whisper-base-cv-highLR-cs / trainer_state.json
sgangireddy's picture
End of training
ac84af0
{
"best_metric": 26.914789782876923,
"best_model_checkpoint": "./checkpoint-5000",
"epoch": 21.0088,
"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": 4.6e-06,
"loss": 1.7084,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.600000000000001e-06,
"loss": 1.0071,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.4599999999999999e-05,
"loss": 0.7307,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.9600000000000002e-05,
"loss": 0.5786,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.46e-05,
"loss": 0.5402,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.96e-05,
"loss": 0.4489,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 3.46e-05,
"loss": 0.3889,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 3.960000000000001e-05,
"loss": 0.382,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.46e-05,
"loss": 0.455,
"step": 225
},
{
"epoch": 1.0,
"learning_rate": 4.96e-05,
"loss": 0.5095,
"step": 250
},
{
"epoch": 1.01,
"learning_rate": 5.4600000000000006e-05,
"loss": 0.4524,
"step": 275
},
{
"epoch": 1.01,
"learning_rate": 5.96e-05,
"loss": 0.3678,
"step": 300
},
{
"epoch": 1.02,
"learning_rate": 6.460000000000001e-05,
"loss": 0.3278,
"step": 325
},
{
"epoch": 1.02,
"learning_rate": 6.96e-05,
"loss": 0.2725,
"step": 350
},
{
"epoch": 1.03,
"learning_rate": 7.46e-05,
"loss": 0.2622,
"step": 375
},
{
"epoch": 1.03,
"learning_rate": 7.960000000000001e-05,
"loss": 0.2167,
"step": 400
},
{
"epoch": 1.04,
"learning_rate": 8.46e-05,
"loss": 0.193,
"step": 425
},
{
"epoch": 1.04,
"learning_rate": 8.960000000000001e-05,
"loss": 0.2222,
"step": 450
},
{
"epoch": 2.0,
"learning_rate": 9.46e-05,
"loss": 0.2821,
"step": 475
},
{
"epoch": 2.01,
"learning_rate": 9.960000000000001e-05,
"loss": 0.2945,
"step": 500
},
{
"epoch": 2.01,
"learning_rate": 9.94888888888889e-05,
"loss": 0.2187,
"step": 525
},
{
"epoch": 2.02,
"learning_rate": 9.893333333333333e-05,
"loss": 0.1994,
"step": 550
},
{
"epoch": 2.02,
"learning_rate": 9.837777777777778e-05,
"loss": 0.1656,
"step": 575
},
{
"epoch": 2.03,
"learning_rate": 9.782222222222223e-05,
"loss": 0.1628,
"step": 600
},
{
"epoch": 2.03,
"learning_rate": 9.726666666666667e-05,
"loss": 0.1278,
"step": 625
},
{
"epoch": 2.04,
"learning_rate": 9.671111111111112e-05,
"loss": 0.115,
"step": 650
},
{
"epoch": 2.04,
"learning_rate": 9.615555555555557e-05,
"loss": 0.1098,
"step": 675
},
{
"epoch": 2.05,
"learning_rate": 9.56e-05,
"loss": 0.139,
"step": 700
},
{
"epoch": 3.0,
"learning_rate": 9.504444444444444e-05,
"loss": 0.1633,
"step": 725
},
{
"epoch": 3.01,
"learning_rate": 9.44888888888889e-05,
"loss": 0.13,
"step": 750
},
{
"epoch": 3.01,
