whisper-large-v2-cantonese / trainer_state.json
simonl0909's picture
End of training
fb4bcbb
raw
history blame
25.9 kB
{
"best_metric": 6.727383120825743,
"best_model_checkpoint": "./models/checkpoint-3000",
"epoch": 66.01,
"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": 0.9855,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.200000000000001e-07,
"loss": 0.387,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.42e-06,
"loss": 0.1816,
"step": 75
},
{
"epoch": 1.0,
"learning_rate": 1.9200000000000003e-06,
"loss": 0.162,
"step": 100
},
{
"epoch": 1.01,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.1352,
"step": 125
},
{
"epoch": 1.01,
"learning_rate": 2.9e-06,
"loss": 0.1054,
"step": 150
},
{
"epoch": 2.0,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.1065,
"step": 175
},
{
"epoch": 2.01,
"learning_rate": 3.900000000000001e-06,
"loss": 0.0835,
"step": 200
},
{
"epoch": 2.02,
"learning_rate": 4.4e-06,
"loss": 0.0658,
"step": 225
},
{
"epoch": 3.0,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.0628,
"step": 250
},
{
"epoch": 3.01,
"learning_rate": 5.400000000000001e-06,
"loss": 0.0514,
"step": 275
},
{
"epoch": 3.02,
"learning_rate": 5.9e-06,
"loss": 0.0414,
"step": 300
},
{
"epoch": 4.0,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.0377,
"step": 325
},
{
"epoch": 4.01,
"learning_rate": 6.9e-06,
"loss": 0.0323,
"step": 350
},
{
"epoch": 4.01,
"learning_rate": 7.4e-06,
"loss": 0.0285,
"step": 375
},
{
"epoch": 5.0,
"learning_rate": 7.9e-06,
"loss": 0.0282,
"step": 400
},
{
"epoch": 5.01,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0221,
"step": 425
},
{
"epoch": 5.01,
"learning_rate": 8.900000000000001e-06,
"loss": 0.0229,
"step": 450
},
{
"epoch": 6.0,
"learning_rate": 9.4e-06,
"loss": 0.0242,
"step": 475
},
{
"epoch": 6.01,
"learning_rate": 9.9e-06,
"loss": 0.0201,
"step": 500
},
{
"epoch": 6.01,
"learning_rate": 9.955555555555556e-06,
"loss": 0.0175,
"step": 525
},
{
"epoch": 7.0,
"learning_rate": 9.9e-06,
"loss": 0.021,
"step": 550
},
{
"epoch": 7.01,
"learning_rate": 9.844444444444446e-06,
"loss": 0.0188,
"step": 575
},
{
"epoch": 7.01,
"learning_rate": 9.78888888888889e-06,
"loss": 0.0145,
"step": 600
},
{
"epoch": 8.01,
"learning_rate": 9.733333333333334e-06,
"loss": 0.0153,
"step": 625
},
{
"epoch": 8.01,
"learning_rate": 9.677777777777778e-06,
"loss": 0.0131,
"step": 650
},
{
"epoch": 8.02,
"learning_rate": 9.622222222222222e-06,
"loss": 0.0117,
"step": 675
},
{
"epoch": 9.01,
"learning_rate": 9.566666666666668e-06,
"loss": 0.0098,
"step": 700
},
{
"epoch": 9.01,
"learning_rate": 9.511111111111112e-06,
"loss": 0.009,
"step": 725
},
{
"epoch": 9.02,
"learning_rate": 9.455555555555557e-06,
"loss": 0.008,
"step": 750
},
{
"epoch": 10.01,
