whisper-base-id-1 / trainer_state.json
arun100's picture
End of training
555f8bb verified
{
"best_metric": 26.607783604747446,
"best_model_checkpoint": "./checkpoint-9500",
"epoch": 95.00725,
"eval_steps": 500,
"global_step": 10000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 2.2999999999999998e-08,
"loss": 1.7225,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 4.8e-08,
"loss": 1.8162,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 7.299999999999999e-08,
"loss": 1.8665,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 9.8e-08,
"loss": 1.6195,
"step": 100
},
{
"epoch": 1.0,
"learning_rate": 1.23e-07,
"loss": 1.6459,
"step": 125
},
{
"epoch": 1.0,
"learning_rate": 1.4799999999999998e-07,
"loss": 1.6399,
"step": 150
},
{
"epoch": 1.01,
"learning_rate": 1.7299999999999997e-07,
"loss": 1.6865,
"step": 175
},
{
"epoch": 1.01,
"learning_rate": 1.98e-07,
"loss": 1.4292,
"step": 200
},
{
"epoch": 2.0,
"learning_rate": 2.23e-07,
"loss": 1.3863,
"step": 225
},
{
"epoch": 2.0,
"learning_rate": 2.48e-07,
"loss": 1.3597,
"step": 250
},
{
"epoch": 2.01,
"learning_rate": 2.73e-07,
"loss": 1.4244,
"step": 275
},
{
"epoch": 2.01,
"learning_rate": 2.98e-07,
"loss": 1.2283,
"step": 300
},
{
"epoch": 3.0,
"learning_rate": 3.23e-07,
"loss": 1.1242,
"step": 325
},
{
"epoch": 3.0,
"learning_rate": 3.4799999999999994e-07,
"loss": 1.0479,
"step": 350
},
{
"epoch": 3.01,
"learning_rate": 3.7299999999999997e-07,
"loss": 1.0434,
"step": 375
},
{
"epoch": 3.01,
"learning_rate": 3.98e-07,
"loss": 0.9246,
"step": 400
},
{
"epoch": 4.0,
"learning_rate": 4.2299999999999996e-07,
"loss": 0.8148,
"step": 425
},
{
"epoch": 4.0,
"learning_rate": 4.48e-07,
"loss": 0.8052,
"step": 450
},
{
"epoch": 4.01,
"learning_rate": 4.7299999999999996e-07,
"loss": 0.8294,
"step": 475
},
{
"epoch": 4.01,
"learning_rate": 4.979999999999999e-07,
"loss": 0.8238,
"step": 500
},
{
"epoch": 4.01,
"eval_loss": 0.6399861574172974,
"eval_runtime": 297.3907,
"eval_samples_per_second": 12.247,
"eval_steps_per_second": 0.383,
"eval_wer": 37.974974698684335,
"step": 500
},
{
"epoch": 5.0,
"learning_rate": 4.987894736842105e-07,
"loss": 0.7435,
"step": 525
},
{
"epoch": 5.0,
"learning_rate": 4.974736842105263e-07,
"loss": 0.7042,
"step": 550
},
{
"epoch": 5.01,
"learning_rate": 4.961578947368421e-07,
"loss": 0.7353,
"step": 575
},
{
"epoch": 5.01,
"learning_rate": 4.948421052631579e-07,
"loss": 0.7831,
"step": 600
},
{
"epoch": 5.01,
"learning_rate": 4.935263157894736e-07,
"loss": 0.6878,
"step": 625
},
{
"epoch": 6.0,
"learning_rate": 4.922105263157894e-07,
"loss": 0.6701,
"step": 650
},
{
"epoch": 6.0,
"learning_rate": 4.908947368421052e-07,
"loss": 0.6667,
"step": 675
},
{
"epoch": 6.01,
"learning_rate": 4.89578947368421e-07,
"loss": 0.725,
"step": 700
},
{
"epoch": 6.01,
"learning_rate": 4.882631578947368e-07,
"loss": 0.6726,
"step": 725
},
{
"epoch": 7.0,
"learning_rate": 4.869473684210526e-07,
"loss": 0.6322,
"step": 750
},
{
"epoch": 7.0,
"learning_rate": 4.856315789473684e-07,
"loss": 0.6211,
"step": 775
},
{
"epoch": 7.01,
"learning_rate": 4.843157894736842e-07,
"loss": 0.7014,
"step": 800
},
{
"epoch": 7.01,
"learning_rate": 4.83e-07,
"loss": 0.6346,
"step": 825
},
{
"epoch": 8.0,
"learning_rate": 4.816842105263158e-07,
"loss": 0.6173,
"step": 850
},
{
"epoch": 8.0,
"learning_rate": 4.803684210526316e-07,
"loss": 0.5881,
"step": 875
},
{
"epoch": 8.01,
"learning_rate": 4.790526315789473e-07,
"loss": 0.6624,
"step": 900
},
{
"epoch": 8.01,
"learning_rate": 4.777368421052631e-07,
"loss": 0.6246,
"step": 925
},
{
"epoch": 9.0,
"learning_rate": 4.764210526315789e-07,
"loss": 0.5866,
"step": 950
},
{
"epoch": 9.0,
"learning_rate": 4.751052631578947e-07,
"loss": 0.5673,
"step": 975
},
{
"epoch": 9.01,
"learning_rate": 4.737894736842105e-07,
"loss": 0.6348,
"step": 1000
},
{
"epoch": 9.01,
"eval_loss": 0.5192969441413879,
"eval_runtime": 270.0894,
"eval_samples_per_second": 13.484,
"eval_steps_per_second": 0.422,
"eval_wer": 32.24767687919772,
"step": 1000
},
{
"epoch": 9.01,
"learning_rate": 4.724736842105263e-07,
"loss": 0.6131,
"step": 1025
},
{
"epoch": 10.0,
"learning_rate": 4.711578947368421e-07,
"loss": 0.5762,
"step": 1050
},
{
"epoch": 10.0,
"learning_rate": 4.698421052631579e-07,
"loss": 0.5529,
"step": 1075
},
{
"epoch": 10.01,
"learning_rate": 4.6852631578947367e-07,
"loss": 0.5852,
"step": 1100
},
{
"epoch": 10.01,
"learning_rate": 4.6721052631578946e-07,
"loss": 0.6196,
"step": 1125
},
{
"epoch": 11.0,
"learning_rate": 4.6589473684210525e-07,
"loss": 0.5637,
"step": 1150
},
{
"epoch": 11.0,
"learning_rate": 4.64578947368421e-07,
"loss": 0.5329,
"step": 1175
},
{
"epoch": 11.01,
"learning_rate": 4.632631578947368e-07,
"loss": 0.5586,
"step": 1200
},
{
"epoch": 11.01,
"learning_rate": 4.6194736842105257e-07,
"loss": 0.6126,
"step": 1225
},
{
"epoch": 11.01,
"learning_rate": 4.606315789473684e-07,
"loss": 0.544,
"step": 1250
},
{
"epoch": 12.0,
"learning_rate": 4.593157894736842e-07,
"loss": 0.5267,
"step": 1275
},
{
"epoch": 12.0,