"learning_rate": 9.393333333333334e-05,
"loss": 0.102,
"step": 775
},
{
"epoch": 3.02,
"learning_rate": 9.337777777777778e-05,
"loss": 0.097,
"step": 800
},
{
"epoch": 3.02,
"learning_rate": 9.282222222222222e-05,
"loss": 0.0864,
"step": 825
},
{
"epoch": 3.03,
"learning_rate": 9.226666666666667e-05,
"loss": 0.0824,
"step": 850
},
{
"epoch": 3.03,
"learning_rate": 9.171111111111112e-05,
"loss": 0.0694,
"step": 875
},
{
"epoch": 3.04,
"learning_rate": 9.115555555555556e-05,
"loss": 0.0603,
"step": 900
},
{
"epoch": 3.04,
"learning_rate": 9.06e-05,
"loss": 0.0704,
"step": 925
},
{
"epoch": 4.0,
"learning_rate": 9.004444444444446e-05,
"loss": 0.0899,
"step": 950
},
{
"epoch": 4.01,
"learning_rate": 8.94888888888889e-05,
"loss": 0.0893,
"step": 975
},
{
"epoch": 4.01,
"learning_rate": 8.893333333333333e-05,
"loss": 0.0606,
"step": 1000
},
{
"epoch": 4.01,
"eval_loss": 0.46788325905799866,
"eval_runtime": 2327.6768,
"eval_samples_per_second": 3.314,
"eval_steps_per_second": 0.104,
"eval_wer": 34.8188799469623,
"step": 1000
},
{
"epoch": 4.02,
"learning_rate": 8.837777777777778e-05,
"loss": 0.0595,
"step": 1025
},
{
"epoch": 4.02,
"learning_rate": 8.782222222222223e-05,
"loss": 0.0596,
"step": 1050
},
{
"epoch": 4.03,
"learning_rate": 8.726666666666667e-05,
"loss": 0.0526,
"step": 1075
},
{
"epoch": 4.03,
"learning_rate": 8.671111111111112e-05,
"loss": 0.0457,
"step": 1100
},
{
"epoch": 4.04,
"learning_rate": 8.615555555555556e-05,
"loss": 0.0402,
"step": 1125
},
{
"epoch": 4.04,
"learning_rate": 8.560000000000001e-05,
"loss": 0.0434,
"step": 1150
},
{
"epoch": 4.05,
"learning_rate": 8.504444444444444e-05,
"loss": 0.0514,
"step": 1175
},
{
"epoch": 5.0,
"learning_rate": 8.44888888888889e-05,
"loss": 0.0677,
"step": 1200
},
{
"epoch": 5.01,
"learning_rate": 8.393333333333335e-05,
"loss": 0.0429,
"step": 1225
},
{
"epoch": 5.01,
"learning_rate": 8.337777777777778e-05,
"loss": 0.0394,
"step": 1250
},
{
"epoch": 5.02,
"learning_rate": 8.282222222222222e-05,
"loss": 0.0402,
"step": 1275
},
{
"epoch": 5.02,
"learning_rate": 8.226666666666667e-05,
"loss": 0.0396,
"step": 1300
},
{
"epoch": 5.03,
"learning_rate": 8.171111111111112e-05,
"loss": 0.0324,
"step": 1325
},
{
"epoch": 5.03,
"learning_rate": 8.115555555555556e-05,
"loss": 0.0307,
"step": 1350
},
{
"epoch": 5.04,
"learning_rate": 8.060000000000001e-05,
"loss": 0.0266,
"step": 1375
},
{
"epoch": 5.04,
"learning_rate": 8.004444444444444e-05,
"loss": 0.0344,
"step": 1400
},
{
"epoch": 6.0,
"learning_rate": 7.94888888888889e-05,
"loss": 0.0479,
"step": 1425
},
{
"epoch": 6.01,
"learning_rate": 7.893333333333333e-05,
"loss": 0.0353,
"step": 1450
},
{
"epoch": 6.01,
"learning_rate": 7.837777777777778e-05,
"loss": 0.031,
"step": 1475
},
{
