"loss": 0.54,
"step": 5640
},
{
"epoch": 81.88,
"learning_rate": 1.6911377245508982e-06,
"loss": 0.4798,
"step": 5650
},
{
"epoch": 82.03,
"learning_rate": 1.6875449101796409e-06,
"loss": 0.5554,
"step": 5660
},
{
"epoch": 82.17,
"learning_rate": 1.6839520958083833e-06,
"loss": 0.493,
"step": 5670
},
{
"epoch": 82.32,
"learning_rate": 1.6803592814371258e-06,
"loss": 0.485,
"step": 5680
},
{
"epoch": 82.46,
"learning_rate": 1.6767664670658685e-06,
"loss": 0.5041,
"step": 5690
},
{
"epoch": 82.6,
"learning_rate": 1.673173652694611e-06,
"loss": 0.4775,
"step": 5700
},
{
"epoch": 82.75,
"learning_rate": 1.6695808383233534e-06,
"loss": 0.5186,
"step": 5710
},
{
"epoch": 82.89,
"learning_rate": 1.6659880239520956e-06,
"loss": 0.4729,
"step": 5720
},
{
"epoch": 83.04,
"learning_rate": 1.6623952095808385e-06,
"loss": 0.4964,
"step": 5730
},
{
"epoch": 83.19,
"learning_rate": 1.658802395209581e-06,
"loss": 0.4981,
"step": 5740
},
{
"epoch": 83.33,
"learning_rate": 1.6552095808383234e-06,
"loss": 0.4994,
"step": 5750
},
{
"epoch": 83.47,
"learning_rate": 1.651616766467066e-06,
"loss": 0.4876,
"step": 5760
},
{
"epoch": 83.62,
"learning_rate": 1.6480239520958085e-06,
"loss": 0.4461,
"step": 5770
},
{
"epoch": 83.76,
"learning_rate": 1.644431137724551e-06,
"loss": 0.4935,
"step": 5780
},
{
"epoch": 83.91,
"learning_rate": 1.6408383233532932e-06,
"loss": 0.4767,
"step": 5790
},
{
"epoch": 84.06,
"learning_rate": 1.6372455089820361e-06,
"loss": 0.4994,
"step": 5800
},
{
"epoch": 84.2,
"learning_rate": 1.6336526946107786e-06,
"loss": 0.4824,
"step": 5810
},
{
"epoch": 84.35,
"learning_rate": 1.6300598802395208e-06,
"loss": 0.4765,
"step": 5820
},
{
"epoch": 84.49,
"learning_rate": 1.6264670658682637e-06,
"loss": 0.5287,
"step": 5830
},
{
"epoch": 84.63,
"learning_rate": 1.6228742514970062e-06,
"loss": 0.4706,
"step": 5840
},
{
"epoch": 84.78,
"learning_rate": 1.6192814371257484e-06,
"loss": 0.4924,
"step": 5850
},
{
"epoch": 84.92,
"learning_rate": 1.6156886227544909e-06,
"loss": 0.4853,
"step": 5860
},
{
"epoch": 85.07,
"learning_rate": 1.6120958083832337e-06,
"loss": 0.4864,
"step": 5870
},
{
"epoch": 85.22,
"learning_rate": 1.6085029940119762e-06,
"loss": 0.494,
"step": 5880
},
{
"epoch": 85.36,
"learning_rate": 1.6049101796407184e-06,
"loss": 0.4671,
"step": 5890
},
{
"epoch": 85.5,
"learning_rate": 1.6013173652694613e-06,
"loss": 0.5037,
"step": 5900
},
{
"epoch": 85.65,
"learning_rate": 1.5977245508982038e-06,
"loss": 0.4588,
"step": 5910
},
{
"epoch": 85.79,
"learning_rate": 1.594131736526946e-06,
"loss": 0.4731,
"step": 5920
},
{
"epoch": 85.94,
"learning_rate": 1.5905389221556885e-06,
"loss": 0.5032,
"step": 5930
},
{
"epoch": 86.09,
"learning_rate": 1.5869461077844314e-06,
"loss": 0.5026,
"step": 5940
},
{
"epoch": 86.23,
"learning_rate": 1.5833532934131736e-06,
"loss": 0.5213,
"step": 5950
},
{
"epoch": 86.37,
"learning_rate": 1.579760479041916e-06,
"loss": 0.4779,
"step": 5960
},
{
"epoch": 86.52,
"learning_rate": 1.576167664670659e-06,
"loss": 0.5014,
"step": 5970
},
{
"epoch": 86.66,
"learning_rate": 1.5725748502994014e-06,
"loss": 0.4869,
"step": 5980
},
{
"epoch": 86.81,
"learning_rate": 1.5689820359281436e-06,
"loss": 0.4932,
"step": 5990
},
{
"epoch": 86.95,
"learning_rate": 1.565389221556886e-06,
"loss": 0.5113,
"step": 6000
},
{
"epoch": 86.95,
"eval_loss": 0.4200317859649658,
"eval_runtime": 81.115,
"eval_samples_per_second": 13.709,
"eval_steps_per_second": 3.427,
"eval_wer": 0.5370403587443946,
"step": 6000
},
{
"epoch": 87.1,
"learning_rate": 1.561796407185629e-06,
"loss": 0.4879,
"step": 6010
},
{
"epoch": 87.24,
"learning_rate": 1.5582035928143712e-06,
"loss": 0.503,
"step": 6020
},
{
"epoch": 87.39,
"learning_rate": 1.5546107784431137e-06,
"loss": 0.4481,
"step": 6030
},
{
"epoch": 87.53,
"learning_rate": 1.5510179640718566e-06,
"loss": 0.4803,
"step": 6040
},
{
"epoch": 87.68,
"learning_rate": 1.5474251497005988e-06,
"loss": 0.4728,
"step": 6050
},
{
"epoch": 87.82,
"learning_rate": 1.5438323353293413e-06,
"loss": 0.4795,
"step": 6060
},
{
"epoch": 87.96,
"learning_rate": 1.5402395209580837e-06,
"loss": 0.4998,
"step": 6070
},
{
"epoch": 88.12,
"learning_rate": 1.5366467065868264e-06,
"loss": 0.4728,
"step": 6080
},
{
"epoch": 88.26,
"learning_rate": 1.5330538922155688e-06,
"loss": 0.5273,
"step": 6090
},
{
"epoch": 88.4,
"learning_rate": 1.5294610778443113e-06,
"loss": 0.4658,
"step": 6100
},
{
"epoch": 88.55,
"learning_rate": 1.5258682634730542e-06,
"loss": 0.4785,
"step": 6110
},
{
"epoch": 88.69,
"learning_rate": 1.5222754491017964e-06,
"loss": 0.4827,
"step": 6120
},
{
"epoch": 88.83,
"learning_rate": 1.5186826347305389e-06,
"loss": 0.4648,
"step": 6130
},
{
"epoch": 88.98,
"learning_rate": 1.5150898203592813e-06,
"loss": 0.4939,
"step": 6140
},
{
"epoch": 89.13,
"learning_rate": 1.511497005988024e-06,
"loss": 0.4796,
"step": 6150
},
{
"epoch": 89.27,
"learning_rate": 1.5079041916167665e-06,
"loss": 0.5026,
"step": 6160
},
{
"epoch": 89.42,
"learning_rate": 1.504311377245509e-06,
"loss": 0.4988,
"step": 6170
},
{
"epoch": 89.56,
"learning_rate": 1.5007185628742516e-06,
"loss": 0.4856,
"step": 6180
},
{
"epoch": 89.71,
"learning_rate": 1.497125748502994e-06,
"loss": 0.4671,
"step": 6190
},
{
"epoch": 89.85,
"learning_rate": 1.4935329341317365e-06,
"loss": 0.4577,
"step": 6200
},
{
"epoch": 89.99,
"learning_rate": 1.4899401197604792e-06,
"loss": 0.5107,
"step": 6210
},
{
"epoch": 90.14,
"learning_rate": 1.4863473053892216e-06,
"loss": 0.4796,
"step": 6220
},
{
"epoch": 90.29,
"learning_rate": 1.482754491017964e-06,
"loss": 0.4787,
"step": 6230
},
{
"epoch": 90.43,
"learning_rate": 1.4791616766467068e-06,
