TheRains's picture
End of training
d8a9ac9
raw
history blame
25.9 kB
{
"best_metric": 14.538762364849322,
"best_model_checkpoint": "./output/small/cv9-special-batch8-small-concat-YT-NotClean/checkpoint-5000",
"epoch": 0.7295010213014298,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.2000000000000006e-07,
"loss": 2.0535,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.200000000000001e-07,
"loss": 1.9823,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.42e-06,
"loss": 1.7383,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 1.9200000000000003e-06,
"loss": 1.5902,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 2.42e-06,
"loss": 1.2048,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 2.92e-06,
"loss": 1.0128,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 3.4200000000000007e-06,
"loss": 0.9557,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 3.920000000000001e-06,
"loss": 0.9149,
"step": 200
},
{
"epoch": 0.03,
"learning_rate": 4.42e-06,
"loss": 0.8862,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 4.92e-06,
"loss": 0.8982,
"step": 250
},
{
"epoch": 0.04,
"learning_rate": 5.420000000000001e-06,
"loss": 0.958,
"step": 275
},
{
"epoch": 0.04,
"learning_rate": 5.92e-06,
"loss": 0.9149,
"step": 300
},
{
"epoch": 0.05,
"learning_rate": 6.42e-06,
"loss": 0.8794,
"step": 325
},
{
"epoch": 0.05,
"learning_rate": 6.92e-06,
"loss": 0.8699,
"step": 350
},
{
"epoch": 0.05,
"learning_rate": 7.420000000000001e-06,
"loss": 1.0492,
"step": 375
},
{
"epoch": 0.06,
"learning_rate": 7.92e-06,
"loss": 0.8222,
"step": 400
},
{
"epoch": 0.06,
"learning_rate": 8.42e-06,
"loss": 0.8455,
"step": 425
},
{
"epoch": 0.07,
"learning_rate": 8.920000000000001e-06,
"loss": 0.828,
"step": 450
},
{
"epoch": 0.07,
"learning_rate": 9.42e-06,
"loss": 0.7183,
"step": 475
},
{
"epoch": 0.07,
"learning_rate": 9.920000000000002e-06,
"loss": 0.7709,
"step": 500
},
{
"epoch": 0.08,
"learning_rate": 9.953333333333333e-06,
"loss": 0.893,
"step": 525
},
{
"epoch": 0.08,
"learning_rate": 9.89777777777778e-06,
"loss": 0.888,
"step": 550
},
{
"epoch": 0.08,
"learning_rate": 9.842222222222223e-06,
"loss": 0.8842,
"step": 575
},
{
"epoch": 0.09,
"learning_rate": 9.786666666666667e-06,
"loss": 0.7789,
"step": 600
},
{
"epoch": 0.09,
"learning_rate": 9.731111111111113e-06,
"loss": 0.7329,
"step": 625
},
{
"epoch": 0.09,
"learning_rate": 9.675555555555555e-06,
"loss": 0.7579,
"step": 650
},
{
"epoch": 0.1,
"learning_rate": 9.620000000000001e-06,
"loss": 0.7731,
"step": 675
},
{
"epoch": 0.1,
"learning_rate": 9.564444444444445e-06,
"loss": 0.7804,
"step": 700
},
{
"epoch": 0.11,
"learning_rate": 9.508888888888889e-06,
"loss": 0.8862,
"step": 725
},
{
"epoch": 0.11,
"learning_rate": 9.453333333333335e-06,
