whisper-small-nl-dy-augmentation / trainer_state.json
wangpuupup's picture
Upload 3 files
c17c480
raw
history blame
53.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 19.0386,
"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": 5e-06,
"loss": 2.5685,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 1e-05,
"loss": 0.9008,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.5e-05,
"loss": 0.6435,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 2e-05,
"loss": 0.5853,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 2.5e-05,
"loss": 0.399,
"step": 125
},
{
"epoch": 0.01,
"learning_rate": 3e-05,
"loss": 0.5344,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 3.5e-05,
"loss": 0.3954,
"step": 175
},
{
"epoch": 0.02,
"learning_rate": 4e-05,
"loss": 0.3485,
"step": 200
},
{
"epoch": 0.02,
"learning_rate": 4.5e-05,
"loss": 0.4053,
"step": 225
},
{
"epoch": 0.03,
"learning_rate": 5e-05,
"loss": 0.3793,
"step": 250
},
{
"epoch": 0.03,
"learning_rate": 5.500000000000001e-05,
"loss": 0.4483,
"step": 275
},
{
"epoch": 0.03,
"learning_rate": 6e-05,
"loss": 0.2139,
"step": 300
},
{
"epoch": 0.03,
"learning_rate": 6.500000000000001e-05,
"loss": 0.2964,
"step": 325
},
{
"epoch": 0.04,
"learning_rate": 7e-05,
"loss": 0.227,
"step": 350
},
{
"epoch": 0.04,
"learning_rate": 7.500000000000001e-05,
"loss": 0.3124,
"step": 375
},
{
"epoch": 0.04,
"learning_rate": 8e-05,
"loss": 0.3604,
"step": 400
},
{
"epoch": 0.04,
"learning_rate": 8.5e-05,
"loss": 0.3426,
"step": 425
},
{
"epoch": 0.04,
"learning_rate": 9e-05,
"loss": 0.3979,
"step": 450
},
{
"epoch": 0.05,
"learning_rate": 9.5e-05,
"loss": 0.1673,
"step": 475
},
{
"epoch": 0.05,
"learning_rate": 0.0001,
"loss": 0.1618,
"step": 500
},
{
"epoch": 0.05,
"eval_loss": 0.3787050247192383,
"eval_runtime": 315.3085,
"eval_samples_per_second": 9.175,
"eval_steps_per_second": 0.289,
"eval_wer": 28.923523015585356,
"step": 500
},
{
"epoch": 1.0,
"learning_rate": 9.973684210526316e-05,
"loss": 0.2244,
"step": 525
},
{
"epoch": 1.0,
"learning_rate": 9.947368421052632e-05,
"loss": 0.3068,
"step": 550
},
{
"epoch": 1.01,
"learning_rate": 9.921052631578947e-05,
"loss": 0.3338,
"step": 575
},
{
"epoch": 1.01,
"learning_rate": 9.894736842105263e-05,
"loss": 0.3125,
"step": 600
},
{
"epoch": 1.01,
"learning_rate": 9.868421052631579e-05,
"loss": 0.3024,
"step": 625
},
{
"epoch": 1.01,
"learning_rate": 9.842105263157894e-05,
"loss": 0.1638,
"step": 650
},
{
"epoch": 1.02,
"learning_rate": 9.815789473684211e-05,
"loss": 0.0933,
"step": 675
},
{
"epoch": 1.02,
"learning_rate": 9.789473684210527e-05,
"loss": 0.1444,
"step": 700
},
{
"epoch": 1.02,
"learning_rate": 9.763157894736843e-05,
"loss": 0.2141,
"step": 725
},
{
"epoch": 1.02,
"learning_rate": 9.736842105263158e-05,
"loss": 0.2199,
"step": 750
},
{
"epoch": 1.03,
"learning_rate": 9.710526315789474e-05,