"loss": 0.4782,
"step": 6240
},
{
"epoch": 90.58,
"learning_rate": 1.4755688622754492e-06,
"loss": 0.4697,
"step": 6250
},
{
"epoch": 90.72,
"learning_rate": 1.4719760479041917e-06,
"loss": 0.4946,
"step": 6260
},
{
"epoch": 90.86,
"learning_rate": 1.4683832335329341e-06,
"loss": 0.4343,
"step": 6270
},
{
"epoch": 91.01,
"learning_rate": 1.4647904191616768e-06,
"loss": 0.5369,
"step": 6280
},
{
"epoch": 91.16,
"learning_rate": 1.461197604790419e-06,
"loss": 0.4424,
"step": 6290
},
{
"epoch": 91.3,
"learning_rate": 1.4576047904191617e-06,
"loss": 0.4452,
"step": 6300
},
{
"epoch": 91.45,
"learning_rate": 1.4540119760479042e-06,
"loss": 0.4837,
"step": 6310
},
{
"epoch": 91.59,
"learning_rate": 1.4504191616766468e-06,
"loss": 0.4635,
"step": 6320
},
{
"epoch": 91.73,
"learning_rate": 1.4468263473053893e-06,
"loss": 0.493,
"step": 6330
},
{
"epoch": 91.88,
"learning_rate": 1.4432335329341317e-06,
"loss": 0.4684,
"step": 6340
},
{
"epoch": 92.03,
"learning_rate": 1.4396407185628744e-06,
"loss": 0.5282,
"step": 6350
},
{
"epoch": 92.17,
"learning_rate": 1.4360479041916167e-06,
"loss": 0.4767,
"step": 6360
},
{
"epoch": 92.32,
"learning_rate": 1.4324550898203593e-06,
"loss": 0.4743,
"step": 6370
},
{
"epoch": 92.46,
"learning_rate": 1.4288622754491018e-06,
"loss": 0.479,
"step": 6380
},
{
"epoch": 92.6,
"learning_rate": 1.4252694610778442e-06,
"loss": 0.4535,
"step": 6390
},
{
"epoch": 92.75,
"learning_rate": 1.421676646706587e-06,
"loss": 0.4942,
"step": 6400
},
{
"epoch": 92.89,
"learning_rate": 1.4180838323353294e-06,
"loss": 0.4388,
"step": 6410
},
{
"epoch": 93.04,
"learning_rate": 1.414491017964072e-06,
"loss": 0.458,
"step": 6420
},
{
"epoch": 93.19,
"learning_rate": 1.4108982035928143e-06,
"loss": 0.5045,
"step": 6430
},
{
"epoch": 93.33,
"learning_rate": 1.407305389221557e-06,
"loss": 0.4553,
"step": 6440
},
{
"epoch": 93.47,
"learning_rate": 1.4037125748502994e-06,
"loss": 0.4839,
"step": 6450
},
{
"epoch": 93.62,
"learning_rate": 1.4001197604790419e-06,
"loss": 0.4398,
"step": 6460
},
{
"epoch": 93.76,
"learning_rate": 1.3965269461077845e-06,
"loss": 0.501,
"step": 6470
},
{
"epoch": 93.91,
"learning_rate": 1.392934131736527e-06,
"loss": 0.468,
"step": 6480
},
{
"epoch": 94.06,
"learning_rate": 1.3893413173652694e-06,
"loss": 0.5142,
"step": 6490
},
{
"epoch": 94.2,
"learning_rate": 1.385748502994012e-06,
"loss": 0.4775,
"step": 6500
},
{
"epoch": 94.2,
"eval_loss": 0.4186658263206482,
"eval_runtime": 80.9451,
"eval_samples_per_second": 13.738,
"eval_steps_per_second": 3.434,
"eval_wer": 0.5334529147982062,
"step": 6500
},
{
"epoch": 94.35,
"learning_rate": 1.3821556886227546e-06,
"loss": 0.4431,
"step": 6510
},
{
"epoch": 94.49,
"learning_rate": 1.378562874251497e-06,
"loss": 0.4771,
"step": 6520
},
{
"epoch": 94.63,
"learning_rate": 1.3749700598802395e-06,
"loss": 0.4696,
"step": 6530
},
{
"epoch": 94.78,
"learning_rate": 1.3713772455089822e-06,
"loss": 0.4699,
"step": 6540
},
{
"epoch": 94.92,
"learning_rate": 1.3677844311377246e-06,
"loss": 0.4793,
"step": 6550
},
{
"epoch": 95.07,
"learning_rate": 1.364191616766467e-06,
"loss": 0.5029,
"step": 6560
},
{
"epoch": 95.22,
"learning_rate": 1.3605988023952095e-06,
"loss": 0.4827,
"step": 6570
},
{
"epoch": 95.36,
"learning_rate": 1.3570059880239522e-06,
"loss": 0.4353,
"step": 6580
},
{
"epoch": 95.5,
"learning_rate": 1.3534131736526946e-06,
"loss": 0.4668,
"step": 6590
},
{
"epoch": 95.65,
"learning_rate": 1.349820359281437e-06,
"loss": 0.4647,
"step": 6600
},
{
"epoch": 95.79,
"learning_rate": 1.3462275449101798e-06,
"loss": 0.441,
"step": 6610
},
{
"epoch": 95.94,
"learning_rate": 1.3426347305389222e-06,
"loss": 0.4617,
"step": 6620
},
{
"epoch": 96.09,
"learning_rate": 1.3390419161676647e-06,
"loss": 0.4927,
"step": 6630
},
{
"epoch": 96.23,
"learning_rate": 1.3354491017964071e-06,
"loss": 0.4926,
"step": 6640
},
{
"epoch": 96.37,
"learning_rate": 1.3318562874251498e-06,
"loss": 0.4581,
"step": 6650
},
{
"epoch": 96.52,
"learning_rate": 1.3282634730538923e-06,
"loss": 0.4901,
"step": 6660
},
{
"epoch": 96.66,
"learning_rate": 1.3246706586826347e-06,
"loss": 0.4429,
"step": 6670
},
{
"epoch": 96.81,
"learning_rate": 1.3210778443113774e-06,
"loss": 0.467,
"step": 6680
},
{
"epoch": 96.95,
"learning_rate": 1.3174850299401199e-06,
"loss": 0.4868,
"step": 6690
},
{
"epoch": 97.1,
"learning_rate": 1.3138922155688623e-06,
"loss": 0.4653,
"step": 6700
},
{
"epoch": 97.24,
"learning_rate": 1.3102994011976048e-06,
"loss": 0.4553,
"step": 6710
},
{
"epoch": 97.39,
"learning_rate": 1.3067065868263474e-06,
"loss": 0.4509,
"step": 6720
},
{
"epoch": 97.53,
"learning_rate": 1.3031137724550897e-06,
"loss": 0.4529,
"step": 6730
},
{
"epoch": 97.68,
"learning_rate": 1.2995209580838323e-06,
"loss": 0.4729,
"step": 6740
},
{
"epoch": 97.82,
"learning_rate": 1.295928143712575e-06,
"loss": 0.4314,
"step": 6750
},
{
"epoch": 97.96,
"learning_rate": 1.2923353293413175e-06,
"loss": 0.4809,
"step": 6760
},
{
"epoch": 98.12,
"learning_rate": 1.28874251497006e-06,
"loss": 0.4623,
"step": 6770
},
{
"epoch": 98.26,
"learning_rate": 1.2851497005988024e-06,
"loss": 0.4736,
"step": 6780
},
{
"epoch": 98.4,
"learning_rate": 1.281556886227545e-06,
"loss": 0.4455,
"step": 6790
},
{
"epoch": 98.55,
"learning_rate": 1.2779640718562873e-06,
"loss": 0.4858,
"step": 6800
},
{
"epoch": 98.69,
"learning_rate": 1.27437125748503e-06,
"loss": 0.4413,
"step": 6810
},
{
"epoch": 98.83,
"learning_rate": 1.2707784431137726e-06,
"loss": 0.4621,
"step": 6820
},
{
"epoch": 98.98,
"learning_rate": 1.2671856287425149e-06,
"loss": 0.5117,
"step": 6830
},
{
"epoch": 99.13,
"learning_rate": 1.2635928143712575e-06,
