whisper-base-eu / trainer_state.json
zuazo's picture
End of training
06e1ba2
{
"best_metric": 25.977155818380655,
"best_model_checkpoint": "./checkpoint-3000",
"epoch": 48.0112,
"eval_steps": 1000,
"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": 1.15e-06,
"loss": 2.1835,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 2.4000000000000003e-06,
"loss": 1.8246,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 3.6499999999999998e-06,
"loss": 1.2675,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.9835,
"step": 100
},
{
"epoch": 1.0,
"learning_rate": 6.15e-06,
"loss": 0.8579,
"step": 125
},
{
"epoch": 1.01,
"learning_rate": 7.4e-06,
"loss": 0.7651,
"step": 150
},
{
"epoch": 1.01,
"learning_rate": 8.65e-06,
"loss": 0.6439,
"step": 175
},
{
"epoch": 1.02,
"learning_rate": 9.900000000000002e-06,
"loss": 0.5563,
"step": 200
},
{
"epoch": 2.0,
"learning_rate": 1.115e-05,
"loss": 0.5319,
"step": 225
},
{
"epoch": 2.01,
"learning_rate": 1.24e-05,
"loss": 0.4959,
"step": 250
},
{
"epoch": 2.01,
"learning_rate": 1.3650000000000001e-05,
"loss": 0.4392,
"step": 275
},
{
"epoch": 2.02,
"learning_rate": 1.49e-05,
"loss": 0.3774,
"step": 300
},
{
"epoch": 3.0,
"learning_rate": 1.6150000000000003e-05,
"loss": 0.3635,
"step": 325
},
{
"epoch": 3.01,
"learning_rate": 1.74e-05,
"loss": 0.3477,
"step": 350
},
{
"epoch": 3.01,
"learning_rate": 1.865e-05,
"loss": 0.3091,
"step": 375
},
{
"epoch": 3.02,
"learning_rate": 1.9900000000000003e-05,
"loss": 0.2645,
"step": 400
},
{
"epoch": 4.0,
"learning_rate": 2.115e-05,
"loss": 0.2522,
"step": 425
},
{
"epoch": 4.01,
"learning_rate": 2.2400000000000002e-05,
"loss": 0.2396,
"step": 450
},
{
"epoch": 4.01,
"learning_rate": 2.365e-05,
"loss": 0.2102,
"step": 475
},
{
"epoch": 4.02,
"learning_rate": 2.4900000000000002e-05,
"loss": 0.1786,
"step": 500
},
{
"epoch": 5.0,
"learning_rate": 2.4872222222222223e-05,
"loss": 0.169,
"step": 525
},
{
"epoch": 5.01,
"learning_rate": 2.4733333333333333e-05,
"loss": 0.1567,
"step": 550
},
{
"epoch": 5.01,
"learning_rate": 2.4594444444444445e-05,
"loss": 0.1345,
"step": 575
},
{
"epoch": 5.02,
"learning_rate": 2.4455555555555558e-05,
"loss": 0.11,
"step": 600
},
{
"epoch": 6.0,
"learning_rate": 2.4316666666666667e-05,
"loss": 0.0998,
"step": 625
},
{
"epoch": 6.01,
"learning_rate": 2.417777777777778e-05,
"loss": 0.0947,
"step": 650
},
{
"epoch": 6.01,
"learning_rate": 2.4038888888888892e-05,
"loss": 0.0833,
"step": 675
},
{
"epoch": 6.02,
"learning_rate": 2.39e-05,
"loss": 0.0664,
"step": 700
},
{
"epoch": 7.0,
"learning_rate": 2.376111111111111e-05,
"loss": 0.0583,
"step": 725
},
{
"epoch": 7.01,
"learning_rate": 2.3622222222222223e-05,
"loss": 0.0543,
"step": 750
},
{
"epoch": 7.01,
"learning_rate": 2.3483333333333336e-05,