"loss": 0.2395,
"step": 775
},
{
"epoch": 1.03,
"learning_rate": 9.68421052631579e-05,
"loss": 0.1447,
"step": 800
},
{
"epoch": 1.03,
"learning_rate": 9.657894736842105e-05,
"loss": 0.0801,
"step": 825
},
{
"epoch": 1.03,
"learning_rate": 9.631578947368421e-05,
"loss": 0.0709,
"step": 850
},
{
"epoch": 1.04,
"learning_rate": 9.605263157894737e-05,
"loss": 0.1415,
"step": 875
},
{
"epoch": 1.04,
"learning_rate": 9.578947368421052e-05,
"loss": 0.1883,
"step": 900
},
{
"epoch": 1.04,
"learning_rate": 9.552631578947369e-05,
"loss": 0.1604,
"step": 925
},
{
"epoch": 1.04,
"learning_rate": 9.526315789473685e-05,
"loss": 0.1998,
"step": 950
},
{
"epoch": 1.05,
"learning_rate": 9.5e-05,
"loss": 0.0909,
"step": 975
},
{
"epoch": 1.05,
"learning_rate": 9.473684210526316e-05,
"loss": 0.0583,
"step": 1000
},
{
"epoch": 1.05,
"eval_loss": 0.3731948435306549,
"eval_runtime": 311.5606,
"eval_samples_per_second": 9.286,
"eval_steps_per_second": 0.292,
"eval_wer": 25.77020659659297,
"step": 1000
},
{
"epoch": 2.0,
"learning_rate": 9.447368421052633e-05,
"loss": 0.0799,
"step": 1025
},
{
"epoch": 2.0,
"learning_rate": 9.421052631578949e-05,
"loss": 0.1263,
"step": 1050
},
{
"epoch": 2.01,
"learning_rate": 9.394736842105264e-05,
"loss": 0.152,
"step": 1075
},
{
"epoch": 2.01,
"learning_rate": 9.36842105263158e-05,
"loss": 0.1448,
"step": 1100
},
{
"epoch": 2.01,
"learning_rate": 9.342105263157896e-05,
"loss": 0.1476,
"step": 1125
},
{
"epoch": 2.01,
"learning_rate": 9.315789473684211e-05,
"loss": 0.0733,
"step": 1150
},
{
"epoch": 2.02,
"learning_rate": 9.289473684210527e-05,
"loss": 0.0288,
"step": 1175
},
{
"epoch": 2.02,
"learning_rate": 9.263157894736843e-05,
"loss": 0.0536,
"step": 1200
},
{
"epoch": 2.02,
"learning_rate": 9.236842105263158e-05,
"loss": 0.0732,
"step": 1225
},
{
"epoch": 2.02,
"learning_rate": 9.210526315789474e-05,
"loss": 0.0858,
"step": 1250
},
{
"epoch": 2.03,
"learning_rate": 9.18421052631579e-05,
"loss": 0.069,
"step": 1275
},
{
"epoch": 2.03,
"learning_rate": 9.157894736842105e-05,
"loss": 0.0694,
"step": 1300
},
{
"epoch": 2.03,
"learning_rate": 9.13263157894737e-05,
"loss": 0.0414,
"step": 1325
},
{
"epoch": 2.03,
"learning_rate": 9.106315789473686e-05,
"loss": 0.0233,
"step": 1350
},
{
"epoch": 2.04,
"learning_rate": 9.080000000000001e-05,
"loss": 0.0622,
"step": 1375
},
{
"epoch": 2.04,
"learning_rate": 9.053684210526317e-05,
"loss": 0.0669,
"step": 1400
},
{
"epoch": 2.04,
"learning_rate": 9.027368421052632e-05,
"loss": 0.0718,
"step": 1425
},
{
"epoch": 2.04,
"learning_rate": 9.001052631578948e-05,
"loss": 0.0733,
"step": 1450
},
{
"epoch": 2.05,
"learning_rate": 8.974736842105264e-05,
"loss": 0.0491,
"step": 1475
},
{
"epoch": 2.05,
"learning_rate": 8.94842105263158e-05,
"loss": 0.0382,
"step": 1500
},
{
"epoch": 2.05,
"eval_loss": 0.4001223146915436,
"eval_runtime": 311.7713,
