whisper-small-chinese_base / trainer_state.json
Jingmiao's picture
End of training
90a30a5
raw
history blame
25.9 kB
{
"best_metric": 14.92818317087243,
"best_model_checkpoint": "./checkpoint-1000",
"epoch": 98.0004,
"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": 4.4e-07,
"loss": 0.7657,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.400000000000001e-07,
"loss": 0.603,
"step": 50
},
{
"epoch": 1.0,
"learning_rate": 1.44e-06,
"loss": 0.4406,
"step": 75
},
{
"epoch": 1.01,
"learning_rate": 1.94e-06,
"loss": 0.3307,
"step": 100
},
{
"epoch": 2.0,
"learning_rate": 2.4400000000000004e-06,
"loss": 0.2685,
"step": 125
},
{
"epoch": 2.01,
"learning_rate": 2.9400000000000002e-06,
"loss": 0.2362,
"step": 150
},
{
"epoch": 3.0,
"learning_rate": 3.44e-06,
"loss": 0.1957,
"step": 175
},
{
"epoch": 3.01,
"learning_rate": 3.94e-06,
"loss": 0.1714,
"step": 200
},
{
"epoch": 4.0,
"learning_rate": 4.440000000000001e-06,
"loss": 0.1397,
"step": 225
},
{
"epoch": 4.01,
"learning_rate": 4.94e-06,
"loss": 0.1208,
"step": 250
},
{
"epoch": 5.0,
"learning_rate": 5.4400000000000004e-06,
"loss": 0.0969,
"step": 275
},
{
"epoch": 5.01,
"learning_rate": 5.94e-06,
"loss": 0.0809,
"step": 300
},
{
"epoch": 6.0,
"learning_rate": 6.440000000000001e-06,
"loss": 0.0621,
"step": 325
},
{
"epoch": 6.01,
"learning_rate": 6.9400000000000005e-06,
"loss": 0.0492,
"step": 350
},
{
"epoch": 7.0,
"learning_rate": 7.440000000000001e-06,
"loss": 0.0372,
"step": 375
},
{
"epoch": 7.01,
"learning_rate": 7.94e-06,
"loss": 0.0284,
"step": 400
},
{
"epoch": 8.0,
"learning_rate": 8.44e-06,
"loss": 0.0236,
"step": 425
},
{
"epoch": 8.01,
"learning_rate": 8.94e-06,
"loss": 0.0167,
"step": 450
},
{
"epoch": 9.0,
"learning_rate": 9.440000000000001e-06,
"loss": 0.0142,
"step": 475
},
{
"epoch": 9.01,
"learning_rate": 9.940000000000001e-06,
"loss": 0.0107,
"step": 500
},
{
"epoch": 10.0,
"learning_rate": 9.951111111111111e-06,
"loss": 0.0085,
"step": 525
},
{
"epoch": 10.01,
"learning_rate": 9.895555555555557e-06,
"loss": 0.0071,
"step": 550
},
{
"epoch": 11.0,
"learning_rate": 9.84e-06,
"loss": 0.0059,
"step": 575
},
{
"epoch": 11.01,
"learning_rate": 9.784444444444445e-06,
"loss": 0.0049,
"step": 600
},
{
"epoch": 12.0,
"learning_rate": 9.72888888888889e-06,
"loss": 0.0042,
"step": 625
},
{
"epoch": 12.01,
"learning_rate": 9.673333333333334e-06,
"loss": 0.0039,
"step": 650
},
{
"epoch": 13.0,
"learning_rate": 9.617777777777778e-06,
"loss": 0.0034,
"step": 675
},
{
"epoch": 13.01,
"learning_rate": 9.562222222222223e-06,
"loss": 0.0031,
"step": 700
},
{
"epoch": 14.0,
"learning_rate": 9.506666666666667e-06,
"loss": 0.0032,
"step": 725
},
{
"epoch": 14.01,
"learning_rate": 9.451111111111112e-06,
"loss": 0.0026,
"step": 750
},