"loss": 0.4415,
"step": 6840
},
{
"epoch": 99.27,
"learning_rate": 1.26e-06,
"loss": 0.4733,
"step": 6850
},
{
"epoch": 99.42,
"learning_rate": 1.2564071856287427e-06,
"loss": 0.4497,
"step": 6860
},
{
"epoch": 99.56,
"learning_rate": 1.252814371257485e-06,
"loss": 0.4563,
"step": 6870
},
{
"epoch": 99.71,
"learning_rate": 1.2492215568862276e-06,
"loss": 0.4589,
"step": 6880
},
{
"epoch": 99.85,
"learning_rate": 1.2456287425149703e-06,
"loss": 0.4441,
"step": 6890
},
{
"epoch": 99.99,
"learning_rate": 1.2420359281437125e-06,
"loss": 0.4774,
"step": 6900
},
{
"epoch": 100.14,
"learning_rate": 1.2384431137724552e-06,
"loss": 0.4602,
"step": 6910
},
{
"epoch": 100.29,
"learning_rate": 1.2348502994011976e-06,
"loss": 0.4538,
"step": 6920
},
{
"epoch": 100.43,
"learning_rate": 1.23125748502994e-06,
"loss": 0.4642,
"step": 6930
},
{
"epoch": 100.58,
"learning_rate": 1.2276646706586825e-06,
"loss": 0.4384,
"step": 6940
},
{
"epoch": 100.72,
"learning_rate": 1.2240718562874252e-06,
"loss": 0.4832,
"step": 6950
},
{
"epoch": 100.86,
"learning_rate": 1.2204790419161677e-06,
"loss": 0.4591,
"step": 6960
},
{
"epoch": 101.01,
"learning_rate": 1.2168862275449101e-06,
"loss": 0.5292,
"step": 6970
},
{
"epoch": 101.16,
"learning_rate": 1.2132934131736528e-06,
"loss": 0.454,
"step": 6980
},
{
"epoch": 101.3,
"learning_rate": 1.2097005988023952e-06,
"loss": 0.47,
"step": 6990
},
{
"epoch": 101.45,
"learning_rate": 1.2061077844311377e-06,
"loss": 0.4898,
"step": 7000
},
{
"epoch": 101.45,
"eval_loss": 0.41909757256507874,
"eval_runtime": 81.7017,
"eval_samples_per_second": 13.61,
"eval_steps_per_second": 3.403,
"eval_wer": 0.5343497757847534,
"step": 7000
},
{
"epoch": 101.59,
"learning_rate": 1.2025149700598802e-06,
"loss": 0.4339,
"step": 7010
},
{
"epoch": 101.73,
"learning_rate": 1.1989221556886228e-06,
"loss": 0.4494,
"step": 7020
},
{
"epoch": 101.88,
"learning_rate": 1.1953293413173653e-06,
"loss": 0.4343,
"step": 7030
},
{
"epoch": 102.03,
"learning_rate": 1.1917365269461077e-06,
"loss": 0.4646,
"step": 7040
},
{
"epoch": 102.17,
"learning_rate": 1.1881437125748504e-06,
"loss": 0.4189,
"step": 7050
},
{
"epoch": 102.32,
"learning_rate": 1.1845508982035929e-06,
"loss": 0.4349,
"step": 7060
},
{
"epoch": 102.46,
"learning_rate": 1.1809580838323353e-06,
"loss": 0.4753,
"step": 7070
},
{
"epoch": 102.6,
"learning_rate": 1.1773652694610778e-06,
"loss": 0.4371,
"step": 7080
},
{
"epoch": 102.75,
"learning_rate": 1.1737724550898205e-06,
"loss": 0.4575,
"step": 7090
},
{
"epoch": 102.89,
"learning_rate": 1.170179640718563e-06,
"loss": 0.4497,
"step": 7100
},
{
"epoch": 103.04,
"learning_rate": 1.1665868263473054e-06,
"loss": 0.5231,
"step": 7110
},
{
"epoch": 103.19,
"learning_rate": 1.162994011976048e-06,
"loss": 0.4492,
"step": 7120
},
{
"epoch": 103.33,
"learning_rate": 1.1594011976047905e-06,
"loss": 0.4519,
"step": 7130
},
{
"epoch": 103.47,
"learning_rate": 1.155808383233533e-06,
"loss": 0.4458,
"step": 7140
},
{
"epoch": 103.62,
"learning_rate": 1.1522155688622754e-06,
"loss": 0.4352,
"step": 7150
},
{
"epoch": 103.76,
"learning_rate": 1.148622754491018e-06,
"loss": 0.468,
"step": 7160
},
{
"epoch": 103.91,
"learning_rate": 1.1450299401197605e-06,
"loss": 0.4469,
"step": 7170
},
{
"epoch": 104.06,
"learning_rate": 1.141437125748503e-06,
"loss": 0.4669,
"step": 7180
},
{
"epoch": 104.2,
"learning_rate": 1.1378443113772457e-06,
"loss": 0.4649,
"step": 7190
},
{
"epoch": 104.35,
"learning_rate": 1.1342514970059881e-06,
"loss": 0.4589,
"step": 7200
},
{
"epoch": 104.49,
"learning_rate": 1.1306586826347306e-06,
"loss": 0.4567,
"step": 7210
},
{
"epoch": 104.63,
"learning_rate": 1.127065868263473e-06,
"loss": 0.4187,
"step": 7220
},
{
"epoch": 104.78,
"learning_rate": 1.1234730538922157e-06,
"loss": 0.4542,
"step": 7230
},
{
"epoch": 104.92,
"learning_rate": 1.1198802395209582e-06,
"loss": 0.4574,
"step": 7240
},
{
"epoch": 105.07,
"learning_rate": 1.1162874251497006e-06,
"loss": 0.4593,
"step": 7250
},
{
"epoch": 105.22,
"learning_rate": 1.1126946107784433e-06,
"loss": 0.4565,
"step": 7260
},
{
"epoch": 105.36,
"learning_rate": 1.1091017964071855e-06,
"loss": 0.4276,
"step": 7270
},
{
"epoch": 105.5,
"learning_rate": 1.1055089820359282e-06,
"loss": 0.4596,
"step": 7280
},
{
"epoch": 105.65,
"learning_rate": 1.1019161676646706e-06,
"loss": 0.4162,
"step": 7290
},
{
"epoch": 105.79,
"learning_rate": 1.0983233532934133e-06,
"loss": 0.4609,
"step": 7300
},
{
"epoch": 105.94,
"learning_rate": 1.0947305389221558e-06,
"loss": 0.4733,
"step": 7310
},
{
"epoch": 106.09,
"learning_rate": 1.0911377245508982e-06,
"loss": 0.4855,
"step": 7320
},
{
"epoch": 106.23,
"learning_rate": 1.087544910179641e-06,
"loss": 0.4568,
"step": 7330
},
{
"epoch": 106.37,
"learning_rate": 1.0839520958083831e-06,
"loss": 0.4387,
"step": 7340
},
{
"epoch": 106.52,
"learning_rate": 1.0803592814371258e-06,
"loss": 0.4659,
"step": 7350
},
{
"epoch": 106.66,
"learning_rate": 1.0767664670658683e-06,
"loss": 0.4215,
"step": 7360
},
{
"epoch": 106.81,
"learning_rate": 1.0731736526946107e-06,
"loss": 0.4399,
"step": 7370
},
{
"epoch": 106.95,
"learning_rate": 1.0695808383233532e-06,
"loss": 0.4673,
"step": 7380
},
{
"epoch": 107.1,
"learning_rate": 1.0659880239520959e-06,
"loss": 0.4391,
"step": 7390
},
{
"epoch": 107.24,
"learning_rate": 1.0623952095808383e-06,
"loss": 0.4501,
"step": 7400
},
{
"epoch": 107.39,
"learning_rate": 1.0588023952095808e-06,
"loss": 0.422,
"step": 7410
},
{
"epoch": 107.53,
"learning_rate": 1.0552095808383234e-06,
"loss": 0.4578,
"step": 7420
},
{
"epoch": 107.68,
"learning_rate": 1.0516167664670659e-06,
"loss": 0.4705,
