whisper-small-ja-elite / trainer_state.json
noflm
End of training
6d501fa
raw
history blame
44.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 188.0036,
"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": 2.2e-06,
"loss": 0.7826,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 4.7e-06,
"loss": 0.5245,
"step": 50
},
{
"epoch": 1.0,
"learning_rate": 7.2000000000000005e-06,
"loss": 0.3614,
"step": 75
},
{
"epoch": 1.0,
"learning_rate": 9.7e-06,
"loss": 0.3264,
"step": 100
},
{
"epoch": 2.0,
"learning_rate": 1e-05,
"loss": 0.1934,
"step": 125
},
{
"epoch": 2.0,
"learning_rate": 1e-05,
"loss": 0.1556,
"step": 150
},
{
"epoch": 3.0,
"learning_rate": 1e-05,
"loss": 0.0975,
"step": 175
},
{
"epoch": 3.0,
"learning_rate": 1e-05,
"loss": 0.0713,
"step": 200
},
{
"epoch": 4.0,
"learning_rate": 1e-05,
"loss": 0.0553,
"step": 225
},
{
"epoch": 4.0,
"learning_rate": 1e-05,
"loss": 0.0363,
"step": 250
},
{
"epoch": 5.0,
"learning_rate": 1e-05,
"loss": 0.027,
"step": 275
},
{
"epoch": 5.0,
"learning_rate": 1e-05,
"loss": 0.0226,
"step": 300
},
{
"epoch": 6.0,
"learning_rate": 1e-05,
"loss": 0.0229,
"step": 325
},
{
"epoch": 6.0,
"learning_rate": 1e-05,
"loss": 0.0111,
"step": 350
},
{
"epoch": 7.0,
"learning_rate": 1e-05,
"loss": 0.0124,
"step": 375
},
{
"epoch": 7.0,
"learning_rate": 1e-05,
"loss": 0.0099,
"step": 400
},
{
"epoch": 8.0,
"learning_rate": 1e-05,
"loss": 0.0088,
"step": 425
},
{
"epoch": 8.0,
"learning_rate": 1e-05,
"loss": 0.0113,
"step": 450
},
{
"epoch": 8.01,
"learning_rate": 1e-05,
"loss": 0.0069,
"step": 475
},
{
"epoch": 9.0,
"learning_rate": 1e-05,
"loss": 0.0053,
"step": 500
},
{
"epoch": 9.0,
"learning_rate": 1e-05,
"loss": 0.0051,
"step": 525
},
{
"epoch": 10.0,
"learning_rate": 1e-05,
"loss": 0.004,
"step": 550
},
{
"epoch": 10.0,
"learning_rate": 1e-05,
"loss": 0.0044,
"step": 575
},
{
"epoch": 11.0,
"learning_rate": 1e-05,
"loss": 0.0023,
"step": 600
},
{
"epoch": 11.0,
"learning_rate": 1e-05,
"loss": 0.0032,
"step": 625
},
{
"epoch": 12.0,
"learning_rate": 1e-05,
"loss": 0.0032,
"step": 650
},
{
"epoch": 12.0,
"learning_rate": 1e-05,
"loss": 0.0065,
"step": 675
},
{
"epoch": 13.0,
"learning_rate": 1e-05,
"loss": 0.0045,
"step": 700
},
{
"epoch": 13.0,
"learning_rate": 1e-05,
"loss": 0.0067,
"step": 725
},
{
"epoch": 14.0,
"learning_rate": 1e-05,
"loss": 0.0094,
"step": 750
},
{
"epoch": 14.0,
"learning_rate": 1e-05,
"loss": 0.0023,
"step": 775
},
{
"epoch": 15.0,
"learning_rate": 1e-05,
"loss": 0.0043,
"step": 800
},
{
"epoch": 15.0,
"learning_rate": 1e-05,
"loss": 0.0089,
"step": 825
},
{
"epoch": 16.0,
"learning_rate": 1e-05,
"loss": 0.0064,
"step": 850
},
{
"epoch": 16.0,
"learning_rate": 1e-05,
"loss": 0.0023,
"step": 875
},
{
"epoch": 16.01,
"learning_rate": 1e-05,
"loss": 0.0049,
"step": 900
},
{
"epoch": 17.0,
"learning_rate": 1e-05,
"loss": 0.0026,
"step": 925
},
{
"epoch": 17.0,
"learning_rate": 1e-05,
"loss": 0.0032,
"step": 950
},
{
"epoch": 18.0,
"learning_rate": 1e-05,
"loss": 0.0013,
"step": 975
},
{
"epoch": 18.0,
"learning_rate": 1e-05,
"loss": 0.0033,
"step": 1000
},
{
"epoch": 18.0,
"eval_loss": 0.6728372573852539,
"eval_runtime": 65.1103,
"eval_samples_per_second": 1.597,
"eval_steps_per_second": 0.2,
"eval_wer": 25.315391084945333,
"step": 1000
},
{
"epoch": 19.0,
"learning_rate": 1e-05,
"loss": 0.0027,
"step": 1025
},
{
"epoch": 19.0,
"learning_rate": 1e-05,
"loss": 0.004,
"step": 1050
},
{
"epoch": 20.0,
"learning_rate": 1e-05,
"loss": 0.0008,
"step": 1075
},
{
"epoch": 20.0,
"learning_rate": 1e-05,
"loss": 0.0012,
"step": 1100
},
{
"epoch": 21.0,
"learning_rate": 1e-05,
"loss": 0.0011,
"step": 1125
},
{
"epoch": 21.0,
"learning_rate": 1e-05,
"loss": 0.0012,
"step": 1150
},
{
"epoch": 22.0,
"learning_rate": 1e-05,
"loss": 0.0008,
"step": 1175
},
{
"epoch": 22.0,
"learning_rate": 1e-05,
"loss": 0.0035,
"step": 1200
},
{
"epoch": 23.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 1225
},
{
"epoch": 23.0,
"learning_rate": 1e-05,
"loss": 0.0008,
"step": 1250
},
{
"epoch": 24.0,
"learning_rate": 1e-05,
"loss": 0.0009,
"step": 1275
},
{
"epoch": 24.0,
"learning_rate": 1e-05,
"loss": 0.0011,
"step": 1300
},
{
"epoch": 24.01,
"learning_rate": 1e-05,
"loss": 0.0011,
"step": 1325
},
{
"epoch": 25.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 1350
},
{
"epoch": 25.0,
"learning_rate": 1e-05,
"loss": 0.0011,
"step": 1375
},
{
"epoch": 26.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 1400
},
{
"epoch": 26.0,
"learning_rate": 1e-05,
"loss": 0.0006,
"step": 1425
},
{
"epoch": 27.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 1450
},
{
"epoch": 27.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 1475