"learning_rate": 4.58e-07,
"loss": 0.5326,
"step": 1300
},
{
"epoch": 12.01,
"learning_rate": 4.566842105263158e-07,
"loss": 0.6053,
"step": 1325
},
{
"epoch": 12.01,
"learning_rate": 4.553684210526316e-07,
"loss": 0.533,
"step": 1350
},
{
"epoch": 13.0,
"learning_rate": 4.5405263157894737e-07,
"loss": 0.5211,
"step": 1375
},
{
"epoch": 13.0,
"learning_rate": 4.5273684210526316e-07,
"loss": 0.5043,
"step": 1400
},
{
"epoch": 13.01,
"learning_rate": 4.514210526315789e-07,
"loss": 0.5782,
"step": 1425
},
{
"epoch": 13.01,
"learning_rate": 4.501052631578947e-07,
"loss": 0.5328,
"step": 1450
},
{
"epoch": 14.0,
"learning_rate": 4.487894736842105e-07,
"loss": 0.5061,
"step": 1475
},
{
"epoch": 14.0,
"learning_rate": 4.4747368421052627e-07,
"loss": 0.4879,
"step": 1500
},
{
"epoch": 14.0,
"eval_loss": 0.48285967111587524,
"eval_runtime": 276.9826,
"eval_samples_per_second": 13.149,
"eval_steps_per_second": 0.412,
"eval_wer": 30.724997699880397,
"step": 1500
},
{
"epoch": 14.01,
"learning_rate": 4.4615789473684206e-07,
"loss": 0.554,
"step": 1525
},
{
"epoch": 14.01,
"learning_rate": 4.4484210526315785e-07,
"loss": 0.5363,
"step": 1550
},
{
"epoch": 15.0,
"learning_rate": 4.4352631578947365e-07,
"loss": 0.4961,
"step": 1575
},
{
"epoch": 15.0,
"learning_rate": 4.422105263157895e-07,
"loss": 0.4813,
"step": 1600
},
{
"epoch": 15.01,
"learning_rate": 4.408947368421053e-07,
"loss": 0.5311,
"step": 1625
},
{
"epoch": 15.01,
"learning_rate": 4.3957894736842107e-07,
"loss": 0.536,
"step": 1650
},
{
"epoch": 16.0,
"learning_rate": 4.382631578947368e-07,
"loss": 0.4911,
"step": 1675
},
{
"epoch": 16.0,
"learning_rate": 4.369473684210526e-07,
"loss": 0.4699,
"step": 1700
},
{
"epoch": 16.01,
"learning_rate": 4.356315789473684e-07,
"loss": 0.493,
"step": 1725
},
{
"epoch": 16.01,
"learning_rate": 4.343157894736842e-07,
"loss": 0.5487,
"step": 1750
},
{
"epoch": 16.01,
"learning_rate": 4.3299999999999997e-07,
"loss": 0.4812,
"step": 1775
},
{
"epoch": 17.0,
"learning_rate": 4.3168421052631576e-07,
"loss": 0.4683,
"step": 1800
},
{
"epoch": 17.0,
"learning_rate": 4.3036842105263155e-07,
"loss": 0.4733,
"step": 1825
},
{
"epoch": 17.01,
"learning_rate": 4.2905263157894735e-07,
"loss": 0.5303,
"step": 1850
},
{
"epoch": 17.01,
"learning_rate": 4.2773684210526314e-07,
"loss": 0.4748,
"step": 1875
},
{
"epoch": 18.0,
"learning_rate": 4.2642105263157893e-07,
"loss": 0.462,
"step": 1900
},
{
"epoch": 18.0,
"learning_rate": 4.251052631578947e-07,
"loss": 0.4471,
"step": 1925
},
{
"epoch": 18.01,
"learning_rate": 4.237894736842105e-07,
"loss": 0.5263,
"step": 1950
},
{
"epoch": 18.01,
"learning_rate": 4.224736842105263e-07,
"loss": 0.4784,
"step": 1975
},
{
"epoch": 19.0,
"learning_rate": 4.211578947368421e-07,
"loss": 0.4518,
"step": 2000
},
{
"epoch": 19.0,
"eval_loss": 0.46445801854133606,
"eval_runtime": 272.5056,
"eval_samples_per_second": 13.365,
"eval_steps_per_second": 0.418,
"eval_wer": 29.703744594718923,
"step": 2000
},
{
"epoch": 19.0,
"learning_rate": 4.198421052631579e-07,
"loss": 0.4344,
"step": 2025
},
{
"epoch": 19.01,
"learning_rate": 4.1852631578947367e-07,
"loss": 0.5051,
"step": 2050
},
{
"epoch": 19.01,
"learning_rate": 4.1721052631578946e-07,
"loss": 0.4771,
"step": 2075
},
{
"epoch": 20.0,
"learning_rate": 4.1589473684210526e-07,
"loss": 0.447,
"step": 2100
},
{
"epoch": 20.0,
"learning_rate": 4.1457894736842105e-07,
"loss": 0.4314,
"step": 2125
},
{
"epoch": 20.01,
"learning_rate": 4.1326315789473684e-07,
"loss": 0.4845,
"step": 2150
},
{
"epoch": 20.01,
"learning_rate": 4.119473684210526e-07,
"loss": 0.4714,
"step": 2175
},
{
"epoch": 21.0,
"learning_rate": 4.1063157894736837e-07,
"loss": 0.4452,
"step": 2200
},
{
"epoch": 21.0,
"learning_rate": 4.0931578947368416e-07,
"loss": 0.4248,
"step": 2225
},
{
"epoch": 21.01,
"learning_rate": 4.0799999999999995e-07,
"loss": 0.4547,
"step": 2250
},
{
"epoch": 21.01,
"learning_rate": 4.066842105263158e-07,
"loss": 0.4868,
"step": 2275
},
{
"epoch": 22.0,
"learning_rate": 4.053684210526316e-07,
"loss": 0.4347,
"step": 2300
},
{
"epoch": 22.0,
"learning_rate": 4.0405263157894737e-07,
"loss": 0.4086,
"step": 2325
},
{
"epoch": 22.01,
"learning_rate": 4.0273684210526316e-07,
"loss": 0.445,
"step": 2350
},
{
"epoch": 22.01,
"learning_rate": 4.0142105263157896e-07,
"loss": 0.4783,
"step": 2375
},
{
"epoch": 22.01,
"learning_rate": 4.0010526315789475e-07,
"loss": 0.4348,
"step": 2400
},
{
"epoch": 23.0,
"learning_rate": 3.987894736842105e-07,
"loss": 0.4127,
"step": 2425
},
{
"epoch": 23.0,
"learning_rate": 3.974736842105263e-07,
"loss": 0.413,
"step": 2450
},
{
"epoch": 23.01,
"learning_rate": 3.9615789473684207e-07,
"loss": 0.4845,
"step": 2475
},
{
"epoch": 23.01,
"learning_rate": 3.9484210526315786e-07,
"loss": 0.4253,
"step": 2500
},
{
"epoch": 23.01,
"eval_loss": 0.4513331651687622,
"eval_runtime": 270.1098,
"eval_samples_per_second": 13.483,
"eval_steps_per_second": 0.422,
"eval_wer": 28.875701536479898,
"step": 2500
},
{
"epoch": 24.0,
"learning_rate": 3.9352631578947365e-07,