{
"epoch": 15.0,
"learning_rate": 9.395555555555556e-06,
"loss": 0.0025,
"step": 775
},
{
"epoch": 15.01,
"learning_rate": 9.340000000000002e-06,
"loss": 0.0021,
"step": 800
},
{
"epoch": 16.0,
"learning_rate": 9.284444444444444e-06,
"loss": 0.002,
"step": 825
},
{
"epoch": 16.01,
"learning_rate": 9.22888888888889e-06,
"loss": 0.0018,
"step": 850
},
{
"epoch": 17.0,
"learning_rate": 9.173333333333334e-06,
"loss": 0.0017,
"step": 875
},
{
"epoch": 17.01,
"learning_rate": 9.117777777777778e-06,
"loss": 0.0016,
"step": 900
},
{
"epoch": 18.0,
"learning_rate": 9.062222222222224e-06,
"loss": 0.0015,
"step": 925
},
{
"epoch": 18.01,
"learning_rate": 9.006666666666666e-06,
"loss": 0.0014,
"step": 950
},
{
"epoch": 19.0,
"learning_rate": 8.951111111111112e-06,
"loss": 0.0013,
"step": 975
},
{
"epoch": 19.01,
"learning_rate": 8.895555555555556e-06,
"loss": 0.0013,
"step": 1000
},
{
"epoch": 19.01,
"eval_loss": 0.29096686840057373,
"eval_runtime": 926.9169,
"eval_samples_per_second": 1.02,
"eval_steps_per_second": 0.032,
"eval_wer": 14.92818317087243,
"step": 1000
},
{
"epoch": 20.0,
"learning_rate": 8.84e-06,
"loss": 0.0012,
"step": 1025
},
{
"epoch": 20.01,
"learning_rate": 8.784444444444446e-06,
"loss": 0.0012,
"step": 1050
},
{
"epoch": 21.0,
"learning_rate": 8.72888888888889e-06,
"loss": 0.0011,
"step": 1075
},
{
"epoch": 21.01,
"learning_rate": 8.673333333333334e-06,
"loss": 0.0011,
"step": 1100
},
{
"epoch": 22.0,
"learning_rate": 8.617777777777778e-06,
"loss": 0.0011,
"step": 1125
},
{
"epoch": 22.01,
"learning_rate": 8.562222222222224e-06,
"loss": 0.001,
"step": 1150
},
{
"epoch": 23.0,
"learning_rate": 8.506666666666668e-06,
"loss": 0.001,
"step": 1175
},
{
"epoch": 23.01,
"learning_rate": 8.451111111111112e-06,
"loss": 0.0009,
"step": 1200
},
{
"epoch": 24.0,
"learning_rate": 8.395555555555557e-06,
"loss": 0.0009,
"step": 1225
},
{
"epoch": 24.01,
"learning_rate": 8.34e-06,
"loss": 0.0009,
"step": 1250
},
{
"epoch": 24.01,
"learning_rate": 8.284444444444446e-06,
"loss": 0.0009,
"step": 1275
},
{
"epoch": 25.0,
"learning_rate": 8.22888888888889e-06,
"loss": 0.0008,
"step": 1300
},
{
"epoch": 25.01,
"learning_rate": 8.173333333333334e-06,
"loss": 0.0008,
"step": 1325
},
{
"epoch": 26.0,
"learning_rate": 8.11777777777778e-06,
"loss": 0.0008,
"step": 1350
},
{
"epoch": 26.01,
"learning_rate": 8.062222222222222e-06,
"loss": 0.0008,
"step": 1375
},
{
"epoch": 27.0,
"learning_rate": 8.006666666666667e-06,
"loss": 0.0007,
"step": 1400
},
{
"epoch": 27.01,
"learning_rate": 7.951111111111111e-06,
"loss": 0.0007,
"step": 1425
},
{
"epoch": 28.0,
"learning_rate": 7.895555555555557e-06,
"loss": 0.0007,
"step": 1450
},
{
"epoch": 28.01,
"learning_rate": 7.840000000000001e-06,
"loss": 0.0007,
"step": 1475
