whisper-large-nob-ncc-s / trainer_state.json
versae's picture
End of training
38469f8
raw
history blame
25.7 kB
{
"best_metric": 12.51522533495737,
"best_model_checkpoint": "./checkpoint-5000",
"epoch": 1.0,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.6000000000000004e-07,
"loss": 1.0966,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.200000000000001e-07,
"loss": 1.0472,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.42e-06,
"loss": 1.0294,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.9200000000000003e-06,
"loss": 0.9092,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.42e-06,
"loss": 0.7806,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.92e-06,
"loss": 0.7536,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 3.4200000000000007e-06,
"loss": 0.7896,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 3.920000000000001e-06,
"loss": 0.7669,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.42e-06,
"loss": 0.7402,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 4.92e-06,
"loss": 0.789,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 5.420000000000001e-06,
"loss": 0.744,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 5.92e-06,
"loss": 0.7739,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 6.42e-06,
"loss": 0.7387,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 6.92e-06,
"loss": 0.7706,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 7.420000000000001e-06,
"loss": 0.7456,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 7.92e-06,
"loss": 0.7411,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 8.42e-06,
"loss": 0.752,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 8.920000000000001e-06,
"loss": 0.7749,
"step": 450
},
{
"epoch": 0.1,
"learning_rate": 9.42e-06,
"loss": 0.7737,
"step": 475
},
{
"epoch": 0.1,
"learning_rate": 9.920000000000002e-06,
"loss": 0.7354,
"step": 500
},
{
"epoch": 0.1,
"learning_rate": 9.953333333333333e-06,
"loss": 0.7525,
"step": 525
},
{
"epoch": 0.11,
"learning_rate": 9.89777777777778e-06,
"loss": 0.7379,
"step": 550
},
{
"epoch": 0.12,
"learning_rate": 9.842222222222223e-06,
"loss": 0.7668,
"step": 575
},
{
"epoch": 0.12,
"learning_rate": 9.786666666666667e-06,
"loss": 0.7776,
"step": 600
},
{
"epoch": 0.12,
"learning_rate": 9.731111111111113e-06,
"loss": 0.7274,
"step": 625
},
{
"epoch": 0.13,
"learning_rate": 9.675555555555555e-06,
"loss": 0.7696,
"step": 650
},
{
"epoch": 0.14,
"learning_rate": 9.620000000000001e-06,
"loss": 0.7645,
"step": 675
},
{
"epoch": 0.14,
"learning_rate": 9.564444444444445e-06,
"loss": 0.6955,
"step": 700
},
{
"epoch": 0.14,
"learning_rate": 9.508888888888889e-06,
"loss": 0.755,
"step": 725
},
{
"epoch": 0.15,
"learning_rate": 9.453333333333335e-06,
"loss": 0.7179,
"step": 750
},
{
"epoch": 0.15,
"learning_rate": 9.397777777777779e-06,
"loss": 0.7662,
"step": 775
},
{
"epoch": 0.16,
"learning_rate": 9.342222222222223e-06,
"loss": 0.7007,
"step": 800
},
{
"epoch": 0.17,
"learning_rate": 9.286666666666667e-06,
"loss": 0.7213,
"step": 825
},
{
"epoch": 0.17,
"learning_rate": 9.231111111111111e-06,
"loss": 0.7328,
"step": 850
},
{
"epoch": 0.17,
"learning_rate": 9.175555555555557e-06,