"loss": 0.804,
"step": 750
},
{
"epoch": 0.11,
"learning_rate": 9.397777777777779e-06,
"loss": 0.8471,
"step": 775
},
{
"epoch": 0.12,
"learning_rate": 9.342222222222223e-06,
"loss": 0.9504,
"step": 800
},
{
"epoch": 0.12,
"learning_rate": 9.286666666666667e-06,
"loss": 0.7592,
"step": 825
},
{
"epoch": 0.12,
"learning_rate": 9.231111111111111e-06,
"loss": 0.7828,
"step": 850
},
{
"epoch": 0.13,
"learning_rate": 9.175555555555557e-06,
"loss": 0.7846,
"step": 875
},
{
"epoch": 0.13,
"learning_rate": 9.12e-06,
"loss": 0.7588,
"step": 900
},
{
"epoch": 0.13,
"learning_rate": 9.064444444444447e-06,
"loss": 0.735,
"step": 925
},
{
"epoch": 0.14,
"learning_rate": 9.008888888888889e-06,
"loss": 0.7392,
"step": 950
},
{
"epoch": 0.14,
"learning_rate": 8.953333333333335e-06,
"loss": 0.627,
"step": 975
},
{
"epoch": 0.15,
"learning_rate": 8.897777777777779e-06,
"loss": 0.8102,
"step": 1000
},
{
"epoch": 0.15,
"eval_loss": 0.3227628469467163,
"eval_runtime": 607.3137,
"eval_samples_per_second": 5.964,
"eval_steps_per_second": 1.492,
"eval_wer": 18.753163100989187,
"step": 1000
},
{
"epoch": 0.15,
"learning_rate": 8.842222222222223e-06,
"loss": 0.8372,
"step": 1025
},
{
"epoch": 0.15,
"learning_rate": 8.786666666666668e-06,
"loss": 0.8668,
"step": 1050
},
{
"epoch": 0.16,
"learning_rate": 8.73111111111111e-06,
"loss": 0.7656,
"step": 1075
},
{
"epoch": 0.16,
"learning_rate": 8.675555555555556e-06,
"loss": 0.7988,
"step": 1100
},
{
"epoch": 0.16,
"learning_rate": 8.62e-06,
"loss": 0.772,
"step": 1125
},
{
"epoch": 0.17,
"learning_rate": 8.564444444444445e-06,
"loss": 0.8093,
"step": 1150
},
{
"epoch": 0.17,
"learning_rate": 8.50888888888889e-06,
"loss": 0.7439,
"step": 1175
},
{
"epoch": 0.18,
"learning_rate": 8.453333333333334e-06,
"loss": 0.7089,
"step": 1200
},
{
"epoch": 0.18,
"learning_rate": 8.397777777777778e-06,
"loss": 0.7016,
"step": 1225
},
{
"epoch": 0.18,
"learning_rate": 8.342222222222222e-06,
"loss": 0.7287,
"step": 1250
},
{
"epoch": 0.19,
"learning_rate": 8.286666666666668e-06,
"loss": 0.7294,
"step": 1275
},
{
"epoch": 0.19,
"learning_rate": 8.231111111111112e-06,
"loss": 0.7996,
"step": 1300
},
{
"epoch": 0.19,
"learning_rate": 8.175555555555556e-06,
"loss": 0.692,
"step": 1325
},
{
"epoch": 0.2,
"learning_rate": 8.120000000000002e-06,
"loss": 0.8012,
"step": 1350
},
{
"epoch": 0.2,
"learning_rate": 8.064444444444444e-06,
"loss": 0.8142,
"step": 1375
},
{
"epoch": 0.2,
"learning_rate": 8.00888888888889e-06,
"loss": 0.7401,
"step": 1400
},
{
"epoch": 0.21,
"learning_rate": 7.953333333333334e-06,
"loss": 0.7315,
"step": 1425
},
{
"epoch": 0.21,
"learning_rate": 7.897777777777778e-06,
"loss": 0.8437,
"step": 1450
},
{
"epoch": 0.22,
"learning_rate": 7.842222222222224e-06,