},
{
"epoch": 29.0,
"learning_rate": 7.784444444444445e-06,
"loss": 0.0007,
"step": 1500
},
{
"epoch": 29.01,
"learning_rate": 7.72888888888889e-06,
"loss": 0.0007,
"step": 1525
},
{
"epoch": 30.0,
"learning_rate": 7.673333333333333e-06,
"loss": 0.0006,
"step": 1550
},
{
"epoch": 30.01,
"learning_rate": 7.617777777777778e-06,
"loss": 0.0006,
"step": 1575
},
{
"epoch": 31.0,
"learning_rate": 7.562222222222223e-06,
"loss": 0.0006,
"step": 1600
},
{
"epoch": 31.01,
"learning_rate": 7.506666666666668e-06,
"loss": 0.0006,
"step": 1625
},
{
"epoch": 32.0,
"learning_rate": 7.451111111111111e-06,
"loss": 0.0006,
"step": 1650
},
{
"epoch": 32.01,
"learning_rate": 7.395555555555556e-06,
"loss": 0.0006,
"step": 1675
},
{
"epoch": 33.0,
"learning_rate": 7.340000000000001e-06,
"loss": 0.0006,
"step": 1700
},
{
"epoch": 33.01,
"learning_rate": 7.284444444444445e-06,
"loss": 0.0005,
"step": 1725
},
{
"epoch": 34.0,
"learning_rate": 7.22888888888889e-06,
"loss": 0.0005,
"step": 1750
},
{
"epoch": 34.01,
"learning_rate": 7.173333333333335e-06,
"loss": 0.0005,
"step": 1775
},
{
"epoch": 35.0,
"learning_rate": 7.117777777777778e-06,
"loss": 0.0005,
"step": 1800
},
{
"epoch": 35.01,
"learning_rate": 7.062222222222223e-06,
"loss": 0.0005,
"step": 1825
},
{
"epoch": 36.0,
"learning_rate": 7.006666666666667e-06,
"loss": 0.0005,
"step": 1850
},
{
"epoch": 36.01,
"learning_rate": 6.951111111111112e-06,
"loss": 0.0005,
"step": 1875
},
{
"epoch": 37.0,
"learning_rate": 6.8955555555555565e-06,
"loss": 0.0005,
"step": 1900
},
{
"epoch": 37.01,
"learning_rate": 6.8400000000000014e-06,
"loss": 0.0005,
"step": 1925
},
{
"epoch": 38.0,
"learning_rate": 6.784444444444445e-06,
"loss": 0.0005,
"step": 1950
},
{
"epoch": 38.01,
"learning_rate": 6.7288888888888895e-06,
"loss": 0.0005,
"step": 1975
},
{
"epoch": 39.0,
"learning_rate": 6.6733333333333335e-06,
"loss": 0.0005,
"step": 2000
},
{
"epoch": 39.0,
"eval_loss": 0.3185093402862549,
"eval_runtime": 943.3048,
"eval_samples_per_second": 1.002,
"eval_steps_per_second": 0.032,
"eval_wer": 15.632686526373327,
"step": 2000
},
{
"epoch": 39.01,
"learning_rate": 6.617777777777778e-06,
"loss": 0.0004,
"step": 2025
},
{
"epoch": 40.0,
"learning_rate": 6.562222222222223e-06,
"loss": 0.0004,
"step": 2050
},
{
"epoch": 40.01,
"learning_rate": 6.5066666666666665e-06,
"loss": 0.0004,
"step": 2075
},
{
"epoch": 41.0,
"learning_rate": 6.451111111111111e-06,
"loss": 0.0004,
"step": 2100
},
{
"epoch": 41.01,
"learning_rate": 6.395555555555556e-06,
"loss": 0.0004,
"step": 2125
},
{
"epoch": 42.0,
"learning_rate": 6.34e-06,
"loss": 0.0004,
"step": 2150
},
{
"epoch": 42.01,
"learning_rate": 6.284444444444445e-06,
"loss": 0.0004,
"step": 2175
},
{
"epoch": 43.0,
"learning_rate": 6.22888888888889e-06,