"loss": 0.0505,
"step": 775
},
{
"epoch": 7.02,
"learning_rate": 2.3344444444444445e-05,
"loss": 0.0403,
"step": 800
},
{
"epoch": 8.0,
"learning_rate": 2.3205555555555555e-05,
"loss": 0.0335,
"step": 825
},
{
"epoch": 8.01,
"learning_rate": 2.3066666666666667e-05,
"loss": 0.034,
"step": 850
},
{
"epoch": 8.01,
"learning_rate": 2.292777777777778e-05,
"loss": 0.0305,
"step": 875
},
{
"epoch": 8.02,
"learning_rate": 2.278888888888889e-05,
"loss": 0.025,
"step": 900
},
{
"epoch": 8.02,
"learning_rate": 2.265e-05,
"loss": 0.0208,
"step": 925
},
{
"epoch": 9.0,
"learning_rate": 2.2511111111111114e-05,
"loss": 0.0233,
"step": 950
},
{
"epoch": 9.01,
"learning_rate": 2.2372222222222224e-05,
"loss": 0.0207,
"step": 975
},
{
"epoch": 9.01,
"learning_rate": 2.2233333333333333e-05,
"loss": 0.0174,
"step": 1000
},
{
"epoch": 9.01,
"eval_loss": 0.4596753418445587,
"eval_runtime": 680.6109,
"eval_samples_per_second": 9.684,
"eval_steps_per_second": 0.151,
"eval_wer": 27.30973307950909,
"step": 1000
},
{
"epoch": 9.02,
"learning_rate": 2.2094444444444445e-05,
"loss": 0.0143,
"step": 1025
},
{
"epoch": 10.0,
"learning_rate": 2.1955555555555558e-05,
"loss": 0.0147,
"step": 1050
},
{
"epoch": 10.01,
"learning_rate": 2.1816666666666667e-05,
"loss": 0.0131,
"step": 1075
},
{
"epoch": 10.01,
"learning_rate": 2.167777777777778e-05,
"loss": 0.0118,
"step": 1100
},
{
"epoch": 10.02,
"learning_rate": 2.153888888888889e-05,
"loss": 0.0101,
"step": 1125
},
{
"epoch": 11.0,
"learning_rate": 2.1400000000000002e-05,
"loss": 0.01,
"step": 1150
},
{
"epoch": 11.01,
"learning_rate": 2.126111111111111e-05,
"loss": 0.0093,
"step": 1175
},
{
"epoch": 11.01,
"learning_rate": 2.1122222222222224e-05,
"loss": 0.0082,
"step": 1200
},
{
"epoch": 11.02,
"learning_rate": 2.0983333333333336e-05,
"loss": 0.0074,
"step": 1225
},
{
"epoch": 12.0,
"learning_rate": 2.0844444444444446e-05,
"loss": 0.0078,
"step": 1250
},
{
"epoch": 12.01,
"learning_rate": 2.0705555555555555e-05,
"loss": 0.0071,
"step": 1275
},
{
"epoch": 12.01,
"learning_rate": 2.0566666666666667e-05,
"loss": 0.006,
"step": 1300
},
{
"epoch": 12.02,
"learning_rate": 2.042777777777778e-05,
"loss": 0.0056,
"step": 1325
},
{
"epoch": 13.0,
"learning_rate": 2.028888888888889e-05,
"loss": 0.0057,
"step": 1350
},
{
"epoch": 13.01,
"learning_rate": 2.0150000000000002e-05,
"loss": 0.0052,
"step": 1375
},
{
"epoch": 13.01,
"learning_rate": 2.001111111111111e-05,
"loss": 0.0046,
"step": 1400
},
{
"epoch": 13.02,
"learning_rate": 1.9872222222222224e-05,
"loss": 0.0041,
"step": 1425
},
{
"epoch": 14.0,
"learning_rate": 1.9733333333333333e-05,
"loss": 0.004,
"step": 1450
},
{
"epoch": 14.01,
"learning_rate": 1.9594444444444446e-05,
"loss": 0.0037,
"step": 1475
},
{
"epoch": 14.01,