"eval_samples_per_second": 9.279,
"eval_steps_per_second": 0.292,
"eval_wer": 25.46212395795578,
"step": 1500
},
{
"epoch": 3.0,
"learning_rate": 8.922105263157895e-05,
"loss": 0.0368,
"step": 1525
},
{
"epoch": 3.0,
"learning_rate": 8.895789473684211e-05,
"loss": 0.0608,
"step": 1550
},
{
"epoch": 3.01,
"learning_rate": 8.869473684210526e-05,
"loss": 0.0706,
"step": 1575
},
{
"epoch": 3.01,
"learning_rate": 8.843157894736842e-05,
"loss": 0.0698,
"step": 1600
},
{
"epoch": 3.01,
"learning_rate": 8.816842105263158e-05,
"loss": 0.0594,
"step": 1625
},
{
"epoch": 3.01,
"learning_rate": 8.790526315789475e-05,
"loss": 0.0434,
"step": 1650
},
{
"epoch": 3.02,
"learning_rate": 8.76421052631579e-05,
"loss": 0.0236,
"step": 1675
},
{
"epoch": 3.02,
"learning_rate": 8.737894736842106e-05,
"loss": 0.0216,
"step": 1700
},
{
"epoch": 3.02,
"learning_rate": 8.711578947368422e-05,
"loss": 0.0294,
"step": 1725
},
{
"epoch": 3.02,
"learning_rate": 8.685263157894737e-05,
"loss": 0.035,
"step": 1750
},
{
"epoch": 3.03,
"learning_rate": 8.658947368421053e-05,
"loss": 0.0371,
"step": 1775
},
{
"epoch": 3.03,
"learning_rate": 8.632631578947369e-05,
"loss": 0.0325,
"step": 1800
},
{
"epoch": 3.03,
"learning_rate": 8.606315789473684e-05,
"loss": 0.0291,
"step": 1825
},
{
"epoch": 3.03,
"learning_rate": 8.58e-05,
"loss": 0.0146,
"step": 1850
},
{
"epoch": 3.04,
"learning_rate": 8.553684210526315e-05,
"loss": 0.0285,
"step": 1875
},
{
"epoch": 3.04,
"learning_rate": 8.527368421052631e-05,
"loss": 0.0388,
"step": 1900
},
{
"epoch": 3.04,
"learning_rate": 8.501052631578947e-05,
"loss": 0.0357,
"step": 1925
},
{
"epoch": 3.04,
"learning_rate": 8.474736842105262e-05,
"loss": 0.0345,
"step": 1950
},
{
"epoch": 3.05,
"learning_rate": 8.44842105263158e-05,
"loss": 0.0323,
"step": 1975
},
{
"epoch": 3.05,
"learning_rate": 8.422105263157895e-05,
"loss": 0.0316,
"step": 2000
},
{
"epoch": 3.05,
"eval_loss": 0.4081331789493561,
"eval_runtime": 319.4309,
"eval_samples_per_second": 9.057,
"eval_steps_per_second": 0.285,
"eval_wer": 24.700978615440377,
"step": 2000
},
{
"epoch": 4.0,
"learning_rate": 8.395789473684211e-05,
"loss": 0.0148,
"step": 2025
},
{
"epoch": 4.0,
"learning_rate": 8.369473684210526e-05,
"loss": 0.0332,
"step": 2050
},
{
"epoch": 4.01,
"learning_rate": 8.343157894736843e-05,
"loss": 0.0359,
"step": 2075
},
{
"epoch": 4.01,
"learning_rate": 8.316842105263159e-05,
"loss": 0.0437,
"step": 2100
},
{
"epoch": 4.01,
"learning_rate": 8.290526315789475e-05,
"loss": 0.0325,
"step": 2125
},
{
"epoch": 4.01,
"learning_rate": 8.26421052631579e-05,
"loss": 0.0318,
"step": 2150
},
{
"epoch": 4.02,
"learning_rate": 8.237894736842106e-05,
"loss": 0.0199,
"step": 2175
},
{
"epoch": 4.02,
"learning_rate": 8.211578947368422e-05,
"loss": 0.0142,
"step": 2200
},
{
"epoch": 4.02,
"learning_rate": 8.185263157894737e-05,
"loss": 0.0129,
"step": 2225
},
{