},
{
"epoch": 28.0,
"learning_rate": 1e-05,
"loss": 0.0005,
"step": 1500
},
{
"epoch": 28.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 1525
},
{
"epoch": 29.0,
"learning_rate": 1e-05,
"loss": 0.0006,
"step": 1550
},
{
"epoch": 29.0,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 1575
},
{
"epoch": 30.0,
"learning_rate": 1e-05,
"loss": 0.0023,
"step": 1600
},
{
"epoch": 30.0,
"learning_rate": 1e-05,
"loss": 0.0014,
"step": 1625
},
{
"epoch": 31.0,
"learning_rate": 1e-05,
"loss": 0.0046,
"step": 1650
},
{
"epoch": 31.0,
"learning_rate": 1e-05,
"loss": 0.0058,
"step": 1675
},
{
"epoch": 32.0,
"learning_rate": 1e-05,
"loss": 0.0036,
"step": 1700
},
{
"epoch": 32.0,
"learning_rate": 1e-05,
"loss": 0.0039,
"step": 1725
},
{
"epoch": 33.0,
"learning_rate": 1e-05,
"loss": 0.0108,
"step": 1750
},
{
"epoch": 33.0,
"learning_rate": 1e-05,
"loss": 0.0058,
"step": 1775
},
{
"epoch": 33.01,
"learning_rate": 1e-05,
"loss": 0.0068,
"step": 1800
},
{
"epoch": 34.0,
"learning_rate": 1e-05,
"loss": 0.0105,
"step": 1825
},
{
"epoch": 34.0,
"learning_rate": 1e-05,
"loss": 0.0156,
"step": 1850
},
{
"epoch": 35.0,
"learning_rate": 1e-05,
"loss": 0.0101,
"step": 1875
},
{
"epoch": 35.0,
"learning_rate": 1e-05,
"loss": 0.0078,
"step": 1900
},
{
"epoch": 36.0,
"learning_rate": 1e-05,
"loss": 0.007,
"step": 1925
},
{
"epoch": 36.0,
"learning_rate": 1e-05,
"loss": 0.0069,
"step": 1950
},
{
"epoch": 37.0,
"learning_rate": 1e-05,
"loss": 0.0031,
"step": 1975
},
{
"epoch": 37.0,
"learning_rate": 1e-05,
"loss": 0.008,
"step": 2000
},
{
"epoch": 37.0,
"eval_loss": 0.6984221339225769,
"eval_runtime": 61.129,
"eval_samples_per_second": 1.701,
"eval_steps_per_second": 0.213,
"eval_wer": 23.38099243061396,
"step": 2000
},
{
"epoch": 38.0,
"learning_rate": 1e-05,
"loss": 0.0031,
"step": 2025
},
{
"epoch": 38.0,
"learning_rate": 1e-05,
"loss": 0.0067,
"step": 2050
},
{
"epoch": 39.0,
"learning_rate": 1e-05,
"loss": 0.0029,
"step": 2075
},
{
"epoch": 39.0,
"learning_rate": 1e-05,
"loss": 0.0024,
"step": 2100
},
{
"epoch": 40.0,
"learning_rate": 1e-05,
"loss": 0.0026,
"step": 2125
},
{
"epoch": 40.0,
"learning_rate": 1e-05,
"loss": 0.0019,
"step": 2150
},
{
"epoch": 41.0,
"learning_rate": 1e-05,
"loss": 0.0024,
"step": 2175
},
{
"epoch": 41.0,
"learning_rate": 1e-05,
"loss": 0.002,
"step": 2200
},
{
"epoch": 41.01,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 2225
},
{
"epoch": 42.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 2250
},
{
"epoch": 42.0,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 2275
},
{
"epoch": 43.0,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 2300
},
{
"epoch": 43.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 2325
},
{
"epoch": 44.0,
"learning_rate": 1e-05,
"loss": 0.0039,
"step": 2350
},
{
"epoch": 44.0,
"learning_rate": 1e-05,
"loss": 0.004,
"step": 2375
},
{
"epoch": 45.0,
"learning_rate": 1e-05,
"loss": 0.0008,
"step": 2400
},
{
"epoch": 45.0,
"learning_rate": 1e-05,
"loss": 0.006,
"step": 2425
},
{
"epoch": 46.0,
"learning_rate": 1e-05,
"loss": 0.0076,
"step": 2450
},
{
"epoch": 46.0,
"learning_rate": 1e-05,
"loss": 0.0061,
"step": 2475
},
{
"epoch": 47.0,
"learning_rate": 1e-05,
"loss": 0.0043,
"step": 2500
},
{
"epoch": 47.0,
"learning_rate": 1e-05,
"loss": 0.0051,
"step": 2525
},
{
"epoch": 48.0,
"learning_rate": 1e-05,
"loss": 0.0077,
"step": 2550
},
{
"epoch": 48.0,
"learning_rate": 1e-05,
"loss": 0.0036,
"step": 2575
},
{
"epoch": 49.0,
"learning_rate": 1e-05,
"loss": 0.0062,
"step": 2600
},
{
"epoch": 49.0,
"learning_rate": 1e-05,
"loss": 0.0108,
"step": 2625
},
{
"epoch": 49.01,
"learning_rate": 1e-05,
"loss": 0.0033,
"step": 2650
},
{
"epoch": 50.0,
"learning_rate": 1e-05,
"loss": 0.0038,
"step": 2675
},
{
"epoch": 50.01,
"learning_rate": 1e-05,
"loss": 0.0024,
"step": 2700
},
{
"epoch": 51.0,
"learning_rate": 1e-05,
"loss": 0.0021,
"step": 2725
},
{
"epoch": 51.0,
"learning_rate": 1e-05,
"loss": 0.003,
"step": 2750
},
{
"epoch": 52.0,
"learning_rate": 1e-05,
"loss": 0.0038,
"step": 2775
},
{
"epoch": 52.0,
"learning_rate": 1e-05,
"loss": 0.0022,
"step": 2800
},
{
"epoch": 53.0,
"learning_rate": 1e-05,
"loss": 0.0008,
"step": 2825
},
{
"epoch": 53.0,
"learning_rate": 1e-05,
"loss": 0.0009,
"step": 2850
},
{
"epoch": 54.0,
"learning_rate": 1e-05,
"loss": 0.001,
"step": 2875
},
{
"epoch": 54.0,
"learning_rate": 1e-05,
"loss": 0.0017,
"step": 2900
},
{
"epoch": 55.0,
"learning_rate": 1e-05,
"loss": 0.0009,
"step": 2925
},
{
"epoch": 55.0,
"learning_rate": 1e-05,
"loss": 0.0033,
"step": 2950
},
{
"epoch": 56.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 2975
},