"step": 7430
},
{
"epoch": 107.82,
"learning_rate": 1.0480239520958083e-06,
"loss": 0.4463,
"step": 7440
},
{
"epoch": 107.96,
"learning_rate": 1.0444311377245508e-06,
"loss": 0.4532,
"step": 7450
},
{
"epoch": 108.12,
"learning_rate": 1.0408383233532935e-06,
"loss": 0.4295,
"step": 7460
},
{
"epoch": 108.26,
"learning_rate": 1.037245508982036e-06,
"loss": 0.4398,
"step": 7470
},
{
"epoch": 108.4,
"learning_rate": 1.0336526946107784e-06,
"loss": 0.4148,
"step": 7480
},
{
"epoch": 108.55,
"learning_rate": 1.030059880239521e-06,
"loss": 0.4626,
"step": 7490
},
{
"epoch": 108.69,
"learning_rate": 1.0264670658682635e-06,
"loss": 0.4601,
"step": 7500
},
{
"epoch": 108.69,
"eval_loss": 0.40853580832481384,
"eval_runtime": 81.6255,
"eval_samples_per_second": 13.623,
"eval_steps_per_second": 3.406,
"eval_wer": 0.5205381165919283,
"step": 7500
},
{
"epoch": 108.83,
"learning_rate": 1.022874251497006e-06,
"loss": 0.4309,
"step": 7510
},
{
"epoch": 108.98,
"learning_rate": 1.0192814371257484e-06,
"loss": 0.4742,
"step": 7520
},
{
"epoch": 109.13,
"learning_rate": 1.015688622754491e-06,
"loss": 0.4287,
"step": 7530
},
{
"epoch": 109.27,
"learning_rate": 1.0120958083832335e-06,
"loss": 0.4517,
"step": 7540
},
{
"epoch": 109.42,
"learning_rate": 1.008502994011976e-06,
"loss": 0.4253,
"step": 7550
},
{
"epoch": 109.56,
"learning_rate": 1.0049101796407187e-06,
"loss": 0.43,
"step": 7560
},
{
"epoch": 109.71,
"learning_rate": 1.0013173652694611e-06,
"loss": 0.4423,
"step": 7570
},
{
"epoch": 109.85,
"learning_rate": 9.977245508982036e-07,
"loss": 0.4176,
"step": 7580
},
{
"epoch": 109.99,
"learning_rate": 9.94131736526946e-07,
"loss": 0.5031,
"step": 7590
},
{
"epoch": 110.14,
"learning_rate": 9.905389221556887e-07,
"loss": 0.4493,
"step": 7600
},
{
"epoch": 110.29,
"learning_rate": 9.869461077844312e-07,
"loss": 0.4616,
"step": 7610
},
{
"epoch": 110.43,
"learning_rate": 9.833532934131736e-07,
"loss": 0.4286,
"step": 7620
},
{
"epoch": 110.58,
"learning_rate": 9.797604790419163e-07,
"loss": 0.4401,
"step": 7630
},
{
"epoch": 110.72,
"learning_rate": 9.761676646706588e-07,
"loss": 0.4855,
"step": 7640
},
{
"epoch": 110.86,
"learning_rate": 9.725748502994012e-07,
"loss": 0.4207,
"step": 7650
},
{
"epoch": 111.01,
"learning_rate": 9.689820359281437e-07,
"loss": 0.4642,
"step": 7660
},
{
"epoch": 111.16,
"learning_rate": 9.653892215568863e-07,
"loss": 0.4306,
"step": 7670
},
{
"epoch": 111.3,
"learning_rate": 9.617964071856288e-07,
"loss": 0.4401,
"step": 7680
},
{
"epoch": 111.45,
"learning_rate": 9.582035928143712e-07,
"loss": 0.47,
"step": 7690
},
{
"epoch": 111.59,
"learning_rate": 9.54610778443114e-07,
"loss": 0.4164,
"step": 7700
},
{
"epoch": 111.73,
"learning_rate": 9.510179640718563e-07,
"loss": 0.4405,
"step": 7710
},
{
"epoch": 111.88,
"learning_rate": 9.474251497005988e-07,
"loss": 0.4264,
"step": 7720
},
{
"epoch": 112.03,
"learning_rate": 9.438323353293413e-07,
"loss": 0.4846,
"step": 7730
},
{
"epoch": 112.17,
"learning_rate": 9.402395209580839e-07,
"loss": 0.419,
"step": 7740
},
{
"epoch": 112.32,
"learning_rate": 9.366467065868264e-07,
"loss": 0.4566,
"step": 7750
},
{
"epoch": 112.46,
"learning_rate": 9.330538922155689e-07,
"loss": 0.4537,
"step": 7760
},
{
"epoch": 112.6,
"learning_rate": 9.294610778443114e-07,
"loss": 0.4189,
"step": 7770
},
{
"epoch": 112.75,
"learning_rate": 9.258682634730539e-07,
"loss": 0.4847,
"step": 7780
},
{
"epoch": 112.89,
"learning_rate": 9.222754491017965e-07,
"loss": 0.4323,
"step": 7790
},
{
"epoch": 113.04,
"learning_rate": 9.186826347305389e-07,
"loss": 0.4771,
"step": 7800
},
{
"epoch": 113.19,
"learning_rate": 9.150898203592815e-07,
"loss": 0.4274,
"step": 7810
},
{
"epoch": 113.33,
"learning_rate": 9.11497005988024e-07,
"loss": 0.4562,
"step": 7820
},
{
"epoch": 113.47,
"learning_rate": 9.079041916167665e-07,
"loss": 0.4898,
"step": 7830
},
{
"epoch": 113.62,
"learning_rate": 9.043113772455091e-07,
"loss": 0.4531,
"step": 7840
},
{
"epoch": 113.76,
"learning_rate": 9.007185628742515e-07,
"loss": 0.4347,
"step": 7850
},
{
"epoch": 113.91,
"learning_rate": 8.971257485029941e-07,
"loss": 0.424,
"step": 7860
},
{
"epoch": 114.06,
"learning_rate": 8.935329341317365e-07,
"loss": 0.4781,
"step": 7870
},
{
"epoch": 114.2,
"learning_rate": 8.899401197604791e-07,
"loss": 0.4423,
"step": 7880
},
{
"epoch": 114.35,
"learning_rate": 8.863473053892217e-07,
"loss": 0.4067,
"step": 7890
},
{
"epoch": 114.49,
"learning_rate": 8.827544910179641e-07,
"loss": 0.4765,
"step": 7900
},
{
"epoch": 114.63,
"learning_rate": 8.791616766467067e-07,
"loss": 0.427,
"step": 7910
},
{
"epoch": 114.78,
"learning_rate": 8.75568862275449e-07,
"loss": 0.4218,
"step": 7920
},
{
"epoch": 114.92,
"learning_rate": 8.719760479041917e-07,
"loss": 0.4216,
"step": 7930
},
{
"epoch": 115.07,
"learning_rate": 8.68383233532934e-07,
"loss": 0.464,
"step": 7940
},
{
"epoch": 115.22,
"learning_rate": 8.647904191616767e-07,
"loss": 0.4488,
"step": 7950
},
{
"epoch": 115.36,
"learning_rate": 8.611976047904193e-07,
"loss": 0.4322,
"step": 7960
},
{
"epoch": 115.5,
"learning_rate": 8.576047904191616e-07,
"loss": 0.4462,
"step": 7970
},
{
"epoch": 115.65,
"learning_rate": 8.540119760479043e-07,
"loss": 0.4091,
"step": 7980
},
{
"epoch": 115.79,
"learning_rate": 8.504191616766466e-07,
"loss": 0.4384,
"step": 7990
},
{
"epoch": 115.94,
"learning_rate": 8.468263473053893e-07,
"loss": 0.4238,
"step": 8000
},
{
"epoch": 115.94,
"eval_loss": 0.4092176854610443,
"eval_runtime": 81.4639,
"eval_samples_per_second": 13.65,
"eval_steps_per_second": 3.413,
"eval_wer": 0.509237668161435,
"step": 8000
},
{
"epoch": 116.09,
"learning_rate": 8.432335329341317e-07,
