whisper_test / trainer_state.json
wangpuupup's picture
Upload 26 files
84b61f5
raw
history blame
53.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 59.0029,
"global_step": 10000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 5.000000000000001e-07,
"loss": 3.0781,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 1.0000000000000002e-06,
"loss": 2.8492,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.5e-06,
"loss": 2.1622,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.3123,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 2.5e-06,
"loss": 0.6802,
"step": 125
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.6133,
"step": 150
},
{
"epoch": 1.0,
"learning_rate": 3.5e-06,
"loss": 0.5776,
"step": 175
},
{
"epoch": 1.0,
"learning_rate": 4.000000000000001e-06,
"loss": 0.5272,
"step": 200
},
{
"epoch": 1.01,
"learning_rate": 4.5e-06,
"loss": 0.5843,
"step": 225
},
{
"epoch": 1.01,
"learning_rate": 5e-06,
"loss": 0.5245,
"step": 250
},
{
"epoch": 1.01,
"learning_rate": 5.500000000000001e-06,
"loss": 0.5909,
"step": 275
},
{
"epoch": 1.01,
"learning_rate": 6e-06,
"loss": 0.3071,
"step": 300
},
{
"epoch": 1.02,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.4596,
"step": 325
},
{
"epoch": 2.0,
"learning_rate": 7e-06,
"loss": 0.3669,
"step": 350
},
{
"epoch": 2.0,
"learning_rate": 7.500000000000001e-06,
"loss": 0.3811,
"step": 375
},
{
"epoch": 2.01,
"learning_rate": 8.000000000000001e-06,
"loss": 0.4295,
"step": 400
},
{
"epoch": 2.01,
"learning_rate": 8.5e-06,
"loss": 0.3843,
"step": 425
},
{
"epoch": 2.01,
"learning_rate": 9e-06,
"loss": 0.4654,
"step": 450
},
{
"epoch": 2.01,
"learning_rate": 9.5e-06,
"loss": 0.1894,
"step": 475
},
{
"epoch": 2.02,
"learning_rate": 1e-05,
"loss": 0.3363,
"step": 500
},
{
"epoch": 2.02,
"eval_loss": 0.376156747341156,
"eval_runtime": 337.4081,
"eval_samples_per_second": 8.574,
"eval_steps_per_second": 0.27,
"eval_wer": 29.793403407031533,
"step": 500
},
{
"epoch": 3.0,
"learning_rate": 9.973684210526316e-06,
"loss": 0.2732,
"step": 525
},
{
"epoch": 3.0,
"learning_rate": 9.947368421052632e-06,
"loss": 0.3217,
"step": 550
},
{
"epoch": 3.01,
"learning_rate": 9.921052631578947e-06,
"loss": 0.3649,
"step": 575
},
{
"epoch": 3.01,
"learning_rate": 9.894736842105264e-06,
"loss": 0.3261,
"step": 600
},
{
"epoch": 3.01,
"learning_rate": 9.868421052631579e-06,
"loss": 0.3428,
"step": 625
},
{
"epoch": 3.01,
"learning_rate": 9.842105263157896e-06,
"loss": 0.1884,
"step": 650
},
{
"epoch": 3.02,
"learning_rate": 9.815789473684212e-06,
"loss": 0.1685,
"step": 675
},
{
"epoch": 4.0,
"learning_rate": 9.789473684210527e-06,
"loss": 0.24,
"step": 700
},
{
"epoch": 4.0,
"learning_rate": 9.763157894736844e-06,
"loss": 0.2742,
"step": 725
},
{
"epoch": 4.01,
"learning_rate": 9.736842105263159e-06,
"loss": 0.3044,
"step": 750
},
{
"epoch": 4.01,
"learning_rate": 9.710526315789474e-06,
"loss": 0.3126,
"step": 775
},
{
"epoch": 4.01,
"learning_rate": 9.68421052631579e-06,
"loss": 0.2322,
"step": 800
},
{
"epoch": 4.01,
"learning_rate": 9.657894736842106e-06,
"loss": 0.1392,
"step": 825
},
{
"epoch": 5.0,
"learning_rate": 9.631578947368422e-06,
"loss": 0.1278,
"step": 850
},
{
"epoch": 5.0,
"learning_rate": 9.605263157894737e-06,
"loss": 0.1932,
"step": 875
},
{
"epoch": 5.01,
"learning_rate": 9.578947368421054e-06,
"loss": 0.2495,
"step": 900
},
{
"epoch": 5.01,
"learning_rate": 9.552631578947369e-06,
"loss": 0.2476,
"step": 925
},
{
"epoch": 5.01,
"learning_rate": 9.526315789473684e-06,
"loss": 0.2918,
"step": 950
},
{
"epoch": 5.01,
"learning_rate": 9.5e-06,
"loss": 0.1333,
"step": 975
},
{
"epoch": 5.02,
"learning_rate": 9.473684210526315e-06,
"loss": 0.0945,
"step": 1000
},
{
"epoch": 5.02,
"eval_loss": 0.3418421149253845,
"eval_runtime": 325.7049,
"eval_samples_per_second": 8.882,
"eval_steps_per_second": 0.279,
"eval_wer": 27.691192461036607,
"step": 1000
},
{
"epoch": 6.0,
"learning_rate": 9.447368421052632e-06,
"loss": 0.109,
"step": 1025
},
{
"epoch": 6.0,
"learning_rate": 9.421052631578949e-06,
"loss": 0.1678,
"step": 1050
},
{
"epoch": 6.01,
"learning_rate": 9.394736842105264e-06,
"loss": 0.225,
"step": 1075
},
{
"epoch": 6.01,
"learning_rate": 9.36842105263158e-06,
"loss": 0.194,
"step": 1100
},
{
"epoch": 6.01,
"learning_rate": 9.342105263157895e-06,
"loss": 0.2468,
"step": 1125
},
{
"epoch": 6.01,
"learning_rate": 9.315789473684212e-06,
"loss": 0.0873,
"step": 1150
},
{
"epoch": 6.02,
"learning_rate": 9.289473684210527e-06,
"loss": 0.0529,
"step": 1175
},
{
"epoch": 7.0,
"learning_rate": 9.263157894736842e-06,
"loss": 0.096,
"step": 1200
},
{
"epoch": 7.0,
"learning_rate": 9.236842105263159e-06,
"loss": 0.149,
"step": 1225
},
{
"epoch": 7.01,
"learning_rate": 9.210526315789474e-06,
"loss": 0.1901,
"step": 1250
},
{
"epoch": 7.01,
"learning_rate": 9.18421052631579e-06,
"loss": 0.1607,
"step": 1275
},
{
"epoch": 7.01,
"learning_rate": 9.157894736842105e-06,
"loss": 0.1828,
"step": 1300
},
{
"epoch": 7.01,
"learning_rate": 9.131578947368422e-06,