{
"epoch": 56.0,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 3000
},
{
"epoch": 56.0,
"eval_loss": 0.748629629611969,
"eval_runtime": 59.9773,
"eval_samples_per_second": 1.734,
"eval_steps_per_second": 0.217,
"eval_wer": 24.47434819175778,
"step": 3000
},
{
"epoch": 57.0,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 3025
},
{
"epoch": 57.0,
"learning_rate": 1e-05,
"loss": 0.001,
"step": 3050
},
{
"epoch": 58.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3075
},
{
"epoch": 58.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3100
},
{
"epoch": 58.01,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 3125
},
{
"epoch": 59.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3150
},
{
"epoch": 59.0,
"learning_rate": 1e-05,
"loss": 0.0005,
"step": 3175
},
{
"epoch": 60.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3200
},
{
"epoch": 60.0,
"learning_rate": 1e-05,
"loss": 0.0006,
"step": 3225
},
{
"epoch": 61.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 3250
},
{
"epoch": 61.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3275
},
{
"epoch": 62.0,
"learning_rate": 1e-05,
"loss": 0.0009,
"step": 3300
},
{
"epoch": 62.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3325
},
{
"epoch": 63.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3350
},
{
"epoch": 63.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3375
},
{
"epoch": 64.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3400
},
{
"epoch": 64.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3425
},
{
"epoch": 65.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3450
},
{
"epoch": 65.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3475
},
{
"epoch": 66.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3500
},
{
"epoch": 66.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3525
},
{
"epoch": 66.01,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 3550
},
{
"epoch": 67.0,
"learning_rate": 1e-05,
"loss": 0.0008,
"step": 3575
},
{
"epoch": 67.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3600
},
{
"epoch": 68.0,
"learning_rate": 1e-05,
"loss": 0.0006,
"step": 3625
},
{
"epoch": 68.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3650
},
{
"epoch": 69.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3675
},
{
"epoch": 69.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 3700
},
{
"epoch": 70.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3725
},
{
"epoch": 70.0,
"learning_rate": 1e-05,
"loss": 0.0004,
"step": 3750
},
{
"epoch": 71.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3775
},
{
"epoch": 71.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 3800
},
{
"epoch": 72.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3825
},
{
"epoch": 72.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 3850
},
{
"epoch": 73.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3875
},
{
"epoch": 73.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3900
},
{
"epoch": 74.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 3925
},
{
"epoch": 74.0,
"learning_rate": 1e-05,
"loss": 0.0002,
"step": 3950
},
{
"epoch": 74.01,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 3975
},
{
"epoch": 75.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 4000
},
{
"epoch": 75.0,
"eval_loss": 0.7752631306648254,
"eval_runtime": 60.3446,
"eval_samples_per_second": 1.723,
"eval_steps_per_second": 0.215,
"eval_wer": 24.47434819175778,
"step": 4000
},
{
"epoch": 75.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 4025
},
{
"epoch": 76.0,
"learning_rate": 1e-05,
"loss": 0.0003,
"step": 4050
},
{
"epoch": 76.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 4075
},
{
"epoch": 77.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 4100
},
{
"epoch": 77.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 4125
},
{
"epoch": 78.0,
"learning_rate": 1e-05,
"loss": 0.0001,
"step": 4150
},
{
"epoch": 78.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4175
},
{
"epoch": 79.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4200
},
{
"epoch": 79.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4225
},
{
"epoch": 80.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4250
},
{
"epoch": 80.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4275
},
{
"epoch": 81.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4300
},
{
"epoch": 81.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4325
},
{
"epoch": 82.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4350
},
{
"epoch": 82.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4375
},
{
"epoch": 83.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4400