"loss": 0.7615,
"step": 875
},
{
"epoch": 0.18,
"learning_rate": 9.12e-06,
"loss": 0.7401,
"step": 900
},
{
"epoch": 0.18,
"learning_rate": 9.064444444444447e-06,
"loss": 0.7413,
"step": 925
},
{
"epoch": 0.19,
"learning_rate": 9.008888888888889e-06,
"loss": 0.7344,
"step": 950
},
{
"epoch": 0.2,
"learning_rate": 8.953333333333335e-06,
"loss": 0.7204,
"step": 975
},
{
"epoch": 0.2,
"learning_rate": 8.897777777777779e-06,
"loss": 0.6892,
"step": 1000
},
{
"epoch": 0.2,
"eval_loss": 0.31772905588150024,
"eval_runtime": 382.7289,
"eval_samples_per_second": 0.421,
"eval_steps_per_second": 0.071,
"eval_wer": 15.10353227771011,
"step": 1000
},
{
"epoch": 0.2,
"learning_rate": 8.842222222222223e-06,
"loss": 0.7361,
"step": 1025
},
{
"epoch": 0.21,
"learning_rate": 8.786666666666668e-06,
"loss": 0.7015,
"step": 1050
},
{
"epoch": 0.21,
"learning_rate": 8.73111111111111e-06,
"loss": 0.7137,
"step": 1075
},
{
"epoch": 0.22,
"learning_rate": 8.675555555555556e-06,
"loss": 0.6902,
"step": 1100
},
{
"epoch": 0.23,
"learning_rate": 8.62e-06,
"loss": 0.7153,
"step": 1125
},
{
"epoch": 0.23,
"learning_rate": 8.566666666666667e-06,
"loss": 0.7062,
"step": 1150
},
{
"epoch": 0.23,
"learning_rate": 8.511111111111113e-06,
"loss": 0.7022,
"step": 1175
},
{
"epoch": 0.24,
"learning_rate": 8.455555555555555e-06,
"loss": 0.7291,
"step": 1200
},
{
"epoch": 0.24,
"learning_rate": 8.400000000000001e-06,
"loss": 0.7372,
"step": 1225
},
{
"epoch": 0.25,
"learning_rate": 8.344444444444445e-06,
"loss": 0.7218,
"step": 1250
},
{
"epoch": 0.26,
"learning_rate": 8.288888888888889e-06,
"loss": 0.7003,
"step": 1275
},
{
"epoch": 0.26,
"learning_rate": 8.233333333333335e-06,
"loss": 0.7167,
"step": 1300
},
{
"epoch": 0.27,
"learning_rate": 8.177777777777779e-06,
"loss": 0.7007,
"step": 1325
},
{
"epoch": 0.27,
"learning_rate": 8.122222222222223e-06,
"loss": 0.712,
"step": 1350
},
{
"epoch": 0.28,
"learning_rate": 8.066666666666667e-06,
"loss": 0.7148,
"step": 1375
},
{
"epoch": 0.28,
"learning_rate": 8.011111111111113e-06,
"loss": 0.7167,
"step": 1400
},
{
"epoch": 0.28,
"learning_rate": 7.955555555555557e-06,
"loss": 0.6501,
"step": 1425
},
{
"epoch": 0.29,
"learning_rate": 7.9e-06,
"loss": 0.6562,
"step": 1450
},
{
"epoch": 0.29,
"learning_rate": 7.844444444444446e-06,
"loss": 0.6613,
"step": 1475
},
{
"epoch": 0.3,
"learning_rate": 7.788888888888889e-06,
"loss": 0.6738,
"step": 1500
},
{
"epoch": 0.3,
"learning_rate": 7.733333333333334e-06,
"loss": 0.7597,
"step": 1525
},
{
"epoch": 0.31,
"learning_rate": 7.677777777777778e-06,
"loss": 0.7037,
"step": 1550
},
{
"epoch": 0.32,
"learning_rate": 7.622222222222223e-06,
"loss": 0.6631,
"step": 1575
},
{
"epoch": 0.32,
"learning_rate": 7.566666666666667e-06,
"loss": 0.7103,
"step": 1600
},
{
"epoch": 0.33,
"learning_rate": 7.511111111111111e-06,
"loss": 0.676,
"step": 1625
},
{
"epoch": 0.33,
"learning_rate": 7.455555555555556e-06,
"loss": 0.7209,
"step": 1650
},
{
"epoch": 0.34,
"learning_rate": 7.4e-06,
"loss": 0.6976,
"step": 1675
},
{
"epoch": 0.34,
"learning_rate": 7.344444444444445e-06,
"loss": 0.6452,
"step": 1700
},
{
"epoch": 0.34,
"learning_rate": 7.28888888888889e-06,
"loss": 0.6783,
