p6_commonvoice_mild / trainer_state.json
honzapucalek's picture
End of training
c4122a4 verified
{
"best_metric": 0.1605595256195834,
"best_model_checkpoint": "./p6_commonvoice_mild/checkpoint-1000",
"epoch": 232.5581395348837,
"eval_steps": 1000,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.16,
"learning_rate": 5.000000000000001e-07,
"loss": 0.3949,
"step": 25
},
{
"epoch": 2.33,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.3249,
"step": 50
},
{
"epoch": 3.49,
"learning_rate": 1.5e-06,
"loss": 0.2254,
"step": 75
},
{
"epoch": 4.65,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.1515,
"step": 100
},
{
"epoch": 5.81,
"learning_rate": 2.5e-06,
"loss": 0.105,
"step": 125
},
{
"epoch": 6.98,
"learning_rate": 3e-06,
"loss": 0.0518,
"step": 150
},
{
"epoch": 8.14,
"learning_rate": 3.5e-06,
"loss": 0.024,
"step": 175
},
{
"epoch": 9.3,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0117,
"step": 200
},
{
"epoch": 10.47,
"learning_rate": 4.5e-06,
"loss": 0.0059,
"step": 225
},
{
"epoch": 11.63,
"learning_rate": 5e-06,
"loss": 0.0052,
"step": 250
},
{
"epoch": 12.79,
"learning_rate": 5.500000000000001e-06,
"loss": 0.003,
"step": 275
},
{
"epoch": 13.95,
"learning_rate": 6e-06,
"loss": 0.0037,
"step": 300
},
{
"epoch": 15.12,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.0057,
"step": 325
},
{
"epoch": 16.28,
"learning_rate": 7e-06,
"loss": 0.0075,
"step": 350
},
{
"epoch": 17.44,
"learning_rate": 7.500000000000001e-06,
"loss": 0.0077,
"step": 375
},
{
"epoch": 18.6,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0063,
"step": 400
},
{
"epoch": 19.77,
"learning_rate": 8.5e-06,
"loss": 0.008,
"step": 425
},
{
"epoch": 20.93,
"learning_rate": 9e-06,
"loss": 0.0061,
"step": 450
},
{
"epoch": 22.09,
"learning_rate": 9.5e-06,
"loss": 0.0103,
"step": 475
},
{
"epoch": 23.26,
"learning_rate": 1e-05,
"loss": 0.0088,
"step": 500
},
{
"epoch": 24.42,
"learning_rate": 9.944444444444445e-06,
"loss": 0.0113,
"step": 525
},
{
"epoch": 25.58,
"learning_rate": 9.88888888888889e-06,
"loss": 0.01,
"step": 550
},
{
"epoch": 26.74,
"learning_rate": 9.833333333333333e-06,
"loss": 0.0054,
"step": 575
},
{
"epoch": 27.91,
"learning_rate": 9.777777777777779e-06,
"loss": 0.0043,
"step": 600
},
{
"epoch": 29.07,
"learning_rate": 9.722222222222223e-06,
"loss": 0.0033,
"step": 625
},
{
"epoch": 30.23,
"learning_rate": 9.666666666666667e-06,
"loss": 0.0018,
"step": 650
},
{
"epoch": 31.4,
"learning_rate": 9.611111111111112e-06,
"loss": 0.0003,
"step": 675
},
{
"epoch": 32.56,
"learning_rate": 9.555555555555556e-06,
"loss": 0.0002,
"step": 700
},
{
"epoch": 33.72,
"learning_rate": 9.5e-06,
"loss": 0.0002,
"step": 725
},
{
"epoch": 34.88,
"learning_rate": 9.444444444444445e-06,