"loss": 0.4297,
"step": 8010
},
{
"epoch": 116.23,
"learning_rate": 8.396407185628742e-07,
"loss": 0.4611,
"step": 8020
},
{
"epoch": 116.37,
"learning_rate": 8.360479041916169e-07,
"loss": 0.4257,
"step": 8030
},
{
"epoch": 116.52,
"learning_rate": 8.324550898203592e-07,
"loss": 0.4358,
"step": 8040
},
{
"epoch": 116.66,
"learning_rate": 8.288622754491019e-07,
"loss": 0.45,
"step": 8050
},
{
"epoch": 116.81,
"learning_rate": 8.252694610778443e-07,
"loss": 0.4278,
"step": 8060
},
{
"epoch": 116.95,
"learning_rate": 8.216766467065868e-07,
"loss": 0.4428,
"step": 8070
},
{
"epoch": 117.1,
"learning_rate": 8.180838323353293e-07,
"loss": 0.4544,
"step": 8080
},
{
"epoch": 117.24,
"learning_rate": 8.144910179640718e-07,
"loss": 0.4593,
"step": 8090
},
{
"epoch": 117.39,
"learning_rate": 8.108982035928145e-07,
"loss": 0.3883,
"step": 8100
},
{
"epoch": 117.53,
"learning_rate": 8.073053892215569e-07,
"loss": 0.4285,
"step": 8110
},
{
"epoch": 117.68,
"learning_rate": 8.037125748502994e-07,
"loss": 0.4802,
"step": 8120
},
{
"epoch": 117.82,
"learning_rate": 8.001197604790419e-07,
"loss": 0.4445,
"step": 8130
},
{
"epoch": 117.96,
"learning_rate": 7.965269461077845e-07,
"loss": 0.4394,
"step": 8140
},
{
"epoch": 118.12,
"learning_rate": 7.929341317365269e-07,
"loss": 0.4368,
"step": 8150
},
{
"epoch": 118.26,
"learning_rate": 7.893413173652695e-07,
"loss": 0.4744,
"step": 8160
},
{
"epoch": 118.4,
"learning_rate": 7.85748502994012e-07,
"loss": 0.4122,
"step": 8170
},
{
"epoch": 118.55,
"learning_rate": 7.821556886227545e-07,
"loss": 0.4165,
"step": 8180
},
{
"epoch": 118.69,
"learning_rate": 7.785628742514971e-07,
"loss": 0.4436,
"step": 8190
},
{
"epoch": 118.83,
"learning_rate": 7.749700598802395e-07,
"loss": 0.4297,
"step": 8200
},
{
"epoch": 118.98,
"learning_rate": 7.713772455089821e-07,
"loss": 0.4567,
"step": 8210
},
{
"epoch": 119.13,
"learning_rate": 7.677844311377245e-07,
"loss": 0.4294,
"step": 8220
},
{
"epoch": 119.27,
"learning_rate": 7.641916167664671e-07,
"loss": 0.4616,
"step": 8230
},
{
"epoch": 119.42,
"learning_rate": 7.605988023952097e-07,
"loss": 0.4375,
"step": 8240
},
{
"epoch": 119.56,
"learning_rate": 7.570059880239521e-07,
"loss": 0.4171,
"step": 8250
},
{
"epoch": 119.71,
"learning_rate": 7.534131736526947e-07,
"loss": 0.4517,
"step": 8260
},
{
"epoch": 119.85,
"learning_rate": 7.498203592814371e-07,
"loss": 0.4208,
"step": 8270
},
{
"epoch": 119.99,
"learning_rate": 7.462275449101797e-07,
"loss": 0.462,
"step": 8280
},
{
"epoch": 120.14,
"learning_rate": 7.426347305389222e-07,
"loss": 0.4442,
"step": 8290
},
{
"epoch": 120.29,
"learning_rate": 7.390419161676647e-07,
"loss": 0.4673,
"step": 8300
},
{
"epoch": 120.43,
"learning_rate": 7.354491017964072e-07,
"loss": 0.4099,
"step": 8310
},
{
"epoch": 120.58,
"learning_rate": 7.318562874251497e-07,
"loss": 0.4243,
"step": 8320
},
{
"epoch": 120.72,
"learning_rate": 7.282634730538922e-07,
"loss": 0.477,
"step": 8330
},
{
"epoch": 120.86,
"learning_rate": 7.246706586826348e-07,
"loss": 0.4151,
"step": 8340
},
{
"epoch": 121.01,
"learning_rate": 7.210778443113773e-07,
"loss": 0.4725,
"step": 8350
},
{
"epoch": 121.16,
"learning_rate": 7.174850299401198e-07,
"loss": 0.4366,
"step": 8360
},
{
"epoch": 121.3,
"learning_rate": 7.138922155688623e-07,
"loss": 0.437,
"step": 8370
},
{
"epoch": 121.45,
"learning_rate": 7.102994011976048e-07,
"loss": 0.4412,
"step": 8380
},
{
"epoch": 121.59,
"learning_rate": 7.067065868263474e-07,
"loss": 0.4253,
"step": 8390
},
{
"epoch": 121.73,
"learning_rate": 7.031137724550898e-07,
"loss": 0.437,
"step": 8400
},
{
"epoch": 121.88,
"learning_rate": 6.995209580838323e-07,
"loss": 0.4177,
"step": 8410
},
{
"epoch": 122.03,
"learning_rate": 6.959281437125749e-07,
"loss": 0.4364,
"step": 8420
},
{
"epoch": 122.17,
"learning_rate": 6.923353293413174e-07,
"loss": 0.4096,
"step": 8430
},
{
"epoch": 122.32,
"learning_rate": 6.8874251497006e-07,
"loss": 0.4474,
"step": 8440
},
{
"epoch": 122.46,
"learning_rate": 6.851497005988024e-07,
"loss": 0.4172,
"step": 8450
},
{
"epoch": 122.6,
"learning_rate": 6.815568862275449e-07,
"loss": 0.4108,
"step": 8460
},
{
"epoch": 122.75,
"learning_rate": 6.779640718562874e-07,
"loss": 0.4479,
"step": 8470
},
{
"epoch": 122.89,
"learning_rate": 6.743712574850299e-07,
"loss": 0.4167,
"step": 8480
},
{
"epoch": 123.04,
"learning_rate": 6.707784431137726e-07,
"loss": 0.4248,
"step": 8490
},
{
"epoch": 123.19,
"learning_rate": 6.67185628742515e-07,
"loss": 0.4313,
"step": 8500
},
{
"epoch": 123.19,
"eval_loss": 0.41111794114112854,
"eval_runtime": 81.5176,
"eval_samples_per_second": 13.641,
"eval_steps_per_second": 3.41,
"eval_wer": 0.5174887892376682,
"step": 8500
},
{
"epoch": 123.33,
"learning_rate": 6.635928143712575e-07,
"loss": 0.4319,
"step": 8510
},
{
"epoch": 123.47,
"learning_rate": 6.6e-07,
"loss": 0.4599,
"step": 8520
},
{
"epoch": 123.62,
"learning_rate": 6.564071856287425e-07,
"loss": 0.4204,
"step": 8530
},
{
"epoch": 123.76,
"learning_rate": 6.528143712574851e-07,
"loss": 0.4711,
"step": 8540
},
{
"epoch": 123.91,
"learning_rate": 6.492215568862275e-07,
"loss": 0.4256,
"step": 8550
},
{
"epoch": 124.06,
"learning_rate": 6.456287425149701e-07,
"loss": 0.4496,
"step": 8560
},
{
"epoch": 124.2,
"learning_rate": 6.420359281437126e-07,
"loss": 0.4358,
"step": 8570
},
{
"epoch": 124.35,
"learning_rate": 6.384431137724551e-07,
"loss": 0.367,
"step": 8580
},
{
"epoch": 124.49,
"learning_rate": 6.348502994011977e-07,
"loss": 0.4442,
"step": 8590
},
{
"epoch": 124.63,
"learning_rate": 6.312574850299401e-07,
"loss": 0.4168,
"step": 8600
},
{
"epoch": 124.78,