},
{
"epoch": 83.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4425
},
{
"epoch": 83.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4450
},
{
"epoch": 84.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4475
},
{
"epoch": 84.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4500
},
{
"epoch": 85.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4525
},
{
"epoch": 85.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4550
},
{
"epoch": 86.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4575
},
{
"epoch": 86.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4600
},
{
"epoch": 87.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4625
},
{
"epoch": 87.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4650
},
{
"epoch": 88.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4675
},
{
"epoch": 88.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4700
},
{
"epoch": 89.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4725
},
{
"epoch": 89.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4750
},
{
"epoch": 90.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4775
},
{
"epoch": 90.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4800
},
{
"epoch": 91.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4825
},
{
"epoch": 91.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4850
},
{
"epoch": 91.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4875
},
{
"epoch": 92.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4900
},
{
"epoch": 92.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4925
},
{
"epoch": 93.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4950
},
{
"epoch": 93.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 4975
},
{
"epoch": 94.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5000
},
{
"epoch": 94.0,
"eval_loss": 0.8013902306556702,
"eval_runtime": 59.4712,
"eval_samples_per_second": 1.749,
"eval_steps_per_second": 0.219,
"eval_wer": 24.053826745164002,
"step": 5000
},
{
"epoch": 94.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5025
},
{
"epoch": 95.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5050
},
{
"epoch": 95.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5075
},
{
"epoch": 96.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5100
},
{
"epoch": 96.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5125
},
{
"epoch": 97.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5150
},
{
"epoch": 97.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5175
},
{
"epoch": 98.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5200
},
{
"epoch": 98.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5225
},
{
"epoch": 99.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5250
},
{
"epoch": 99.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5275
},
{
"epoch": 99.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5300
},
{
"epoch": 100.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5325
},
{
"epoch": 100.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5350
},
{
"epoch": 101.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5375
},
{
"epoch": 101.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5400
},
{
"epoch": 102.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5425
},
{
"epoch": 102.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5450
},
{
"epoch": 103.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5475
},
{
"epoch": 103.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5500
},
{
"epoch": 104.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5525
},
{
"epoch": 104.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5550
},
{
"epoch": 105.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5575
},
{
"epoch": 105.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5600
},
{
"epoch": 106.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5625
},
{
"epoch": 106.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5650
},
{
"epoch": 107.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5675
},
{
"epoch": 107.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5700
},
{
"epoch": 108.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5725
},
{
"epoch": 108.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5750
},
{
"epoch": 108.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5775
},
{
"epoch": 109.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5800
},
{
"epoch": 109.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5825
},
{
"epoch": 110.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5850
},
{