"loss": 0.0002,
"step": 750
},
{
"epoch": 36.05,
"learning_rate": 9.38888888888889e-06,
"loss": 0.0002,
"step": 775
},
{
"epoch": 37.21,
"learning_rate": 9.333333333333334e-06,
"loss": 0.0001,
"step": 800
},
{
"epoch": 38.37,
"learning_rate": 9.277777777777778e-06,
"loss": 0.0001,
"step": 825
},
{
"epoch": 39.53,
"learning_rate": 9.222222222222224e-06,
"loss": 0.0001,
"step": 850
},
{
"epoch": 40.7,
"learning_rate": 9.166666666666666e-06,
"loss": 0.0001,
"step": 875
},
{
"epoch": 41.86,
"learning_rate": 9.111111111111112e-06,
"loss": 0.0001,
"step": 900
},
{
"epoch": 43.02,
"learning_rate": 9.055555555555556e-06,
"loss": 0.0001,
"step": 925
},
{
"epoch": 44.19,
"learning_rate": 9e-06,
"loss": 0.0001,
"step": 950
},
{
"epoch": 45.35,
"learning_rate": 8.944444444444446e-06,
"loss": 0.0001,
"step": 975
},
{
"epoch": 46.51,
"learning_rate": 8.888888888888888e-06,
"loss": 0.0001,
"step": 1000
},
{
"epoch": 46.51,
"eval_loss": 0.5155884027481079,
"eval_runtime": 260.6477,
"eval_samples_per_second": 1.147,
"eval_steps_per_second": 0.146,
"eval_wer": 0.1605595256195834,
"step": 1000
},
{
"epoch": 47.67,
"learning_rate": 8.833333333333334e-06,
"loss": 0.0001,
"step": 1025
},
{
"epoch": 48.84,
"learning_rate": 8.777777777777778e-06,
"loss": 0.0001,
"step": 1050
},
{
"epoch": 50.0,
"learning_rate": 8.722222222222224e-06,
"loss": 0.0001,
"step": 1075
},
{
"epoch": 51.16,
"learning_rate": 8.666666666666668e-06,
"loss": 0.0001,
"step": 1100
},
{
"epoch": 52.33,
"learning_rate": 8.611111111111112e-06,
"loss": 0.0001,
"step": 1125
},
{
"epoch": 53.49,
"learning_rate": 8.555555555555556e-06,
"loss": 0.0001,
"step": 1150
},
{
"epoch": 54.65,
"learning_rate": 8.5e-06,
"loss": 0.0001,
"step": 1175
},
{
"epoch": 55.81,
"learning_rate": 8.444444444444446e-06,
"loss": 0.0001,
"step": 1200
},
{
"epoch": 56.98,
"learning_rate": 8.38888888888889e-06,
"loss": 0.0001,
"step": 1225
},
{
"epoch": 58.14,
"learning_rate": 8.333333333333334e-06,
"loss": 0.0001,
"step": 1250
},
{
"epoch": 59.3,
"learning_rate": 8.277777777777778e-06,
"loss": 0.0001,
"step": 1275
},
{
"epoch": 60.47,
"learning_rate": 8.222222222222222e-06,
"loss": 0.0001,
"step": 1300
},
{
"epoch": 61.63,
"learning_rate": 8.166666666666668e-06,
"loss": 0.0001,
"step": 1325
},
{
"epoch": 62.79,
"learning_rate": 8.111111111111112e-06,
"loss": 0.0001,
"step": 1350
},
{
"epoch": 63.95,
"learning_rate": 8.055555555555557e-06,
"loss": 0.0001,
"step": 1375
},
{
"epoch": 65.12,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0001,
"step": 1400
},
{
"epoch": 66.28,
"learning_rate": 7.944444444444445e-06,
"loss": 0.0001,
"step": 1425
},
{
"epoch": 67.44,
"learning_rate": 7.88888888888889e-06,
"loss": 0.0001,
"step": 1450