"loss": 0.0553,
"step": 1325
},
{
"epoch": 7.02,
"learning_rate": 9.105263157894739e-06,
"loss": 0.0232,
"step": 1350
},
{
"epoch": 8.0,
"learning_rate": 9.078947368421054e-06,
"loss": 0.0956,
"step": 1375
},
{
"epoch": 8.0,
"learning_rate": 9.05263157894737e-06,
"loss": 0.1245,
"step": 1400
},
{
"epoch": 8.01,
"learning_rate": 9.026315789473685e-06,
"loss": 0.1506,
"step": 1425
},
{
"epoch": 8.01,
"learning_rate": 9e-06,
"loss": 0.1453,
"step": 1450
},
{
"epoch": 8.01,
"learning_rate": 8.973684210526317e-06,
"loss": 0.1084,
"step": 1475
},
{
"epoch": 8.01,
"learning_rate": 8.947368421052632e-06,
"loss": 0.0332,
"step": 1500
},
{
"epoch": 8.01,
"eval_loss": 0.33526483178138733,
"eval_runtime": 320.57,
"eval_samples_per_second": 9.025,
"eval_steps_per_second": 0.284,
"eval_wer": 26.1689017760058,
"step": 1500
},
{
"epoch": 9.0,
"learning_rate": 8.921052631578949e-06,
"loss": 0.0268,
"step": 1525
},
{
"epoch": 9.0,
"learning_rate": 8.894736842105264e-06,
"loss": 0.0712,
"step": 1550
},
{
"epoch": 9.01,
"learning_rate": 8.86842105263158e-06,
"loss": 0.1149,
"step": 1575
},
{
"epoch": 9.01,
"learning_rate": 8.842105263157895e-06,
"loss": 0.1063,
"step": 1600
},
{
"epoch": 9.01,
"learning_rate": 8.81578947368421e-06,
"loss": 0.1176,
"step": 1625
},
{
"epoch": 9.01,
"learning_rate": 8.789473684210527e-06,
"loss": 0.0602,
"step": 1650
},
{
"epoch": 9.02,
"learning_rate": 8.763157894736842e-06,
"loss": 0.0184,
"step": 1675
},
{
"epoch": 10.0,
"learning_rate": 8.736842105263158e-06,
"loss": 0.0263,
"step": 1700
},
{
"epoch": 10.0,
"learning_rate": 8.710526315789475e-06,
"loss": 0.0568,
"step": 1725
},
{
"epoch": 10.01,
"learning_rate": 8.68421052631579e-06,
"loss": 0.0881,
"step": 1750
},
{
"epoch": 10.01,
"learning_rate": 8.657894736842107e-06,
"loss": 0.0684,
"step": 1775
},
{
"epoch": 10.01,
"learning_rate": 8.631578947368422e-06,
"loss": 0.0895,
"step": 1800
},
{
"epoch": 10.01,
"learning_rate": 8.605263157894738e-06,
"loss": 0.029,
"step": 1825
},
{
"epoch": 10.02,
"learning_rate": 8.578947368421053e-06,
"loss": 0.0099,
"step": 1850
},
{
"epoch": 11.0,
"learning_rate": 8.552631578947368e-06,
"loss": 0.0267,
"step": 1875
},
{
"epoch": 11.0,
"learning_rate": 8.526315789473685e-06,
"loss": 0.0406,
"step": 1900
},
{
"epoch": 11.01,
"learning_rate": 8.5e-06,
"loss": 0.0656,
"step": 1925
},
{
"epoch": 11.01,
"learning_rate": 8.473684210526317e-06,
"loss": 0.0459,
"step": 1950
},
{
"epoch": 11.01,
"learning_rate": 8.447368421052632e-06,
"loss": 0.0526,
"step": 1975
},
{
"epoch": 11.01,
"learning_rate": 8.421052631578948e-06,
"loss": 0.0147,
"step": 2000
},
{
"epoch": 11.01,
"eval_loss": 0.34762969613075256,
"eval_runtime": 321.1139,
"eval_samples_per_second": 9.009,
"eval_steps_per_second": 0.283,
"eval_wer": 26.132656759695543,
"step": 2000
},
{
"epoch": 11.02,
"learning_rate": 8.394736842105263e-06,
"loss": 0.0055,
"step": 2025
},
{
"epoch": 12.0,
"learning_rate": 8.36842105263158e-06,
"loss": 0.022,
"step": 2050
},
{
"epoch": 12.0,
"learning_rate": 8.342105263157897e-06,
"loss": 0.0299,
"step": 2075
},
{
"epoch": 12.01,
"learning_rate": 8.315789473684212e-06,
"loss": 0.0439,
"step": 2100
},
{
"epoch": 12.01,
"learning_rate": 8.289473684210526e-06,
"loss": 0.0345,
"step": 2125
},
{
"epoch": 12.01,
"learning_rate": 8.263157894736843e-06,
"loss": 0.0266,
"step": 2150
},
{
"epoch": 12.01,
"learning_rate": 8.236842105263158e-06,
"loss": 0.0096,
"step": 2175
},
{
"epoch": 13.0,
"learning_rate": 8.210526315789475e-06,
"loss": 0.006,
"step": 2200
},
{
"epoch": 13.0,
"learning_rate": 8.18421052631579e-06,
"loss": 0.0143,
"step": 2225
},
{
"epoch": 13.01,
"learning_rate": 8.157894736842106e-06,
"loss": 0.0226,
"step": 2250
},
{
"epoch": 13.01,
"learning_rate": 8.131578947368421e-06,
"loss": 0.0239,
"step": 2275
},
{
"epoch": 13.01,
"learning_rate": 8.105263157894736e-06,
"loss": 0.0252,
"step": 2300
},
{
"epoch": 13.01,
"learning_rate": 8.078947368421053e-06,
"loss": 0.0137,
"step": 2325
},
{
"epoch": 13.02,
"learning_rate": 8.052631578947368e-06,
"loss": 0.0061,
"step": 2350
},
{
"epoch": 14.0,
"learning_rate": 8.026315789473685e-06,
"loss": 0.0068,
"step": 2375
},
{
"epoch": 14.0,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0102,
"step": 2400
},
{
"epoch": 14.01,
"learning_rate": 7.973684210526316e-06,
"loss": 0.0159,
"step": 2425
},
{
"epoch": 14.01,
"learning_rate": 7.947368421052633e-06,
"loss": 0.0147,
"step": 2450
},
{
"epoch": 14.01,
"learning_rate": 7.921052631578948e-06,
"loss": 0.0151,
"step": 2475
},
{
"epoch": 14.01,
"learning_rate": 7.894736842105265e-06,
"loss": 0.0071,
"step": 2500
},
{
"epoch": 14.01,
"eval_loss": 0.36233454942703247,
"eval_runtime": 324.1926,
"eval_samples_per_second": 8.924,
"eval_steps_per_second": 0.281,
"eval_wer": 25.933309169989126,
"step": 2500
},
{
"epoch": 14.02,
"learning_rate": 7.86842105263158e-06,
"loss": 0.004,
"step": 2525
},
{
"epoch": 15.0,
"learning_rate": 7.842105263157895e-06,
"loss": 0.0056,
"step": 2550
},
{
"epoch": 15.0,