"loss": 0.7634,
"step": 1475
},
{
"epoch": 0.22,
"learning_rate": 7.786666666666666e-06,
"loss": 0.6829,
"step": 1500
},
{
"epoch": 0.22,
"learning_rate": 7.731111111111112e-06,
"loss": 0.6732,
"step": 1525
},
{
"epoch": 0.23,
"learning_rate": 7.675555555555556e-06,
"loss": 0.7316,
"step": 1550
},
{
"epoch": 0.23,
"learning_rate": 7.620000000000001e-06,
"loss": 0.7191,
"step": 1575
},
{
"epoch": 0.23,
"learning_rate": 7.564444444444446e-06,
"loss": 0.7026,
"step": 1600
},
{
"epoch": 0.24,
"learning_rate": 7.50888888888889e-06,
"loss": 0.7266,
"step": 1625
},
{
"epoch": 0.24,
"learning_rate": 7.453333333333334e-06,
"loss": 0.7509,
"step": 1650
},
{
"epoch": 0.24,
"learning_rate": 7.3977777777777786e-06,
"loss": 0.7587,
"step": 1675
},
{
"epoch": 0.25,
"learning_rate": 7.342222222222223e-06,
"loss": 0.672,
"step": 1700
},
{
"epoch": 0.25,
"learning_rate": 7.2866666666666675e-06,
"loss": 0.6921,
"step": 1725
},
{
"epoch": 0.26,
"learning_rate": 7.231111111111112e-06,
"loss": 0.7775,
"step": 1750
},
{
"epoch": 0.26,
"learning_rate": 7.1755555555555556e-06,
"loss": 0.7402,
"step": 1775
},
{
"epoch": 0.26,
"learning_rate": 7.1200000000000004e-06,
"loss": 0.7578,
"step": 1800
},
{
"epoch": 0.27,
"learning_rate": 7.0644444444444445e-06,
"loss": 0.7355,
"step": 1825
},
{
"epoch": 0.27,
"learning_rate": 7.008888888888889e-06,
"loss": 0.8153,
"step": 1850
},
{
"epoch": 0.27,
"learning_rate": 6.953333333333334e-06,
"loss": 0.7923,
"step": 1875
},
{
"epoch": 0.28,
"learning_rate": 6.897777777777779e-06,
"loss": 0.6245,
"step": 1900
},
{
"epoch": 0.28,
"learning_rate": 6.842222222222222e-06,
"loss": 0.7447,
"step": 1925
},
{
"epoch": 0.28,
"learning_rate": 6.786666666666667e-06,
"loss": 0.7685,
"step": 1950
},
{
"epoch": 0.29,
"learning_rate": 6.731111111111111e-06,
"loss": 0.7267,
"step": 1975
},
{
"epoch": 0.29,
"learning_rate": 6.675555555555556e-06,
"loss": 0.6991,
"step": 2000
},
{
"epoch": 0.29,
"eval_loss": 0.2806273400783539,
"eval_runtime": 591.6859,
"eval_samples_per_second": 6.121,
"eval_steps_per_second": 1.531,
"eval_wer": 16.65516448125144,
"step": 2000
},
{
"epoch": 0.3,
"learning_rate": 6.620000000000001e-06,
"loss": 0.713,
"step": 2025
},
{
"epoch": 0.3,
"learning_rate": 6.564444444444446e-06,
"loss": 0.6476,
"step": 2050
},
{
"epoch": 0.3,
"learning_rate": 6.508888888888889e-06,
"loss": 0.661,
"step": 2075
},
{
"epoch": 0.31,
"learning_rate": 6.453333333333334e-06,
"loss": 0.7321,
"step": 2100
},
{
"epoch": 0.31,
"learning_rate": 6.397777777777778e-06,
"loss": 0.6634,
"step": 2125
},
{
"epoch": 0.31,
"learning_rate": 6.342222222222223e-06,
"loss": 0.6978,
"step": 2150
},
{
"epoch": 0.32,
"learning_rate": 6.286666666666668e-06,
"loss": 0.8157,
"step": 2175
},
{
"epoch": 0.32,