"epoch": 6.02,
"learning_rate": 7.782222222222223e-05,
"loss": 0.0297,
"step": 1500
},
{
"epoch": 6.02,
"learning_rate": 7.726666666666667e-05,
"loss": 0.0306,
"step": 1525
},
{
"epoch": 6.03,
"learning_rate": 7.671111111111111e-05,
"loss": 0.0267,
"step": 1550
},
{
"epoch": 6.03,
"learning_rate": 7.615555555555556e-05,
"loss": 0.0229,
"step": 1575
},
{
"epoch": 6.04,
"learning_rate": 7.560000000000001e-05,
"loss": 0.0177,
"step": 1600
},
{
"epoch": 6.04,
"learning_rate": 7.504444444444444e-05,
"loss": 0.0232,
"step": 1625
},
{
"epoch": 6.05,
"learning_rate": 7.44888888888889e-05,
"loss": 0.0273,
"step": 1650
},
{
"epoch": 7.0,
"learning_rate": 7.393333333333333e-05,
"loss": 0.0327,
"step": 1675
},
{
"epoch": 7.01,
"learning_rate": 7.337777777777778e-05,
"loss": 0.0218,
"step": 1700
},
{
"epoch": 7.01,
"learning_rate": 7.282222222222222e-05,
"loss": 0.0213,
"step": 1725
},
{
"epoch": 7.02,
"learning_rate": 7.226666666666667e-05,
"loss": 0.0198,
"step": 1750
},
{
"epoch": 7.02,
"learning_rate": 7.171111111111112e-05,
"loss": 0.0199,
"step": 1775
},
{
"epoch": 7.03,
"learning_rate": 7.115555555555556e-05,
"loss": 0.0168,
"step": 1800
},
{
"epoch": 7.03,
"learning_rate": 7.06e-05,
"loss": 0.0163,
"step": 1825
},
{
"epoch": 7.04,
"learning_rate": 7.004444444444445e-05,
"loss": 0.0144,
"step": 1850
},
{
"epoch": 7.04,
"learning_rate": 6.94888888888889e-05,
"loss": 0.0183,
"step": 1875
},
{
"epoch": 8.0,
"learning_rate": 6.893333333333333e-05,
"loss": 0.0224,
"step": 1900
},
{
"epoch": 8.01,
"learning_rate": 6.837777777777778e-05,
"loss": 0.0209,
"step": 1925
},
{
"epoch": 8.01,
"learning_rate": 6.782222222222222e-05,
"loss": 0.0151,
"step": 1950
},
{
"epoch": 8.02,
"learning_rate": 6.726666666666667e-05,
"loss": 0.0164,
"step": 1975
},
{
"epoch": 8.02,
"learning_rate": 6.671111111111111e-05,
"loss": 0.0153,
"step": 2000
},
{
"epoch": 8.02,
"eval_loss": 0.5124102234840393,
"eval_runtime": 2312.8,
"eval_samples_per_second": 3.335,
"eval_steps_per_second": 0.105,
"eval_wer": 32.445074676341136,
"step": 2000
},
{
"epoch": 8.03,
"learning_rate": 6.615555555555556e-05,
"loss": 0.0132,
"step": 2025
},
{
"epoch": 8.03,
"learning_rate": 6.560000000000001e-05,
"loss": 0.0122,
"step": 2050
},
{
"epoch": 8.04,
"learning_rate": 6.504444444444445e-05,
"loss": 0.0104,
"step": 2075
},
{
"epoch": 8.04,
"learning_rate": 6.448888888888888e-05,
"loss": 0.0115,
"step": 2100
},
{
"epoch": 9.0,
"learning_rate": 6.393333333333333e-05,
"loss": 0.0132,
"step": 2125
},
{
"epoch": 9.01,
"learning_rate": 6.337777777777778e-05,
"loss": 0.0167,
"step": 2150
},
{
"epoch": 9.01,
"learning_rate": 6.282222222222222e-05,
"loss": 0.0113,
"step": 2175
},
{
"epoch": 9.02,
"learning_rate": 6.226666666666667e-05,
"loss": 0.011,