"learning_rate": 7.815789473684211e-06,
"loss": 0.0071,
"step": 2575
},
{
"epoch": 15.01,
"learning_rate": 7.789473684210526e-06,
"loss": 0.0105,
"step": 2600
},
{
"epoch": 15.01,
"learning_rate": 7.763157894736843e-06,
"loss": 0.0086,
"step": 2625
},
{
"epoch": 15.01,
"learning_rate": 7.736842105263158e-06,
"loss": 0.0083,
"step": 2650
},
{
"epoch": 15.01,
"learning_rate": 7.710526315789474e-06,
"loss": 0.0042,
"step": 2675
},
{
"epoch": 15.02,
"learning_rate": 7.68421052631579e-06,
"loss": 0.0029,
"step": 2700
},
{
"epoch": 16.0,
"learning_rate": 7.657894736842106e-06,
"loss": 0.005,
"step": 2725
},
{
"epoch": 16.0,
"learning_rate": 7.631578947368423e-06,
"loss": 0.0081,
"step": 2750
},
{
"epoch": 16.01,
"learning_rate": 7.605263157894738e-06,
"loss": 0.0081,
"step": 2775
},
{
"epoch": 16.01,
"learning_rate": 7.578947368421054e-06,
"loss": 0.0061,
"step": 2800
},
{
"epoch": 16.01,
"learning_rate": 7.552631578947369e-06,
"loss": 0.005,
"step": 2825
},
{
"epoch": 16.01,
"learning_rate": 7.526315789473685e-06,
"loss": 0.004,
"step": 2850
},
{
"epoch": 17.0,
"learning_rate": 7.500000000000001e-06,
"loss": 0.0026,
"step": 2875
},
{
"epoch": 17.0,
"learning_rate": 7.473684210526316e-06,
"loss": 0.0047,
"step": 2900
},
{
"epoch": 17.01,
"learning_rate": 7.447368421052632e-06,
"loss": 0.0046,
"step": 2925
},
{
"epoch": 17.01,
"learning_rate": 7.421052631578948e-06,
"loss": 0.0068,
"step": 2950
},
{
"epoch": 17.01,
"learning_rate": 7.3947368421052635e-06,
"loss": 0.0059,
"step": 2975
},
{
"epoch": 17.01,
"learning_rate": 7.368421052631579e-06,
"loss": 0.0034,
"step": 3000
},
{
"epoch": 17.01,
"eval_loss": 0.3788630962371826,
"eval_runtime": 321.5331,
"eval_samples_per_second": 8.998,
"eval_steps_per_second": 0.283,
"eval_wer": 25.20840884378398,
"step": 3000
},
{
"epoch": 17.02,
"learning_rate": 7.342105263157895e-06,
"loss": 0.0033,
"step": 3025
},
{
"epoch": 18.0,
"learning_rate": 7.315789473684212e-06,
"loss": 0.0024,
"step": 3050
},
{
"epoch": 18.0,
"learning_rate": 7.289473684210528e-06,
"loss": 0.0035,
"step": 3075
},
{
"epoch": 18.01,
"learning_rate": 7.263157894736843e-06,
"loss": 0.0038,
"step": 3100
},
{
"epoch": 18.01,
"learning_rate": 7.236842105263158e-06,
"loss": 0.0044,
"step": 3125
},
{
"epoch": 18.01,
"learning_rate": 7.210526315789474e-06,
"loss": 0.0044,
"step": 3150
},
{
"epoch": 18.01,
"learning_rate": 7.18421052631579e-06,
"loss": 0.0029,
"step": 3175
},
{
"epoch": 18.02,
"learning_rate": 7.157894736842106e-06,
"loss": 0.0026,
"step": 3200
},
{
"epoch": 19.0,
"learning_rate": 7.131578947368422e-06,
"loss": 0.0028,
"step": 3225
},
{
"epoch": 19.0,
"learning_rate": 7.1052631578947375e-06,
"loss": 0.0039,
"step": 3250
},
{
"epoch": 19.01,
"learning_rate": 7.078947368421053e-06,
"loss": 0.0045,
"step": 3275
},
{
"epoch": 19.01,
"learning_rate": 7.052631578947369e-06,
"loss": 0.0032,
"step": 3300
},
{
"epoch": 19.01,
"learning_rate": 7.026315789473684e-06,
"loss": 0.0052,
"step": 3325
},
{
"epoch": 19.01,
"learning_rate": 7e-06,
"loss": 0.0027,
"step": 3350
},
{
"epoch": 19.02,
"learning_rate": 6.973684210526316e-06,
"loss": 0.0021,
"step": 3375
},
{
"epoch": 20.0,
"learning_rate": 6.947368421052632e-06,
"loss": 0.003,
"step": 3400
},
{
"epoch": 20.0,
"learning_rate": 6.921052631578948e-06,
"loss": 0.0024,
"step": 3425
},
{
"epoch": 20.01,
"learning_rate": 6.894736842105264e-06,
"loss": 0.0028,
"step": 3450
},
{
"epoch": 20.01,
"learning_rate": 6.86842105263158e-06,
"loss": 0.0041,
"step": 3475
},
{
"epoch": 20.01,
"learning_rate": 6.842105263157896e-06,
"loss": 0.0024,
"step": 3500
},
{
"epoch": 20.01,
"eval_loss": 0.3826642334461212,
"eval_runtime": 318.6137,
"eval_samples_per_second": 9.08,
"eval_steps_per_second": 0.286,
"eval_wer": 24.864081188836533,
"step": 3500
},
{
"epoch": 20.01,
"learning_rate": 6.8157894736842115e-06,
"loss": 0.0019,
"step": 3525
},
{
"epoch": 21.0,
"learning_rate": 6.789473684210527e-06,
"loss": 0.0015,
"step": 3550
},
{
"epoch": 21.0,
"learning_rate": 6.763157894736842e-06,
"loss": 0.0018,
"step": 3575
},
{
"epoch": 21.01,
"learning_rate": 6.736842105263158e-06,
"loss": 0.0027,
"step": 3600
},
{
"epoch": 21.01,
"learning_rate": 6.710526315789474e-06,
"loss": 0.0021,
"step": 3625
},
{
"epoch": 21.01,
"learning_rate": 6.68421052631579e-06,
"loss": 0.0031,
"step": 3650
},
{
"epoch": 21.01,
"learning_rate": 6.6578947368421055e-06,
"loss": 0.002,
"step": 3675
},
{
"epoch": 21.02,
"learning_rate": 6.631578947368421e-06,
"loss": 0.0016,
"step": 3700
},
{
"epoch": 22.0,
"learning_rate": 6.605263157894738e-06,
"loss": 0.0016,
"step": 3725
},
{
"epoch": 22.0,
"learning_rate": 6.578947368421054e-06,
"loss": 0.0027,
"step": 3750
},
{
"epoch": 22.01,
"learning_rate": 6.55263157894737e-06,
"loss": 0.0031,
"step": 3775
},
{
"epoch": 22.01,
"learning_rate": 6.526315789473685e-06,
"loss": 0.0026,
"step": 3800
},
{
"epoch": 22.01,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.0021,
"step": 3825
},
{
"epoch": 22.01,
"learning_rate": 6.473684210526316e-06,