"epoch": 110.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5875
},
{
"epoch": 111.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5900
},
{
"epoch": 111.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5925
},
{
"epoch": 112.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5950
},
{
"epoch": 112.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 5975
},
{
"epoch": 113.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6000
},
{
"epoch": 113.0,
"eval_loss": 0.8243789672851562,
"eval_runtime": 59.3452,
"eval_samples_per_second": 1.752,
"eval_steps_per_second": 0.219,
"eval_wer": 24.390243902439025,
"step": 6000
},
{
"epoch": 113.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6025
},
{
"epoch": 114.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6050
},
{
"epoch": 114.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6075
},
{
"epoch": 115.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6100
},
{
"epoch": 115.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6125
},
{
"epoch": 116.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6150
},
{
"epoch": 116.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6175
},
{
"epoch": 116.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6200
},
{
"epoch": 117.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6225
},
{
"epoch": 117.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6250
},
{
"epoch": 118.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6275
},
{
"epoch": 118.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6300
},
{
"epoch": 119.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6325
},
{
"epoch": 119.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6350
},
{
"epoch": 120.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6375
},
{
"epoch": 120.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6400
},
{
"epoch": 121.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6425
},
{
"epoch": 121.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6450
},
{
"epoch": 122.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6475
},
{
"epoch": 122.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6500
},
{
"epoch": 123.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6525
},
{
"epoch": 123.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6550
},
{
"epoch": 124.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6575
},
{
"epoch": 124.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6600
},
{
"epoch": 124.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6625
},
{
"epoch": 125.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6650
},
{
"epoch": 125.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6675
},
{
"epoch": 126.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6700
},
{
"epoch": 126.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6725
},
{
"epoch": 127.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6750
},
{
"epoch": 127.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6775
},
{
"epoch": 128.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6800
},
{
"epoch": 128.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6825
},
{
"epoch": 129.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6850
},
{
"epoch": 129.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6875
},
{
"epoch": 130.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6900
},
{
"epoch": 130.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6925
},
{
"epoch": 131.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6950
},
{
"epoch": 131.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 6975
},
{
"epoch": 132.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7000
},
{
"epoch": 132.0,
"eval_loss": 0.8468073606491089,
"eval_runtime": 58.9515,
"eval_samples_per_second": 1.764,
"eval_steps_per_second": 0.221,
"eval_wer": 23.80151387720774,
"step": 7000
},
{
"epoch": 132.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7025
},
{
"epoch": 133.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7050
},
{
"epoch": 133.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7075
},
{
"epoch": 133.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7100
},
{
"epoch": 134.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7125
},
{
"epoch": 134.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7150
},
{
"epoch": 135.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7175
},
{
"epoch": 135.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7200
},
{
"epoch": 136.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7225
},
{
"epoch": 136.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7250
},
{