"learning_rate": 6.276646706586827e-07,
"loss": 0.444,
"step": 8610
},
{
"epoch": 124.92,
"learning_rate": 6.240718562874251e-07,
"loss": 0.4369,
"step": 8620
},
{
"epoch": 125.07,
"learning_rate": 6.204790419161677e-07,
"loss": 0.4294,
"step": 8630
},
{
"epoch": 125.22,
"learning_rate": 6.168862275449103e-07,
"loss": 0.4691,
"step": 8640
},
{
"epoch": 125.36,
"learning_rate": 6.132934131736527e-07,
"loss": 0.4,
"step": 8650
},
{
"epoch": 125.5,
"learning_rate": 6.097005988023953e-07,
"loss": 0.4567,
"step": 8660
},
{
"epoch": 125.65,
"learning_rate": 6.061077844311377e-07,
"loss": 0.4344,
"step": 8670
},
{
"epoch": 125.79,
"learning_rate": 6.025149700598802e-07,
"loss": 0.4259,
"step": 8680
},
{
"epoch": 125.94,
"learning_rate": 5.989221556886228e-07,
"loss": 0.4076,
"step": 8690
},
{
"epoch": 126.09,
"learning_rate": 5.953293413173652e-07,
"loss": 0.4284,
"step": 8700
},
{
"epoch": 126.23,
"learning_rate": 5.917365269461079e-07,
"loss": 0.4718,
"step": 8710
},
{
"epoch": 126.37,
"learning_rate": 5.881437125748503e-07,
"loss": 0.4171,
"step": 8720
},
{
"epoch": 126.52,
"learning_rate": 5.845508982035928e-07,
"loss": 0.4398,
"step": 8730
},
{
"epoch": 126.66,
"learning_rate": 5.809580838323354e-07,
"loss": 0.4109,
"step": 8740
},
{
"epoch": 126.81,
"learning_rate": 5.773652694610778e-07,
"loss": 0.4151,
"step": 8750
},
{
"epoch": 126.95,
"learning_rate": 5.737724550898204e-07,
"loss": 0.4295,
"step": 8760
},
{
"epoch": 127.1,
"learning_rate": 5.701796407185628e-07,
"loss": 0.4181,
"step": 8770
},
{
"epoch": 127.24,
"learning_rate": 5.665868263473054e-07,
"loss": 0.4569,
"step": 8780
},
{
"epoch": 127.39,
"learning_rate": 5.62994011976048e-07,
"loss": 0.4241,
"step": 8790
},
{
"epoch": 127.53,
"learning_rate": 5.594011976047904e-07,
"loss": 0.4396,
"step": 8800
},
{
"epoch": 127.68,
"learning_rate": 5.55808383233533e-07,
"loss": 0.4506,
"step": 8810
},
{
"epoch": 127.82,
"learning_rate": 5.522155688622754e-07,
"loss": 0.4116,
"step": 8820
},
{
"epoch": 127.96,
"learning_rate": 5.48622754491018e-07,
"loss": 0.4466,
"step": 8830
},
{
"epoch": 128.12,
"learning_rate": 5.450299401197605e-07,
"loss": 0.4179,
"step": 8840
},
{
"epoch": 128.26,
"learning_rate": 5.41437125748503e-07,
"loss": 0.4358,
"step": 8850
},
{
"epoch": 128.4,
"learning_rate": 5.378443113772456e-07,
"loss": 0.437,
"step": 8860
},
{
"epoch": 128.55,
"learning_rate": 5.34251497005988e-07,
"loss": 0.4238,
"step": 8870
},
{
"epoch": 128.69,
"learning_rate": 5.306586826347306e-07,
"loss": 0.4265,
"step": 8880
},
{
"epoch": 128.83,
"learning_rate": 5.270658682634731e-07,
"loss": 0.4087,
"step": 8890
},
{
"epoch": 128.98,
"learning_rate": 5.234730538922155e-07,
"loss": 0.4167,
"step": 8900
},
{
"epoch": 129.13,
"learning_rate": 5.198802395209581e-07,
"loss": 0.4044,
"step": 8910
},
{
"epoch": 129.27,
"learning_rate": 5.162874251497006e-07,
"loss": 0.4681,
"step": 8920
},
{
"epoch": 129.42,
"learning_rate": 5.126946107784432e-07,
"loss": 0.4098,
"step": 8930
},
{
"epoch": 129.56,
"learning_rate": 5.091017964071857e-07,
"loss": 0.4331,
"step": 8940
},
{
"epoch": 129.71,
"learning_rate": 5.055089820359281e-07,
"loss": 0.4334,
"step": 8950
},
{
"epoch": 129.85,
"learning_rate": 5.019161676646707e-07,
"loss": 0.3982,
"step": 8960
},
{
"epoch": 129.99,
"learning_rate": 4.983233532934131e-07,
"loss": 0.4254,
"step": 8970
},
{
"epoch": 130.14,
"learning_rate": 4.947305389221557e-07,
"loss": 0.4162,
"step": 8980
},
{
"epoch": 130.29,
"learning_rate": 4.911377245508983e-07,
"loss": 0.4427,
"step": 8990
},
{
"epoch": 130.43,
"learning_rate": 4.875449101796407e-07,
"loss": 0.4222,
"step": 9000
},
{
"epoch": 130.43,
"eval_loss": 0.4150010645389557,
"eval_runtime": 81.6728,
"eval_samples_per_second": 13.615,
"eval_steps_per_second": 3.404,
"eval_wer": 0.5137219730941704,
"step": 9000
},
{
"epoch": 130.58,
"learning_rate": 4.839520958083833e-07,
"loss": 0.4335,
"step": 9010
},
{
"epoch": 130.72,
"learning_rate": 4.803592814371257e-07,
"loss": 0.4265,
"step": 9020
},
{
"epoch": 130.86,
"learning_rate": 4.7676646706586824e-07,
"loss": 0.3998,
"step": 9030
},
{
"epoch": 131.01,
"learning_rate": 4.7317365269461075e-07,
"loss": 0.4751,
"step": 9040
},
{
"epoch": 131.16,
"learning_rate": 4.6958083832335326e-07,
"loss": 0.4236,
"step": 9050
},
{
"epoch": 131.3,
"learning_rate": 4.659880239520959e-07,
"loss": 0.4266,
"step": 9060
},
{
"epoch": 131.45,
"learning_rate": 4.623952095808384e-07,
"loss": 0.4566,
"step": 9070
},
{
"epoch": 131.59,
"learning_rate": 4.5880239520958084e-07,
"loss": 0.4017,
"step": 9080
},
{
"epoch": 131.73,
"learning_rate": 4.5520958083832335e-07,
"loss": 0.4419,
"step": 9090
},
{
"epoch": 131.88,
"learning_rate": 4.5161676646706586e-07,
"loss": 0.4006,
"step": 9100
},
{
"epoch": 132.03,
"learning_rate": 4.4802395209580837e-07,
"loss": 0.4588,
"step": 9110
},
{
"epoch": 132.17,
"learning_rate": 4.444311377245509e-07,
"loss": 0.4224,
"step": 9120
},
{
"epoch": 132.32,
"learning_rate": 4.4083832335329345e-07,
"loss": 0.4283,
"step": 9130
},
{
"epoch": 132.46,
"learning_rate": 4.3724550898203596e-07,
"loss": 0.4116,
"step": 9140
},
{
"epoch": 132.6,
"learning_rate": 4.3365269461077847e-07,
"loss": 0.3906,
"step": 9150
},
{
"epoch": 132.75,
"learning_rate": 4.30059880239521e-07,
"loss": 0.4578,
"step": 9160
},
{
"epoch": 132.89,
"learning_rate": 4.264670658682635e-07,
"loss": 0.4218,
"step": 9170
},
{
"epoch": 133.04,
"learning_rate": 4.22874251497006e-07,
"loss": 0.4549,
"step": 9180
},
{
"epoch": 133.19,
"learning_rate": 4.192814371257485e-07,
"loss": 0.4071,
"step": 9190
},
{
"epoch": 133.33,
"learning_rate": 4.15688622754491e-07,
"loss": 0.427,
"step": 9200
},
{
"epoch": 133.47,