"loss": 0.0012,
"step": 3850
},
{
"epoch": 22.02,
"learning_rate": 6.447368421052632e-06,
"loss": 0.0015,
"step": 3875
},
{
"epoch": 23.0,
"learning_rate": 6.421052631578948e-06,
"loss": 0.0012,
"step": 3900
},
{
"epoch": 23.0,
"learning_rate": 6.394736842105264e-06,
"loss": 0.0012,
"step": 3925
},
{
"epoch": 23.01,
"learning_rate": 6.3684210526315795e-06,
"loss": 0.0027,
"step": 3950
},
{
"epoch": 23.01,
"learning_rate": 6.342105263157895e-06,
"loss": 0.0019,
"step": 3975
},
{
"epoch": 23.01,
"learning_rate": 6.31578947368421e-06,
"loss": 0.0026,
"step": 4000
},
{
"epoch": 23.01,
"eval_loss": 0.3877031207084656,
"eval_runtime": 319.9159,
"eval_samples_per_second": 9.043,
"eval_steps_per_second": 0.284,
"eval_wer": 25.31714389271475,
"step": 4000
},
{
"epoch": 23.01,
"learning_rate": 6.289473684210526e-06,
"loss": 0.0011,
"step": 4025
},
{
"epoch": 23.02,
"learning_rate": 6.263157894736842e-06,
"loss": 0.0014,
"step": 4050
},
{
"epoch": 24.0,
"learning_rate": 6.236842105263159e-06,
"loss": 0.0014,
"step": 4075
},
{
"epoch": 24.0,
"learning_rate": 6.2105263157894745e-06,
"loss": 0.0015,
"step": 4100
},
{
"epoch": 24.01,
"learning_rate": 6.18421052631579e-06,
"loss": 0.0023,
"step": 4125
},
{
"epoch": 24.01,
"learning_rate": 6.157894736842106e-06,
"loss": 0.002,
"step": 4150
},
{
"epoch": 24.01,
"learning_rate": 6.131578947368422e-06,
"loss": 0.0013,
"step": 4175
},
{
"epoch": 24.01,
"learning_rate": 6.105263157894738e-06,
"loss": 0.0012,
"step": 4200
},
{
"epoch": 24.02,
"learning_rate": 6.0789473684210535e-06,
"loss": 0.0012,
"step": 4225
},
{
"epoch": 25.0,
"learning_rate": 6.0526315789473685e-06,
"loss": 0.0015,
"step": 4250
},
{
"epoch": 25.0,
"learning_rate": 6.026315789473684e-06,
"loss": 0.0012,
"step": 4275
},
{
"epoch": 25.01,
"learning_rate": 6e-06,
"loss": 0.0012,
"step": 4300
},
{
"epoch": 25.01,
"learning_rate": 5.973684210526316e-06,
"loss": 0.0024,
"step": 4325
},
{
"epoch": 25.01,
"learning_rate": 5.947368421052632e-06,
"loss": 0.0012,
"step": 4350
},
{
"epoch": 25.02,
"learning_rate": 5.921052631578948e-06,
"loss": 0.0011,
"step": 4375
},
{
"epoch": 26.0,
"learning_rate": 5.8947368421052634e-06,
"loss": 0.001,
"step": 4400
},
{
"epoch": 26.0,
"learning_rate": 5.86842105263158e-06,
"loss": 0.0017,
"step": 4425
},
{
"epoch": 26.01,
"learning_rate": 5.842105263157896e-06,
"loss": 0.0022,
"step": 4450
},
{
"epoch": 26.01,
"learning_rate": 5.815789473684212e-06,
"loss": 0.0015,
"step": 4475
},
{
"epoch": 26.01,
"learning_rate": 5.789473684210527e-06,
"loss": 0.0021,
"step": 4500
},
{
"epoch": 26.01,
"eval_loss": 0.39328622817993164,
"eval_runtime": 321.0452,
"eval_samples_per_second": 9.011,
"eval_steps_per_second": 0.283,
"eval_wer": 25.425878941645525,
"step": 4500
},
{
"epoch": 26.01,
"learning_rate": 5.7631578947368425e-06,
"loss": 0.0011,
"step": 4525
},
{
"epoch": 26.02,
"learning_rate": 5.736842105263158e-06,
"loss": 0.0011,
"step": 4550
},
{
"epoch": 27.0,
"learning_rate": 5.710526315789474e-06,
"loss": 0.001,
"step": 4575
},
{
"epoch": 27.0,
"learning_rate": 5.68421052631579e-06,
"loss": 0.0009,
"step": 4600
},
{
"epoch": 27.01,
"learning_rate": 5.657894736842106e-06,
"loss": 0.0008,
"step": 4625
},
{
"epoch": 27.01,
"learning_rate": 5.631578947368422e-06,
"loss": 0.0008,
"step": 4650
},
{
"epoch": 27.01,
"learning_rate": 5.605263157894737e-06,
"loss": 0.0016,
"step": 4675
},
{
"epoch": 27.01,
"learning_rate": 5.578947368421052e-06,
"loss": 0.0006,
"step": 4700
},
{
"epoch": 27.02,
"learning_rate": 5.552631578947368e-06,
"loss": 0.001,
"step": 4725
},
{
"epoch": 28.0,
"learning_rate": 5.526315789473685e-06,
"loss": 0.0007,
"step": 4750
},
{
"epoch": 28.0,
"learning_rate": 5.500000000000001e-06,
"loss": 0.0007,
"step": 4775
},
{
"epoch": 28.01,
"learning_rate": 5.4736842105263165e-06,
"loss": 0.0008,
"step": 4800
},
{
"epoch": 28.01,
"learning_rate": 5.447368421052632e-06,
"loss": 0.0008,
"step": 4825
},
{
"epoch": 28.01,
"learning_rate": 5.421052631578948e-06,
"loss": 0.0007,
"step": 4850
},
{
"epoch": 28.01,
"learning_rate": 5.394736842105264e-06,
"loss": 0.0007,
"step": 4875
},
{
"epoch": 28.02,
"learning_rate": 5.36842105263158e-06,
"loss": 0.0008,
"step": 4900
},
{
"epoch": 29.0,
"learning_rate": 5.342105263157895e-06,
"loss": 0.0006,
"step": 4925
},
{
"epoch": 29.0,
"learning_rate": 5.315789473684211e-06,
"loss": 0.0006,
"step": 4950
},
{
"epoch": 29.01,
"learning_rate": 5.289473684210526e-06,
"loss": 0.0014,
"step": 4975
},
{
"epoch": 29.01,
"learning_rate": 5.263157894736842e-06,
"loss": 0.0014,
"step": 5000
},
{
"epoch": 29.01,
"eval_loss": 0.3941277861595154,
"eval_runtime": 326.072,
"eval_samples_per_second": 8.872,
"eval_steps_per_second": 0.279,
"eval_wer": 25.099673794853206,
"step": 5000
},
{
"epoch": 29.01,
"learning_rate": 5.236842105263158e-06,
"loss": 0.0006,
"step": 5025
},
{
"epoch": 29.01,
"learning_rate": 5.210526315789474e-06,
"loss": 0.0008,
"step": 5050
},
{
"epoch": 30.0,
"learning_rate": 5.18421052631579e-06,
"loss": 0.0007,