"loss": 0.4112,
"step": 2525
},
{
"epoch": 24.0,
"learning_rate": 3.9221052631578944e-07,
"loss": 0.3978,
"step": 2550
},
{
"epoch": 24.01,
"learning_rate": 3.9089473684210523e-07,
"loss": 0.4728,
"step": 2575
},
{
"epoch": 24.01,
"learning_rate": 3.895789473684211e-07,
"loss": 0.4234,
"step": 2600
},
{
"epoch": 25.0,
"learning_rate": 3.8826315789473686e-07,
"loss": 0.4057,
"step": 2625
},
{
"epoch": 25.0,
"learning_rate": 3.8694736842105266e-07,
"loss": 0.3916,
"step": 2650
},
{
"epoch": 25.01,
"learning_rate": 3.856315789473684e-07,
"loss": 0.4582,
"step": 2675
},
{
"epoch": 25.01,
"learning_rate": 3.843157894736842e-07,
"loss": 0.4305,
"step": 2700
},
{
"epoch": 26.0,
"learning_rate": 3.83e-07,
"loss": 0.3989,
"step": 2725
},
{
"epoch": 26.0,
"learning_rate": 3.8168421052631577e-07,
"loss": 0.3844,
"step": 2750
},
{
"epoch": 26.01,
"learning_rate": 3.8036842105263156e-07,
"loss": 0.4324,
"step": 2775
},
{
"epoch": 26.01,
"learning_rate": 3.7905263157894735e-07,
"loss": 0.4384,
"step": 2800
},
{
"epoch": 27.0,
"learning_rate": 3.7773684210526314e-07,
"loss": 0.3972,
"step": 2825
},
{
"epoch": 27.0,
"learning_rate": 3.7642105263157893e-07,
"loss": 0.3816,
"step": 2850
},
{
"epoch": 27.01,
"learning_rate": 3.751052631578947e-07,
"loss": 0.407,
"step": 2875
},
{
"epoch": 27.01,
"learning_rate": 3.737894736842105e-07,
"loss": 0.4353,
"step": 2900
},
{
"epoch": 27.01,
"learning_rate": 3.7247368421052625e-07,
"loss": 0.4011,
"step": 2925
},
{
"epoch": 28.0,
"learning_rate": 3.711578947368421e-07,
"loss": 0.3719,
"step": 2950
},
{
"epoch": 28.0,
"learning_rate": 3.698421052631579e-07,
"loss": 0.3946,
"step": 2975
},
{
"epoch": 28.01,
"learning_rate": 3.685263157894737e-07,
"loss": 0.4471,
"step": 3000
},
{
"epoch": 28.01,
"eval_loss": 0.4408997595310211,
"eval_runtime": 269.0858,
"eval_samples_per_second": 13.535,
"eval_steps_per_second": 0.424,
"eval_wer": 28.093660870365262,
"step": 3000
},
{
"epoch": 28.01,
"learning_rate": 3.6721052631578947e-07,
"loss": 0.3836,
"step": 3025
},
{
"epoch": 29.0,
"learning_rate": 3.6589473684210526e-07,
"loss": 0.3761,
"step": 3050
},
{
"epoch": 29.0,
"learning_rate": 3.6457894736842105e-07,
"loss": 0.37,
"step": 3075
},
{
"epoch": 29.01,
"learning_rate": 3.6326315789473684e-07,
"loss": 0.4441,
"step": 3100
},
{
"epoch": 29.01,
"learning_rate": 3.6194736842105263e-07,
"loss": 0.3872,
"step": 3125
},
{
"epoch": 30.0,
"learning_rate": 3.606315789473684e-07,
"loss": 0.3702,
"step": 3150
},
{
"epoch": 30.0,
"learning_rate": 3.5931578947368416e-07,
"loss": 0.3579,
"step": 3175
},
{
"epoch": 30.01,
"learning_rate": 3.5799999999999995e-07,
"loss": 0.4379,
"step": 3200
},
{
"epoch": 30.01,
"learning_rate": 3.5668421052631574e-07,
"loss": 0.3795,
"step": 3225
},
{
"epoch": 31.0,
"learning_rate": 3.5536842105263153e-07,
"loss": 0.3763,
"step": 3250
},
{
"epoch": 31.0,
"learning_rate": 3.540526315789473e-07,
"loss": 0.3547,
"step": 3275
},
{
"epoch": 31.01,
"learning_rate": 3.5273684210526317e-07,
"loss": 0.412,
"step": 3300
},
{
"epoch": 31.01,
"learning_rate": 3.5142105263157896e-07,
"loss": 0.3898,
"step": 3325
},
{
"epoch": 32.0,
"learning_rate": 3.5010526315789475e-07,
"loss": 0.3672,
"step": 3350
},
{
"epoch": 32.0,
"learning_rate": 3.4878947368421054e-07,
"loss": 0.3489,
"step": 3375
},
{
"epoch": 32.01,
"learning_rate": 3.4747368421052633e-07,
"loss": 0.3835,
"step": 3400
},
{
"epoch": 32.01,
"learning_rate": 3.4615789473684207e-07,
"loss": 0.4061,
"step": 3425
},
{
"epoch": 33.0,
"learning_rate": 3.4484210526315786e-07,
"loss": 0.3685,
"step": 3450
},
{
"epoch": 33.0,
"learning_rate": 3.4352631578947365e-07,
"loss": 0.3429,
"step": 3475
},
{
"epoch": 33.01,
"learning_rate": 3.4221052631578944e-07,
"loss": 0.3713,
"step": 3500
},
{
"epoch": 33.01,
"eval_loss": 0.43465471267700195,
"eval_runtime": 267.8336,
"eval_samples_per_second": 13.598,
"eval_steps_per_second": 0.426,
"eval_wer": 27.78544484313184,
"step": 3500
},
{
"epoch": 33.01,
"learning_rate": 3.4089473684210523e-07,
"loss": 0.4016,
"step": 3525
},
{
"epoch": 33.01,
"learning_rate": 3.39578947368421e-07,
"loss": 0.3692,
"step": 3550
},
{
"epoch": 34.0,
"learning_rate": 3.382631578947368e-07,
"loss": 0.3389,
"step": 3575
},
{
"epoch": 34.0,
"learning_rate": 3.369473684210526e-07,
"loss": 0.3501,
"step": 3600
},
{
"epoch": 34.01,
"learning_rate": 3.3563157894736845e-07,
"loss": 0.4038,
"step": 3625
},
{
"epoch": 34.01,
"learning_rate": 3.3431578947368424e-07,
"loss": 0.3605,
"step": 3650
},
{
"epoch": 35.0,
"learning_rate": 3.33e-07,
"loss": 0.3527,
"step": 3675
},
{
"epoch": 35.0,
"learning_rate": 3.3168421052631577e-07,
"loss": 0.3306,
"step": 3700
},
{
"epoch": 35.01,
"learning_rate": 3.3036842105263156e-07,
"loss": 0.4099,
"step": 3725
},
{
"epoch": 35.01,
"learning_rate": 3.2905263157894735e-07,
"loss": 0.3496,
"step": 3750
},
{
"epoch": 36.0,
"learning_rate": 3.2773684210526314e-07,
"loss": 0.3451,
"step": 3775
},
{
"epoch": 36.0,
"learning_rate": 3.2642105263157893e-07,
"loss": 0.3313,
"step": 3800
},
{
"epoch": 36.01,