"learning_rate": 9.4e-06,
"loss": 0.0089,
"step": 775
},
{
"epoch": 10.01,
"learning_rate": 9.344444444444446e-06,
"loss": 0.0081,
"step": 800
},
{
"epoch": 10.02,
"learning_rate": 9.28888888888889e-06,
"loss": 0.0056,
"step": 825
},
{
"epoch": 11.01,
"learning_rate": 9.233333333333334e-06,
"loss": 0.0076,
"step": 850
},
{
"epoch": 11.01,
"learning_rate": 9.17777777777778e-06,
"loss": 0.0072,
"step": 875
},
{
"epoch": 11.02,
"learning_rate": 9.122222222222223e-06,
"loss": 0.0034,
"step": 900
},
{
"epoch": 12.01,
"learning_rate": 9.066666666666667e-06,
"loss": 0.0026,
"step": 925
},
{
"epoch": 12.01,
"learning_rate": 9.011111111111111e-06,
"loss": 0.0023,
"step": 950
},
{
"epoch": 12.02,
"learning_rate": 8.955555555555555e-06,
"loss": 0.0026,
"step": 975
},
{
"epoch": 13.01,
"learning_rate": 8.900000000000001e-06,
"loss": 0.0032,
"step": 1000
},
{
"epoch": 13.01,
"eval_cer": 6.856911556365109,
"eval_loss": 0.23175829648971558,
"eval_runtime": 2336.026,
"eval_samples_per_second": 1.044,
"eval_steps_per_second": 0.261,
"step": 1000
},
{
"epoch": 13.01,
"learning_rate": 8.844444444444445e-06,
"loss": 0.0028,
"step": 1025
},
{
"epoch": 13.02,
"learning_rate": 8.788888888888891e-06,
"loss": 0.0035,
"step": 1050
},
{
"epoch": 14.01,
"learning_rate": 8.733333333333333e-06,
"loss": 0.0026,
"step": 1075
},
{
"epoch": 14.01,
"learning_rate": 8.677777777777779e-06,
"loss": 0.0018,
"step": 1100
},
{
"epoch": 14.02,
"learning_rate": 8.622222222222223e-06,
"loss": 0.001,
"step": 1125
},
{
"epoch": 15.01,
"learning_rate": 8.566666666666667e-06,
"loss": 0.0013,
"step": 1150
},
{
"epoch": 15.01,
"learning_rate": 8.511111111111113e-06,
"loss": 0.0017,
"step": 1175
},
{
"epoch": 15.02,
"learning_rate": 8.455555555555555e-06,
"loss": 0.0015,
"step": 1200
},
{
"epoch": 16.0,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0015,
"step": 1225
},
{
"epoch": 16.01,
"learning_rate": 8.344444444444445e-06,
"loss": 0.0007,
"step": 1250
},
{
"epoch": 16.02,
"learning_rate": 8.288888888888889e-06,
"loss": 0.0008,
"step": 1275
},
{
"epoch": 17.0,
"learning_rate": 8.233333333333335e-06,
"loss": 0.0007,
"step": 1300
},
{
"epoch": 17.01,
"learning_rate": 8.177777777777779e-06,
"loss": 0.0004,
"step": 1325
},
{
"epoch": 17.02,
"learning_rate": 8.122222222222223e-06,
"loss": 0.0006,
"step": 1350
},
{
"epoch": 18.0,
"learning_rate": 8.066666666666667e-06,
"loss": 0.0013,
"step": 1375
},
{
"epoch": 18.01,
"learning_rate": 8.011111111111113e-06,
"loss": 0.0006,
"step": 1400
},
{
"epoch": 18.02,
"learning_rate": 7.955555555555557e-06,
"loss": 0.0013,
"step": 1425
},
{
"epoch": 19.0,
"learning_rate": 7.9e-06,
"loss": 0.0017,
"step": 1450
},
{
"epoch": 19.01,
"learning_rate": 7.844444444444446e-06,
"loss": 0.0011,
"step": 1475
},
{