"step": 5075
},
{
"epoch": 30.0,
"learning_rate": 5.157894736842106e-06,
"loss": 0.0006,
"step": 5100
},
{
"epoch": 30.01,
"learning_rate": 5.131578947368422e-06,
"loss": 0.0006,
"step": 5125
},
{
"epoch": 30.01,
"learning_rate": 5.105263157894738e-06,
"loss": 0.0006,
"step": 5150
},
{
"epoch": 30.01,
"learning_rate": 5.078947368421053e-06,
"loss": 0.0008,
"step": 5175
},
{
"epoch": 30.01,
"learning_rate": 5.052631578947369e-06,
"loss": 0.0005,
"step": 5200
},
{
"epoch": 30.02,
"learning_rate": 5.026315789473685e-06,
"loss": 0.0008,
"step": 5225
},
{
"epoch": 31.0,
"learning_rate": 5e-06,
"loss": 0.0005,
"step": 5250
},
{
"epoch": 31.0,
"learning_rate": 4.973684210526316e-06,
"loss": 0.0006,
"step": 5275
},
{
"epoch": 31.01,
"learning_rate": 4.947368421052632e-06,
"loss": 0.0005,
"step": 5300
},
{
"epoch": 31.01,
"learning_rate": 4.921052631578948e-06,
"loss": 0.0011,
"step": 5325
},
{
"epoch": 31.01,
"learning_rate": 4.894736842105264e-06,
"loss": 0.001,
"step": 5350
},
{
"epoch": 31.01,
"learning_rate": 4.8684210526315795e-06,
"loss": 0.0004,
"step": 5375
},
{
"epoch": 31.02,
"learning_rate": 4.842105263157895e-06,
"loss": 0.0008,
"step": 5400
},
{
"epoch": 32.0,
"learning_rate": 4.815789473684211e-06,
"loss": 0.0005,
"step": 5425
},
{
"epoch": 32.0,
"learning_rate": 4.789473684210527e-06,
"loss": 0.0005,
"step": 5450
},
{
"epoch": 32.01,
"learning_rate": 4.763157894736842e-06,
"loss": 0.0005,
"step": 5475
},
{
"epoch": 32.01,
"learning_rate": 4.736842105263158e-06,
"loss": 0.0008,
"step": 5500
},
{
"epoch": 32.01,
"eval_loss": 0.40144988894462585,
"eval_runtime": 321.1338,
"eval_samples_per_second": 9.009,
"eval_steps_per_second": 0.283,
"eval_wer": 25.099673794853206,
"step": 5500
},
{
"epoch": 32.01,
"learning_rate": 4.710526315789474e-06,
"loss": 0.0005,
"step": 5525
},
{
"epoch": 32.01,
"learning_rate": 4.68421052631579e-06,
"loss": 0.0005,
"step": 5550
},
{
"epoch": 32.02,
"learning_rate": 4.657894736842106e-06,
"loss": 0.0007,
"step": 5575
},
{
"epoch": 33.0,
"learning_rate": 4.631578947368421e-06,
"loss": 0.0004,
"step": 5600
},
{
"epoch": 33.0,
"learning_rate": 4.605263157894737e-06,
"loss": 0.0004,
"step": 5625
},
{
"epoch": 33.01,
"learning_rate": 4.578947368421053e-06,
"loss": 0.0005,
"step": 5650
},
{
"epoch": 33.01,
"learning_rate": 4.552631578947369e-06,
"loss": 0.0007,
"step": 5675
},
{
"epoch": 33.01,
"learning_rate": 4.526315789473685e-06,
"loss": 0.0004,
"step": 5700
},
{
"epoch": 33.01,
"learning_rate": 4.5e-06,
"loss": 0.0006,
"step": 5725
},
{
"epoch": 34.0,
"learning_rate": 4.473684210526316e-06,
"loss": 0.0005,
"step": 5750
},
{
"epoch": 34.0,
"learning_rate": 4.447368421052632e-06,
"loss": 0.0004,
"step": 5775
},
{
"epoch": 34.01,
"learning_rate": 4.4210526315789476e-06,
"loss": 0.0004,
"step": 5800
},
{
"epoch": 34.01,
"learning_rate": 4.394736842105263e-06,
"loss": 0.0006,
"step": 5825
},
{
"epoch": 34.01,
"learning_rate": 4.368421052631579e-06,
"loss": 0.0005,
"step": 5850
},
{
"epoch": 34.01,
"learning_rate": 4.342105263157895e-06,
"loss": 0.0003,
"step": 5875
},
{
"epoch": 34.02,
"learning_rate": 4.315789473684211e-06,
"loss": 0.0006,
"step": 5900
},
{
"epoch": 35.0,
"learning_rate": 4.289473684210527e-06,
"loss": 0.0004,
"step": 5925
},
{
"epoch": 35.0,
"learning_rate": 4.2631578947368425e-06,
"loss": 0.0004,
"step": 5950
},
{
"epoch": 35.01,
"learning_rate": 4.236842105263158e-06,
"loss": 0.0004,
"step": 5975
},
{
"epoch": 35.01,
"learning_rate": 4.210526315789474e-06,
"loss": 0.0004,
"step": 6000
},
{
"epoch": 35.01,
"eval_loss": 0.40347200632095337,
"eval_runtime": 319.8618,
"eval_samples_per_second": 9.045,
"eval_steps_per_second": 0.284,
"eval_wer": 24.827836172526276,
"step": 6000
},
{
"epoch": 35.01,
"learning_rate": 4.18421052631579e-06,
"loss": 0.0004,
"step": 6025
},
{
"epoch": 35.01,
"learning_rate": 4.157894736842106e-06,
"loss": 0.0003,
"step": 6050
},
{
"epoch": 35.02,
"learning_rate": 4.1315789473684216e-06,
"loss": 0.0006,
"step": 6075
},
{
"epoch": 36.0,
"learning_rate": 4.105263157894737e-06,
"loss": 0.0004,
"step": 6100
},
{
"epoch": 36.0,
"learning_rate": 4.078947368421053e-06,
"loss": 0.0004,
"step": 6125
},
{
"epoch": 36.01,
"learning_rate": 4.052631578947368e-06,
"loss": 0.0003,
"step": 6150
},
{
"epoch": 36.01,
"learning_rate": 4.026315789473684e-06,
"loss": 0.0004,
"step": 6175
},
{
"epoch": 36.01,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0004,
"step": 6200
},
{
"epoch": 36.01,
"learning_rate": 3.9736842105263165e-06,
"loss": 0.0003,
"step": 6225
},
{
"epoch": 36.02,
"learning_rate": 3.947368421052632e-06,
"loss": 0.0005,
"step": 6250
},
{
"epoch": 37.0,
"learning_rate": 3.921052631578947e-06,
"loss": 0.0003,
"step": 6275
},
{
"epoch": 37.0,
"learning_rate": 3.894736842105263e-06,
"loss": 0.0003,
"step": 6300
},
{
"epoch": 37.01,
"learning_rate": 3.868421052631579e-06,
"loss": 0.0003,
"step": 6325
},
{
"epoch": 37.01,
"learning_rate": 3.842105263157895e-06,
"loss": 0.0003,
"step": 6350
},
{
"epoch": 37.01,