"learning_rate": 4.120958083832336e-07,
"loss": 0.436,
"step": 9210
},
{
"epoch": 133.62,
"learning_rate": 4.085029940119761e-07,
"loss": 0.4129,
"step": 9220
},
{
"epoch": 133.76,
"learning_rate": 4.049101796407186e-07,
"loss": 0.417,
"step": 9230
},
{
"epoch": 133.91,
"learning_rate": 4.013173652694611e-07,
"loss": 0.4136,
"step": 9240
},
{
"epoch": 134.06,
"learning_rate": 3.9772455089820356e-07,
"loss": 0.4519,
"step": 9250
},
{
"epoch": 134.2,
"learning_rate": 3.9413173652694607e-07,
"loss": 0.4484,
"step": 9260
},
{
"epoch": 134.35,
"learning_rate": 3.905389221556886e-07,
"loss": 0.4252,
"step": 9270
},
{
"epoch": 134.49,
"learning_rate": 3.869461077844312e-07,
"loss": 0.4513,
"step": 9280
},
{
"epoch": 134.63,
"learning_rate": 3.833532934131737e-07,
"loss": 0.3864,
"step": 9290
},
{
"epoch": 134.78,
"learning_rate": 3.7976047904191616e-07,
"loss": 0.4181,
"step": 9300
},
{
"epoch": 134.92,
"learning_rate": 3.761676646706587e-07,
"loss": 0.3996,
"step": 9310
},
{
"epoch": 135.07,
"learning_rate": 3.725748502994012e-07,
"loss": 0.4413,
"step": 9320
},
{
"epoch": 135.22,
"learning_rate": 3.6898203592814375e-07,
"loss": 0.4278,
"step": 9330
},
{
"epoch": 135.36,
"learning_rate": 3.6538922155688626e-07,
"loss": 0.4141,
"step": 9340
},
{
"epoch": 135.5,
"learning_rate": 3.6179640718562877e-07,
"loss": 0.431,
"step": 9350
},
{
"epoch": 135.65,
"learning_rate": 3.582035928143712e-07,
"loss": 0.4222,
"step": 9360
},
{
"epoch": 135.79,
"learning_rate": 3.546107784431138e-07,
"loss": 0.4311,
"step": 9370
},
{
"epoch": 135.94,
"learning_rate": 3.510179640718563e-07,
"loss": 0.4138,
"step": 9380
},
{
"epoch": 136.09,
"learning_rate": 3.474251497005988e-07,
"loss": 0.4316,
"step": 9390
},
{
"epoch": 136.23,
"learning_rate": 3.4383233532934137e-07,
"loss": 0.4308,
"step": 9400
},
{
"epoch": 136.37,
"learning_rate": 3.402395209580838e-07,
"loss": 0.4151,
"step": 9410
},
{
"epoch": 136.52,
"learning_rate": 3.3664670658682633e-07,
"loss": 0.4248,
"step": 9420
},
{
"epoch": 136.66,
"learning_rate": 3.3305389221556884e-07,
"loss": 0.4265,
"step": 9430
},
{
"epoch": 136.81,
"learning_rate": 3.294610778443114e-07,
"loss": 0.4208,
"step": 9440
},
{
"epoch": 136.95,
"learning_rate": 3.258682634730539e-07,
"loss": 0.423,
"step": 9450
},
{
"epoch": 137.1,
"learning_rate": 3.222754491017964e-07,
"loss": 0.4365,
"step": 9460
},
{
"epoch": 137.24,
"learning_rate": 3.1868263473053894e-07,
"loss": 0.4305,
"step": 9470
},
{
"epoch": 137.39,
"learning_rate": 3.1508982035928145e-07,
"loss": 0.4057,
"step": 9480
},
{
"epoch": 137.53,
"learning_rate": 3.1149700598802396e-07,
"loss": 0.4096,
"step": 9490
},
{
"epoch": 137.68,
"learning_rate": 3.0790419161676646e-07,
"loss": 0.4322,
"step": 9500
},
{
"epoch": 137.68,
"eval_loss": 0.419295072555542,
"eval_runtime": 81.0597,
"eval_samples_per_second": 13.718,
"eval_steps_per_second": 3.43,
"eval_wer": 0.5146188340807175,
"step": 9500
},
{
"epoch": 137.82,
"learning_rate": 3.0431137724550903e-07,
"loss": 0.4039,
"step": 9510
},
{
"epoch": 137.96,
"learning_rate": 3.007185628742515e-07,
"loss": 0.4436,
"step": 9520
},
{
"epoch": 138.12,
"learning_rate": 2.97125748502994e-07,
"loss": 0.4218,
"step": 9530
},
{
"epoch": 138.26,
"learning_rate": 2.9353293413173656e-07,
"loss": 0.4455,
"step": 9540
},
{
"epoch": 138.4,
"learning_rate": 2.8994011976047907e-07,
"loss": 0.4125,
"step": 9550
},
{
"epoch": 138.55,
"learning_rate": 2.863473053892216e-07,
"loss": 0.4273,
"step": 9560
},
{
"epoch": 138.69,
"learning_rate": 2.827544910179641e-07,
"loss": 0.449,
"step": 9570
},
{
"epoch": 138.83,
"learning_rate": 2.791616766467066e-07,
"loss": 0.422,
"step": 9580
},
{
"epoch": 138.98,
"learning_rate": 2.755688622754491e-07,
"loss": 0.4238,
"step": 9590
},
{
"epoch": 139.13,
"learning_rate": 2.719760479041916e-07,
"loss": 0.432,
"step": 9600
},
{
"epoch": 139.27,
"learning_rate": 2.683832335329341e-07,
"loss": 0.4525,
"step": 9610
},
{
"epoch": 139.42,
"learning_rate": 2.647904191616767e-07,
"loss": 0.4024,
"step": 9620
},
{
"epoch": 139.56,
"learning_rate": 2.6119760479041914e-07,
"loss": 0.4106,
"step": 9630
},
{
"epoch": 139.71,
"learning_rate": 2.5760479041916165e-07,
"loss": 0.4289,
"step": 9640
},
{
"epoch": 139.85,
"learning_rate": 2.540119760479042e-07,
"loss": 0.4254,
"step": 9650
},
{
"epoch": 139.99,
"learning_rate": 2.5041916167664673e-07,
"loss": 0.452,
"step": 9660
},
{
"epoch": 140.14,
"learning_rate": 2.4682634730538924e-07,
"loss": 0.4301,
"step": 9670
},
{
"epoch": 140.29,
"learning_rate": 2.4323353293413175e-07,
"loss": 0.4289,
"step": 9680
},
{
"epoch": 140.43,
"learning_rate": 2.3964071856287426e-07,
"loss": 0.4122,
"step": 9690
},
{
"epoch": 140.58,
"learning_rate": 2.3604790419161677e-07,
"loss": 0.4272,
"step": 9700
},
{
"epoch": 140.72,
"learning_rate": 2.3245508982035928e-07,
"loss": 0.4299,
"step": 9710
},
{
"epoch": 140.86,
"learning_rate": 2.288622754491018e-07,
"loss": 0.3917,
"step": 9720
},
{
"epoch": 141.01,
"learning_rate": 2.2526946107784432e-07,
"loss": 0.4517,
"step": 9730
},
{
"epoch": 141.16,
"learning_rate": 2.2167664670658683e-07,
"loss": 0.4028,
"step": 9740
},
{
"epoch": 141.3,
"learning_rate": 2.1808383233532934e-07,
"loss": 0.4209,
"step": 9750
},
{
"epoch": 141.45,
"learning_rate": 2.1449101796407188e-07,
"loss": 0.4519,
"step": 9760
},
{
"epoch": 141.59,
"learning_rate": 2.108982035928144e-07,
"loss": 0.3883,
"step": 9770
},
{
"epoch": 141.73,
"learning_rate": 2.0730538922155687e-07,
"loss": 0.4381,
"step": 9780
},
{
"epoch": 141.88,
"learning_rate": 2.0371257485029943e-07,
"loss": 0.3989,
"step": 9790
},
{
"epoch": 142.03,
"learning_rate": 2.0011976047904194e-07,