},
{
"epoch": 68.6,
"learning_rate": 7.833333333333333e-06,
"loss": 0.0001,
"step": 1475
},
{
"epoch": 69.77,
"learning_rate": 7.77777777777778e-06,
"loss": 0.0001,
"step": 1500
},
{
"epoch": 70.93,
"learning_rate": 7.722222222222223e-06,
"loss": 0.0,
"step": 1525
},
{
"epoch": 72.09,
"learning_rate": 7.666666666666667e-06,
"loss": 0.0,
"step": 1550
},
{
"epoch": 73.26,
"learning_rate": 7.611111111111111e-06,
"loss": 0.0,
"step": 1575
},
{
"epoch": 74.42,
"learning_rate": 7.555555555555556e-06,
"loss": 0.0,
"step": 1600
},
{
"epoch": 75.58,
"learning_rate": 7.500000000000001e-06,
"loss": 0.0,
"step": 1625
},
{
"epoch": 76.74,
"learning_rate": 7.444444444444445e-06,
"loss": 0.0,
"step": 1650
},
{
"epoch": 77.91,
"learning_rate": 7.38888888888889e-06,
"loss": 0.0,
"step": 1675
},
{
"epoch": 79.07,
"learning_rate": 7.333333333333333e-06,
"loss": 0.0,
"step": 1700
},
{
"epoch": 80.23,
"learning_rate": 7.277777777777778e-06,
"loss": 0.0,
"step": 1725
},
{
"epoch": 81.4,
"learning_rate": 7.222222222222223e-06,
"loss": 0.0,
"step": 1750
},
{
"epoch": 82.56,
"learning_rate": 7.166666666666667e-06,
"loss": 0.0,
"step": 1775
},
{
"epoch": 83.72,
"learning_rate": 7.111111111111112e-06,
"loss": 0.0,
"step": 1800
},
{
"epoch": 84.88,
"learning_rate": 7.055555555555557e-06,
"loss": 0.0,
"step": 1825
},
{
"epoch": 86.05,
"learning_rate": 7e-06,
"loss": 0.0,
"step": 1850
},
{
"epoch": 87.21,
"learning_rate": 6.944444444444445e-06,
"loss": 0.0,
"step": 1875
},
{
"epoch": 88.37,
"learning_rate": 6.88888888888889e-06,
"loss": 0.0,
"step": 1900
},
{
"epoch": 89.53,
"learning_rate": 6.833333333333334e-06,
"loss": 0.0,
"step": 1925
},
{
"epoch": 90.7,
"learning_rate": 6.777777777777779e-06,
"loss": 0.0,
"step": 1950
},
{
"epoch": 91.86,
"learning_rate": 6.7222222222222235e-06,
"loss": 0.0,
"step": 1975
},
{
"epoch": 93.02,
"learning_rate": 6.666666666666667e-06,
"loss": 0.0,
"step": 2000
},
{
"epoch": 93.02,
"eval_loss": 0.5643903613090515,
"eval_runtime": 259.7049,
"eval_samples_per_second": 1.151,
"eval_steps_per_second": 0.146,
"eval_wer": 0.15843089554508136,
"step": 2000
},
{
"epoch": 94.19,
"learning_rate": 6.6111111111111115e-06,
"loss": 0.0,
"step": 2025
},
{
"epoch": 95.35,
"learning_rate": 6.555555555555556e-06,
"loss": 0.0,
"step": 2050
},
{
"epoch": 96.51,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.0,
"step": 2075
},
{
"epoch": 97.67,
"learning_rate": 6.444444444444445e-06,
"loss": 0.0,
"step": 2100
},
{
"epoch": 98.84,
"learning_rate": 6.3888888888888885e-06,
"loss": 0.0,
"step": 2125
},
{
"epoch": 100.0,
"learning_rate": 6.333333333333333e-06,
"loss": 0.0,
"step": 2150
},
{
"epoch": 101.16,
"learning_rate": 6.277777777777778e-06,
"loss": 0.0,
"step": 2175
},
{