"learning_rate": 3.815789473684211e-06,
"loss": 0.0003,
"step": 6375
},
{
"epoch": 37.01,
"learning_rate": 3.789473684210527e-06,
"loss": 0.0004,
"step": 6400
},
{
"epoch": 38.0,
"learning_rate": 3.7631578947368426e-06,
"loss": 0.0004,
"step": 6425
},
{
"epoch": 38.0,
"learning_rate": 3.736842105263158e-06,
"loss": 0.0003,
"step": 6450
},
{
"epoch": 38.01,
"learning_rate": 3.710526315789474e-06,
"loss": 0.0003,
"step": 6475
},
{
"epoch": 38.01,
"learning_rate": 3.6842105263157896e-06,
"loss": 0.0003,
"step": 6500
},
{
"epoch": 38.01,
"eval_loss": 0.40795087814331055,
"eval_runtime": 321.9686,
"eval_samples_per_second": 8.985,
"eval_steps_per_second": 0.283,
"eval_wer": 24.91844871330192,
"step": 6500
},
{
"epoch": 38.01,
"learning_rate": 3.657894736842106e-06,
"loss": 0.0003,
"step": 6525
},
{
"epoch": 38.01,
"learning_rate": 3.6315789473684217e-06,
"loss": 0.0003,
"step": 6550
},
{
"epoch": 38.02,
"learning_rate": 3.605263157894737e-06,
"loss": 0.0005,
"step": 6575
},
{
"epoch": 39.0,
"learning_rate": 3.578947368421053e-06,
"loss": 0.0003,
"step": 6600
},
{
"epoch": 39.0,
"learning_rate": 3.5526315789473687e-06,
"loss": 0.0003,
"step": 6625
},
{
"epoch": 39.01,
"learning_rate": 3.5263157894736846e-06,
"loss": 0.0003,
"step": 6650
},
{
"epoch": 39.01,
"learning_rate": 3.5e-06,
"loss": 0.0003,
"step": 6675
},
{
"epoch": 39.01,
"learning_rate": 3.473684210526316e-06,
"loss": 0.0003,
"step": 6700
},
{
"epoch": 39.01,
"learning_rate": 3.447368421052632e-06,
"loss": 0.0002,
"step": 6725
},
{
"epoch": 39.02,
"learning_rate": 3.421052631578948e-06,
"loss": 0.0005,
"step": 6750
},
{
"epoch": 40.0,
"learning_rate": 3.3947368421052636e-06,
"loss": 0.0003,
"step": 6775
},
{
"epoch": 40.0,
"learning_rate": 3.368421052631579e-06,
"loss": 0.0003,
"step": 6800
},
{
"epoch": 40.01,
"learning_rate": 3.342105263157895e-06,
"loss": 0.0003,
"step": 6825
},
{
"epoch": 40.01,
"learning_rate": 3.3157894736842107e-06,
"loss": 0.0003,
"step": 6850
},
{
"epoch": 40.01,
"learning_rate": 3.289473684210527e-06,
"loss": 0.0003,
"step": 6875
},
{
"epoch": 40.01,
"learning_rate": 3.2631578947368423e-06,
"loss": 0.0002,
"step": 6900
},
{
"epoch": 40.02,
"learning_rate": 3.236842105263158e-06,
"loss": 0.0005,
"step": 6925
},
{
"epoch": 41.0,
"learning_rate": 3.210526315789474e-06,
"loss": 0.0003,
"step": 6950
},
{
"epoch": 41.0,
"learning_rate": 3.1842105263157898e-06,
"loss": 0.0002,
"step": 6975
},
{
"epoch": 41.01,
"learning_rate": 3.157894736842105e-06,
"loss": 0.0003,
"step": 7000
},
{
"epoch": 41.01,
"eval_loss": 0.41202512383461,
"eval_runtime": 320.9926,
"eval_samples_per_second": 9.013,
"eval_steps_per_second": 0.283,
"eval_wer": 24.80971366437115,
"step": 7000
},
{
"epoch": 41.01,
"learning_rate": 3.131578947368421e-06,
"loss": 0.0003,
"step": 7025
},
{
"epoch": 41.01,
"learning_rate": 3.1052631578947372e-06,
"loss": 0.0002,
"step": 7050
},
{
"epoch": 41.01,
"learning_rate": 3.078947368421053e-06,
"loss": 0.0003,
"step": 7075
},
{
"epoch": 42.0,
"learning_rate": 3.052631578947369e-06,
"loss": 0.0003,
"step": 7100
},
{
"epoch": 42.0,
"learning_rate": 3.0263157894736843e-06,
"loss": 0.0002,
"step": 7125
},
{
"epoch": 42.01,
"learning_rate": 3e-06,
"loss": 0.0002,
"step": 7150
},
{
"epoch": 42.01,
"learning_rate": 2.973684210526316e-06,
"loss": 0.0002,
"step": 7175
},
{
"epoch": 42.01,
"learning_rate": 2.9473684210526317e-06,
"loss": 0.0003,
"step": 7200
},
{
"epoch": 42.01,
"learning_rate": 2.921052631578948e-06,
"loss": 0.0002,
"step": 7225
},
{
"epoch": 42.02,
"learning_rate": 2.8947368421052634e-06,
"loss": 0.0004,
"step": 7250
},
{
"epoch": 43.0,
"learning_rate": 2.868421052631579e-06,
"loss": 0.0003,
"step": 7275
},
{
"epoch": 43.0,
"learning_rate": 2.842105263157895e-06,
"loss": 0.0002,
"step": 7300
},
{
"epoch": 43.01,
"learning_rate": 2.815789473684211e-06,
"loss": 0.0002,
"step": 7325
},
{
"epoch": 43.01,
"learning_rate": 2.789473684210526e-06,
"loss": 0.0002,
"step": 7350
},
{
"epoch": 43.01,
"learning_rate": 2.7631578947368424e-06,
"loss": 0.0002,
"step": 7375
},
{
"epoch": 43.01,
"learning_rate": 2.7368421052631583e-06,
"loss": 0.0002,
"step": 7400
},
{
"epoch": 43.02,
"learning_rate": 2.710526315789474e-06,
"loss": 0.0004,
"step": 7425
},
{
"epoch": 44.0,
"learning_rate": 2.68421052631579e-06,
"loss": 0.0002,
"step": 7450
},
{
"epoch": 44.0,
"learning_rate": 2.6578947368421053e-06,
"loss": 0.0002,
"step": 7475
},
{
"epoch": 44.01,
"learning_rate": 2.631578947368421e-06,
"loss": 0.0002,
"step": 7500
},
{
"epoch": 44.01,
"eval_loss": 0.41507139801979065,
"eval_runtime": 322.4996,
"eval_samples_per_second": 8.971,
"eval_steps_per_second": 0.282,
"eval_wer": 24.610366074664732,
"step": 7500
},
{
"epoch": 44.01,
"learning_rate": 2.605263157894737e-06,
"loss": 0.0002,
"step": 7525
},
{
"epoch": 44.01,
"learning_rate": 2.578947368421053e-06,
"loss": 0.0002,
"step": 7550
},
{
"epoch": 44.01,
"learning_rate": 2.552631578947369e-06,
"loss": 0.0002,
"step": 7575
},
{
"epoch": 44.02,
"learning_rate": 2.5263157894736844e-06,