"epoch": 4.02,
"learning_rate": 8.158947368421053e-05,
"loss": 0.018,
"step": 2250
},
{
"epoch": 4.03,
"learning_rate": 8.132631578947369e-05,
"loss": 0.0195,
"step": 2275
},
{
"epoch": 4.03,
"learning_rate": 8.106315789473684e-05,
"loss": 0.0202,
"step": 2300
},
{
"epoch": 4.03,
"learning_rate": 8.080000000000001e-05,
"loss": 0.0193,
"step": 2325
},
{
"epoch": 4.03,
"learning_rate": 8.053684210526317e-05,
"loss": 0.0088,
"step": 2350
},
{
"epoch": 4.04,
"learning_rate": 8.027368421052633e-05,
"loss": 0.0139,
"step": 2375
},
{
"epoch": 4.04,
"learning_rate": 8.001052631578948e-05,
"loss": 0.0179,
"step": 2400
},
{
"epoch": 4.04,
"learning_rate": 7.974736842105264e-05,
"loss": 0.0195,
"step": 2425
},
{
"epoch": 4.04,
"learning_rate": 7.94842105263158e-05,
"loss": 0.0224,
"step": 2450
},
{
"epoch": 4.05,
"learning_rate": 7.922105263157895e-05,
"loss": 0.0207,
"step": 2475
},
{
"epoch": 4.05,
"learning_rate": 7.895789473684211e-05,
"loss": 0.0169,
"step": 2500
},
{
"epoch": 4.05,
"eval_loss": 0.4324879050254822,
"eval_runtime": 313.5877,
"eval_samples_per_second": 9.225,
"eval_steps_per_second": 0.29,
"eval_wer": 24.193548387096776,
"step": 2500
},
{
"epoch": 4.05,
"learning_rate": 7.869473684210526e-05,
"loss": 0.0068,
"step": 2525
},
{
"epoch": 5.0,
"learning_rate": 7.843157894736842e-05,
"loss": 0.0206,
"step": 2550
},
{
"epoch": 5.0,
"learning_rate": 7.816842105263158e-05,
"loss": 0.0219,
"step": 2575
},
{
"epoch": 5.01,
"learning_rate": 7.790526315789473e-05,
"loss": 0.0208,
"step": 2600
},
{
"epoch": 5.01,
"learning_rate": 7.764210526315789e-05,
"loss": 0.0244,
"step": 2625
},
{
"epoch": 5.01,
"learning_rate": 7.737894736842105e-05,
"loss": 0.0136,
"step": 2650
},
{
"epoch": 5.01,
"learning_rate": 7.711578947368422e-05,
"loss": 0.0135,
"step": 2675
},
{
"epoch": 5.02,
"learning_rate": 7.685263157894737e-05,
"loss": 0.0048,
"step": 2700
},
{
"epoch": 5.02,
"learning_rate": 7.658947368421053e-05,
"loss": 0.0104,
"step": 2725
},
{
"epoch": 5.02,
"learning_rate": 7.632631578947369e-05,
"loss": 0.0108,
"step": 2750
},
{
"epoch": 5.02,
"learning_rate": 7.606315789473684e-05,
"loss": 0.0123,
"step": 2775
},
{
"epoch": 5.03,
"learning_rate": 7.58e-05,
"loss": 0.0066,
"step": 2800
},
{
"epoch": 5.03,
"learning_rate": 7.553684210526316e-05,
"loss": 0.0088,
"step": 2825
},
{
"epoch": 5.03,
"learning_rate": 7.527368421052631e-05,
"loss": 0.0057,
"step": 2850
},
{
"epoch": 5.03,
"learning_rate": 7.501052631578947e-05,
"loss": 0.0069,
"step": 2875
},
{
"epoch": 5.04,
"learning_rate": 7.474736842105264e-05,
"loss": 0.015,
"step": 2900
},
{
"epoch": 5.04,
"learning_rate": 7.44842105263158e-05,
"loss": 0.0114,
"step": 2925
},
{
"epoch": 5.04,
"learning_rate": 7.422105263157895e-05,
"loss": 0.0096,
"step": 2950
},
{
"epoch": 5.04,
"learning_rate": 7.395789473684211e-05,
"loss": 0.0105,
"step": 2975
},
{
"epoch": 5.05,