"loss": 0.0004,
"step": 2200
},
{
"epoch": 43.01,
"learning_rate": 6.173333333333333e-06,
"loss": 0.0003,
"step": 2225
},
{
"epoch": 44.0,
"learning_rate": 6.117777777777778e-06,
"loss": 0.0003,
"step": 2250
},
{
"epoch": 44.01,
"learning_rate": 6.062222222222223e-06,
"loss": 0.0003,
"step": 2275
},
{
"epoch": 45.0,
"learning_rate": 6.006666666666667e-06,
"loss": 0.0003,
"step": 2300
},
{
"epoch": 45.01,
"learning_rate": 5.951111111111112e-06,
"loss": 0.0003,
"step": 2325
},
{
"epoch": 46.0,
"learning_rate": 5.895555555555557e-06,
"loss": 0.0003,
"step": 2350
},
{
"epoch": 46.01,
"learning_rate": 5.84e-06,
"loss": 0.0003,
"step": 2375
},
{
"epoch": 47.0,
"learning_rate": 5.784444444444445e-06,
"loss": 0.0003,
"step": 2400
},
{
"epoch": 47.01,
"learning_rate": 5.72888888888889e-06,
"loss": 0.0003,
"step": 2425
},
{
"epoch": 48.0,
"learning_rate": 5.673333333333334e-06,
"loss": 0.0003,
"step": 2450
},
{
"epoch": 48.01,
"learning_rate": 5.617777777777779e-06,
"loss": 0.0003,
"step": 2475
},
{
"epoch": 49.0,
"learning_rate": 5.562222222222222e-06,
"loss": 0.0003,
"step": 2500
},
{
"epoch": 49.01,
"learning_rate": 5.506666666666667e-06,
"loss": 0.0003,
"step": 2525
},
{
"epoch": 49.01,
"learning_rate": 5.451111111111112e-06,
"loss": 0.0003,
"step": 2550
},
{
"epoch": 50.01,
"learning_rate": 5.3955555555555565e-06,
"loss": 0.0003,
"step": 2575
},
{
"epoch": 50.01,
"learning_rate": 5.3400000000000005e-06,
"loss": 0.0003,
"step": 2600
},
{
"epoch": 51.0,
"learning_rate": 5.2844444444444454e-06,
"loss": 0.0003,
"step": 2625
},
{
"epoch": 51.01,
"learning_rate": 5.228888888888889e-06,
"loss": 0.0003,
"step": 2650
},
{
"epoch": 52.0,
"learning_rate": 5.1733333333333335e-06,
"loss": 0.0003,
"step": 2675
},
{
"epoch": 52.01,
"learning_rate": 5.117777777777778e-06,
"loss": 0.0003,
"step": 2700
},
{
"epoch": 53.0,
"learning_rate": 5.062222222222222e-06,
"loss": 0.0003,
"step": 2725
},
{
"epoch": 53.01,
"learning_rate": 5.006666666666667e-06,
"loss": 0.0003,
"step": 2750
},
{
"epoch": 54.0,
"learning_rate": 4.951111111111111e-06,
"loss": 0.0003,
"step": 2775
},
{
"epoch": 54.01,
"learning_rate": 4.895555555555556e-06,
"loss": 0.0003,
"step": 2800
},
{
"epoch": 55.0,
"learning_rate": 4.84e-06,
"loss": 0.0002,
"step": 2825
},
{
"epoch": 55.01,
"learning_rate": 4.784444444444445e-06,
"loss": 0.0002,
"step": 2850
},
{
"epoch": 56.0,
"learning_rate": 4.728888888888889e-06,
"loss": 0.0002,
"step": 2875
},
{
"epoch": 56.01,
"learning_rate": 4.673333333333333e-06,
"loss": 0.0002,
"step": 2900
},
{
"epoch": 57.0,
"learning_rate": 4.617777777777778e-06,
"loss": 0.0002,
"step": 2925
},
{
"epoch": 57.01,
"learning_rate": 4.562222222222222e-06,
"loss": 0.0002,
"step": 2950
},
{
"epoch": 58.0,