"loss": 0.0004,
"step": 7600
},
{
"epoch": 45.0,
"learning_rate": 2.5e-06,
"loss": 0.0002,
"step": 7625
},
{
"epoch": 45.0,
"learning_rate": 2.473684210526316e-06,
"loss": 0.0002,
"step": 7650
},
{
"epoch": 45.01,
"learning_rate": 2.447368421052632e-06,
"loss": 0.0002,
"step": 7675
},
{
"epoch": 45.01,
"learning_rate": 2.4210526315789477e-06,
"loss": 0.0002,
"step": 7700
},
{
"epoch": 45.01,
"learning_rate": 2.3947368421052635e-06,
"loss": 0.0002,
"step": 7725
},
{
"epoch": 45.01,
"learning_rate": 2.368421052631579e-06,
"loss": 0.0003,
"step": 7750
},
{
"epoch": 46.0,
"learning_rate": 2.342105263157895e-06,
"loss": 0.0003,
"step": 7775
},
{
"epoch": 46.0,
"learning_rate": 2.3157894736842105e-06,
"loss": 0.0002,
"step": 7800
},
{
"epoch": 46.01,
"learning_rate": 2.2894736842105263e-06,
"loss": 0.0002,
"step": 7825
},
{
"epoch": 46.01,
"learning_rate": 2.2631578947368426e-06,
"loss": 0.0002,
"step": 7850
},
{
"epoch": 46.01,
"learning_rate": 2.236842105263158e-06,
"loss": 0.0002,
"step": 7875
},
{
"epoch": 46.01,
"learning_rate": 2.2105263157894738e-06,
"loss": 0.0002,
"step": 7900
},
{
"epoch": 46.02,
"learning_rate": 2.1842105263157896e-06,
"loss": 0.0003,
"step": 7925
},
{
"epoch": 47.0,
"learning_rate": 2.1578947368421054e-06,
"loss": 0.0002,
"step": 7950
},
{
"epoch": 47.0,
"learning_rate": 2.1315789473684212e-06,
"loss": 0.0002,
"step": 7975
},
{
"epoch": 47.01,
"learning_rate": 2.105263157894737e-06,
"loss": 0.0002,
"step": 8000
},
{
"epoch": 47.01,
"eval_loss": 0.4176327586174011,
"eval_runtime": 320.9516,
"eval_samples_per_second": 9.014,
"eval_steps_per_second": 0.284,
"eval_wer": 24.39289597680319,
"step": 8000
},
{
"epoch": 47.01,
"learning_rate": 2.078947368421053e-06,
"loss": 0.0002,
"step": 8025
},
{
"epoch": 47.01,
"learning_rate": 2.0526315789473687e-06,
"loss": 0.0002,
"step": 8050
},
{
"epoch": 47.01,
"learning_rate": 2.026315789473684e-06,
"loss": 0.0002,
"step": 8075
},
{
"epoch": 47.02,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.0003,
"step": 8100
},
{
"epoch": 48.0,
"learning_rate": 1.973684210526316e-06,
"loss": 0.0002,
"step": 8125
},
{
"epoch": 48.0,
"learning_rate": 1.9473684210526315e-06,
"loss": 0.0002,
"step": 8150
},
{
"epoch": 48.01,
"learning_rate": 1.9210526315789474e-06,
"loss": 0.0002,
"step": 8175
},
{
"epoch": 48.01,
"learning_rate": 1.8947368421052634e-06,
"loss": 0.0002,
"step": 8200
},
{
"epoch": 48.01,
"learning_rate": 1.868421052631579e-06,
"loss": 0.0002,
"step": 8225
},
{
"epoch": 48.01,
"learning_rate": 1.8421052631578948e-06,
"loss": 0.0002,
"step": 8250
},
{
"epoch": 48.02,
"learning_rate": 1.8157894736842109e-06,
"loss": 0.0003,
"step": 8275
},
{
"epoch": 49.0,
"learning_rate": 1.7894736842105265e-06,
"loss": 0.0002,
"step": 8300
},
{
"epoch": 49.0,
"learning_rate": 1.7631578947368423e-06,
"loss": 0.0002,
"step": 8325
},
{
"epoch": 49.01,
"learning_rate": 1.736842105263158e-06,
"loss": 0.0002,
"step": 8350
},
{
"epoch": 49.01,
"learning_rate": 1.710526315789474e-06,
"loss": 0.0002,
"step": 8375
},
{
"epoch": 49.01,
"learning_rate": 1.6842105263157895e-06,
"loss": 0.0002,
"step": 8400
},
{
"epoch": 49.01,
"learning_rate": 1.6578947368421053e-06,
"loss": 0.0002,
"step": 8425
},
{
"epoch": 49.02,
"learning_rate": 1.6315789473684212e-06,
"loss": 0.0003,
"step": 8450
},
{
"epoch": 50.0,
"learning_rate": 1.605263157894737e-06,
"loss": 0.0002,
"step": 8475
},
{
"epoch": 50.01,
"learning_rate": 1.5789473684210526e-06,
"loss": 0.0002,
"step": 8500
},
{
"epoch": 50.01,
"eval_loss": 0.4200422167778015,
"eval_runtime": 319.2495,
"eval_samples_per_second": 9.062,
"eval_steps_per_second": 0.285,
"eval_wer": 24.51975353388909,
"step": 8500
},
{
"epoch": 50.01,
"learning_rate": 1.5526315789473686e-06,
"loss": 0.0002,
"step": 8525
},
{
"epoch": 50.01,
"learning_rate": 1.5263157894736844e-06,
"loss": 0.0002,
"step": 8550
},
{
"epoch": 50.01,
"learning_rate": 1.5e-06,
"loss": 0.0001,
"step": 8575
},
{
"epoch": 50.02,
"learning_rate": 1.4736842105263159e-06,
"loss": 0.0003,
"step": 8600
},
{
"epoch": 51.0,
"learning_rate": 1.4473684210526317e-06,
"loss": 0.0002,
"step": 8625
},
{
"epoch": 51.0,
"learning_rate": 1.4210526315789475e-06,
"loss": 0.0001,
"step": 8650
},
{
"epoch": 51.01,
"learning_rate": 1.394736842105263e-06,
"loss": 0.0002,
"step": 8675
},
{
"epoch": 51.01,
"learning_rate": 1.3684210526315791e-06,
"loss": 0.0001,
"step": 8700
},
{
"epoch": 51.01,
"learning_rate": 1.342105263157895e-06,
"loss": 0.0002,
"step": 8725
},
{
"epoch": 51.01,
"learning_rate": 1.3157894736842106e-06,
"loss": 0.0001,
"step": 8750
},
{
"epoch": 51.02,
"learning_rate": 1.2894736842105266e-06,
"loss": 0.0003,
"step": 8775
},
{
"epoch": 52.0,
"learning_rate": 1.2631578947368422e-06,
"loss": 0.0002,
"step": 8800
},
{
"epoch": 52.0,
"learning_rate": 1.236842105263158e-06,
"loss": 0.0001,
"step": 8825
},
{
"epoch": 52.01,
"learning_rate": 1.2105263157894738e-06,
"loss": 0.0002,
"step": 8850
},
{
"epoch": 52.01,
"learning_rate": 1.1842105263157894e-06,