"learning_rate": 1.9455555555555558e-05,
"loss": 0.0034,
"step": 1500
},
{
"epoch": 14.02,
"learning_rate": 1.9316666666666668e-05,
"loss": 0.003,
"step": 1525
},
{
"epoch": 15.0,
"learning_rate": 1.9177777777777777e-05,
"loss": 0.0029,
"step": 1550
},
{
"epoch": 15.01,
"learning_rate": 1.903888888888889e-05,
"loss": 0.0028,
"step": 1575
},
{
"epoch": 15.01,
"learning_rate": 1.8900000000000002e-05,
"loss": 0.0027,
"step": 1600
},
{
"epoch": 15.02,
"learning_rate": 1.876111111111111e-05,
"loss": 0.0024,
"step": 1625
},
{
"epoch": 16.0,
"learning_rate": 1.8622222222222224e-05,
"loss": 0.0023,
"step": 1650
},
{
"epoch": 16.01,
"learning_rate": 1.8483333333333333e-05,
"loss": 0.0023,
"step": 1675
},
{
"epoch": 16.01,
"learning_rate": 1.8344444444444446e-05,
"loss": 0.0022,
"step": 1700
},
{
"epoch": 16.02,
"learning_rate": 1.8205555555555555e-05,
"loss": 0.0021,
"step": 1725
},
{
"epoch": 16.02,
"learning_rate": 1.8066666666666668e-05,
"loss": 0.002,
"step": 1750
},
{
"epoch": 17.0,
"learning_rate": 1.792777777777778e-05,
"loss": 0.002,
"step": 1775
},
{
"epoch": 17.01,
"learning_rate": 1.778888888888889e-05,
"loss": 0.0019,
"step": 1800
},
{
"epoch": 17.01,
"learning_rate": 1.765e-05,
"loss": 0.0018,
"step": 1825
},
{
"epoch": 17.02,
"learning_rate": 1.751111111111111e-05,
"loss": 0.0017,
"step": 1850
},
{
"epoch": 18.0,
"learning_rate": 1.7372222222222224e-05,
"loss": 0.0017,
"step": 1875
},
{
"epoch": 18.01,
"learning_rate": 1.7233333333333333e-05,
"loss": 0.0017,
"step": 1900
},
{
"epoch": 18.01,
"learning_rate": 1.7094444444444446e-05,
"loss": 0.0017,
"step": 1925
},
{
"epoch": 18.02,
"learning_rate": 1.6955555555555555e-05,
"loss": 0.0015,
"step": 1950
},
{
"epoch": 19.0,
"learning_rate": 1.6816666666666668e-05,
"loss": 0.0016,
"step": 1975
},
{
"epoch": 19.01,
"learning_rate": 1.6677777777777777e-05,
"loss": 0.0016,
"step": 2000
},
{
"epoch": 19.01,
"eval_loss": 0.5159559845924377,
"eval_runtime": 669.0954,
"eval_samples_per_second": 9.851,
"eval_steps_per_second": 0.154,
"eval_wer": 26.01968487990603,
"step": 2000
},
{
"epoch": 19.01,
"learning_rate": 1.653888888888889e-05,
"loss": 0.0015,
"step": 2025
},
{
"epoch": 19.02,
"learning_rate": 1.6400000000000002e-05,
"loss": 0.0014,
"step": 2050
},
{
"epoch": 20.0,
"learning_rate": 1.626111111111111e-05,
"loss": 0.0014,
"step": 2075
},
{
"epoch": 20.01,
"learning_rate": 1.612222222222222e-05,
"loss": 0.0014,
"step": 2100
},
{
"epoch": 20.01,
"learning_rate": 1.5983333333333333e-05,
"loss": 0.0014,
"step": 2125
},
{
"epoch": 20.02,
"learning_rate": 1.5844444444444446e-05,
"loss": 0.0013,
"step": 2150
},
{
"epoch": 21.0,
"learning_rate": 1.5705555555555555e-05,
"loss": 0.0013,
"step": 2175
},
{
"epoch": 21.01,
"learning_rate": 1.5566666666666668e-05,
"loss": 0.0013,