"learning_rate": 4.506666666666667e-06,
"loss": 0.0002,
"step": 2975
},
{
"epoch": 58.01,
"learning_rate": 4.451111111111112e-06,
"loss": 0.0002,
"step": 3000
},
{
"epoch": 58.01,
"eval_loss": 0.3369024991989136,
"eval_runtime": 953.919,
"eval_samples_per_second": 0.991,
"eval_steps_per_second": 0.031,
"eval_wer": 15.696456226655734,
"step": 3000
},
{
"epoch": 59.0,
"learning_rate": 4.395555555555556e-06,
"loss": 0.0002,
"step": 3025
},
{
"epoch": 59.01,
"learning_rate": 4.34e-06,
"loss": 0.0002,
"step": 3050
},
{
"epoch": 60.0,
"learning_rate": 4.284444444444445e-06,
"loss": 0.0002,
"step": 3075
},
{
"epoch": 60.01,
"learning_rate": 4.228888888888889e-06,
"loss": 0.0002,
"step": 3100
},
{
"epoch": 61.0,
"learning_rate": 4.173333333333334e-06,
"loss": 0.0002,
"step": 3125
},
{
"epoch": 61.01,
"learning_rate": 4.117777777777779e-06,
"loss": 0.0002,
"step": 3150
},
{
"epoch": 62.0,
"learning_rate": 4.062222222222223e-06,
"loss": 0.0002,
"step": 3175
},
{
"epoch": 62.01,
"learning_rate": 4.006666666666667e-06,
"loss": 0.0002,
"step": 3200
},
{
"epoch": 63.0,
"learning_rate": 3.951111111111112e-06,
"loss": 0.0002,
"step": 3225
},
{
"epoch": 63.01,
"learning_rate": 3.895555555555556e-06,
"loss": 0.0002,
"step": 3250
},
{
"epoch": 64.0,
"learning_rate": 3.8400000000000005e-06,
"loss": 0.0002,
"step": 3275
},
{
"epoch": 64.01,
"learning_rate": 3.784444444444445e-06,
"loss": 0.0002,
"step": 3300
},
{
"epoch": 65.0,
"learning_rate": 3.728888888888889e-06,
"loss": 0.0002,
"step": 3325
},
{
"epoch": 65.01,
"learning_rate": 3.673333333333334e-06,
"loss": 0.0002,
"step": 3350
},
{
"epoch": 66.0,
"learning_rate": 3.617777777777778e-06,
"loss": 0.0002,
"step": 3375
},
{
"epoch": 66.01,
"learning_rate": 3.5622222222222224e-06,
"loss": 0.0002,
"step": 3400
},
{
"epoch": 67.0,
"learning_rate": 3.5066666666666673e-06,
"loss": 0.0002,
"step": 3425
},
{
"epoch": 67.01,
"learning_rate": 3.4511111111111113e-06,
"loss": 0.0002,
"step": 3450
},
{
"epoch": 68.0,
"learning_rate": 3.3955555555555558e-06,
"loss": 0.0002,
"step": 3475
},
{
"epoch": 68.01,
"learning_rate": 3.3400000000000006e-06,
"loss": 0.0002,
"step": 3500
},
{
"epoch": 69.0,
"learning_rate": 3.2844444444444447e-06,
"loss": 0.0002,
"step": 3525
},
{
"epoch": 69.01,
"learning_rate": 3.228888888888889e-06,
"loss": 0.0002,
"step": 3550
},
{
"epoch": 70.0,
"learning_rate": 3.173333333333334e-06,
"loss": 0.0002,
"step": 3575
},
{
"epoch": 70.01,
"learning_rate": 3.117777777777778e-06,
"loss": 0.0002,
"step": 3600
},
{
"epoch": 71.0,
"learning_rate": 3.0622222222222225e-06,
"loss": 0.0002,
"step": 3625
},
{
"epoch": 71.01,
"learning_rate": 3.0066666666666674e-06,
"loss": 0.0002,
"step": 3650
},
{
"epoch": 72.0,
"learning_rate": 2.9511111111111114e-06,