"epoch": 19.02,
"learning_rate": 7.788888888888889e-06,
"loss": 0.0006,
"step": 1500
},
{
"epoch": 20.0,
"learning_rate": 7.733333333333334e-06,
"loss": 0.0018,
"step": 1525
},
{
"epoch": 20.01,
"learning_rate": 7.677777777777778e-06,
"loss": 0.0008,
"step": 1550
},
{
"epoch": 20.02,
"learning_rate": 7.622222222222223e-06,
"loss": 0.0011,
"step": 1575
},
{
"epoch": 21.0,
"learning_rate": 7.566666666666667e-06,
"loss": 0.0009,
"step": 1600
},
{
"epoch": 21.01,
"learning_rate": 7.511111111111111e-06,
"loss": 0.0012,
"step": 1625
},
{
"epoch": 21.02,
"learning_rate": 7.455555555555556e-06,
"loss": 0.0015,
"step": 1650
},
{
"epoch": 22.0,
"learning_rate": 7.4e-06,
"loss": 0.0013,
"step": 1675
},
{
"epoch": 22.01,
"learning_rate": 7.344444444444445e-06,
"loss": 0.0017,
"step": 1700
},
{
"epoch": 22.02,
"learning_rate": 7.28888888888889e-06,
"loss": 0.0018,
"step": 1725
},
{
"epoch": 23.0,
"learning_rate": 7.233333333333334e-06,
"loss": 0.0022,
"step": 1750
},
{
"epoch": 23.01,
"learning_rate": 7.177777777777778e-06,
"loss": 0.0017,
"step": 1775
},
{
"epoch": 23.02,
"learning_rate": 7.122222222222222e-06,
"loss": 0.0019,
"step": 1800
},
{
"epoch": 24.0,
"learning_rate": 7.066666666666667e-06,
"loss": 0.0052,
"step": 1825
},
{
"epoch": 24.01,
"learning_rate": 7.011111111111112e-06,
"loss": 0.0059,
"step": 1850
},
{
"epoch": 24.02,
"learning_rate": 6.955555555555557e-06,
"loss": 0.0041,
"step": 1875
},
{
"epoch": 25.0,
"learning_rate": 6.9e-06,
"loss": 0.004,
"step": 1900
},
{
"epoch": 25.01,
"learning_rate": 6.844444444444445e-06,
"loss": 0.0026,
"step": 1925
},
{
"epoch": 25.02,
"learning_rate": 6.788888888888889e-06,
"loss": 0.003,
"step": 1950
},
{
"epoch": 26.0,
"learning_rate": 6.733333333333334e-06,
"loss": 0.0037,
"step": 1975
},
{
"epoch": 26.01,
"learning_rate": 6.677777777777779e-06,
"loss": 0.002,
"step": 2000
},
{
"epoch": 26.01,
"eval_cer": 7.152398299939283,
"eval_loss": 0.2404206544160843,
"eval_runtime": 2331.1686,
"eval_samples_per_second": 1.046,
"eval_steps_per_second": 0.262,
"step": 2000
},
{
"epoch": 26.02,
"learning_rate": 6.6222222222222236e-06,
"loss": 0.0014,
"step": 2025
},
{
"epoch": 27.0,
"learning_rate": 6.566666666666667e-06,
"loss": 0.0024,
"step": 2050
},
{
"epoch": 27.01,
"learning_rate": 6.511111111111112e-06,
"loss": 0.001,
"step": 2075
},
{
"epoch": 27.02,
"learning_rate": 6.455555555555556e-06,
"loss": 0.0009,
"step": 2100
},
{
"epoch": 28.0,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.0005,
"step": 2125
},
{
"epoch": 28.01,
"learning_rate": 6.3444444444444454e-06,
"loss": 0.0003,
"step": 2150
},
{
"epoch": 28.02,
"learning_rate": 6.28888888888889e-06,
"loss": 0.0005,
"step": 2175
},
{
"epoch": 29.0,
"learning_rate": 6.2333333333333335e-06,
"loss": 0.0002,