"loss": 0.0001,
"step": 8875
},
{
"epoch": 52.01,
"learning_rate": 1.1578947368421053e-06,
"loss": 0.0002,
"step": 8900
},
{
"epoch": 52.01,
"learning_rate": 1.1315789473684213e-06,
"loss": 0.0001,
"step": 8925
},
{
"epoch": 52.02,
"learning_rate": 1.1052631578947369e-06,
"loss": 0.0003,
"step": 8950
},
{
"epoch": 53.0,
"learning_rate": 1.0789473684210527e-06,
"loss": 0.0001,
"step": 8975
},
{
"epoch": 53.0,
"learning_rate": 1.0526315789473685e-06,
"loss": 0.0001,
"step": 9000
},
{
"epoch": 53.0,
"eval_loss": 0.42296102643013,
"eval_runtime": 320.1088,
"eval_samples_per_second": 9.038,
"eval_steps_per_second": 0.284,
"eval_wer": 24.51975353388909,
"step": 9000
},
{
"epoch": 53.01,
"learning_rate": 1.0263157894736843e-06,
"loss": 0.0001,
"step": 9025
},
{
"epoch": 53.01,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.0001,
"step": 9050
},
{
"epoch": 53.01,
"learning_rate": 9.736842105263158e-07,
"loss": 0.0001,
"step": 9075
},
{
"epoch": 53.01,
"learning_rate": 9.473684210526317e-07,
"loss": 0.0002,
"step": 9100
},
{
"epoch": 53.02,
"learning_rate": 9.210526315789474e-07,
"loss": 0.0002,
"step": 9125
},
{
"epoch": 54.0,
"learning_rate": 8.947368421052632e-07,
"loss": 0.0001,
"step": 9150
},
{
"epoch": 54.0,
"learning_rate": 8.68421052631579e-07,
"loss": 0.0001,
"step": 9175
},
{
"epoch": 54.01,
"learning_rate": 8.421052631578948e-07,
"loss": 0.0001,
"step": 9200
},
{
"epoch": 54.01,
"learning_rate": 8.157894736842106e-07,
"loss": 0.0001,
"step": 9225
},
{
"epoch": 54.01,
"learning_rate": 7.894736842105263e-07,
"loss": 0.0001,
"step": 9250
},
{
"epoch": 54.01,
"learning_rate": 7.631578947368422e-07,
"loss": 0.0002,
"step": 9275
},
{
"epoch": 55.0,
"learning_rate": 7.368421052631579e-07,
"loss": 0.0002,
"step": 9300
},
{
"epoch": 55.0,
"learning_rate": 7.105263157894737e-07,
"loss": 0.0001,
"step": 9325
},
{
"epoch": 55.01,
"learning_rate": 6.842105263157896e-07,
"loss": 0.0001,
"step": 9350
},
{
"epoch": 55.01,
"learning_rate": 6.578947368421053e-07,
"loss": 0.0001,
"step": 9375
},
{
"epoch": 55.01,
"learning_rate": 6.315789473684211e-07,
"loss": 0.0001,
"step": 9400
},
{
"epoch": 55.01,
"learning_rate": 6.052631578947369e-07,
"loss": 0.0001,
"step": 9425
},
{
"epoch": 55.02,
"learning_rate": 5.789473684210526e-07,
"loss": 0.0003,
"step": 9450
},
{
"epoch": 56.0,
"learning_rate": 5.526315789473684e-07,
"loss": 0.0002,
"step": 9475
},
{
"epoch": 56.0,
"learning_rate": 5.263157894736843e-07,
"loss": 0.0001,
"step": 9500
},
{
"epoch": 56.0,
"eval_loss": 0.42520004510879517,
"eval_runtime": 323.521,
"eval_samples_per_second": 8.942,
"eval_steps_per_second": 0.281,
"eval_wer": 24.429140993113446,
"step": 9500
},
{
"epoch": 56.01,
"learning_rate": 5.000000000000001e-07,
"loss": 0.0001,
"step": 9525
},
{
"epoch": 56.01,
"learning_rate": 4.7368421052631585e-07,
"loss": 0.0001,
"step": 9550
},
{
"epoch": 56.01,
"learning_rate": 4.473684210526316e-07,
"loss": 0.0001,
"step": 9575
},
{
"epoch": 56.01,
"learning_rate": 4.210526315789474e-07,
"loss": 0.0001,
"step": 9600
},
{
"epoch": 56.02,
"learning_rate": 3.9473684210526315e-07,
"loss": 0.0003,
"step": 9625
},
{
"epoch": 57.0,
"learning_rate": 3.6842105263157896e-07,
"loss": 0.0001,
"step": 9650
},
{
"epoch": 57.0,
"learning_rate": 3.421052631578948e-07,
"loss": 0.0001,
"step": 9675
},
{
"epoch": 57.01,
"learning_rate": 3.1578947368421055e-07,
"loss": 0.0001,
"step": 9700
},
{
"epoch": 57.01,
"learning_rate": 2.894736842105263e-07,
"loss": 0.0001,
"step": 9725
},
{
"epoch": 57.01,
"learning_rate": 2.6315789473684213e-07,
"loss": 0.0001,
"step": 9750
},
{
"epoch": 57.01,
"learning_rate": 2.3684210526315792e-07,
"loss": 0.0002,
"step": 9775
},
{
"epoch": 57.02,
"learning_rate": 2.105263157894737e-07,
"loss": 0.0002,
"step": 9800
},
{
"epoch": 58.0,
"learning_rate": 1.8421052631578948e-07,
"loss": 0.0001,
"step": 9825
},
{
"epoch": 58.0,
"learning_rate": 1.5789473684210527e-07,
"loss": 0.0001,
"step": 9850
},
{
"epoch": 58.01,
"learning_rate": 1.3157894736842107e-07,
"loss": 0.0001,
"step": 9875
},
{
"epoch": 58.01,
"learning_rate": 1.0526315789473685e-07,
"loss": 0.0001,
"step": 9900
},
{
"epoch": 58.01,
"learning_rate": 7.894736842105264e-08,
"loss": 0.0001,
"step": 9925
},
{
"epoch": 58.01,
"learning_rate": 5.263157894736842e-08,
"loss": 0.0002,
"step": 9950
},
{
"epoch": 59.0,
"learning_rate": 2.631578947368421e-08,
"loss": 0.0002,
"step": 9975
},
{
"epoch": 59.0,
"learning_rate": 0.0,
"loss": 0.0001,
"step": 10000
},
{
"epoch": 59.0,
"eval_loss": 0.4241657555103302,
"eval_runtime": 322.2889,
"eval_samples_per_second": 8.976,
"eval_steps_per_second": 0.282,
"eval_wer": 24.555998550199348,
"step": 10000
},
{
"epoch": 59.0,
"step": 10000,
"total_flos": 1.84268992739328e+20,
"train_loss": 0.06514070921628735,
"train_runtime": 43812.9748,
"train_samples_per_second": 14.608,
"train_steps_per_second": 0.228
}
],
"max_steps": 10000,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.84268992739328e+20,
"trial_name": null,
"trial_params": null
}