"step": 2200
},
{
"epoch": 9.02,
"learning_rate": 6.171111111111112e-05,
"loss": 0.0108,
"step": 2225
},
{
"epoch": 9.03,
"learning_rate": 6.115555555555556e-05,
"loss": 0.0106,
"step": 2250
},
{
"epoch": 9.03,
"learning_rate": 6.06e-05,
"loss": 0.0088,
"step": 2275
},
{
"epoch": 9.04,
"learning_rate": 6.0044444444444446e-05,
"loss": 0.0079,
"step": 2300
},
{
"epoch": 9.04,
"learning_rate": 5.94888888888889e-05,
"loss": 0.0085,
"step": 2325
},
{
"epoch": 9.05,
"learning_rate": 5.893333333333334e-05,
"loss": 0.0088,
"step": 2350
},
{
"epoch": 10.0,
"learning_rate": 5.837777777777778e-05,
"loss": 0.011,
"step": 2375
},
{
"epoch": 10.01,
"learning_rate": 5.782222222222222e-05,
"loss": 0.0105,
"step": 2400
},
{
"epoch": 10.01,
"learning_rate": 5.726666666666667e-05,
"loss": 0.0074,
"step": 2425
},
{
"epoch": 10.02,
"learning_rate": 5.6711111111111116e-05,
"loss": 0.007,
"step": 2450
},
{
"epoch": 10.02,
"learning_rate": 5.615555555555556e-05,
"loss": 0.0074,
"step": 2475
},
{
"epoch": 10.03,
"learning_rate": 5.560000000000001e-05,
"loss": 0.0068,
"step": 2500
},
{
"epoch": 10.03,
"learning_rate": 5.504444444444444e-05,
"loss": 0.0054,
"step": 2525
},
{
"epoch": 10.04,
"learning_rate": 5.448888888888889e-05,
"loss": 0.006,
"step": 2550
},
{
"epoch": 10.04,
"learning_rate": 5.3933333333333334e-05,
"loss": 0.005,
"step": 2575
},
{
"epoch": 11.0,
"learning_rate": 5.3377777777777785e-05,
"loss": 0.0057,
"step": 2600
},
{
"epoch": 11.01,
"learning_rate": 5.282222222222223e-05,
"loss": 0.0068,
"step": 2625
},
{
"epoch": 11.01,
"learning_rate": 5.2266666666666665e-05,
"loss": 0.0065,
"step": 2650
},
{
"epoch": 11.02,
"learning_rate": 5.171111111111111e-05,
"loss": 0.0049,
"step": 2675
},
{
"epoch": 11.02,
"learning_rate": 5.115555555555556e-05,
"loss": 0.003,
"step": 2700
},
{
"epoch": 11.03,
"learning_rate": 5.0600000000000003e-05,
"loss": 0.0047,
"step": 2725
},
{
"epoch": 11.03,
"learning_rate": 5.004444444444445e-05,
"loss": 0.0033,
"step": 2750
},
{
"epoch": 11.04,
"learning_rate": 4.948888888888889e-05,
"loss": 0.0033,
"step": 2775
},
{
"epoch": 11.04,
"learning_rate": 4.8933333333333335e-05,
"loss": 0.0027,
"step": 2800
},
{
"epoch": 11.05,
"learning_rate": 4.837777777777778e-05,
"loss": 0.0035,
"step": 2825
},
{
"epoch": 12.0,
"learning_rate": 4.782222222222222e-05,
"loss": 0.004,
"step": 2850
},
{
"epoch": 12.01,
"learning_rate": 4.726666666666667e-05,
"loss": 0.0042,
"step": 2875
},
{
"epoch": 12.01,
"learning_rate": 4.671111111111111e-05,
"loss": 0.003,
"step": 2900
},
{
"epoch": 12.02,
"learning_rate": 4.615555555555556e-05,
"loss": 0.0024,
"step": 2925
},
{
"epoch": 12.02,
"learning_rate": 4.5600000000000004e-05,
"loss": 0.0026,
"step": 2950
},
{
"epoch": 12.03,
"learning_rate": 4.504444444444445e-05,