"epoch": 102.33,
"learning_rate": 6.222222222222223e-06,
"loss": 0.0,
"step": 2200
},
{
"epoch": 103.49,
"learning_rate": 6.166666666666667e-06,
"loss": 0.0,
"step": 2225
},
{
"epoch": 104.65,
"learning_rate": 6.111111111111112e-06,
"loss": 0.0,
"step": 2250
},
{
"epoch": 105.81,
"learning_rate": 6.055555555555555e-06,
"loss": 0.0,
"step": 2275
},
{
"epoch": 106.98,
"learning_rate": 6e-06,
"loss": 0.0,
"step": 2300
},
{
"epoch": 108.14,
"learning_rate": 5.944444444444445e-06,
"loss": 0.0,
"step": 2325
},
{
"epoch": 109.3,
"learning_rate": 5.88888888888889e-06,
"loss": 0.0,
"step": 2350
},
{
"epoch": 110.47,
"learning_rate": 5.833333333333334e-06,
"loss": 0.0,
"step": 2375
},
{
"epoch": 111.63,
"learning_rate": 5.777777777777778e-06,
"loss": 0.0,
"step": 2400
},
{
"epoch": 112.79,
"learning_rate": 5.722222222222222e-06,
"loss": 0.0,
"step": 2425
},
{
"epoch": 113.95,
"learning_rate": 5.666666666666667e-06,
"loss": 0.0,
"step": 2450
},
{
"epoch": 115.12,
"learning_rate": 5.611111111111112e-06,
"loss": 0.0,
"step": 2475
},
{
"epoch": 116.28,
"learning_rate": 5.555555555555557e-06,
"loss": 0.0,
"step": 2500
},
{
"epoch": 117.44,
"learning_rate": 5.500000000000001e-06,
"loss": 0.0,
"step": 2525
},
{
"epoch": 118.6,
"learning_rate": 5.444444444444445e-06,
"loss": 0.0,
"step": 2550
},
{
"epoch": 119.77,
"learning_rate": 5.388888888888889e-06,
"loss": 0.0,
"step": 2575
},
{
"epoch": 120.93,
"learning_rate": 5.333333333333334e-06,
"loss": 0.0,
"step": 2600
},
{
"epoch": 122.09,
"learning_rate": 5.2777777777777785e-06,
"loss": 0.0,
"step": 2625
},
{
"epoch": 123.26,
"learning_rate": 5.2222222222222226e-06,
"loss": 0.0,
"step": 2650
},
{
"epoch": 124.42,
"learning_rate": 5.1666666666666675e-06,
"loss": 0.0,
"step": 2675
},
{
"epoch": 125.58,
"learning_rate": 5.1111111111111115e-06,
"loss": 0.0,
"step": 2700
},
{
"epoch": 126.74,
"learning_rate": 5.0555555555555555e-06,
"loss": 0.0,
"step": 2725
},
{
"epoch": 127.91,
"learning_rate": 5e-06,
"loss": 0.0,
"step": 2750
},
{
"epoch": 129.07,
"learning_rate": 4.944444444444445e-06,
"loss": 0.0,
"step": 2775
},
{
"epoch": 130.23,
"learning_rate": 4.888888888888889e-06,
"loss": 0.0,
"step": 2800
},
{
"epoch": 131.4,
"learning_rate": 4.833333333333333e-06,
"loss": 0.0,
"step": 2825
},
{
"epoch": 132.56,
"learning_rate": 4.777777777777778e-06,
"loss": 0.0,
"step": 2850
},
{
"epoch": 133.72,
"learning_rate": 4.722222222222222e-06,
"loss": 0.0,
"step": 2875
},
{
"epoch": 134.88,
"learning_rate": 4.666666666666667e-06,
"loss": 0.0,
"step": 2900
},
{
"epoch": 136.05,
"learning_rate": 4.611111111111112e-06,
"loss": 0.0,
"step": 2925
},
{
"epoch": 137.21,
"learning_rate": 4.555555555555556e-06,
"loss": 0.0,
"step": 2950
},
{
"epoch": 138.37,