"step": 2200
},
{
"epoch": 29.01,
"learning_rate": 6.177777777777778e-06,
"loss": 0.0002,
"step": 2225
},
{
"epoch": 29.02,
"learning_rate": 6.1222222222222224e-06,
"loss": 0.0003,
"step": 2250
},
{
"epoch": 30.0,
"learning_rate": 6.066666666666667e-06,
"loss": 0.0002,
"step": 2275
},
{
"epoch": 30.01,
"learning_rate": 6.011111111111112e-06,
"loss": 0.0002,
"step": 2300
},
{
"epoch": 30.02,
"learning_rate": 5.955555555555555e-06,
"loss": 0.0002,
"step": 2325
},
{
"epoch": 31.0,
"learning_rate": 5.9e-06,
"loss": 0.0002,
"step": 2350
},
{
"epoch": 31.01,
"learning_rate": 5.844444444444445e-06,
"loss": 0.0001,
"step": 2375
},
{
"epoch": 31.02,
"learning_rate": 5.788888888888889e-06,
"loss": 0.0001,
"step": 2400
},
{
"epoch": 32.01,
"learning_rate": 5.733333333333334e-06,
"loss": 0.0001,
"step": 2425
},
{
"epoch": 32.01,
"learning_rate": 5.677777777777779e-06,
"loss": 0.0001,
"step": 2450
},
{
"epoch": 32.02,
"learning_rate": 5.622222222222222e-06,
"loss": 0.0001,
"step": 2475
},
{
"epoch": 33.01,
"learning_rate": 5.566666666666667e-06,
"loss": 0.0001,
"step": 2500
},
{
"epoch": 33.01,
"learning_rate": 5.511111111111112e-06,
"loss": 0.0001,
"step": 2525
},
{
"epoch": 33.02,
"learning_rate": 5.455555555555556e-06,
"loss": 0.0001,
"step": 2550
},
{
"epoch": 34.01,
"learning_rate": 5.400000000000001e-06,
"loss": 0.0001,
"step": 2575
},
{
"epoch": 34.01,
"learning_rate": 5.344444444444446e-06,
"loss": 0.0001,
"step": 2600
},
{
"epoch": 34.02,
"learning_rate": 5.288888888888889e-06,
"loss": 0.0001,
"step": 2625
},
{
"epoch": 35.01,
"learning_rate": 5.233333333333334e-06,
"loss": 0.0001,
"step": 2650
},
{
"epoch": 35.01,
"learning_rate": 5.177777777777779e-06,
"loss": 0.0001,
"step": 2675
},
{
"epoch": 35.02,
"learning_rate": 5.122222222222223e-06,
"loss": 0.0001,
"step": 2700
},
{
"epoch": 36.01,
"learning_rate": 5.0666666666666676e-06,
"loss": 0.0001,
"step": 2725
},
{
"epoch": 36.01,
"learning_rate": 5.011111111111111e-06,
"loss": 0.0001,
"step": 2750
},
{
"epoch": 36.02,
"learning_rate": 4.9555555555555565e-06,
"loss": 0.0001,
"step": 2775
},
{
"epoch": 37.01,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.0001,
"step": 2800
},
{
"epoch": 37.01,
"learning_rate": 4.8444444444444446e-06,
"loss": 0.0001,
"step": 2825
},
{
"epoch": 37.02,
"learning_rate": 4.7888888888888894e-06,
"loss": 0.0001,
"step": 2850
},
{
"epoch": 38.01,
"learning_rate": 4.7333333333333335e-06,
"loss": 0.0001,
"step": 2875
},
{
"epoch": 38.01,
"learning_rate": 4.677777777777778e-06,
"loss": 0.0001,
"step": 2900
},
{
"epoch": 38.02,
"learning_rate": 4.622222222222222e-06,
"loss": 0.0001,
"step": 2925
},
{
"epoch": 39.01,
"learning_rate": 4.566666666666667e-06,
"loss": 0.0001,
"step": 2950
},
{
"epoch": 39.01,
"learning_rate": 4.511111111111111e-06,