"step": 1725
},
{
"epoch": 0.35,
"learning_rate": 7.233333333333334e-06,
"loss": 0.6861,
"step": 1750
},
{
"epoch": 0.35,
"learning_rate": 7.177777777777778e-06,
"loss": 0.6507,
"step": 1775
},
{
"epoch": 0.36,
"learning_rate": 7.122222222222222e-06,
"loss": 0.6878,
"step": 1800
},
{
"epoch": 0.36,
"learning_rate": 7.066666666666667e-06,
"loss": 0.6879,
"step": 1825
},
{
"epoch": 0.37,
"learning_rate": 7.011111111111112e-06,
"loss": 0.6633,
"step": 1850
},
{
"epoch": 0.38,
"learning_rate": 6.955555555555557e-06,
"loss": 0.6557,
"step": 1875
},
{
"epoch": 0.38,
"learning_rate": 6.9e-06,
"loss": 0.6748,
"step": 1900
},
{
"epoch": 0.39,
"learning_rate": 6.844444444444445e-06,
"loss": 0.7066,
"step": 1925
},
{
"epoch": 0.39,
"learning_rate": 6.788888888888889e-06,
"loss": 0.6731,
"step": 1950
},
{
"epoch": 0.4,
"learning_rate": 6.733333333333334e-06,
"loss": 0.6031,
"step": 1975
},
{
"epoch": 0.4,
"learning_rate": 6.677777777777779e-06,
"loss": 0.6782,
"step": 2000
},
{
"epoch": 0.4,
"eval_loss": 0.30329471826553345,
"eval_runtime": 378.3807,
"eval_samples_per_second": 0.425,
"eval_steps_per_second": 0.071,
"eval_wer": 13.45919610231425,
"step": 2000
},
{
"epoch": 0.41,
"learning_rate": 6.6222222222222236e-06,
"loss": 0.6905,
"step": 2025
},
{
"epoch": 0.41,
"learning_rate": 6.566666666666667e-06,
"loss": 0.6515,
"step": 2050
},
{
"epoch": 0.41,
"learning_rate": 6.511111111111112e-06,
"loss": 0.6758,
"step": 2075
},
{
"epoch": 0.42,
"learning_rate": 6.455555555555556e-06,
"loss": 0.6798,
"step": 2100
},
{
"epoch": 0.42,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.6526,
"step": 2125
},
{
"epoch": 0.43,
"learning_rate": 6.3444444444444454e-06,
"loss": 0.6411,
"step": 2150
},
{
"epoch": 0.43,
"learning_rate": 6.28888888888889e-06,
"loss": 0.6341,
"step": 2175
},
{
"epoch": 0.44,
"learning_rate": 6.2333333333333335e-06,
"loss": 0.6493,
"step": 2200
},
{
"epoch": 0.45,
"learning_rate": 6.177777777777778e-06,
"loss": 0.6378,
"step": 2225
},
{
"epoch": 0.45,
"learning_rate": 6.1222222222222224e-06,
"loss": 0.646,
"step": 2250
},
{
"epoch": 0.46,
"learning_rate": 6.066666666666667e-06,
"loss": 0.6662,
"step": 2275
},
{
"epoch": 0.46,
"learning_rate": 6.011111111111112e-06,
"loss": 0.629,
"step": 2300
},
{
"epoch": 0.47,
"learning_rate": 5.955555555555555e-06,
"loss": 0.6247,
"step": 2325
},
{
"epoch": 0.47,
"learning_rate": 5.9e-06,
"loss": 0.6436,
"step": 2350
},
{
"epoch": 0.47,
"learning_rate": 5.844444444444445e-06,
"loss": 0.6471,
"step": 2375
},
{
"epoch": 0.48,
"learning_rate": 5.788888888888889e-06,
"loss": 0.6451,
"step": 2400
},
{
"epoch": 0.48,
"learning_rate": 5.733333333333334e-06,
"loss": 0.6251,
"step": 2425
},
{
"epoch": 0.49,
"learning_rate": 5.677777777777779e-06,
"loss": 0.655,
"step": 2450
},
{
"epoch": 0.49,
"learning_rate": 5.622222222222222e-06,
"loss": 0.6378,
"step": 2475
},
{
"epoch": 0.5,
"learning_rate": 5.566666666666667e-06,
"loss": 0.6501,
"step": 2500
},
{
"epoch": 0.51,
"learning_rate": 5.511111111111112e-06,
"loss": 0.6147,
"step": 2525
},
{
"epoch": 0.51,
"learning_rate": 5.455555555555556e-06,
"loss": 0.6282,
"step": 2550
},
{
"epoch": 0.52,
"learning_rate": 5.400000000000001e-06,
"loss": 0.6338,