"step": 2200
},
{
"epoch": 21.01,
"learning_rate": 1.542777777777778e-05,
"loss": 0.0012,
"step": 2225
},
{
"epoch": 21.02,
"learning_rate": 1.528888888888889e-05,
"loss": 0.0012,
"step": 2250
},
{
"epoch": 22.0,
"learning_rate": 1.515e-05,
"loss": 0.0012,
"step": 2275
},
{
"epoch": 22.01,
"learning_rate": 1.5011111111111112e-05,
"loss": 0.0012,
"step": 2300
},
{
"epoch": 22.01,
"learning_rate": 1.4872222222222224e-05,
"loss": 0.0011,
"step": 2325
},
{
"epoch": 22.02,
"learning_rate": 1.4733333333333335e-05,
"loss": 0.0011,
"step": 2350
},
{
"epoch": 23.0,
"learning_rate": 1.4594444444444444e-05,
"loss": 0.0011,
"step": 2375
},
{
"epoch": 23.01,
"learning_rate": 1.4455555555555555e-05,
"loss": 0.0011,
"step": 2400
},
{
"epoch": 23.01,
"learning_rate": 1.4316666666666668e-05,
"loss": 0.0011,
"step": 2425
},
{
"epoch": 23.02,
"learning_rate": 1.4177777777777779e-05,
"loss": 0.001,
"step": 2450
},
{
"epoch": 24.0,
"learning_rate": 1.403888888888889e-05,
"loss": 0.001,
"step": 2475
},
{
"epoch": 24.01,
"learning_rate": 1.3900000000000002e-05,
"loss": 0.001,
"step": 2500
},
{
"epoch": 24.01,
"learning_rate": 1.376111111111111e-05,
"loss": 0.001,
"step": 2525
},
{
"epoch": 24.02,
"learning_rate": 1.3622222222222223e-05,
"loss": 0.001,
"step": 2550
},
{
"epoch": 24.02,
"learning_rate": 1.3483333333333334e-05,
"loss": 0.0009,
"step": 2575
},
{
"epoch": 25.0,
"learning_rate": 1.3344444444444446e-05,
"loss": 0.0009,
"step": 2600
},
{
"epoch": 25.01,
"learning_rate": 1.3205555555555557e-05,
"loss": 0.0009,
"step": 2625
},
{
"epoch": 25.02,
"learning_rate": 1.3066666666666666e-05,
"loss": 0.0009,
"step": 2650
},
{
"epoch": 25.02,
"learning_rate": 1.2927777777777777e-05,
"loss": 0.0009,
"step": 2675
},
{
"epoch": 26.0,
"learning_rate": 1.278888888888889e-05,
"loss": 0.0009,
"step": 2700
},
{
"epoch": 26.01,
"learning_rate": 1.2650000000000001e-05,
"loss": 0.0009,
"step": 2725
},
{
"epoch": 26.01,
"learning_rate": 1.2511111111111112e-05,
"loss": 0.0009,
"step": 2750
},
{
"epoch": 26.02,
"learning_rate": 1.2372222222222223e-05,
"loss": 0.0008,
"step": 2775
},
{
"epoch": 27.0,
"learning_rate": 1.2233333333333334e-05,
"loss": 0.0008,
"step": 2800
},
{
"epoch": 27.01,
"learning_rate": 1.2094444444444445e-05,
"loss": 0.0008,
"step": 2825
},
{
"epoch": 27.01,
"learning_rate": 1.1955555555555556e-05,
"loss": 0.0008,
"step": 2850
},
{
"epoch": 27.02,
"learning_rate": 1.1816666666666668e-05,
"loss": 0.0008,
"step": 2875
},
{
"epoch": 28.0,
"learning_rate": 1.1677777777777777e-05,
"loss": 0.0008,
"step": 2900
},
{
"epoch": 28.01,
"learning_rate": 1.153888888888889e-05,
"loss": 0.0008,
"step": 2925
},
{
"epoch": 28.01,
"learning_rate": 1.1400000000000001e-05,
"loss": 0.0008,
"step": 2950
},
{
"epoch": 28.02,