"learning_rate": 3.251052631578947e-07,
"loss": 0.3868,
"step": 3825
},
{
"epoch": 36.01,
"learning_rate": 3.237894736842105e-07,
"loss": 0.3619,
"step": 3850
},
{
"epoch": 37.0,
"learning_rate": 3.224736842105263e-07,
"loss": 0.3382,
"step": 3875
},
{
"epoch": 37.0,
"learning_rate": 3.211578947368421e-07,
"loss": 0.3225,
"step": 3900
},
{
"epoch": 37.01,
"learning_rate": 3.1984210526315783e-07,
"loss": 0.3713,
"step": 3925
},
{
"epoch": 37.01,
"learning_rate": 3.185263157894736e-07,
"loss": 0.3739,
"step": 3950
},
{
"epoch": 38.0,
"learning_rate": 3.1721052631578947e-07,
"loss": 0.3376,
"step": 3975
},
{
"epoch": 38.0,
"learning_rate": 3.1589473684210526e-07,
"loss": 0.3233,
"step": 4000
},
{
"epoch": 38.0,
"eval_loss": 0.43065741658210754,
"eval_runtime": 268.7558,
"eval_samples_per_second": 13.551,
"eval_steps_per_second": 0.424,
"eval_wer": 27.638237188333793,
"step": 4000
},
{
"epoch": 38.01,
"learning_rate": 3.1457894736842105e-07,
"loss": 0.3438,
"step": 4025
},
{
"epoch": 38.01,
"learning_rate": 3.1326315789473684e-07,
"loss": 0.3801,
"step": 4050
},
{
"epoch": 38.01,
"learning_rate": 3.1194736842105263e-07,
"loss": 0.34,
"step": 4075
},
{
"epoch": 39.0,
"learning_rate": 3.106315789473684e-07,
"loss": 0.3193,
"step": 4100
},
{
"epoch": 39.0,
"learning_rate": 3.0936842105263157e-07,
"loss": 0.3308,
"step": 4125
},
{
"epoch": 39.01,
"learning_rate": 3.0805263157894736e-07,
"loss": 0.3835,
"step": 4150
},
{
"epoch": 39.01,
"learning_rate": 3.0673684210526315e-07,
"loss": 0.3343,
"step": 4175
},
{
"epoch": 40.0,
"learning_rate": 3.054210526315789e-07,
"loss": 0.3197,
"step": 4200
},
{
"epoch": 40.0,
"learning_rate": 3.041052631578947e-07,
"loss": 0.3168,
"step": 4225
},
{
"epoch": 40.01,
"learning_rate": 3.027894736842105e-07,
"loss": 0.3843,
"step": 4250
},
{
"epoch": 40.01,
"learning_rate": 3.014736842105263e-07,
"loss": 0.3263,
"step": 4275
},
{
"epoch": 41.0,
"learning_rate": 3.001578947368421e-07,
"loss": 0.3279,
"step": 4300
},
{
"epoch": 41.0,
"learning_rate": 2.988421052631579e-07,
"loss": 0.3039,
"step": 4325
},
{
"epoch": 41.01,
"learning_rate": 2.975263157894737e-07,
"loss": 0.3731,
"step": 4350
},
{
"epoch": 41.01,
"learning_rate": 2.962105263157895e-07,
"loss": 0.3322,
"step": 4375
},
{
"epoch": 42.0,
"learning_rate": 2.9489473684210527e-07,
"loss": 0.3242,
"step": 4400
},
{
"epoch": 42.0,
"learning_rate": 2.9357894736842106e-07,
"loss": 0.3005,
"step": 4425
},
{
"epoch": 42.01,
"learning_rate": 2.922631578947368e-07,
"loss": 0.3512,
"step": 4450
},
{
"epoch": 42.01,
"learning_rate": 2.909473684210526e-07,
"loss": 0.3442,
"step": 4475
},
{
"epoch": 43.0,
"learning_rate": 2.896315789473684e-07,
"loss": 0.3152,
"step": 4500
},
{
"epoch": 43.0,
"eval_loss": 0.42795851826667786,
"eval_runtime": 269.4089,
"eval_samples_per_second": 13.518,
"eval_steps_per_second": 0.423,
"eval_wer": 27.532431686447694,
"step": 4500
},
{
"epoch": 43.0,
"learning_rate": 2.8831578947368417e-07,
"loss": 0.3043,
"step": 4525
},
{
"epoch": 43.01,
"learning_rate": 2.8699999999999996e-07,
"loss": 0.3352,
"step": 4550
},
{
"epoch": 43.01,
"learning_rate": 2.8568421052631575e-07,
"loss": 0.349,
"step": 4575
},
{
"epoch": 44.0,
"learning_rate": 2.843684210526316e-07,
"loss": 0.3155,
"step": 4600
},
{
"epoch": 44.0,
"learning_rate": 2.830526315789474e-07,
"loss": 0.2993,
"step": 4625
},
{
"epoch": 44.01,
"learning_rate": 2.817368421052632e-07,
"loss": 0.3198,
"step": 4650
},
{
"epoch": 44.01,
"learning_rate": 2.8042105263157897e-07,
"loss": 0.3554,
"step": 4675
},
{
"epoch": 44.01,
"learning_rate": 2.791052631578947e-07,
"loss": 0.3107,
"step": 4700
},
{
"epoch": 45.0,
"learning_rate": 2.777894736842105e-07,
"loss": 0.2961,
"step": 4725
},
{
"epoch": 45.0,
"learning_rate": 2.764736842105263e-07,
"loss": 0.3121,
"step": 4750
},
{
"epoch": 45.01,
"learning_rate": 2.751578947368421e-07,
"loss": 0.3532,
"step": 4775
},
{
"epoch": 45.01,
"learning_rate": 2.7384210526315787e-07,
"loss": 0.3139,
"step": 4800
},
{
"epoch": 46.0,
"learning_rate": 2.7252631578947366e-07,
"loss": 0.2993,
"step": 4825
},
{
"epoch": 46.0,
"learning_rate": 2.7121052631578945e-07,
"loss": 0.2978,
"step": 4850
},
{
"epoch": 46.01,
"learning_rate": 2.6989473684210524e-07,
"loss": 0.3465,
"step": 4875
},
{
"epoch": 46.01,
"learning_rate": 2.6857894736842104e-07,
"loss": 0.3137,
"step": 4900
},
{
"epoch": 47.0,
"learning_rate": 2.672631578947369e-07,
"loss": 0.2984,
"step": 4925
},
{
"epoch": 47.0,
"learning_rate": 2.659473684210526e-07,
"loss": 0.2842,
"step": 4950
},
{
"epoch": 47.01,
"learning_rate": 2.646315789473684e-07,
"loss": 0.3396,
"step": 4975
},
{
"epoch": 47.01,
"learning_rate": 2.633157894736842e-07,
"loss": 0.3152,
"step": 5000
},
{
"epoch": 47.01,
"eval_loss": 0.42454764246940613,
"eval_runtime": 267.1822,
"eval_samples_per_second": 13.631,
"eval_steps_per_second": 0.427,
"eval_wer": 27.21961542000184,
"step": 5000
},
{
"epoch": 48.0,
"learning_rate": 2.62e-07,
"loss": 0.3029,
"step": 5025
},
{
"epoch": 48.0,
"learning_rate": 2.606842105263158e-07,