"learning_rate": 6.231111111111111e-06,
"loss": 0.7218,
"step": 2200
},
{
"epoch": 0.32,
"learning_rate": 6.175555555555556e-06,
"loss": 0.7102,
"step": 2225
},
{
"epoch": 0.33,
"learning_rate": 6.120000000000001e-06,
"loss": 0.6498,
"step": 2250
},
{
"epoch": 0.33,
"learning_rate": 6.064444444444445e-06,
"loss": 0.6101,
"step": 2275
},
{
"epoch": 0.34,
"learning_rate": 6.00888888888889e-06,
"loss": 0.7386,
"step": 2300
},
{
"epoch": 0.34,
"learning_rate": 5.9533333333333345e-06,
"loss": 0.6093,
"step": 2325
},
{
"epoch": 0.34,
"learning_rate": 5.897777777777778e-06,
"loss": 0.6496,
"step": 2350
},
{
"epoch": 0.35,
"learning_rate": 5.844444444444445e-06,
"loss": 0.7292,
"step": 2375
},
{
"epoch": 0.35,
"learning_rate": 5.788888888888889e-06,
"loss": 0.702,
"step": 2400
},
{
"epoch": 0.35,
"learning_rate": 5.733333333333334e-06,
"loss": 0.6179,
"step": 2425
},
{
"epoch": 0.36,
"learning_rate": 5.677777777777779e-06,
"loss": 0.7102,
"step": 2450
},
{
"epoch": 0.36,
"learning_rate": 5.622222222222222e-06,
"loss": 0.7102,
"step": 2475
},
{
"epoch": 0.36,
"learning_rate": 5.566666666666667e-06,
"loss": 0.695,
"step": 2500
},
{
"epoch": 0.37,
"learning_rate": 5.511111111111112e-06,
"loss": 0.6628,
"step": 2525
},
{
"epoch": 0.37,
"learning_rate": 5.455555555555556e-06,
"loss": 0.7201,
"step": 2550
},
{
"epoch": 0.38,
"learning_rate": 5.400000000000001e-06,
"loss": 0.7195,
"step": 2575
},
{
"epoch": 0.38,
"learning_rate": 5.344444444444446e-06,
"loss": 0.7224,
"step": 2600
},
{
"epoch": 0.38,
"learning_rate": 5.288888888888889e-06,
"loss": 0.6303,
"step": 2625
},
{
"epoch": 0.39,
"learning_rate": 5.233333333333334e-06,
"loss": 0.7031,
"step": 2650
},
{
"epoch": 0.39,
"learning_rate": 5.177777777777779e-06,
"loss": 0.7739,
"step": 2675
},
{
"epoch": 0.39,
"learning_rate": 5.122222222222223e-06,
"loss": 0.5961,
"step": 2700
},
{
"epoch": 0.4,
"learning_rate": 5.0666666666666676e-06,
"loss": 0.6193,
"step": 2725
},
{
"epoch": 0.4,
"learning_rate": 5.011111111111111e-06,
"loss": 0.7679,
"step": 2750
},
{
"epoch": 0.4,
"learning_rate": 4.9555555555555565e-06,
"loss": 0.698,
"step": 2775
},
{
"epoch": 0.41,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.5985,
"step": 2800
},
{
"epoch": 0.41,
"learning_rate": 4.8444444444444446e-06,
"loss": 0.5928,
"step": 2825
},
{
"epoch": 0.42,
"learning_rate": 4.7888888888888894e-06,
"loss": 0.6955,
"step": 2850
},
{
"epoch": 0.42,
"learning_rate": 4.7333333333333335e-06,
"loss": 0.6749,
"step": 2875
},
{
"epoch": 0.42,
"learning_rate": 4.677777777777778e-06,
"loss": 0.6241,
"step": 2900
},
{
"epoch": 0.43,
"learning_rate": 4.622222222222222e-06,
"loss": 0.7025,
"step": 2925
},
{
"epoch": 0.43,
"learning_rate": 4.566666666666667e-06,
"loss": 0.6894,