"epoch": 137.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7275
},
{
"epoch": 137.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7300
},
{
"epoch": 138.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7325
},
{
"epoch": 138.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7350
},
{
"epoch": 139.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7375
},
{
"epoch": 139.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7400
},
{
"epoch": 140.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7425
},
{
"epoch": 140.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7450
},
{
"epoch": 141.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7475
},
{
"epoch": 141.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7500
},
{
"epoch": 141.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7525
},
{
"epoch": 142.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7550
},
{
"epoch": 142.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7575
},
{
"epoch": 143.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7600
},
{
"epoch": 143.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7625
},
{
"epoch": 144.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7650
},
{
"epoch": 144.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7675
},
{
"epoch": 145.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7700
},
{
"epoch": 145.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7725
},
{
"epoch": 146.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7750
},
{
"epoch": 146.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7775
},
{
"epoch": 147.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7800
},
{
"epoch": 147.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7825
},
{
"epoch": 148.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7850
},
{
"epoch": 148.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7875
},
{
"epoch": 149.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7900
},
{
"epoch": 149.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7925
},
{
"epoch": 149.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7950
},
{
"epoch": 150.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 7975
},
{
"epoch": 150.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8000
},
{
"epoch": 150.0,
"eval_loss": 0.869913637638092,
"eval_runtime": 62.6359,
"eval_samples_per_second": 1.66,
"eval_steps_per_second": 0.208,
"eval_wer": 23.465096719932717,
"step": 8000
},
{
"epoch": 151.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8025
},
{
"epoch": 151.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8050
},
{
"epoch": 152.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8075
},
{
"epoch": 152.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8100
},
{
"epoch": 153.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8125
},
{
"epoch": 153.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8150
},
{
"epoch": 154.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8175
},
{
"epoch": 154.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8200
},
{
"epoch": 155.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8225
},
{
"epoch": 155.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8250
},
{
"epoch": 156.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8275
},
{
"epoch": 156.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8300
},
{
"epoch": 157.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8325
},
{
"epoch": 157.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8350
},
{
"epoch": 158.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8375
},
{
"epoch": 158.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8400
},
{
"epoch": 158.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8425
},
{
"epoch": 159.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8450
},
{
"epoch": 159.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8475
},
{
"epoch": 160.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8500
},
{
"epoch": 160.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8525
},
{
"epoch": 161.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8550
},
{
"epoch": 161.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8575
},
{
"epoch": 162.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8600
},
{
"epoch": 162.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8625
},
{
"epoch": 163.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8650
},
{
"epoch": 163.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8675
},
{
"epoch": 164.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8700
},
{
"epoch": 164.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8725