"learning_rate": 4.5e-06,
"loss": 0.0,
"step": 2975
},
{
"epoch": 139.53,
"learning_rate": 4.444444444444444e-06,
"loss": 0.0,
"step": 3000
},
{
"epoch": 139.53,
"eval_loss": 0.5888092517852783,
"eval_runtime": 260.1228,
"eval_samples_per_second": 1.149,
"eval_steps_per_second": 0.146,
"eval_wer": 0.15767067051847347,
"step": 3000
},
{
"epoch": 140.7,
"learning_rate": 4.388888888888889e-06,
"loss": 0.0,
"step": 3025
},
{
"epoch": 141.86,
"learning_rate": 4.333333333333334e-06,
"loss": 0.0,
"step": 3050
},
{
"epoch": 143.02,
"learning_rate": 4.277777777777778e-06,
"loss": 0.0,
"step": 3075
},
{
"epoch": 144.19,
"learning_rate": 4.222222222222223e-06,
"loss": 0.0,
"step": 3100
},
{
"epoch": 145.35,
"learning_rate": 4.166666666666667e-06,
"loss": 0.0,
"step": 3125
},
{
"epoch": 146.51,
"learning_rate": 4.111111111111111e-06,
"loss": 0.0,
"step": 3150
},
{
"epoch": 147.67,
"learning_rate": 4.055555555555556e-06,
"loss": 0.0,
"step": 3175
},
{
"epoch": 148.84,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0,
"step": 3200
},
{
"epoch": 150.0,
"learning_rate": 3.944444444444445e-06,
"loss": 0.0,
"step": 3225
},
{
"epoch": 151.16,
"learning_rate": 3.88888888888889e-06,
"loss": 0.0,
"step": 3250
},
{
"epoch": 152.33,
"learning_rate": 3.833333333333334e-06,
"loss": 0.0,
"step": 3275
},
{
"epoch": 153.49,
"learning_rate": 3.777777777777778e-06,
"loss": 0.0,
"step": 3300
},
{
"epoch": 154.65,
"learning_rate": 3.7222222222222225e-06,
"loss": 0.0,
"step": 3325
},
{
"epoch": 155.81,
"learning_rate": 3.6666666666666666e-06,
"loss": 0.0,
"step": 3350
},
{
"epoch": 156.98,
"learning_rate": 3.6111111111111115e-06,
"loss": 0.0,
"step": 3375
},
{
"epoch": 158.14,
"learning_rate": 3.555555555555556e-06,
"loss": 0.0,
"step": 3400
},
{
"epoch": 159.3,
"learning_rate": 3.5e-06,
"loss": 0.0,
"step": 3425
},
{
"epoch": 160.47,
"learning_rate": 3.444444444444445e-06,
"loss": 0.0,
"step": 3450
},
{
"epoch": 161.63,
"learning_rate": 3.3888888888888893e-06,
"loss": 0.0,
"step": 3475
},
{
"epoch": 162.79,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.0,
"step": 3500
},
{
"epoch": 163.95,
"learning_rate": 3.277777777777778e-06,
"loss": 0.0,
"step": 3525
},
{
"epoch": 165.12,
"learning_rate": 3.2222222222222227e-06,
"loss": 0.0,
"step": 3550
},
{
"epoch": 166.28,
"learning_rate": 3.1666666666666667e-06,
"loss": 0.0,
"step": 3575
},
{
"epoch": 167.44,
"learning_rate": 3.1111111111111116e-06,
"loss": 0.0,
"step": 3600
},
{
"epoch": 168.6,
"learning_rate": 3.055555555555556e-06,
"loss": 0.0,
"step": 3625
},
{
"epoch": 169.77,
"learning_rate": 3e-06,
"loss": 0.0,
"step": 3650
},
{
"epoch": 170.93,
"learning_rate": 2.944444444444445e-06,
"loss": 0.0,
"step": 3675
},
{
"epoch": 172.09,