"loss": 0.0017,
"step": 2975
},
{
"epoch": 12.03,
"learning_rate": 4.448888888888889e-05,
"loss": 0.0018,
"step": 3000
},
{
"epoch": 12.03,
"eval_loss": 0.4890795648097992,
"eval_runtime": 2290.1296,
"eval_samples_per_second": 3.368,
"eval_steps_per_second": 0.106,
"eval_wer": 29.00683228669822,
"step": 3000
},
{
"epoch": 12.04,
"learning_rate": 4.3933333333333335e-05,
"loss": 0.0016,
"step": 3025
},
{
"epoch": 12.04,
"learning_rate": 4.337777777777778e-05,
"loss": 0.0013,
"step": 3050
},
{
"epoch": 13.0,
"learning_rate": 4.282222222222222e-05,
"loss": 0.0022,
"step": 3075
},
{
"epoch": 13.01,
"learning_rate": 4.226666666666667e-05,
"loss": 0.0028,
"step": 3100
},
{
"epoch": 13.01,
"learning_rate": 4.171111111111111e-05,
"loss": 0.0021,
"step": 3125
},
{
"epoch": 13.02,
"learning_rate": 4.115555555555556e-05,
"loss": 0.0014,
"step": 3150
},
{
"epoch": 13.02,
"learning_rate": 4.0600000000000004e-05,
"loss": 0.0019,
"step": 3175
},
{
"epoch": 13.03,
"learning_rate": 4.004444444444445e-05,
"loss": 0.0024,
"step": 3200
},
{
"epoch": 13.03,
"learning_rate": 3.948888888888889e-05,
"loss": 0.0014,
"step": 3225
},
{
"epoch": 13.04,
"learning_rate": 3.8933333333333336e-05,
"loss": 0.0011,
"step": 3250
},
{
"epoch": 13.04,
"learning_rate": 3.837777777777778e-05,
"loss": 0.0007,
"step": 3275
},
{
"epoch": 13.05,
"learning_rate": 3.782222222222222e-05,
"loss": 0.0018,
"step": 3300
},
{
"epoch": 14.0,
"learning_rate": 3.726666666666667e-05,
"loss": 0.0013,
"step": 3325
},
{
"epoch": 14.01,
"learning_rate": 3.671111111111111e-05,
"loss": 0.0009,
"step": 3350
},
{
"epoch": 14.01,
"learning_rate": 3.615555555555556e-05,
"loss": 0.0009,
"step": 3375
},
{
"epoch": 14.02,
"learning_rate": 3.56e-05,
"loss": 0.0006,
"step": 3400
},
{
"epoch": 14.02,
"learning_rate": 3.504444444444445e-05,
"loss": 0.001,
"step": 3425
},
{
"epoch": 14.03,
"learning_rate": 3.448888888888889e-05,
"loss": 0.0005,
"step": 3450
},
{
"epoch": 14.03,
"learning_rate": 3.3933333333333336e-05,
"loss": 0.0006,
"step": 3475
},
{
"epoch": 14.04,
"learning_rate": 3.337777777777778e-05,
"loss": 0.0004,
"step": 3500
},
{
"epoch": 14.04,
"learning_rate": 3.2822222222222223e-05,
"loss": 0.0006,
"step": 3525
},
{
"epoch": 15.0,
"learning_rate": 3.226666666666667e-05,
"loss": 0.0005,
"step": 3550
},
{
"epoch": 15.01,
"learning_rate": 3.171111111111111e-05,
"loss": 0.001,
"step": 3575
},
{
"epoch": 15.01,
"learning_rate": 3.1155555555555555e-05,
"loss": 0.0007,
"step": 3600
},
{
"epoch": 15.02,
"learning_rate": 3.06e-05,
"loss": 0.0005,
"step": 3625
},
{
"epoch": 15.02,
"learning_rate": 3.004444444444445e-05,
"loss": 0.0005,
"step": 3650
},
{
"epoch": 15.03,
"learning_rate": 2.948888888888889e-05,
"loss": 0.0007,
"step": 3675