"loss": 0.0002,
"step": 3675
},
{
"epoch": 72.01,
"learning_rate": 2.895555555555556e-06,
"loss": 0.0002,
"step": 3700
},
{
"epoch": 73.0,
"learning_rate": 2.84e-06,
"loss": 0.0002,
"step": 3725
},
{
"epoch": 73.01,
"learning_rate": 2.784444444444445e-06,
"loss": 0.0002,
"step": 3750
},
{
"epoch": 74.0,
"learning_rate": 2.7288888888888893e-06,
"loss": 0.0002,
"step": 3775
},
{
"epoch": 74.01,
"learning_rate": 2.6733333333333333e-06,
"loss": 0.0002,
"step": 3800
},
{
"epoch": 74.01,
"learning_rate": 2.617777777777778e-06,
"loss": 0.0002,
"step": 3825
},
{
"epoch": 75.0,
"learning_rate": 2.5622222222222226e-06,
"loss": 0.0002,
"step": 3850
},
{
"epoch": 75.01,
"learning_rate": 2.5066666666666667e-06,
"loss": 0.0002,
"step": 3875
},
{
"epoch": 76.0,
"learning_rate": 2.451111111111111e-06,
"loss": 0.0002,
"step": 3900
},
{
"epoch": 76.01,
"learning_rate": 2.3955555555555556e-06,
"loss": 0.0002,
"step": 3925
},
{
"epoch": 77.0,
"learning_rate": 2.3400000000000005e-06,
"loss": 0.0002,
"step": 3950
},
{
"epoch": 77.01,
"learning_rate": 2.2844444444444445e-06,
"loss": 0.0002,
"step": 3975
},
{
"epoch": 78.0,
"learning_rate": 2.228888888888889e-06,
"loss": 0.0002,
"step": 4000
},
{
"epoch": 78.0,
"eval_loss": 0.34801387786865234,
"eval_runtime": 942.5876,
"eval_samples_per_second": 1.003,
"eval_steps_per_second": 0.032,
"eval_wer": 16.066927818772587,
"step": 4000
},
{
"epoch": 78.01,
"learning_rate": 2.1733333333333334e-06,
"loss": 0.0002,
"step": 4025
},
{
"epoch": 79.0,
"learning_rate": 2.117777777777778e-06,
"loss": 0.0002,
"step": 4050
},
{
"epoch": 79.01,
"learning_rate": 2.0622222222222223e-06,
"loss": 0.0002,
"step": 4075
},
{
"epoch": 80.0,
"learning_rate": 2.006666666666667e-06,
"loss": 0.0002,
"step": 4100
},
{
"epoch": 80.01,
"learning_rate": 1.9511111111111113e-06,
"loss": 0.0002,
"step": 4125
},
{
"epoch": 81.0,
"learning_rate": 1.8955555555555557e-06,
"loss": 0.0001,
"step": 4150
},
{
"epoch": 81.01,
"learning_rate": 1.8400000000000002e-06,
"loss": 0.0001,
"step": 4175
},
{
"epoch": 82.0,
"learning_rate": 1.7844444444444444e-06,
"loss": 0.0001,
"step": 4200
},
{
"epoch": 82.01,
"learning_rate": 1.728888888888889e-06,
"loss": 0.0001,
"step": 4225
},
{
"epoch": 83.0,
"learning_rate": 1.6733333333333335e-06,
"loss": 0.0001,
"step": 4250
},
{
"epoch": 83.01,
"learning_rate": 1.6177777777777778e-06,
"loss": 0.0001,
"step": 4275
},
{
"epoch": 84.0,
"learning_rate": 1.5622222222222225e-06,
"loss": 0.0001,
"step": 4300
},
{
"epoch": 84.01,
"learning_rate": 1.506666666666667e-06,
"loss": 0.0001,
"step": 4325
},
{
"epoch": 85.0,
"learning_rate": 1.4511111111111112e-06,
"loss": 0.0001,
"step": 4350
},
{
"epoch": 85.01,
"learning_rate": 1.3955555555555556e-06,
"loss": 0.0001,
"step": 4375
},
{