"step": 2950
},
{
"epoch": 0.43,
"learning_rate": 4.511111111111111e-06,
"loss": 0.728,
"step": 2975
},
{
"epoch": 0.44,
"learning_rate": 4.455555555555555e-06,
"loss": 0.6588,
"step": 3000
},
{
"epoch": 0.44,
"eval_loss": 0.2646391689777374,
"eval_runtime": 588.0604,
"eval_samples_per_second": 6.159,
"eval_steps_per_second": 1.541,
"eval_wer": 14.883827927306188,
"step": 3000
},
{
"epoch": 0.44,
"learning_rate": 4.4e-06,
"loss": 0.622,
"step": 3025
},
{
"epoch": 0.44,
"learning_rate": 4.344444444444445e-06,
"loss": 0.7499,
"step": 3050
},
{
"epoch": 0.45,
"learning_rate": 4.288888888888889e-06,
"loss": 0.6773,
"step": 3075
},
{
"epoch": 0.45,
"learning_rate": 4.233333333333334e-06,
"loss": 0.6319,
"step": 3100
},
{
"epoch": 0.46,
"learning_rate": 4.177777777777778e-06,
"loss": 0.718,
"step": 3125
},
{
"epoch": 0.46,
"learning_rate": 4.122222222222222e-06,
"loss": 0.712,
"step": 3150
},
{
"epoch": 0.46,
"learning_rate": 4.066666666666667e-06,
"loss": 0.6022,
"step": 3175
},
{
"epoch": 0.47,
"learning_rate": 4.011111111111111e-06,
"loss": 0.7059,
"step": 3200
},
{
"epoch": 0.47,
"learning_rate": 3.955555555555556e-06,
"loss": 0.6357,
"step": 3225
},
{
"epoch": 0.47,
"learning_rate": 3.900000000000001e-06,
"loss": 0.6207,
"step": 3250
},
{
"epoch": 0.48,
"learning_rate": 3.844444444444445e-06,
"loss": 0.6044,
"step": 3275
},
{
"epoch": 0.48,
"learning_rate": 3.7888888888888893e-06,
"loss": 0.7074,
"step": 3300
},
{
"epoch": 0.49,
"learning_rate": 3.7333333333333337e-06,
"loss": 0.7237,
"step": 3325
},
{
"epoch": 0.49,
"learning_rate": 3.6777777777777778e-06,
"loss": 0.5633,
"step": 3350
},
{
"epoch": 0.49,
"learning_rate": 3.6222222222222226e-06,
"loss": 0.6134,
"step": 3375
},
{
"epoch": 0.5,
"learning_rate": 3.566666666666667e-06,
"loss": 0.6621,
"step": 3400
},
{
"epoch": 0.5,
"learning_rate": 3.511111111111111e-06,
"loss": 0.5838,
"step": 3425
},
{
"epoch": 0.5,
"learning_rate": 3.455555555555556e-06,
"loss": 0.6785,
"step": 3450
},
{
"epoch": 0.51,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.6518,
"step": 3475
},
{
"epoch": 0.51,
"learning_rate": 3.3444444444444445e-06,
"loss": 0.7239,
"step": 3500
},
{
"epoch": 0.51,
"learning_rate": 3.2888888888888894e-06,
"loss": 0.6214,
"step": 3525
},
{
"epoch": 0.52,
"learning_rate": 3.2333333333333334e-06,
"loss": 0.6181,
"step": 3550
},
{
"epoch": 0.52,
"learning_rate": 3.177777777777778e-06,
"loss": 0.712,
"step": 3575
},
{
"epoch": 0.53,
"learning_rate": 3.1222222222222228e-06,
"loss": 0.6462,
"step": 3600
},
{
"epoch": 0.53,
"learning_rate": 3.066666666666667e-06,
"loss": 0.6712,
"step": 3625
},
{
"epoch": 0.53,
"learning_rate": 3.0111111111111113e-06,
"loss": 0.6999,
"step": 3650
},
{
"epoch": 0.54,
"learning_rate": 2.955555555555556e-06,