"learning_rate": 2.888888888888889e-06,
"loss": 0.0,
"step": 3700
},
{
"epoch": 173.26,
"learning_rate": 2.8333333333333335e-06,
"loss": 0.0,
"step": 3725
},
{
"epoch": 174.42,
"learning_rate": 2.7777777777777783e-06,
"loss": 0.0,
"step": 3750
},
{
"epoch": 175.58,
"learning_rate": 2.7222222222222224e-06,
"loss": 0.0,
"step": 3775
},
{
"epoch": 176.74,
"learning_rate": 2.666666666666667e-06,
"loss": 0.0,
"step": 3800
},
{
"epoch": 177.91,
"learning_rate": 2.6111111111111113e-06,
"loss": 0.0,
"step": 3825
},
{
"epoch": 179.07,
"learning_rate": 2.5555555555555557e-06,
"loss": 0.0,
"step": 3850
},
{
"epoch": 180.23,
"learning_rate": 2.5e-06,
"loss": 0.0,
"step": 3875
},
{
"epoch": 181.4,
"learning_rate": 2.4444444444444447e-06,
"loss": 0.0,
"step": 3900
},
{
"epoch": 182.56,
"learning_rate": 2.388888888888889e-06,
"loss": 0.0,
"step": 3925
},
{
"epoch": 183.72,
"learning_rate": 2.3333333333333336e-06,
"loss": 0.0,
"step": 3950
},
{
"epoch": 184.88,
"learning_rate": 2.277777777777778e-06,
"loss": 0.0,
"step": 3975
},
{
"epoch": 186.05,
"learning_rate": 2.222222222222222e-06,
"loss": 0.0,
"step": 4000
},
{
"epoch": 186.05,
"eval_loss": 0.6043856739997864,
"eval_runtime": 260.6973,
"eval_samples_per_second": 1.147,
"eval_steps_per_second": 0.146,
"eval_wer": 0.15919112057168922,
"step": 4000
},
{
"epoch": 187.21,
"learning_rate": 2.166666666666667e-06,
"loss": 0.0,
"step": 4025
},
{
"epoch": 188.37,
"learning_rate": 2.1111111111111114e-06,
"loss": 0.0,
"step": 4050
},
{
"epoch": 189.53,
"learning_rate": 2.0555555555555555e-06,
"loss": 0.0,
"step": 4075
},
{
"epoch": 190.7,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.0,
"step": 4100
},
{
"epoch": 191.86,
"learning_rate": 1.944444444444445e-06,
"loss": 0.0,
"step": 4125
},
{
"epoch": 193.02,
"learning_rate": 1.888888888888889e-06,
"loss": 0.0,
"step": 4150
},
{
"epoch": 194.19,
"learning_rate": 1.8333333333333333e-06,
"loss": 0.0,
"step": 4175
},
{
"epoch": 195.35,
"learning_rate": 1.777777777777778e-06,
"loss": 0.0,
"step": 4200
},
{
"epoch": 196.51,
"learning_rate": 1.7222222222222224e-06,
"loss": 0.0,
"step": 4225
},
{
"epoch": 197.67,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.0,
"step": 4250
},
{
"epoch": 198.84,
"learning_rate": 1.6111111111111113e-06,
"loss": 0.0,
"step": 4275
},
{
"epoch": 200.0,
"learning_rate": 1.5555555555555558e-06,
"loss": 0.0,
"step": 4300
},
{
"epoch": 201.16,
"learning_rate": 1.5e-06,
"loss": 0.0,
"step": 4325
},
{
"epoch": 202.33,
"learning_rate": 1.4444444444444445e-06,
"loss": 0.0,
"step": 4350
},
{
"epoch": 203.49,
"learning_rate": 1.3888888888888892e-06,
"loss": 0.0,
"step": 4375
},
{
"epoch": 204.65,
"learning_rate": 1.3333333333333334e-06,
"loss": 0.0,
"step": 4400