"learning_rate": 1.1261111111111112e-05,
"loss": 0.0007,
"step": 2975
},
{
"epoch": 29.0,
"learning_rate": 1.1122222222222223e-05,
"loss": 0.0007,
"step": 3000
},
{
"epoch": 29.0,
"eval_loss": 0.5520183444023132,
"eval_runtime": 676.133,
"eval_samples_per_second": 9.748,
"eval_steps_per_second": 0.152,
"eval_wer": 25.977155818380655,
"step": 3000
},
{
"epoch": 29.01,
"learning_rate": 1.0983333333333334e-05,
"loss": 0.0007,
"step": 3025
},
{
"epoch": 29.01,
"learning_rate": 1.0844444444444445e-05,
"loss": 0.0007,
"step": 3050
},
{
"epoch": 29.02,
"learning_rate": 1.0705555555555556e-05,
"loss": 0.0007,
"step": 3075
},
{
"epoch": 30.0,
"learning_rate": 1.0566666666666668e-05,
"loss": 0.0007,
"step": 3100
},
{
"epoch": 30.01,
"learning_rate": 1.0427777777777778e-05,
"loss": 0.0007,
"step": 3125
},
{
"epoch": 30.01,
"learning_rate": 1.028888888888889e-05,
"loss": 0.0007,
"step": 3150
},
{
"epoch": 30.02,
"learning_rate": 1.0150000000000001e-05,
"loss": 0.0007,
"step": 3175
},
{
"epoch": 31.0,
"learning_rate": 1.0011111111111112e-05,
"loss": 0.0007,
"step": 3200
},
{
"epoch": 31.01,
"learning_rate": 9.872222222222223e-06,
"loss": 0.0007,
"step": 3225
},
{
"epoch": 31.01,
"learning_rate": 9.733333333333334e-06,
"loss": 0.0007,
"step": 3250
},
{
"epoch": 31.02,
"learning_rate": 9.594444444444445e-06,
"loss": 0.0006,
"step": 3275
},
{
"epoch": 32.0,
"learning_rate": 9.455555555555556e-06,
"loss": 0.0006,
"step": 3300
},
{
"epoch": 32.01,
"learning_rate": 9.316666666666667e-06,
"loss": 0.0006,
"step": 3325
},
{
"epoch": 32.01,
"learning_rate": 9.177777777777778e-06,
"loss": 0.0006,
"step": 3350
},
{
"epoch": 32.02,
"learning_rate": 9.03888888888889e-06,
"loss": 0.0006,
"step": 3375
},
{
"epoch": 33.0,
"learning_rate": 8.9e-06,
"loss": 0.0006,
"step": 3400
},
{
"epoch": 33.01,
"learning_rate": 8.761111111111112e-06,
"loss": 0.0006,
"step": 3425
},
{
"epoch": 33.01,
"learning_rate": 8.622222222222223e-06,
"loss": 0.0006,
"step": 3450
},
{
"epoch": 33.02,
"learning_rate": 8.483333333333334e-06,
"loss": 0.0006,
"step": 3475
},
{
"epoch": 33.02,
"learning_rate": 8.344444444444445e-06,
"loss": 0.0006,
"step": 3500
},
{
"epoch": 34.0,
"learning_rate": 8.205555555555556e-06,
"loss": 0.0006,
"step": 3525
},
{
"epoch": 34.01,
"learning_rate": 8.066666666666667e-06,
"loss": 0.0006,
"step": 3550
},
{
"epoch": 34.01,
"learning_rate": 7.927777777777778e-06,
"loss": 0.0006,
"step": 3575
},
{
"epoch": 34.02,
"learning_rate": 7.788888888888889e-06,
"loss": 0.0005,
"step": 3600
},
{
"epoch": 35.0,
"learning_rate": 7.65e-06,
"loss": 0.0006,
"step": 3625
},
{
"epoch": 35.01,
"learning_rate": 7.511111111111112e-06,
"loss": 0.0006,
"step": 3650
},
{
"epoch": 35.01,
"learning_rate": 7.372222222222222e-06,
"loss": 0.0005,
"step": 3675
},