"loss": 0.2838,
"step": 5050
},
{
"epoch": 48.01,
"learning_rate": 2.5936842105263157e-07,
"loss": 0.3264,
"step": 5075
},
{
"epoch": 48.01,
"learning_rate": 2.5805263157894736e-07,
"loss": 0.3249,
"step": 5100
},
{
"epoch": 49.0,
"learning_rate": 2.5673684210526315e-07,
"loss": 0.2944,
"step": 5125
},
{
"epoch": 49.0,
"learning_rate": 2.5542105263157894e-07,
"loss": 0.2796,
"step": 5150
},
{
"epoch": 49.01,
"learning_rate": 2.5410526315789474e-07,
"loss": 0.3093,
"step": 5175
},
{
"epoch": 49.01,
"learning_rate": 2.527894736842105e-07,
"loss": 0.3354,
"step": 5200
},
{
"epoch": 49.01,
"learning_rate": 2.5147368421052626e-07,
"loss": 0.2965,
"step": 5225
},
{
"epoch": 50.0,
"learning_rate": 2.5015789473684206e-07,
"loss": 0.2776,
"step": 5250
},
{
"epoch": 50.01,
"learning_rate": 2.488421052631579e-07,
"loss": 0.2965,
"step": 5275
},
{
"epoch": 50.01,
"learning_rate": 2.475263157894737e-07,
"loss": 0.3361,
"step": 5300
},
{
"epoch": 50.01,
"learning_rate": 2.4621052631578943e-07,
"loss": 0.2956,
"step": 5325
},
{
"epoch": 51.0,
"learning_rate": 2.4489473684210527e-07,
"loss": 0.2787,
"step": 5350
},
{
"epoch": 51.0,
"learning_rate": 2.4357894736842106e-07,
"loss": 0.283,
"step": 5375
},
{
"epoch": 51.01,
"learning_rate": 2.4226315789473685e-07,
"loss": 0.3343,
"step": 5400
},
{
"epoch": 51.01,
"learning_rate": 2.4094736842105265e-07,
"loss": 0.2927,
"step": 5425
},
{
"epoch": 52.0,
"learning_rate": 2.396315789473684e-07,
"loss": 0.2812,
"step": 5450
},
{
"epoch": 52.0,
"learning_rate": 2.383157894736842e-07,
"loss": 0.2714,
"step": 5475
},
{
"epoch": 52.01,
"learning_rate": 2.3699999999999996e-07,
"loss": 0.333,
"step": 5500
},
{
"epoch": 52.01,
"eval_loss": 0.42268866300582886,
"eval_runtime": 269.9173,
"eval_samples_per_second": 13.493,
"eval_steps_per_second": 0.422,
"eval_wer": 26.99420369859233,
"step": 5500
},
{
"epoch": 52.01,
"learning_rate": 2.3568421052631578e-07,
"loss": 0.2938,
"step": 5525
},
{
"epoch": 53.0,
"learning_rate": 2.3436842105263157e-07,
"loss": 0.2837,
"step": 5550
},
{
"epoch": 53.0,
"learning_rate": 2.3305263157894736e-07,
"loss": 0.2645,
"step": 5575
},
{
"epoch": 53.01,
"learning_rate": 2.3173684210526316e-07,
"loss": 0.3201,
"step": 5600
},
{
"epoch": 53.01,
"learning_rate": 2.3042105263157892e-07,
"loss": 0.3024,
"step": 5625
},
{
"epoch": 54.0,
"learning_rate": 2.291052631578947e-07,
"loss": 0.2817,
"step": 5650
},
{
"epoch": 54.0,
"learning_rate": 2.277894736842105e-07,
"loss": 0.2586,
"step": 5675
},
{
"epoch": 54.01,
"learning_rate": 2.2647368421052632e-07,
"loss": 0.3051,
"step": 5700
},
{
"epoch": 54.01,
"learning_rate": 2.251578947368421e-07,
"loss": 0.3171,
"step": 5725
},
{
"epoch": 55.0,
"learning_rate": 2.238421052631579e-07,
"loss": 0.2812,
"step": 5750
},
{
"epoch": 55.0,
"learning_rate": 2.2252631578947367e-07,
"loss": 0.2621,
"step": 5775
},
{
"epoch": 55.01,
"learning_rate": 2.2121052631578946e-07,
"loss": 0.2892,
"step": 5800
},
{
"epoch": 55.01,
"learning_rate": 2.1989473684210525e-07,
"loss": 0.3119,
"step": 5825
},
{
"epoch": 55.01,
"learning_rate": 2.1857894736842104e-07,
"loss": 0.2844,
"step": 5850
},
{
"epoch": 56.0,
"learning_rate": 2.1726315789473686e-07,
"loss": 0.258,
"step": 5875
},
{
"epoch": 56.0,
"learning_rate": 2.1594736842105262e-07,
"loss": 0.2751,
"step": 5900
},
{
"epoch": 56.01,
"learning_rate": 2.146315789473684e-07,
"loss": 0.3205,
"step": 5925
},
{
"epoch": 56.01,
"learning_rate": 2.133157894736842e-07,
"loss": 0.28,
"step": 5950
},
{
"epoch": 57.0,
"learning_rate": 2.12e-07,
"loss": 0.2694,
"step": 5975
},
{
"epoch": 57.0,
"learning_rate": 2.1068421052631576e-07,
"loss": 0.257,
"step": 6000
},
{
"epoch": 57.0,
"eval_loss": 0.4216955006122589,
"eval_runtime": 269.5343,
"eval_samples_per_second": 13.512,
"eval_steps_per_second": 0.423,
"eval_wer": 26.962002024105253,
"step": 6000
},
{
"epoch": 57.01,
"learning_rate": 2.0936842105263157e-07,
"loss": 0.3208,
"step": 6025
},
{
"epoch": 57.01,
"learning_rate": 2.0805263157894737e-07,
"loss": 0.2829,
"step": 6050
},
{
"epoch": 58.0,
"learning_rate": 2.0673684210526316e-07,
"loss": 0.2649,
"step": 6075
},
{
"epoch": 58.0,
"learning_rate": 2.0542105263157895e-07,
"loss": 0.2552,
"step": 6100
},
{
"epoch": 58.01,
"learning_rate": 2.041052631578947e-07,
"loss": 0.3103,
"step": 6125
},
{
"epoch": 58.01,
"learning_rate": 2.027894736842105e-07,
"loss": 0.2794,
"step": 6150
},
{
"epoch": 59.0,
"learning_rate": 2.014736842105263e-07,
"loss": 0.273,
"step": 6175
},
{
"epoch": 59.0,
"learning_rate": 2.001578947368421e-07,
"loss": 0.2494,
"step": 6200
},
{
"epoch": 59.01,
"learning_rate": 1.988421052631579e-07,
"loss": 0.3029,
"step": 6225
},
{
"epoch": 59.01,
"learning_rate": 1.975263157894737e-07,
"loss": 0.2865,
"step": 6250
},
{
"epoch": 60.0,
"learning_rate": 1.962631578947368e-07,
"loss": 0.2712,
"step": 6275
},
{
"epoch": 60.0,
"learning_rate": 1.949473684210526e-07,
"loss": 0.2504,
"step": 6300
},
{
"epoch": 60.01,
"learning_rate": 1.9363157894736842e-07,
"loss": 0.2789,
"step": 6325
},
{