},
{
"epoch": 205.81,
"learning_rate": 1.2777777777777779e-06,
"loss": 0.0,
"step": 4425
},
{
"epoch": 206.98,
"learning_rate": 1.2222222222222223e-06,
"loss": 0.0,
"step": 4450
},
{
"epoch": 208.14,
"learning_rate": 1.1666666666666668e-06,
"loss": 0.0,
"step": 4475
},
{
"epoch": 209.3,
"learning_rate": 1.111111111111111e-06,
"loss": 0.0,
"step": 4500
},
{
"epoch": 210.47,
"learning_rate": 1.0555555555555557e-06,
"loss": 0.0,
"step": 4525
},
{
"epoch": 211.63,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.0,
"step": 4550
},
{
"epoch": 212.79,
"learning_rate": 9.444444444444445e-07,
"loss": 0.0,
"step": 4575
},
{
"epoch": 213.95,
"learning_rate": 8.88888888888889e-07,
"loss": 0.0,
"step": 4600
},
{
"epoch": 215.12,
"learning_rate": 8.333333333333333e-07,
"loss": 0.0,
"step": 4625
},
{
"epoch": 216.28,
"learning_rate": 7.777777777777779e-07,
"loss": 0.0,
"step": 4650
},
{
"epoch": 217.44,
"learning_rate": 7.222222222222222e-07,
"loss": 0.0,
"step": 4675
},
{
"epoch": 218.6,
"learning_rate": 6.666666666666667e-07,
"loss": 0.0,
"step": 4700
},
{
"epoch": 219.77,
"learning_rate": 6.111111111111112e-07,
"loss": 0.0,
"step": 4725
},
{
"epoch": 220.93,
"learning_rate": 5.555555555555555e-07,
"loss": 0.0,
"step": 4750
},
{
"epoch": 222.09,
"learning_rate": 5.000000000000001e-07,
"loss": 0.0,
"step": 4775
},
{
"epoch": 223.26,
"learning_rate": 4.444444444444445e-07,
"loss": 0.0,
"step": 4800
},
{
"epoch": 224.42,
"learning_rate": 3.8888888888888895e-07,
"loss": 0.0,
"step": 4825
},
{
"epoch": 225.58,
"learning_rate": 3.3333333333333335e-07,
"loss": 0.0,
"step": 4850
},
{
"epoch": 226.74,
"learning_rate": 2.7777777777777776e-07,
"loss": 0.0,
"step": 4875
},
{
"epoch": 227.91,
"learning_rate": 2.2222222222222224e-07,
"loss": 0.0,
"step": 4900
},
{
"epoch": 229.07,
"learning_rate": 1.6666666666666668e-07,
"loss": 0.0,
"step": 4925
},
{
"epoch": 230.23,
"learning_rate": 1.1111111111111112e-07,
"loss": 0.0,
"step": 4950
},
{
"epoch": 231.4,
"learning_rate": 5.555555555555556e-08,
"loss": 0.0,
"step": 4975
},
{
"epoch": 232.56,
"learning_rate": 0.0,
"loss": 0.0,
"step": 5000
},
{
"epoch": 232.56,
"eval_loss": 0.6097880005836487,
"eval_runtime": 258.7688,
"eval_samples_per_second": 1.155,
"eval_steps_per_second": 0.147,
"eval_wer": 0.15919112057168922,
"step": 5000
},
{
"epoch": 232.56,
"step": 5000,
"total_flos": 2.678587664891904e+20,
"train_loss": 0.00705241062908899,
"train_runtime": 28602.098,
"train_samples_per_second": 2.797,
"train_steps_per_second": 0.175
}
],
"logging_steps": 25,
"max_steps": 5000,
"num_input_tokens_seen": 0,
"num_train_epochs": 239,
"save_steps": 1000,
"total_flos": 2.678587664891904e+20,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}