"loss": 0.0001,
"step": 2975
},
{
"epoch": 39.02,
"learning_rate": 4.455555555555555e-06,
"loss": 0.0001,
"step": 3000
},
{
"epoch": 39.02,
"eval_cer": 6.727383120825743,
"eval_loss": 0.280737042427063,
"eval_runtime": 2341.0522,
"eval_samples_per_second": 1.041,
"eval_steps_per_second": 0.261,
"step": 3000
},
{
"epoch": 40.01,
"learning_rate": 4.4e-06,
"loss": 0.0001,
"step": 3025
},
{
"epoch": 40.01,
"learning_rate": 4.344444444444445e-06,
"loss": 0.0001,
"step": 3050
},
{
"epoch": 40.02,
"learning_rate": 4.288888888888889e-06,
"loss": 0.0001,
"step": 3075
},
{
"epoch": 41.01,
"learning_rate": 4.233333333333334e-06,
"loss": 0.0001,
"step": 3100
},
{
"epoch": 41.01,
"learning_rate": 4.177777777777778e-06,
"loss": 0.0001,
"step": 3125
},
{
"epoch": 41.02,
"learning_rate": 4.122222222222222e-06,
"loss": 0.0001,
"step": 3150
},
{
"epoch": 42.01,
"learning_rate": 4.066666666666667e-06,
"loss": 0.0001,
"step": 3175
},
{
"epoch": 42.01,
"learning_rate": 4.011111111111111e-06,
"loss": 0.0001,
"step": 3200
},
{
"epoch": 42.02,
"learning_rate": 3.955555555555556e-06,
"loss": 0.0001,
"step": 3225
},
{
"epoch": 43.01,
"learning_rate": 3.900000000000001e-06,
"loss": 0.0001,
"step": 3250
},
{
"epoch": 43.01,
"learning_rate": 3.844444444444445e-06,
"loss": 0.0001,
"step": 3275
},
{
"epoch": 43.02,
"learning_rate": 3.7888888888888893e-06,
"loss": 0.0001,
"step": 3300
},
{
"epoch": 44.01,
"learning_rate": 3.7333333333333337e-06,
"loss": 0.0001,
"step": 3325
},
{
"epoch": 44.01,
"learning_rate": 3.6777777777777778e-06,
"loss": 0.0001,
"step": 3350
},
{
"epoch": 44.02,
"learning_rate": 3.6222222222222226e-06,
"loss": 0.0001,
"step": 3375
},
{
"epoch": 45.01,
"learning_rate": 3.566666666666667e-06,
"loss": 0.0001,
"step": 3400
},
{
"epoch": 45.01,
"learning_rate": 3.511111111111111e-06,
"loss": 0.0001,
"step": 3425
},
{
"epoch": 45.02,
"learning_rate": 3.455555555555556e-06,
"loss": 0.0001,
"step": 3450
},
{
"epoch": 46.01,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.0001,
"step": 3475
},
{
"epoch": 46.01,
"learning_rate": 3.3444444444444445e-06,
"loss": 0.0001,
"step": 3500
},
{
"epoch": 46.02,
"learning_rate": 3.2888888888888894e-06,
"loss": 0.0001,
"step": 3525
},
{
"epoch": 47.01,
"learning_rate": 3.2333333333333334e-06,
"loss": 0.0001,
"step": 3550
},
{
"epoch": 47.01,
"learning_rate": 3.177777777777778e-06,
"loss": 0.0001,
"step": 3575
},
{
"epoch": 47.02,
"learning_rate": 3.1222222222222228e-06,
"loss": 0.0001,
"step": 3600
},
{
"epoch": 48.01,
"learning_rate": 3.066666666666667e-06,
"loss": 0.0001,
"step": 3625
},
{
"epoch": 48.01,
"learning_rate": 3.0111111111111113e-06,
"loss": 0.0001,
"step": 3650
},
{
"epoch": 48.02,
"learning_rate": 2.955555555555556e-06,
"loss": 0.0001,