"loss": 0.6998,
"step": 3675
},
{
"epoch": 0.54,
"learning_rate": 2.9e-06,
"loss": 0.6752,
"step": 3700
},
{
"epoch": 0.54,
"learning_rate": 2.8444444444444446e-06,
"loss": 0.611,
"step": 3725
},
{
"epoch": 0.55,
"learning_rate": 2.788888888888889e-06,
"loss": 0.7246,
"step": 3750
},
{
"epoch": 0.55,
"learning_rate": 2.7333333333333336e-06,
"loss": 0.6551,
"step": 3775
},
{
"epoch": 0.55,
"learning_rate": 2.677777777777778e-06,
"loss": 0.7257,
"step": 3800
},
{
"epoch": 0.56,
"learning_rate": 2.6222222222222225e-06,
"loss": 0.6496,
"step": 3825
},
{
"epoch": 0.56,
"learning_rate": 2.566666666666667e-06,
"loss": 0.6711,
"step": 3850
},
{
"epoch": 0.57,
"learning_rate": 2.5111111111111114e-06,
"loss": 0.683,
"step": 3875
},
{
"epoch": 0.57,
"learning_rate": 2.455555555555556e-06,
"loss": 0.6637,
"step": 3900
},
{
"epoch": 0.57,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.5956,
"step": 3925
},
{
"epoch": 0.58,
"learning_rate": 2.3444444444444448e-06,
"loss": 0.6124,
"step": 3950
},
{
"epoch": 0.58,
"learning_rate": 2.2888888888888892e-06,
"loss": 0.6389,
"step": 3975
},
{
"epoch": 0.58,
"learning_rate": 2.2333333333333333e-06,
"loss": 0.6162,
"step": 4000
},
{
"epoch": 0.58,
"eval_loss": 0.254045695066452,
"eval_runtime": 634.854,
"eval_samples_per_second": 5.705,
"eval_steps_per_second": 1.427,
"eval_wer": 14.626178974005061,
"step": 4000
},
{
"epoch": 0.59,
"learning_rate": 2.1777777777777777e-06,
"loss": 0.6488,
"step": 4025
},
{
"epoch": 0.59,
"learning_rate": 2.1222222222222226e-06,
"loss": 0.6883,
"step": 4050
},
{
"epoch": 0.59,
"learning_rate": 2.0666666666666666e-06,
"loss": 0.4776,
"step": 4075
},
{
"epoch": 0.6,
"learning_rate": 2.011111111111111e-06,
"loss": 0.5422,
"step": 4100
},
{
"epoch": 0.6,
"learning_rate": 1.955555555555556e-06,
"loss": 0.5351,
"step": 4125
},
{
"epoch": 0.61,
"learning_rate": 1.9000000000000002e-06,
"loss": 0.654,
"step": 4150
},
{
"epoch": 0.61,
"learning_rate": 1.8444444444444445e-06,
"loss": 0.5341,
"step": 4175
},
{
"epoch": 0.61,
"learning_rate": 1.788888888888889e-06,
"loss": 0.6017,
"step": 4200
},
{
"epoch": 0.62,
"learning_rate": 1.7333333333333336e-06,
"loss": 0.6563,
"step": 4225
},
{
"epoch": 0.62,
"learning_rate": 1.6777777777777779e-06,
"loss": 0.6088,
"step": 4250
},
{
"epoch": 0.62,
"learning_rate": 1.6222222222222223e-06,
"loss": 0.554,
"step": 4275
},
{
"epoch": 0.63,
"learning_rate": 1.566666666666667e-06,
"loss": 0.6509,
"step": 4300
},
{
"epoch": 0.63,
"learning_rate": 1.5111111111111112e-06,
"loss": 0.6028,
"step": 4325
},
{
"epoch": 0.63,
"learning_rate": 1.4555555555555557e-06,
"loss": 0.6169,
"step": 4350
},
{
"epoch": 0.64,
"learning_rate": 1.4000000000000001e-06,
"loss": 0.6355,
"step": 4375
},