"step": 2575
},
{
"epoch": 0.52,
"learning_rate": 5.344444444444446e-06,
"loss": 0.6172,
"step": 2600
},
{
"epoch": 0.53,
"learning_rate": 5.288888888888889e-06,
"loss": 0.6414,
"step": 2625
},
{
"epoch": 0.53,
"learning_rate": 5.233333333333334e-06,
"loss": 0.6364,
"step": 2650
},
{
"epoch": 0.54,
"learning_rate": 5.177777777777779e-06,
"loss": 0.6032,
"step": 2675
},
{
"epoch": 0.54,
"learning_rate": 5.122222222222223e-06,
"loss": 0.6074,
"step": 2700
},
{
"epoch": 0.55,
"learning_rate": 5.0666666666666676e-06,
"loss": 0.6007,
"step": 2725
},
{
"epoch": 0.55,
"learning_rate": 5.011111111111111e-06,
"loss": 0.6551,
"step": 2750
},
{
"epoch": 0.56,
"learning_rate": 4.9555555555555565e-06,
"loss": 0.6185,
"step": 2775
},
{
"epoch": 0.56,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.6533,
"step": 2800
},
{
"epoch": 0.56,
"learning_rate": 4.8444444444444446e-06,
"loss": 0.621,
"step": 2825
},
{
"epoch": 0.57,
"learning_rate": 4.7888888888888894e-06,
"loss": 0.5842,
"step": 2850
},
{
"epoch": 0.57,
"learning_rate": 4.7333333333333335e-06,
"loss": 0.6183,
"step": 2875
},
{
"epoch": 0.58,
"learning_rate": 4.677777777777778e-06,
"loss": 0.6241,
"step": 2900
},
{
"epoch": 0.58,
"learning_rate": 4.622222222222222e-06,
"loss": 0.6324,
"step": 2925
},
{
"epoch": 0.59,
"learning_rate": 4.566666666666667e-06,
"loss": 0.6348,
"step": 2950
},
{
"epoch": 0.59,
"learning_rate": 4.511111111111111e-06,
"loss": 0.5805,
"step": 2975
},
{
"epoch": 0.6,
"learning_rate": 4.455555555555555e-06,
"loss": 0.6317,
"step": 3000
},
{
"epoch": 0.6,
"eval_loss": 0.29093676805496216,
"eval_runtime": 381.6236,
"eval_samples_per_second": 0.422,
"eval_steps_per_second": 0.071,
"eval_wer": 13.763702801461633,
"step": 3000
},
{
"epoch": 0.6,
"learning_rate": 4.4e-06,
"loss": 0.6437,
"step": 3025
},
{
"epoch": 0.61,
"learning_rate": 4.344444444444445e-06,
"loss": 0.5674,
"step": 3050
},
{
"epoch": 0.61,
"learning_rate": 4.288888888888889e-06,
"loss": 0.6217,
"step": 3075
},
{
"epoch": 0.62,
"learning_rate": 4.233333333333334e-06,
"loss": 0.6181,
"step": 3100
},
{
"epoch": 0.62,
"learning_rate": 4.177777777777778e-06,
"loss": 0.586,
"step": 3125
},
{
"epoch": 0.63,
"learning_rate": 4.122222222222222e-06,
"loss": 0.6244,
"step": 3150
},
{
"epoch": 0.64,
"learning_rate": 4.066666666666667e-06,
"loss": 0.5894,
"step": 3175
},
{
"epoch": 0.64,
"learning_rate": 4.011111111111111e-06,
"loss": 0.6249,
"step": 3200
},
{
"epoch": 0.65,
"learning_rate": 3.955555555555556e-06,
"loss": 0.6617,
"step": 3225
},
{
"epoch": 0.65,
"learning_rate": 3.900000000000001e-06,
"loss": 0.5786,
"step": 3250
},
{
"epoch": 0.66,
"learning_rate": 3.844444444444445e-06,
"loss": 0.6574,
"step": 3275
},
{
"epoch": 0.66,
"learning_rate": 3.7888888888888893e-06,
"loss": 0.6017,
"step": 3300
},
{
"epoch": 0.67,
"learning_rate": 3.7333333333333337e-06,
"loss": 0.5849,
"step": 3325
},
{
"epoch": 0.67,
"learning_rate": 3.6777777777777778e-06,
"loss": 0.652,
"step": 3350
},
{
"epoch": 0.68,
"learning_rate": 3.6222222222222226e-06,
"loss": 0.5876,
"step": 3375
},
{
"epoch": 0.68,
"learning_rate": 3.566666666666667e-06,
"loss": 0.5792,
"step": 3400
},
{
"epoch": 0.69,