"step": 3675
},
{
"epoch": 49.01,
"learning_rate": 2.9e-06,
"loss": 0.0001,
"step": 3700
},
{
"epoch": 49.01,
"learning_rate": 2.8444444444444446e-06,
"loss": 0.0001,
"step": 3725
},
{
"epoch": 49.02,
"learning_rate": 2.788888888888889e-06,
"loss": 0.0001,
"step": 3750
},
{
"epoch": 50.01,
"learning_rate": 2.7333333333333336e-06,
"loss": 0.0001,
"step": 3775
},
{
"epoch": 50.01,
"learning_rate": 2.677777777777778e-06,
"loss": 0.0001,
"step": 3800
},
{
"epoch": 50.02,
"learning_rate": 2.6222222222222225e-06,
"loss": 0.0001,
"step": 3825
},
{
"epoch": 51.01,
"learning_rate": 2.566666666666667e-06,
"loss": 0.0001,
"step": 3850
},
{
"epoch": 51.01,
"learning_rate": 2.5111111111111114e-06,
"loss": 0.0001,
"step": 3875
},
{
"epoch": 51.02,
"learning_rate": 2.455555555555556e-06,
"loss": 0.0001,
"step": 3900
},
{
"epoch": 52.01,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.0001,
"step": 3925
},
{
"epoch": 52.01,
"learning_rate": 2.3444444444444448e-06,
"loss": 0.0001,
"step": 3950
},
{
"epoch": 52.02,
"learning_rate": 2.2888888888888892e-06,
"loss": 0.0001,
"step": 3975
},
{
"epoch": 53.01,
"learning_rate": 2.2333333333333333e-06,
"loss": 0.0001,
"step": 4000
},
{
"epoch": 53.01,
"eval_cer": 6.751669702489374,
"eval_loss": 0.2911806106567383,
"eval_runtime": 2342.7634,
"eval_samples_per_second": 1.041,
"eval_steps_per_second": 0.26,
"step": 4000
},
{
"epoch": 53.01,
"learning_rate": 2.1777777777777777e-06,
"loss": 0.0001,
"step": 4025
},
{
"epoch": 53.02,
"learning_rate": 2.1222222222222226e-06,
"loss": 0.0001,
"step": 4050
},
{
"epoch": 54.01,
"learning_rate": 2.0666666666666666e-06,
"loss": 0.0001,
"step": 4075
},
{
"epoch": 54.01,
"learning_rate": 2.011111111111111e-06,
"loss": 0.0001,
"step": 4100
},
{
"epoch": 54.02,
"learning_rate": 1.955555555555556e-06,
"loss": 0.0001,
"step": 4125
},
{
"epoch": 55.01,
"learning_rate": 1.9000000000000002e-06,
"loss": 0.0001,
"step": 4150
},
{
"epoch": 55.01,
"learning_rate": 1.8444444444444445e-06,
"loss": 0.0001,
"step": 4175
},
{
"epoch": 55.02,
"learning_rate": 1.788888888888889e-06,
"loss": 0.0001,
"step": 4200
},
{
"epoch": 56.01,
"learning_rate": 1.7333333333333336e-06,
"loss": 0.0001,
"step": 4225
},
{
"epoch": 56.01,
"learning_rate": 1.6777777777777779e-06,
"loss": 0.0001,
"step": 4250
},
{
"epoch": 56.02,
"learning_rate": 1.6222222222222223e-06,
"loss": 0.0001,
"step": 4275
},
{
"epoch": 57.01,
"learning_rate": 1.566666666666667e-06,
"loss": 0.0001,
"step": 4300
},
{
"epoch": 57.01,
"learning_rate": 1.5111111111111112e-06,
"loss": 0.0001,
"step": 4325
},
{
"epoch": 57.02,
"learning_rate": 1.4555555555555557e-06,
"loss": 0.0001,
"step": 4350
},
{
"epoch": 58.01,
"learning_rate": 1.4000000000000001e-06,
"loss": 0.0001,
"step": 4375
},
{
"epoch": 58.01,