},
{
"epoch": 15.03,
"learning_rate": 2.8933333333333333e-05,
"loss": 0.0003,
"step": 3700
},
{
"epoch": 15.04,
"learning_rate": 2.837777777777778e-05,
"loss": 0.0003,
"step": 3725
},
{
"epoch": 15.04,
"learning_rate": 2.782222222222222e-05,
"loss": 0.0003,
"step": 3750
},
{
"epoch": 15.05,
"learning_rate": 2.7266666666666668e-05,
"loss": 0.0005,
"step": 3775
},
{
"epoch": 16.0,
"learning_rate": 2.6711111111111115e-05,
"loss": 0.0004,
"step": 3800
},
{
"epoch": 16.01,
"learning_rate": 2.6155555555555555e-05,
"loss": 0.0004,
"step": 3825
},
{
"epoch": 16.01,
"learning_rate": 2.5600000000000002e-05,
"loss": 0.0003,
"step": 3850
},
{
"epoch": 16.02,
"learning_rate": 2.504444444444445e-05,
"loss": 0.0004,
"step": 3875
},
{
"epoch": 16.02,
"learning_rate": 2.448888888888889e-05,
"loss": 0.0005,
"step": 3900
},
{
"epoch": 16.03,
"learning_rate": 2.3933333333333337e-05,
"loss": 0.0003,
"step": 3925
},
{
"epoch": 16.03,
"learning_rate": 2.337777777777778e-05,
"loss": 0.0003,
"step": 3950
},
{
"epoch": 16.04,
"learning_rate": 2.282222222222222e-05,
"loss": 0.0003,
"step": 3975
},
{
"epoch": 16.04,
"learning_rate": 2.2266666666666668e-05,
"loss": 0.0004,
"step": 4000
},
{
"epoch": 16.04,
"eval_loss": 0.480441689491272,
"eval_runtime": 2248.4312,
"eval_samples_per_second": 3.431,
"eval_steps_per_second": 0.108,
"eval_wer": 27.121047494521278,
"step": 4000
},
{
"epoch": 17.0,
"learning_rate": 2.1711111111111112e-05,
"loss": 0.0003,
"step": 4025
},
{
"epoch": 17.01,
"learning_rate": 2.1155555555555556e-05,
"loss": 0.0003,
"step": 4050
},
{
"epoch": 17.01,
"learning_rate": 2.06e-05,
"loss": 0.0003,
"step": 4075
},
{
"epoch": 17.02,
"learning_rate": 2.0044444444444446e-05,
"loss": 0.0003,
"step": 4100
},
{
"epoch": 17.02,
"learning_rate": 1.948888888888889e-05,
"loss": 0.0003,
"step": 4125
},
{
"epoch": 17.03,
"learning_rate": 1.8933333333333334e-05,
"loss": 0.0002,
"step": 4150
},
{
"epoch": 17.03,
"learning_rate": 1.837777777777778e-05,
"loss": 0.0002,
"step": 4175
},
{
"epoch": 17.04,
"learning_rate": 1.7822222222222225e-05,
"loss": 0.0002,
"step": 4200
},
{
"epoch": 17.04,
"learning_rate": 1.726666666666667e-05,
"loss": 0.0002,
"step": 4225
},
{
"epoch": 18.0,
"learning_rate": 1.6711111111111112e-05,
"loss": 0.0003,
"step": 4250
},
{
"epoch": 18.01,
"learning_rate": 1.6155555555555556e-05,
"loss": 0.0003,
"step": 4275
},
{
"epoch": 18.01,
"learning_rate": 1.56e-05,
"loss": 0.0003,
"step": 4300
},
{
"epoch": 18.02,
"learning_rate": 1.5044444444444445e-05,
"loss": 0.0002,
"step": 4325
},
{
"epoch": 18.02,
"learning_rate": 1.448888888888889e-05,
"loss": 0.0002,
"step": 4350
},
{
"epoch": 18.03,
"learning_rate": 1.3933333333333334e-05,
"loss": 0.0002,
"step": 4375
},
{
"epoch": 18.03,