{
"epoch": 35.02,
"learning_rate": 7.233333333333333e-06,
"loss": 0.0005,
"step": 3700
},
{
"epoch": 36.0,
"learning_rate": 7.094444444444445e-06,
"loss": 0.0005,
"step": 3725
},
{
"epoch": 36.01,
"learning_rate": 6.955555555555555e-06,
"loss": 0.0005,
"step": 3750
},
{
"epoch": 36.01,
"learning_rate": 6.816666666666667e-06,
"loss": 0.0005,
"step": 3775
},
{
"epoch": 36.02,
"learning_rate": 6.677777777777779e-06,
"loss": 0.0005,
"step": 3800
},
{
"epoch": 37.0,
"learning_rate": 6.538888888888889e-06,
"loss": 0.0005,
"step": 3825
},
{
"epoch": 37.01,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.0005,
"step": 3850
},
{
"epoch": 37.01,
"learning_rate": 6.261111111111112e-06,
"loss": 0.0005,
"step": 3875
},
{
"epoch": 37.02,
"learning_rate": 6.1222222222222224e-06,
"loss": 0.0005,
"step": 3900
},
{
"epoch": 38.0,
"learning_rate": 5.983333333333334e-06,
"loss": 0.0005,
"step": 3925
},
{
"epoch": 38.01,
"learning_rate": 5.844444444444445e-06,
"loss": 0.0005,
"step": 3950
},
{
"epoch": 38.01,
"learning_rate": 5.705555555555555e-06,
"loss": 0.0005,
"step": 3975
},
{
"epoch": 38.02,
"learning_rate": 5.566666666666667e-06,
"loss": 0.0005,
"step": 4000
},
{
"epoch": 38.02,
"eval_loss": 0.572807788848877,
"eval_runtime": 669.1023,
"eval_samples_per_second": 9.851,
"eval_steps_per_second": 0.154,
"eval_wer": 26.14524687107619,
"step": 4000
},
{
"epoch": 39.0,
"learning_rate": 5.427777777777778e-06,
"loss": 0.0005,
"step": 4025
},
{
"epoch": 39.01,
"learning_rate": 5.288888888888889e-06,
"loss": 0.0005,
"step": 4050
},
{
"epoch": 39.01,
"learning_rate": 5.15e-06,
"loss": 0.0005,
"step": 4075
},
{
"epoch": 39.02,
"learning_rate": 5.011111111111112e-06,
"loss": 0.0005,
"step": 4100
},
{
"epoch": 40.0,
"learning_rate": 4.8722222222222225e-06,
"loss": 0.0005,
"step": 4125
},
{
"epoch": 40.01,
"learning_rate": 4.7333333333333335e-06,
"loss": 0.0005,
"step": 4150
},
{
"epoch": 40.01,
"learning_rate": 4.594444444444445e-06,
"loss": 0.0005,
"step": 4175
},
{
"epoch": 40.02,
"learning_rate": 4.455555555555556e-06,
"loss": 0.0004,
"step": 4200
},
{
"epoch": 41.0,
"learning_rate": 4.316666666666667e-06,
"loss": 0.0004,
"step": 4225
},
{
"epoch": 41.01,
"learning_rate": 4.177777777777778e-06,
"loss": 0.0004,
"step": 4250
},
{
"epoch": 41.01,
"learning_rate": 4.038888888888889e-06,
"loss": 0.0005,
"step": 4275
},
{
"epoch": 41.02,
"learning_rate": 3.9e-06,
"loss": 0.0004,
"step": 4300
},
{
"epoch": 41.02,
"learning_rate": 3.7611111111111113e-06,
"loss": 0.0004,
"step": 4325
},
{
"epoch": 42.0,
"learning_rate": 3.6222222222222226e-06,
"loss": 0.0004,
"step": 4350
},
{
"epoch": 42.01,
"learning_rate": 3.4833333333333336e-06,
"loss": 0.0004,
"step": 4375
},
{
"epoch": 42.01,
"learning_rate": 3.3444444444444445e-06,