"learning_rate": 3.511111111111111e-06,
"loss": 0.64,
"step": 3425
},
{
"epoch": 0.69,
"learning_rate": 3.455555555555556e-06,
"loss": 0.5855,
"step": 3450
},
{
"epoch": 0.69,
"learning_rate": 3.4022222222222222e-06,
"loss": 0.6089,
"step": 3475
},
{
"epoch": 0.7,
"learning_rate": 3.346666666666667e-06,
"loss": 0.6211,
"step": 3500
},
{
"epoch": 0.7,
"learning_rate": 3.2911111111111116e-06,
"loss": 0.6219,
"step": 3525
},
{
"epoch": 0.71,
"learning_rate": 3.2355555555555556e-06,
"loss": 0.6342,
"step": 3550
},
{
"epoch": 0.71,
"learning_rate": 3.1800000000000005e-06,
"loss": 0.5594,
"step": 3575
},
{
"epoch": 0.72,
"learning_rate": 3.124444444444445e-06,
"loss": 0.5864,
"step": 3600
},
{
"epoch": 0.72,
"learning_rate": 3.068888888888889e-06,
"loss": 0.6102,
"step": 3625
},
{
"epoch": 0.73,
"learning_rate": 3.013333333333334e-06,
"loss": 0.5731,
"step": 3650
},
{
"epoch": 0.73,
"learning_rate": 2.957777777777778e-06,
"loss": 0.6105,
"step": 3675
},
{
"epoch": 0.74,
"learning_rate": 2.904444444444445e-06,
"loss": 0.6447,
"step": 3700
},
{
"epoch": 0.74,
"learning_rate": 2.8488888888888894e-06,
"loss": 0.5659,
"step": 3725
},
{
"epoch": 0.75,
"learning_rate": 2.7933333333333334e-06,
"loss": 0.6003,
"step": 3750
},
{
"epoch": 0.76,
"learning_rate": 2.7377777777777783e-06,
"loss": 0.5652,
"step": 3775
},
{
"epoch": 0.76,
"learning_rate": 2.6822222222222223e-06,
"loss": 0.5832,
"step": 3800
},
{
"epoch": 0.77,
"learning_rate": 2.6266666666666668e-06,
"loss": 0.5754,
"step": 3825
},
{
"epoch": 0.77,
"learning_rate": 2.5711111111111112e-06,
"loss": 0.5698,
"step": 3850
},
{
"epoch": 0.78,
"learning_rate": 2.5155555555555557e-06,
"loss": 0.6056,
"step": 3875
},
{
"epoch": 0.78,
"learning_rate": 2.46e-06,
"loss": 0.6033,
"step": 3900
},
{
"epoch": 0.79,
"learning_rate": 2.4044444444444446e-06,
"loss": 0.6017,
"step": 3925
},
{
"epoch": 0.79,
"learning_rate": 2.348888888888889e-06,
"loss": 0.5924,
"step": 3950
},
{
"epoch": 0.8,
"learning_rate": 2.2933333333333335e-06,
"loss": 0.577,
"step": 3975
},
{
"epoch": 0.8,
"learning_rate": 2.237777777777778e-06,
"loss": 0.5609,
"step": 4000
},
{
"epoch": 0.8,
"eval_loss": 0.2802838683128357,
"eval_runtime": 383.0412,
"eval_samples_per_second": 0.42,
"eval_steps_per_second": 0.07,
"eval_wer": 12.667478684531059,
"step": 4000
},
{
"epoch": 0.81,
"learning_rate": 2.1822222222222225e-06,
"loss": 0.5854,
"step": 4025
},
{
"epoch": 0.81,
"learning_rate": 2.126666666666667e-06,
"loss": 0.555,
"step": 4050
},
{
"epoch": 0.81,
"learning_rate": 2.0711111111111114e-06,
"loss": 0.5628,
"step": 4075
},
{
"epoch": 0.82,
"learning_rate": 2.0155555555555554e-06,
"loss": 0.603,
"step": 4100
},
{
"epoch": 0.82,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.6071,
"step": 4125
},
{
"epoch": 0.83,
"learning_rate": 1.9044444444444445e-06,
"loss": 0.5964,
"step": 4150
},
{
"epoch": 0.83,
"learning_rate": 1.848888888888889e-06,
"loss": 0.6204,
"step": 4175
},
{
"epoch": 0.84,
"learning_rate": 1.7933333333333337e-06,
"loss": 0.6015,
"step": 4200
},
{
"epoch": 0.84,
"learning_rate": 1.737777777777778e-06,
"loss": 0.5609,
"step": 4225
},
{
"epoch": 0.85,
"learning_rate": 1.6822222222222224e-06,
"loss": 0.6031,
"step": 4250