"learning_rate": 1.3377777777777778e-05,
"loss": 0.0002,
"step": 4400
},
{
"epoch": 18.04,
"learning_rate": 1.2822222222222222e-05,
"loss": 0.0002,
"step": 4425
},
{
"epoch": 18.04,
"learning_rate": 1.2266666666666667e-05,
"loss": 0.0002,
"step": 4450
},
{
"epoch": 18.05,
"learning_rate": 1.1711111111111111e-05,
"loss": 0.0002,
"step": 4475
},
{
"epoch": 19.0,
"learning_rate": 1.1155555555555556e-05,
"loss": 0.0003,
"step": 4500
},
{
"epoch": 19.01,
"learning_rate": 1.06e-05,
"loss": 0.0003,
"step": 4525
},
{
"epoch": 19.01,
"learning_rate": 1.0044444444444446e-05,
"loss": 0.0002,
"step": 4550
},
{
"epoch": 19.02,
"learning_rate": 9.48888888888889e-06,
"loss": 0.0002,
"step": 4575
},
{
"epoch": 19.02,
"learning_rate": 8.933333333333333e-06,
"loss": 0.0002,
"step": 4600
},
{
"epoch": 19.03,
"learning_rate": 8.377777777777779e-06,
"loss": 0.0002,
"step": 4625
},
{
"epoch": 19.03,
"learning_rate": 7.822222222222222e-06,
"loss": 0.0002,
"step": 4650
},
{
"epoch": 19.04,
"learning_rate": 7.266666666666668e-06,
"loss": 0.0002,
"step": 4675
},
{
"epoch": 19.04,
"learning_rate": 6.711111111111111e-06,
"loss": 0.0002,
"step": 4700
},
{
"epoch": 20.0,
"learning_rate": 6.155555555555556e-06,
"loss": 0.0002,
"step": 4725
},
{
"epoch": 20.01,
"learning_rate": 5.600000000000001e-06,
"loss": 0.0002,
"step": 4750
},
{
"epoch": 20.01,
"learning_rate": 5.044444444444444e-06,
"loss": 0.0002,
"step": 4775
},
{
"epoch": 20.02,
"learning_rate": 4.488888888888889e-06,
"loss": 0.0002,
"step": 4800
},
{
"epoch": 20.02,
"learning_rate": 3.9333333333333335e-06,
"loss": 0.0002,
"step": 4825
},
{
"epoch": 20.03,
"learning_rate": 3.3777777777777777e-06,
"loss": 0.0002,
"step": 4850
},
{
"epoch": 20.03,
"learning_rate": 2.8222222222222223e-06,
"loss": 0.0002,
"step": 4875
},
{
"epoch": 20.04,
"learning_rate": 2.266666666666667e-06,
"loss": 0.0002,
"step": 4900
},
{
"epoch": 20.04,
"learning_rate": 1.7111111111111112e-06,
"loss": 0.0002,
"step": 4925
},
{
"epoch": 20.05,
"learning_rate": 1.1555555555555556e-06,
"loss": 0.0002,
"step": 4950
},
{
"epoch": 21.0,
"learning_rate": 6.000000000000001e-07,
"loss": 0.0002,
"step": 4975
},
{
"epoch": 21.01,
"learning_rate": 4.444444444444445e-08,
"loss": 0.0002,
"step": 5000
},
{
"epoch": 21.01,
"eval_loss": 0.4828060269355774,
"eval_runtime": 2204.7463,
"eval_samples_per_second": 3.499,
"eval_steps_per_second": 0.11,
"eval_wer": 26.914789782876923,
"step": 5000
},
{
"epoch": 21.01,
"step": 5000,
"total_flos": 2.073068169265152e+19,
"train_loss": 0.06790026845252141,
"train_runtime": 85623.5177,
"train_samples_per_second": 3.737,
"train_steps_per_second": 0.058
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 2.073068169265152e+19,
"trial_name": null,
"trial_params": null
}