"loss": 0.0004,
"step": 4400
},
{
"epoch": 42.02,
"learning_rate": 3.2055555555555555e-06,
"loss": 0.0004,
"step": 4425
},
{
"epoch": 43.0,
"learning_rate": 3.066666666666667e-06,
"loss": 0.0004,
"step": 4450
},
{
"epoch": 43.01,
"learning_rate": 2.9277777777777777e-06,
"loss": 0.0004,
"step": 4475
},
{
"epoch": 43.01,
"learning_rate": 2.788888888888889e-06,
"loss": 0.0004,
"step": 4500
},
{
"epoch": 43.02,
"learning_rate": 2.65e-06,
"loss": 0.0004,
"step": 4525
},
{
"epoch": 44.0,
"learning_rate": 2.5111111111111114e-06,
"loss": 0.0004,
"step": 4550
},
{
"epoch": 44.01,
"learning_rate": 2.3722222222222223e-06,
"loss": 0.0004,
"step": 4575
},
{
"epoch": 44.01,
"learning_rate": 2.2333333333333333e-06,
"loss": 0.0004,
"step": 4600
},
{
"epoch": 44.02,
"learning_rate": 2.0944444444444446e-06,
"loss": 0.0004,
"step": 4625
},
{
"epoch": 45.0,
"learning_rate": 1.9555555555555556e-06,
"loss": 0.0004,
"step": 4650
},
{
"epoch": 45.01,
"learning_rate": 1.816666666666667e-06,
"loss": 0.0004,
"step": 4675
},
{
"epoch": 45.01,
"learning_rate": 1.6777777777777779e-06,
"loss": 0.0004,
"step": 4700
},
{
"epoch": 45.02,
"learning_rate": 1.538888888888889e-06,
"loss": 0.0004,
"step": 4725
},
{
"epoch": 46.0,
"learning_rate": 1.4000000000000001e-06,
"loss": 0.0004,
"step": 4750
},
{
"epoch": 46.01,
"learning_rate": 1.261111111111111e-06,
"loss": 0.0004,
"step": 4775
},
{
"epoch": 46.01,
"learning_rate": 1.1222222222222222e-06,
"loss": 0.0004,
"step": 4800
},
{
"epoch": 46.02,
"learning_rate": 9.833333333333334e-07,
"loss": 0.0004,
"step": 4825
},
{
"epoch": 47.0,
"learning_rate": 8.444444444444444e-07,
"loss": 0.0004,
"step": 4850
},
{
"epoch": 47.01,
"learning_rate": 7.055555555555556e-07,
"loss": 0.0004,
"step": 4875
},
{
"epoch": 47.01,
"learning_rate": 5.666666666666667e-07,
"loss": 0.0004,
"step": 4900
},
{
"epoch": 47.02,
"learning_rate": 4.277777777777778e-07,
"loss": 0.0004,
"step": 4925
},
{
"epoch": 48.0,
"learning_rate": 2.888888888888889e-07,
"loss": 0.0004,
"step": 4950
},
{
"epoch": 48.01,
"learning_rate": 1.5000000000000002e-07,
"loss": 0.0004,
"step": 4975
},
{
"epoch": 48.01,
"learning_rate": 1.1111111111111112e-08,
"loss": 0.0004,
"step": 5000
},
{
"epoch": 48.01,
"eval_loss": 0.5817816853523254,
"eval_runtime": 676.2223,
"eval_samples_per_second": 9.747,
"eval_steps_per_second": 0.152,
"eval_wer": 26.22017902709709,
"step": 5000
},
{
"epoch": 48.01,
"step": 5000,
"total_flos": 4.150417098276864e+19,
"train_loss": 0.07347059211265296,
"train_runtime": 65946.469,
"train_samples_per_second": 9.705,
"train_steps_per_second": 0.076
}
],
"logging_steps": 25,
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"save_steps": 1000,
"total_flos": 4.150417098276864e+19,
"trial_name": null,
"trial_params": null
}