"epoch": 60.01,
"learning_rate": 1.923157894736842e-07,
"loss": 0.3024,
"step": 6350
},
{
"epoch": 61.0,
"learning_rate": 1.91e-07,
"loss": 0.2681,
"step": 6375
},
{
"epoch": 61.0,
"learning_rate": 1.8968421052631577e-07,
"loss": 0.2486,
"step": 6400
},
{
"epoch": 61.01,
"learning_rate": 1.8836842105263156e-07,
"loss": 0.2696,
"step": 6425
},
{
"epoch": 61.01,
"learning_rate": 1.8705263157894735e-07,
"loss": 0.3003,
"step": 6450
},
{
"epoch": 61.01,
"learning_rate": 1.8573684210526314e-07,
"loss": 0.2716,
"step": 6475
},
{
"epoch": 62.0,
"learning_rate": 1.8442105263157896e-07,
"loss": 0.25,
"step": 6500
},
{
"epoch": 62.0,
"eval_loss": 0.42138177156448364,
"eval_runtime": 269.9205,
"eval_samples_per_second": 13.493,
"eval_steps_per_second": 0.422,
"eval_wer": 26.814794369307204,
"step": 6500
},
{
"epoch": 62.0,
"learning_rate": 1.8310526315789472e-07,
"loss": 0.2535,
"step": 6525
},
{
"epoch": 62.01,
"learning_rate": 1.817894736842105e-07,
"loss": 0.3058,
"step": 6550
},
{
"epoch": 62.01,
"learning_rate": 1.804736842105263e-07,
"loss": 0.2721,
"step": 6575
},
{
"epoch": 63.0,
"learning_rate": 1.791578947368421e-07,
"loss": 0.2511,
"step": 6600
},
{
"epoch": 63.0,
"learning_rate": 1.7784210526315789e-07,
"loss": 0.2466,
"step": 6625
},
{
"epoch": 63.01,
"learning_rate": 1.7652631578947368e-07,
"loss": 0.3013,
"step": 6650
},
{
"epoch": 63.01,
"learning_rate": 1.7521052631578947e-07,
"loss": 0.2675,
"step": 6675
},
{
"epoch": 64.0,
"learning_rate": 1.7389473684210526e-07,
"loss": 0.2626,
"step": 6700
},
{
"epoch": 64.0,
"learning_rate": 1.7257894736842105e-07,
"loss": 0.2373,
"step": 6725
},
{
"epoch": 64.01,
"learning_rate": 1.7126315789473684e-07,
"loss": 0.2953,
"step": 6750
},
{
"epoch": 64.01,
"learning_rate": 1.699473684210526e-07,
"loss": 0.2742,
"step": 6775
},
{
"epoch": 65.0,
"learning_rate": 1.686315789473684e-07,
"loss": 0.2545,
"step": 6800
},
{
"epoch": 65.0,
"learning_rate": 1.673157894736842e-07,
"loss": 0.2389,
"step": 6825
},
{
"epoch": 65.01,
"learning_rate": 1.66e-07,
"loss": 0.2817,
"step": 6850
},
{
"epoch": 65.01,
"learning_rate": 1.646842105263158e-07,
"loss": 0.2805,
"step": 6875
},
{
"epoch": 66.0,
"learning_rate": 1.6336842105263156e-07,
"loss": 0.2575,
"step": 6900
},
{
"epoch": 66.0,
"learning_rate": 1.6205263157894735e-07,
"loss": 0.2403,
"step": 6925
},
{
"epoch": 66.01,
"learning_rate": 1.6073684210526314e-07,
"loss": 0.2616,
"step": 6950
},
{
"epoch": 66.01,
"learning_rate": 1.5942105263157893e-07,
"loss": 0.2919,
"step": 6975
},
{
"epoch": 66.01,
"learning_rate": 1.5810526315789475e-07,
"loss": 0.2587,
"step": 7000
},
{
"epoch": 66.01,
"eval_loss": 0.4206458032131195,
"eval_runtime": 275.0064,
"eval_samples_per_second": 13.243,
"eval_steps_per_second": 0.415,
"eval_wer": 26.754991259545495,
"step": 7000
},
{
"epoch": 67.0,
"learning_rate": 1.5678947368421051e-07,
"loss": 0.2443,
"step": 7025
},
{
"epoch": 67.0,
"learning_rate": 1.554736842105263e-07,
"loss": 0.2499,
"step": 7050
},
{
"epoch": 67.01,
"learning_rate": 1.541578947368421e-07,
"loss": 0.2868,
"step": 7075
},
{
"epoch": 67.01,
"learning_rate": 1.528421052631579e-07,
"loss": 0.2624,
"step": 7100
},
{
"epoch": 68.0,
"learning_rate": 1.5152631578947368e-07,
"loss": 0.2426,
"step": 7125
},
{
"epoch": 68.0,
"learning_rate": 1.5021052631578944e-07,
"loss": 0.2435,
"step": 7150
},
{
"epoch": 68.01,
"learning_rate": 1.4889473684210526e-07,
"loss": 0.2894,
"step": 7175
},
{
"epoch": 68.01,
"learning_rate": 1.4757894736842105e-07,
"loss": 0.2589,
"step": 7200
},
{
"epoch": 69.0,
"learning_rate": 1.4626315789473684e-07,
"loss": 0.2434,
"step": 7225
},
{
"epoch": 69.0,
"learning_rate": 1.4494736842105263e-07,
"loss": 0.2352,
"step": 7250
},
{
"epoch": 69.01,
"learning_rate": 1.436315789473684e-07,
"loss": 0.2855,
"step": 7275
},
{
"epoch": 69.01,
"learning_rate": 1.423157894736842e-07,
"loss": 0.2677,
"step": 7300
},
{
"epoch": 70.0,
"learning_rate": 1.4099999999999998e-07,
"loss": 0.2434,
"step": 7325
},
{
"epoch": 70.0,
"learning_rate": 1.396842105263158e-07,
"loss": 0.2307,
"step": 7350
},
{
"epoch": 70.01,
"learning_rate": 1.383684210526316e-07,
"loss": 0.2778,
"step": 7375
},
{
"epoch": 70.01,
"learning_rate": 1.3705263157894735e-07,
"loss": 0.2674,
"step": 7400
},
{
"epoch": 71.0,
"learning_rate": 1.3573684210526314e-07,
"loss": 0.2508,
"step": 7425
},
{
"epoch": 71.0,
"learning_rate": 1.3442105263157893e-07,
"loss": 0.2273,
"step": 7450
},
{
"epoch": 71.01,
"learning_rate": 1.3310526315789472e-07,
"loss": 0.2658,
"step": 7475
},
{
"epoch": 71.01,
"learning_rate": 1.3178947368421054e-07,
"loss": 0.2765,
"step": 7500
},
{
"epoch": 71.01,
"eval_loss": 0.41979241371154785,
"eval_runtime": 273.8283,
"eval_samples_per_second": 13.3,
"eval_steps_per_second": 0.416,
"eval_wer": 26.699788388996225,
"step": 7500
},
{
"epoch": 72.0,
"learning_rate": 1.304736842105263e-07,
"loss": 0.2499,
"step": 7525
},
{
"epoch": 72.0,
"learning_rate": 1.291578947368421e-07,
"loss": 0.2308,
"step": 7550
},
{
"epoch": 72.01,
"learning_rate": 1.278421052631579e-07,