"epoch": 86.0,
"learning_rate": 1.34e-06,
"loss": 0.0001,
"step": 4400
},
{
"epoch": 86.01,
"learning_rate": 1.2844444444444445e-06,
"loss": 0.0001,
"step": 4425
},
{
"epoch": 87.0,
"learning_rate": 1.228888888888889e-06,
"loss": 0.0001,
"step": 4450
},
{
"epoch": 87.01,
"learning_rate": 1.1733333333333335e-06,
"loss": 0.0001,
"step": 4475
},
{
"epoch": 88.0,
"learning_rate": 1.117777777777778e-06,
"loss": 0.0001,
"step": 4500
},
{
"epoch": 88.01,
"learning_rate": 1.0622222222222222e-06,
"loss": 0.0001,
"step": 4525
},
{
"epoch": 89.0,
"learning_rate": 1.0066666666666668e-06,
"loss": 0.0001,
"step": 4550
},
{
"epoch": 89.01,
"learning_rate": 9.511111111111111e-07,
"loss": 0.0001,
"step": 4575
},
{
"epoch": 90.0,
"learning_rate": 8.955555555555557e-07,
"loss": 0.0001,
"step": 4600
},
{
"epoch": 90.01,
"learning_rate": 8.400000000000001e-07,
"loss": 0.0001,
"step": 4625
},
{
"epoch": 91.0,
"learning_rate": 7.844444444444445e-07,
"loss": 0.0001,
"step": 4650
},
{
"epoch": 91.01,
"learning_rate": 7.28888888888889e-07,
"loss": 0.0001,
"step": 4675
},
{
"epoch": 92.0,
"learning_rate": 6.733333333333334e-07,
"loss": 0.0001,
"step": 4700
},
{
"epoch": 92.01,
"learning_rate": 6.177777777777778e-07,
"loss": 0.0001,
"step": 4725
},
{
"epoch": 93.0,
"learning_rate": 5.622222222222223e-07,
"loss": 0.0001,
"step": 4750
},
{
"epoch": 93.01,
"learning_rate": 5.066666666666667e-07,
"loss": 0.0001,
"step": 4775
},
{
"epoch": 94.0,
"learning_rate": 4.511111111111111e-07,
"loss": 0.0001,
"step": 4800
},
{
"epoch": 94.01,
"learning_rate": 3.9555555555555557e-07,
"loss": 0.0001,
"step": 4825
},
{
"epoch": 95.0,
"learning_rate": 3.4000000000000003e-07,
"loss": 0.0001,
"step": 4850
},
{
"epoch": 95.01,
"learning_rate": 2.844444444444445e-07,
"loss": 0.0001,
"step": 4875
},
{
"epoch": 96.0,
"learning_rate": 2.2888888888888892e-07,
"loss": 0.0001,
"step": 4900
},
{
"epoch": 96.01,
"learning_rate": 1.7333333333333335e-07,
"loss": 0.0001,
"step": 4925
},
{
"epoch": 97.0,
"learning_rate": 1.1777777777777778e-07,
"loss": 0.0001,
"step": 4950
},
{
"epoch": 97.01,
"learning_rate": 6.222222222222223e-08,
"loss": 0.0001,
"step": 4975
},
{
"epoch": 98.0,
"learning_rate": 6.666666666666667e-09,
"loss": 0.0001,
"step": 5000
},
{
"epoch": 98.0,
"eval_loss": 0.3536686599254608,
"eval_runtime": 954.4933,
"eval_samples_per_second": 0.99,
"eval_steps_per_second": 0.031,
"eval_wer": 16.05478120919498,
"step": 5000
},
{
"epoch": 98.0,
"step": 5000,
"total_flos": 9.172513797931008e+19,
"train_loss": 0.019050344833452255,
"train_runtime": 38721.8075,
"train_samples_per_second": 8.264,
"train_steps_per_second": 0.129
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 9.172513797931008e+19,
"trial_name": null,
"trial_params": null
}