"learning_rate": 1.3444444444444446e-06,
"loss": 0.0001,
"step": 4400
},
{
"epoch": 58.02,
"learning_rate": 1.288888888888889e-06,
"loss": 0.0001,
"step": 4425
},
{
"epoch": 59.01,
"learning_rate": 1.2333333333333335e-06,
"loss": 0.0001,
"step": 4450
},
{
"epoch": 59.01,
"learning_rate": 1.1777777777777778e-06,
"loss": 0.0001,
"step": 4475
},
{
"epoch": 59.02,
"learning_rate": 1.1222222222222222e-06,
"loss": 0.0001,
"step": 4500
},
{
"epoch": 60.01,
"learning_rate": 1.066666666666667e-06,
"loss": 0.0001,
"step": 4525
},
{
"epoch": 60.01,
"learning_rate": 1.0111111111111111e-06,
"loss": 0.0001,
"step": 4550
},
{
"epoch": 60.02,
"learning_rate": 9.555555555555556e-07,
"loss": 0.0001,
"step": 4575
},
{
"epoch": 61.01,
"learning_rate": 9.000000000000001e-07,
"loss": 0.0001,
"step": 4600
},
{
"epoch": 61.01,
"learning_rate": 8.444444444444445e-07,
"loss": 0.0001,
"step": 4625
},
{
"epoch": 61.02,
"learning_rate": 7.888888888888889e-07,
"loss": 0.0001,
"step": 4650
},
{
"epoch": 62.01,
"learning_rate": 7.333333333333334e-07,
"loss": 0.0001,
"step": 4675
},
{
"epoch": 62.01,
"learning_rate": 6.777777777777779e-07,
"loss": 0.0001,
"step": 4700
},
{
"epoch": 62.02,
"learning_rate": 6.222222222222223e-07,
"loss": 0.0001,
"step": 4725
},
{
"epoch": 63.01,
"learning_rate": 5.666666666666667e-07,
"loss": 0.0001,
"step": 4750
},
{
"epoch": 63.01,
"learning_rate": 5.111111111111112e-07,
"loss": 0.0001,
"step": 4775
},
{
"epoch": 63.02,
"learning_rate": 4.5555555555555563e-07,
"loss": 0.0001,
"step": 4800
},
{
"epoch": 64.0,
"learning_rate": 4.0000000000000003e-07,
"loss": 0.0001,
"step": 4825
},
{
"epoch": 64.01,
"learning_rate": 3.444444444444445e-07,
"loss": 0.0,
"step": 4850
},
{
"epoch": 64.02,
"learning_rate": 2.888888888888889e-07,
"loss": 0.0001,
"step": 4875
},
{
"epoch": 65.0,
"learning_rate": 2.3333333333333336e-07,
"loss": 0.0001,
"step": 4900
},
{
"epoch": 65.01,
"learning_rate": 1.777777777777778e-07,
"loss": 0.0001,
"step": 4925
},
{
"epoch": 65.02,
"learning_rate": 1.2222222222222225e-07,
"loss": 0.0001,
"step": 4950
},
{
"epoch": 66.0,
"learning_rate": 6.666666666666668e-08,
"loss": 0.0001,
"step": 4975
},
{
"epoch": 66.01,
"learning_rate": 1.1111111111111112e-08,
"loss": 0.0,
"step": 5000
},
{
"epoch": 66.01,
"eval_cer": 6.76381299332119,
"eval_loss": 0.2956882417201996,
"eval_runtime": 2363.6982,
"eval_samples_per_second": 1.031,
"eval_steps_per_second": 0.258,
"step": 5000
},
{
"epoch": 66.01,
"step": 5000,
"total_flos": 6.847396395761664e+20,
"train_loss": 0.01433451753302943,
"train_runtime": 108259.9519,
"train_samples_per_second": 2.956,
"train_steps_per_second": 0.046
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 6.847396395761664e+20,
"trial_name": null,
"trial_params": null
}