"loss": 0.254,
"step": 7575
},
{
"epoch": 72.01,
"learning_rate": 1.2652631578947368e-07,
"loss": 0.274,
"step": 7600
},
{
"epoch": 72.01,
"learning_rate": 1.2521052631578947e-07,
"loss": 0.2542,
"step": 7625
},
{
"epoch": 73.0,
"learning_rate": 1.2389473684210526e-07,
"loss": 0.2302,
"step": 7650
},
{
"epoch": 73.0,
"learning_rate": 1.2257894736842105e-07,
"loss": 0.2407,
"step": 7675
},
{
"epoch": 73.01,
"learning_rate": 1.2126315789473684e-07,
"loss": 0.2825,
"step": 7700
},
{
"epoch": 73.01,
"learning_rate": 1.1994736842105263e-07,
"loss": 0.2546,
"step": 7725
},
{
"epoch": 74.0,
"learning_rate": 1.1863157894736841e-07,
"loss": 0.2307,
"step": 7750
},
{
"epoch": 74.0,
"learning_rate": 1.173157894736842e-07,
"loss": 0.235,
"step": 7775
},
{
"epoch": 74.01,
"learning_rate": 1.16e-07,
"loss": 0.2821,
"step": 7800
},
{
"epoch": 74.01,
"learning_rate": 1.1468421052631578e-07,
"loss": 0.2483,
"step": 7825
},
{
"epoch": 75.0,
"learning_rate": 1.1336842105263158e-07,
"loss": 0.2387,
"step": 7850
},
{
"epoch": 75.0,
"learning_rate": 1.1205263157894735e-07,
"loss": 0.2274,
"step": 7875
},
{
"epoch": 75.01,
"learning_rate": 1.1073684210526316e-07,
"loss": 0.273,
"step": 7900
},
{
"epoch": 75.01,
"learning_rate": 1.0942105263157895e-07,
"loss": 0.2584,
"step": 7925
},
{
"epoch": 76.0,
"learning_rate": 1.0810526315789473e-07,
"loss": 0.2394,
"step": 7950
},
{
"epoch": 76.0,
"learning_rate": 1.0678947368421053e-07,
"loss": 0.2237,
"step": 7975
},
{
"epoch": 76.01,
"learning_rate": 1.0547368421052631e-07,
"loss": 0.2664,
"step": 8000
},
{
"epoch": 76.01,
"eval_loss": 0.4198013246059418,
"eval_runtime": 274.2514,
"eval_samples_per_second": 13.28,
"eval_steps_per_second": 0.416,
"eval_wer": 26.621584322384763,
"step": 8000
},
{
"epoch": 76.01,
"learning_rate": 1.041578947368421e-07,
"loss": 0.2647,
"step": 8025
},
{
"epoch": 77.0,
"learning_rate": 1.028421052631579e-07,
"loss": 0.2419,
"step": 8050
},
{
"epoch": 77.0,
"learning_rate": 1.0152631578947368e-07,
"loss": 0.2202,
"step": 8075
},
{
"epoch": 77.01,
"learning_rate": 1.0021052631578947e-07,
"loss": 0.2496,
"step": 8100
},
{
"epoch": 77.01,
"learning_rate": 9.889473684210525e-08,
"loss": 0.2719,
"step": 8125
},
{
"epoch": 77.01,
"learning_rate": 9.757894736842105e-08,
"loss": 0.2449,
"step": 8150
},
{
"epoch": 78.0,
"learning_rate": 9.626315789473684e-08,
"loss": 0.2229,
"step": 8175
},
{
"epoch": 78.0,
"learning_rate": 9.494736842105262e-08,
"loss": 0.2395,
"step": 8200
},
{
"epoch": 78.01,
"learning_rate": 9.363157894736843e-08,
"loss": 0.276,
"step": 8225
},
{
"epoch": 78.01,
"learning_rate": 9.23157894736842e-08,
"loss": 0.2425,
"step": 8250
},
{
"epoch": 79.0,
"learning_rate": 9.1e-08,
"loss": 0.2317,
"step": 8275
},
{
"epoch": 79.0,
"learning_rate": 8.968421052631579e-08,
"loss": 0.2338,
"step": 8300
},
{
"epoch": 79.01,
"learning_rate": 8.836842105263158e-08,
"loss": 0.2721,
"step": 8325
},
{
"epoch": 79.01,
"learning_rate": 8.705263157894737e-08,
"loss": 0.2434,
"step": 8350
},
{
"epoch": 80.0,
"learning_rate": 8.573684210526314e-08,
"loss": 0.2325,
"step": 8375
},
{
"epoch": 80.0,
"learning_rate": 8.442105263157895e-08,
"loss": 0.2227,
"step": 8400
},
{
"epoch": 80.01,
"learning_rate": 8.310526315789474e-08,
"loss": 0.2711,
"step": 8425
},
{
"epoch": 80.01,
"learning_rate": 8.178947368421052e-08,
"loss": 0.2464,
"step": 8450
},
{
"epoch": 81.0,
"learning_rate": 8.047368421052632e-08,
"loss": 0.2333,
"step": 8475
},
{
"epoch": 81.0,
"learning_rate": 7.91578947368421e-08,
"loss": 0.223,
"step": 8500
},
{
"epoch": 81.0,
"eval_loss": 0.4198976457118988,
"eval_runtime": 275.2042,
"eval_samples_per_second": 13.234,
"eval_steps_per_second": 0.414,
"eval_wer": 26.644585518446963,
"step": 8500
},
{
"epoch": 81.01,
"learning_rate": 7.784210526315789e-08,
"loss": 0.2604,
"step": 8525
},
{
"epoch": 81.01,
"learning_rate": 7.652631578947368e-08,
"loss": 0.2553,
"step": 8550
},
{
"epoch": 82.0,
"learning_rate": 7.521052631578947e-08,
"loss": 0.2346,
"step": 8575
},
{
"epoch": 82.0,
"learning_rate": 7.389473684210526e-08,
"loss": 0.2213,
"step": 8600
},
{
"epoch": 82.01,
"learning_rate": 7.257894736842104e-08,
"loss": 0.2514,
"step": 8625
},
{
"epoch": 82.01,
"learning_rate": 7.126315789473684e-08,
"loss": 0.2606,
"step": 8650
},
{
"epoch": 83.0,
"learning_rate": 6.994736842105264e-08,
"loss": 0.2402,
"step": 8675
},
{
"epoch": 83.0,
"learning_rate": 6.863157894736841e-08,
"loss": 0.2145,
"step": 8700
},
{
"epoch": 83.01,
"learning_rate": 6.73157894736842e-08,
"loss": 0.2445,
"step": 8725
},
{
"epoch": 83.01,
"learning_rate": 6.6e-08,
"loss": 0.266,
"step": 8750
},
{
"epoch": 83.01,
"learning_rate": 6.468421052631579e-08,
"loss": 0.2387,
"step": 8775
},
{
"epoch": 84.0,
"learning_rate": 6.336842105263158e-08,
"loss": 0.2232,
"step": 8800
},
{
"epoch": 84.0,
"learning_rate": 6.205263157894737e-08,
"loss": 0.2283,
"step": 8825
},
{
"epoch": 84.01,
"learning_rate": 6.073684210526316e-08,
"loss": 0.272,
"step": 8850
},
{
"epoch": 84.01,
"learning_rate": 5.9421052631578943e-08,