},
{
"epoch": 165.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8750
},
{
"epoch": 165.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8775
},
{
"epoch": 166.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8800
},
{
"epoch": 166.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8825
},
{
"epoch": 166.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8850
},
{
"epoch": 167.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8875
},
{
"epoch": 167.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8900
},
{
"epoch": 168.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8925
},
{
"epoch": 168.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8950
},
{
"epoch": 169.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 8975
},
{
"epoch": 169.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9000
},
{
"epoch": 169.0,
"eval_loss": 0.8935946822166443,
"eval_runtime": 61.1473,
"eval_samples_per_second": 1.701,
"eval_steps_per_second": 0.213,
"eval_wer": 23.21278385197645,
"step": 9000
},
{
"epoch": 170.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9025
},
{
"epoch": 170.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9050
},
{
"epoch": 171.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9075
},
{
"epoch": 171.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9100
},
{
"epoch": 172.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9125
},
{
"epoch": 172.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9150
},
{
"epoch": 173.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9175
},
{
"epoch": 173.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9200
},
{
"epoch": 174.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9225
},
{
"epoch": 174.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9250
},
{
"epoch": 174.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9275
},
{
"epoch": 175.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9300
},
{
"epoch": 175.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9325
},
{
"epoch": 176.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9350
},
{
"epoch": 176.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9375
},
{
"epoch": 177.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9400
},
{
"epoch": 177.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9425
},
{
"epoch": 178.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9450
},
{
"epoch": 178.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9475
},
{
"epoch": 179.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9500
},
{
"epoch": 179.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9525
},
{
"epoch": 180.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9550
},
{
"epoch": 180.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9575
},
{
"epoch": 181.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9600
},
{
"epoch": 181.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9625
},
{
"epoch": 182.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9650
},
{
"epoch": 182.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9675
},
{
"epoch": 183.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9700
},
{
"epoch": 183.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9725
},
{
"epoch": 183.01,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9750
},
{
"epoch": 184.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9775
},
{
"epoch": 184.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9800
},
{
"epoch": 185.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9825
},
{
"epoch": 185.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9850
},
{
"epoch": 186.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9875
},
{
"epoch": 186.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9900
},
{
"epoch": 187.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9925
},
{
"epoch": 187.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9950
},
{
"epoch": 188.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 9975
},
{
"epoch": 188.0,
"learning_rate": 1e-05,
"loss": 0.0,
"step": 10000
},
{
"epoch": 188.0,
"eval_loss": 0.9179897308349609,
"eval_runtime": 62.8354,
"eval_samples_per_second": 1.655,
"eval_steps_per_second": 0.207,
"eval_wer": 23.296888141295206,
"step": 10000
},
{
"epoch": 188.0,
"step": 10000,
"total_flos": 4.601090188541952e+19,
"train_loss": 0.007761261822304368,
"train_runtime": 43699.0959,
"train_samples_per_second": 3.661,
"train_steps_per_second": 0.229
}
],
"max_steps": 10000,
"num_train_epochs": 9223372036854775807,
"total_flos": 4.601090188541952e+19,
"trial_name": null,
"trial_params": null
}