},
{
"epoch": 0.85,
"learning_rate": 1.6266666666666666e-06,
"loss": 0.601,
"step": 4275
},
{
"epoch": 0.86,
"learning_rate": 1.5711111111111113e-06,
"loss": 0.5936,
"step": 4300
},
{
"epoch": 0.86,
"learning_rate": 1.5155555555555558e-06,
"loss": 0.5775,
"step": 4325
},
{
"epoch": 0.87,
"learning_rate": 1.46e-06,
"loss": 0.5828,
"step": 4350
},
{
"epoch": 0.88,
"learning_rate": 1.4044444444444447e-06,
"loss": 0.5625,
"step": 4375
},
{
"epoch": 0.88,
"learning_rate": 1.3488888888888891e-06,
"loss": 0.5913,
"step": 4400
},
{
"epoch": 0.89,
"learning_rate": 1.2933333333333334e-06,
"loss": 0.6003,
"step": 4425
},
{
"epoch": 0.89,
"learning_rate": 1.2377777777777778e-06,
"loss": 0.6098,
"step": 4450
},
{
"epoch": 0.9,
"learning_rate": 1.1822222222222223e-06,
"loss": 0.5843,
"step": 4475
},
{
"epoch": 0.9,
"learning_rate": 1.1266666666666667e-06,
"loss": 0.6121,
"step": 4500
},
{
"epoch": 0.91,
"learning_rate": 1.0711111111111112e-06,
"loss": 0.6102,
"step": 4525
},
{
"epoch": 0.91,
"learning_rate": 1.0155555555555557e-06,
"loss": 0.565,
"step": 4550
},
{
"epoch": 0.92,
"learning_rate": 9.600000000000001e-07,
"loss": 0.5881,
"step": 4575
},
{
"epoch": 0.92,
"learning_rate": 9.044444444444445e-07,
"loss": 0.6131,
"step": 4600
},
{
"epoch": 0.93,
"learning_rate": 8.488888888888889e-07,
"loss": 0.5857,
"step": 4625
},
{
"epoch": 0.93,
"learning_rate": 7.933333333333335e-07,
"loss": 0.583,
"step": 4650
},
{
"epoch": 0.94,
"learning_rate": 7.377777777777779e-07,
"loss": 0.5698,
"step": 4675
},
{
"epoch": 0.94,
"learning_rate": 6.822222222222223e-07,
"loss": 0.561,
"step": 4700
},
{
"epoch": 0.94,
"learning_rate": 6.266666666666667e-07,
"loss": 0.5673,
"step": 4725
},
{
"epoch": 0.95,
"learning_rate": 5.711111111111111e-07,
"loss": 0.5863,
"step": 4750
},
{
"epoch": 0.95,
"learning_rate": 5.155555555555556e-07,
"loss": 0.5884,
"step": 4775
},
{
"epoch": 0.96,
"learning_rate": 4.6000000000000004e-07,
"loss": 0.5923,
"step": 4800
},
{
"epoch": 0.96,
"learning_rate": 4.0444444444444445e-07,
"loss": 0.5995,
"step": 4825
},
{
"epoch": 0.97,
"learning_rate": 3.488888888888889e-07,
"loss": 0.5717,
"step": 4850
},
{
"epoch": 0.97,
"learning_rate": 2.9333333333333337e-07,
"loss": 0.5667,
"step": 4875
},
{
"epoch": 0.98,
"learning_rate": 2.3777777777777777e-07,
"loss": 0.5785,
"step": 4900
},
{
"epoch": 0.98,
"learning_rate": 1.8222222222222226e-07,
"loss": 0.6161,
"step": 4925
},
{
"epoch": 0.99,
"learning_rate": 1.2666666666666666e-07,
"loss": 0.5518,
"step": 4950
},
{
"epoch": 0.99,
"learning_rate": 7.111111111111112e-08,
"loss": 0.6018,
"step": 4975
},
{
"epoch": 1.0,
"learning_rate": 1.5555555555555557e-08,
"loss": 0.5726,
"step": 5000
},
{
"epoch": 1.0,
"eval_loss": 0.27755075693130493,
"eval_runtime": 386.9803,
"eval_samples_per_second": 0.416,
"eval_steps_per_second": 0.07,
"eval_wer": 12.51522533495737,
"step": 5000
},
{
"epoch": 1.0,
"step": 5000,
"total_flos": 1.2739025557174354e+20,
"train_loss": 0.6583859201431275,
"train_runtime": 31239.5975,
"train_samples_per_second": 1.921,
"train_steps_per_second": 0.16
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.2739025557174354e+20,
"trial_name": null,
"trial_params": null
}