"loss": 0.2404,
"step": 8875
},
{
"epoch": 85.0,
"learning_rate": 5.8105263157894734e-08,
"loss": 0.2254,
"step": 8900
},
{
"epoch": 85.0,
"learning_rate": 5.6789473684210525e-08,
"loss": 0.2185,
"step": 8925
},
{
"epoch": 85.01,
"learning_rate": 5.547368421052631e-08,
"loss": 0.2691,
"step": 8950
},
{
"epoch": 85.01,
"learning_rate": 5.41578947368421e-08,
"loss": 0.2427,
"step": 8975
},
{
"epoch": 86.0,
"learning_rate": 5.284210526315789e-08,
"loss": 0.2309,
"step": 9000
},
{
"epoch": 86.0,
"eval_loss": 0.4199080169200897,
"eval_runtime": 273.9651,
"eval_samples_per_second": 13.294,
"eval_steps_per_second": 0.416,
"eval_wer": 26.653785996871836,
"step": 9000
},
{
"epoch": 86.0,
"learning_rate": 5.152631578947368e-08,
"loss": 0.2141,
"step": 9025
},
{
"epoch": 86.01,
"learning_rate": 5.021052631578947e-08,
"loss": 0.2636,
"step": 9050
},
{
"epoch": 86.01,
"learning_rate": 4.889473684210526e-08,
"loss": 0.2499,
"step": 9075
},
{
"epoch": 87.0,
"learning_rate": 4.757894736842105e-08,
"loss": 0.2293,
"step": 9100
},
{
"epoch": 87.0,
"learning_rate": 4.631578947368421e-08,
"loss": 0.2155,
"step": 9125
},
{
"epoch": 87.01,
"learning_rate": 4.5e-08,
"loss": 0.2538,
"step": 9150
},
{
"epoch": 87.01,
"learning_rate": 4.368421052631579e-08,
"loss": 0.2573,
"step": 9175
},
{
"epoch": 88.0,
"learning_rate": 4.2368421052631575e-08,
"loss": 0.2352,
"step": 9200
},
{
"epoch": 88.0,
"learning_rate": 4.1052631578947365e-08,
"loss": 0.2161,
"step": 9225
},
{
"epoch": 88.01,
"learning_rate": 3.9736842105263156e-08,
"loss": 0.2418,
"step": 9250
},
{
"epoch": 88.01,
"learning_rate": 3.842105263157895e-08,
"loss": 0.2608,
"step": 9275
},
{
"epoch": 88.01,
"learning_rate": 3.710526315789474e-08,
"loss": 0.2367,
"step": 9300
},
{
"epoch": 89.0,
"learning_rate": 3.578947368421052e-08,
"loss": 0.2161,
"step": 9325
},
{
"epoch": 89.0,
"learning_rate": 3.4473684210526313e-08,
"loss": 0.2328,
"step": 9350
},
{
"epoch": 89.01,
"learning_rate": 3.3157894736842104e-08,
"loss": 0.2615,
"step": 9375
},
{
"epoch": 89.01,
"learning_rate": 3.1842105263157895e-08,
"loss": 0.238,
"step": 9400
},
{
"epoch": 90.0,
"learning_rate": 3.052631578947368e-08,
"loss": 0.223,
"step": 9425
},
{
"epoch": 90.0,
"learning_rate": 2.9210526315789474e-08,
"loss": 0.221,
"step": 9450
},
{
"epoch": 90.01,
"learning_rate": 2.789473684210526e-08,
"loss": 0.2699,
"step": 9475
},
{
"epoch": 90.01,
"learning_rate": 2.6578947368421052e-08,
"loss": 0.233,
"step": 9500
},
{
"epoch": 90.01,
"eval_loss": 0.41979506611824036,
"eval_runtime": 273.0629,
"eval_samples_per_second": 13.338,
"eval_steps_per_second": 0.417,
"eval_wer": 26.607783604747446,
"step": 9500
},
{
"epoch": 91.0,
"learning_rate": 2.526315789473684e-08,
"loss": 0.2299,
"step": 9525
},
{
"epoch": 91.0,
"learning_rate": 2.394736842105263e-08,
"loss": 0.2155,
"step": 9550
},
{
"epoch": 91.01,
"learning_rate": 2.2631578947368422e-08,
"loss": 0.265,
"step": 9575
},
{
"epoch": 91.01,
"learning_rate": 2.131578947368421e-08,
"loss": 0.2429,
"step": 9600
},
{
"epoch": 92.0,
"learning_rate": 2e-08,
"loss": 0.229,
"step": 9625
},
{
"epoch": 92.0,
"learning_rate": 1.8684210526315788e-08,
"loss": 0.2157,
"step": 9650
},
{
"epoch": 92.01,
"learning_rate": 1.736842105263158e-08,
"loss": 0.2621,
"step": 9675
},
{
"epoch": 92.01,
"learning_rate": 1.6052631578947367e-08,
"loss": 0.241,
"step": 9700
},
{
"epoch": 93.0,
"learning_rate": 1.4736842105263158e-08,
"loss": 0.2305,
"step": 9725
},
{
"epoch": 93.0,
"learning_rate": 1.3421052631578947e-08,
"loss": 0.2128,
"step": 9750
},
{
"epoch": 93.01,
"learning_rate": 1.2105263157894736e-08,
"loss": 0.2451,
"step": 9775
},
{
"epoch": 93.01,
"learning_rate": 1.0789473684210525e-08,
"loss": 0.2615,
"step": 9800
},
{
"epoch": 94.0,
"learning_rate": 9.473684210526316e-09,
"loss": 0.2299,
"step": 9825
},
{
"epoch": 94.0,
"learning_rate": 8.157894736842106e-09,
"loss": 0.217,
"step": 9850
},
{
"epoch": 94.01,
"learning_rate": 6.842105263157895e-09,
"loss": 0.2343,
"step": 9875
},
{
"epoch": 94.01,
"learning_rate": 5.526315789473683e-09,
"loss": 0.2592,
"step": 9900
},
{
"epoch": 94.01,
"learning_rate": 4.210526315789473e-09,
"loss": 0.2389,
"step": 9925
},
{
"epoch": 95.0,
"learning_rate": 2.894736842105263e-09,
"loss": 0.2161,
"step": 9950
},
{
"epoch": 95.0,
"learning_rate": 1.5789473684210525e-09,
"loss": 0.2254,
"step": 9975
},
{
"epoch": 95.01,
"learning_rate": 2.631578947368421e-10,
"loss": 0.2647,
"step": 10000
},
{
"epoch": 95.01,
"eval_loss": 0.4198123514652252,
"eval_runtime": 274.1405,
"eval_samples_per_second": 13.285,
"eval_steps_per_second": 0.416,
"eval_wer": 26.621584322384763,
"step": 10000
},
{
"epoch": 95.01,
"step": 10000,
"total_flos": 4.14611039453184e+19,
"train_loss": 0.3942032409667969,
"train_runtime": 37077.489,
"train_samples_per_second": 17.261,
"train_steps_per_second": 0.27
}
],
"logging_steps": 25,
"max_steps": 10000,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 4.14611039453184e+19,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}