"loss": 0.4381,
"step": 9800
},
{
"epoch": 142.17,
"learning_rate": 1.9652694610778443e-07,
"loss": 0.4062,
"step": 9810
},
{
"epoch": 142.32,
"learning_rate": 1.9293413173652694e-07,
"loss": 0.4328,
"step": 9820
},
{
"epoch": 142.46,
"learning_rate": 1.8934131736526947e-07,
"loss": 0.4234,
"step": 9830
},
{
"epoch": 142.6,
"learning_rate": 1.8574850299401198e-07,
"loss": 0.3766,
"step": 9840
},
{
"epoch": 142.75,
"learning_rate": 1.821556886227545e-07,
"loss": 0.4531,
"step": 9850
},
{
"epoch": 142.89,
"learning_rate": 1.78562874251497e-07,
"loss": 0.4087,
"step": 9860
},
{
"epoch": 143.04,
"learning_rate": 1.7497005988023954e-07,
"loss": 0.4333,
"step": 9870
},
{
"epoch": 143.19,
"learning_rate": 1.7137724550898205e-07,
"loss": 0.4115,
"step": 9880
},
{
"epoch": 143.33,
"learning_rate": 1.6778443113772456e-07,
"loss": 0.4098,
"step": 9890
},
{
"epoch": 143.47,
"learning_rate": 1.6419161676646707e-07,
"loss": 0.4257,
"step": 9900
},
{
"epoch": 143.62,
"learning_rate": 1.605988023952096e-07,
"loss": 0.4237,
"step": 9910
},
{
"epoch": 143.76,
"learning_rate": 1.5700598802395209e-07,
"loss": 0.4396,
"step": 9920
},
{
"epoch": 143.91,
"learning_rate": 1.5341317365269462e-07,
"loss": 0.4159,
"step": 9930
},
{
"epoch": 144.06,
"learning_rate": 1.4982035928143713e-07,
"loss": 0.454,
"step": 9940
},
{
"epoch": 144.2,
"learning_rate": 1.4622754491017964e-07,
"loss": 0.4233,
"step": 9950
},
{
"epoch": 144.35,
"learning_rate": 1.4263473053892218e-07,
"loss": 0.4169,
"step": 9960
},
{
"epoch": 144.49,
"learning_rate": 1.3904191616766466e-07,
"loss": 0.4761,
"step": 9970
},
{
"epoch": 144.63,
"learning_rate": 1.354491017964072e-07,
"loss": 0.4147,
"step": 9980
},
{
"epoch": 144.78,
"learning_rate": 1.318562874251497e-07,
"loss": 0.4092,
"step": 9990
},
{
"epoch": 144.92,
"learning_rate": 1.2826347305389222e-07,
"loss": 0.3993,
"step": 10000
},
{
"epoch": 144.92,
"eval_loss": 0.4131280779838562,
"eval_runtime": 83.3021,
"eval_samples_per_second": 13.349,
"eval_steps_per_second": 3.337,
"eval_wer": 0.5103139013452915,
"step": 10000
},
{
"epoch": 145.07,
"learning_rate": 1.2467065868263473e-07,
"loss": 0.4419,
"step": 10010
},
{
"epoch": 145.22,
"learning_rate": 1.2107784431137726e-07,
"loss": 0.4307,
"step": 10020
},
{
"epoch": 145.36,
"learning_rate": 1.1748502994011976e-07,
"loss": 0.3922,
"step": 10030
},
{
"epoch": 145.5,
"learning_rate": 1.1389221556886228e-07,
"loss": 0.4411,
"step": 10040
},
{
"epoch": 145.65,
"learning_rate": 1.102994011976048e-07,
"loss": 0.415,
"step": 10050
},
{
"epoch": 145.79,
"learning_rate": 1.067065868263473e-07,
"loss": 0.4081,
"step": 10060
},
{
"epoch": 145.94,
"learning_rate": 1.0311377245508982e-07,
"loss": 0.4245,
"step": 10070
},
{
"epoch": 146.09,
"learning_rate": 9.952095808383233e-08,
"loss": 0.4253,
"step": 10080
},
{
"epoch": 146.23,
"learning_rate": 9.592814371257486e-08,
"loss": 0.4344,
"step": 10090
},
{
"epoch": 146.37,
"learning_rate": 9.233532934131737e-08,
"loss": 0.3836,
"step": 10100
},
{
"epoch": 146.52,
"learning_rate": 8.874251497005988e-08,
"loss": 0.4533,
"step": 10110
},
{
"epoch": 146.66,
"learning_rate": 8.51497005988024e-08,
"loss": 0.4043,
"step": 10120
},
{
"epoch": 146.81,
"learning_rate": 8.155688622754491e-08,
"loss": 0.3657,
"step": 10130
},
{
"epoch": 146.95,
"learning_rate": 7.796407185628742e-08,
"loss": 0.443,
"step": 10140
},
{
"epoch": 147.1,
"learning_rate": 7.437125748502994e-08,
"loss": 0.4252,
"step": 10150
},
{
"epoch": 147.24,
"learning_rate": 7.077844311377247e-08,
"loss": 0.4494,
"step": 10160
},
{
"epoch": 147.39,
"learning_rate": 6.718562874251498e-08,
"loss": 0.4119,
"step": 10170
},
{
"epoch": 147.53,
"learning_rate": 6.359281437125748e-08,
"loss": 0.438,
"step": 10180
},
{
"epoch": 147.68,
"learning_rate": 6.000000000000001e-08,
"loss": 0.4132,
"step": 10190
},
{
"epoch": 147.82,
"learning_rate": 5.640718562874252e-08,
"loss": 0.4163,
"step": 10200
},
{
"epoch": 147.96,
"learning_rate": 5.281437125748503e-08,
"loss": 0.453,
"step": 10210
},
{
"epoch": 148.12,
"learning_rate": 4.9221556886227544e-08,
"loss": 0.4142,
"step": 10220
},
{
"epoch": 148.26,
"learning_rate": 4.562874251497006e-08,
"loss": 0.4451,
"step": 10230
},
{
"epoch": 148.4,
"learning_rate": 4.2035928143712576e-08,
"loss": 0.4084,
"step": 10240
},
{
"epoch": 148.55,
"learning_rate": 3.844311377245509e-08,
"loss": 0.4284,
"step": 10250
},
{
"epoch": 148.69,
"learning_rate": 3.48502994011976e-08,
"loss": 0.4211,
"step": 10260
},
{
"epoch": 148.83,
"learning_rate": 3.125748502994012e-08,
"loss": 0.4105,
"step": 10270
},
{
"epoch": 148.98,
"learning_rate": 2.7664670658682635e-08,
"loss": 0.4207,
"step": 10280
},
{
"epoch": 149.13,
"learning_rate": 2.4071856287425148e-08,
"loss": 0.4493,
"step": 10290
},
{
"epoch": 149.27,
"learning_rate": 2.0479041916167665e-08,
"loss": 0.4296,
"step": 10300
},
{
"epoch": 149.42,
"learning_rate": 1.6886227544910178e-08,
"loss": 0.4007,
"step": 10310
},
{
"epoch": 149.56,
"learning_rate": 1.3293413173652696e-08,
"loss": 0.427,
"step": 10320
},
{
"epoch": 149.71,
"learning_rate": 9.70059880239521e-09,
"loss": 0.4267,
"step": 10330
},
{
"epoch": 149.85,
"learning_rate": 6.107784431137725e-09,
"loss": 0.4272,
"step": 10340
},
{
"epoch": 149.99,
"learning_rate": 2.5149700598802395e-09,
"loss": 0.4254,
"step": 10350
},
{
"epoch": 149.99,
"step": 10350,
"total_flos": 4.832513992657894e+20,
"train_loss": 1.2498395964719247,
"train_runtime": 108547.6376,
"train_samples_per_second": 4.61,
"train_steps_per_second": 0.095
}
],
"max_steps": 10350,
"num_train_epochs": 150,
"total_flos": 4.832513992657894e+20,
"trial_name": null,
"trial_params": null
}