{
"epoch": 0.64,
"learning_rate": 1.3444444444444446e-06,
"loss": 0.6243,
"step": 4400
},
{
"epoch": 0.65,
"learning_rate": 1.288888888888889e-06,
"loss": 0.6761,
"step": 4425
},
{
"epoch": 0.65,
"learning_rate": 1.2333333333333335e-06,
"loss": 0.6025,
"step": 4450
},
{
"epoch": 0.65,
"learning_rate": 1.1777777777777778e-06,
"loss": 0.6682,
"step": 4475
},
{
"epoch": 0.66,
"learning_rate": 1.1222222222222222e-06,
"loss": 0.6594,
"step": 4500
},
{
"epoch": 0.66,
"learning_rate": 1.066666666666667e-06,
"loss": 0.6569,
"step": 4525
},
{
"epoch": 0.66,
"learning_rate": 1.0111111111111111e-06,
"loss": 0.5149,
"step": 4550
},
{
"epoch": 0.67,
"learning_rate": 9.555555555555556e-07,
"loss": 0.5708,
"step": 4575
},
{
"epoch": 0.67,
"learning_rate": 9.000000000000001e-07,
"loss": 0.6673,
"step": 4600
},
{
"epoch": 0.67,
"learning_rate": 8.444444444444445e-07,
"loss": 0.5476,
"step": 4625
},
{
"epoch": 0.68,
"learning_rate": 7.888888888888889e-07,
"loss": 0.6071,
"step": 4650
},
{
"epoch": 0.68,
"learning_rate": 7.333333333333334e-07,
"loss": 0.6548,
"step": 4675
},
{
"epoch": 0.69,
"learning_rate": 6.777777777777779e-07,
"loss": 0.6399,
"step": 4700
},
{
"epoch": 0.69,
"learning_rate": 6.222222222222223e-07,
"loss": 0.6747,
"step": 4725
},
{
"epoch": 0.69,
"learning_rate": 5.666666666666667e-07,
"loss": 0.5189,
"step": 4750
},
{
"epoch": 0.7,
"learning_rate": 5.111111111111112e-07,
"loss": 0.5984,
"step": 4775
},
{
"epoch": 0.7,
"learning_rate": 4.5555555555555563e-07,
"loss": 0.6844,
"step": 4800
},
{
"epoch": 0.7,
"learning_rate": 4.0000000000000003e-07,
"loss": 0.6189,
"step": 4825
},
{
"epoch": 0.71,
"learning_rate": 3.444444444444445e-07,
"loss": 0.5675,
"step": 4850
},
{
"epoch": 0.71,
"learning_rate": 2.888888888888889e-07,
"loss": 0.6328,
"step": 4875
},
{
"epoch": 0.71,
"learning_rate": 2.3333333333333336e-07,
"loss": 0.644,
"step": 4900
},
{
"epoch": 0.72,
"learning_rate": 1.777777777777778e-07,
"loss": 0.5618,
"step": 4925
},
{
"epoch": 0.72,
"learning_rate": 1.2222222222222225e-07,
"loss": 0.5793,
"step": 4950
},
{
"epoch": 0.73,
"learning_rate": 6.666666666666668e-08,
"loss": 0.5976,
"step": 4975
},
{
"epoch": 0.73,
"learning_rate": 1.1111111111111112e-08,
"loss": 0.5996,
"step": 5000
},
{
"epoch": 0.73,
"eval_loss": 0.2466590255498886,
"eval_runtime": 595.0979,
"eval_samples_per_second": 6.086,
"eval_steps_per_second": 1.522,
"eval_wer": 14.538762364849322,
"step": 5000
},
{
"epoch": 0.73,
"step": 5000,
"total_flos": 1.15434160128e+19,
"train_loss": 0.7296087381362915,
"train_runtime": 7698.7874,
"train_samples_per_second": 5.196,
"train_steps_per_second": 0.649
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 1.15434160128e+19,
"trial_name": null,
"trial_params": null
}