"learning_rate": 7.369473684210528e-05,
"loss": 0.0153,
"step": 3000
},
{
"epoch": 5.05,
"eval_loss": 0.4324972629547119,
"eval_runtime": 328.2162,
"eval_samples_per_second": 8.814,
"eval_steps_per_second": 0.277,
"eval_wer": 33.41790503805726,
"step": 3000
},
{
"epoch": 5.05,
"learning_rate": 7.343157894736843e-05,
"loss": 0.0055,
"step": 3025
},
{
"epoch": 6.0,
"learning_rate": 7.316842105263159e-05,
"loss": 0.0118,
"step": 3050
},
{
"epoch": 6.0,
"learning_rate": 7.290526315789475e-05,
"loss": 0.0162,
"step": 3075
},
{
"epoch": 6.01,
"learning_rate": 7.26421052631579e-05,
"loss": 0.016,
"step": 3100
},
{
"epoch": 6.01,
"learning_rate": 7.237894736842106e-05,
"loss": 0.0159,
"step": 3125
},
{
"epoch": 6.01,
"learning_rate": 7.211578947368422e-05,
"loss": 0.0175,
"step": 3150
},
{
"epoch": 6.01,
"learning_rate": 7.185263157894737e-05,
"loss": 0.0144,
"step": 3175
},
{
"epoch": 6.02,
"learning_rate": 7.158947368421053e-05,
"loss": 0.0051,
"step": 3200
},
{
"epoch": 6.02,
"learning_rate": 7.132631578947369e-05,
"loss": 0.0076,
"step": 3225
},
{
"epoch": 6.02,
"learning_rate": 7.106315789473684e-05,
"loss": 0.0058,
"step": 3250
},
{
"epoch": 6.02,
"learning_rate": 7.08e-05,
"loss": 0.0092,
"step": 3275
},
{
"epoch": 6.03,
"learning_rate": 7.053684210526316e-05,
"loss": 0.0082,
"step": 3300
},
{
"epoch": 6.03,
"learning_rate": 7.027368421052631e-05,
"loss": 0.0079,
"step": 3325
},
{
"epoch": 6.03,
"learning_rate": 7.001052631578948e-05,
"loss": 0.0055,
"step": 3350
},
{
"epoch": 6.03,
"learning_rate": 6.974736842105264e-05,
"loss": 0.005,
"step": 3375
},
{
"epoch": 6.04,
"learning_rate": 6.94842105263158e-05,
"loss": 0.0077,
"step": 3400
},
{
"epoch": 6.04,
"learning_rate": 6.922105263157895e-05,
"loss": 0.0065,
"step": 3425
},
{
"epoch": 6.04,
"learning_rate": 6.895789473684211e-05,
"loss": 0.0092,
"step": 3450
},
{
"epoch": 6.04,
"learning_rate": 6.869473684210527e-05,
"loss": 0.0085,
"step": 3475
},
{
"epoch": 6.05,
"learning_rate": 6.843157894736842e-05,
"loss": 0.0074,
"step": 3500
},
{
"epoch": 6.05,
"eval_loss": 0.43665018677711487,
"eval_runtime": 324.495,
"eval_samples_per_second": 8.915,
"eval_steps_per_second": 0.28,
"eval_wer": 23.939833272924975,
"step": 3500
},
{
"epoch": 6.05,
"learning_rate": 6.816842105263158e-05,
"loss": 0.0079,
"step": 3525
},
{
"epoch": 7.0,
"learning_rate": 6.790526315789473e-05,
"loss": 0.0071,
"step": 3550
},
{
"epoch": 7.0,
"learning_rate": 6.764210526315789e-05,
"loss": 0.0083,
"step": 3575
},
{
"epoch": 7.01,
"learning_rate": 6.737894736842105e-05,
"loss": 0.0089,
"step": 3600
},
{
"epoch": 7.01,
"learning_rate": 6.71157894736842e-05,
"loss": 0.0109,
"step": 3625
},
{
"epoch": 7.01,
"learning_rate": 6.685263157894736e-05,
"loss": 0.0102,
"step": 3650
},
{
"epoch": 7.01,
"learning_rate": 6.658947368421053e-05,
"loss": 0.01,
"step": 3675
},
{
"epoch": 7.02,
"learning_rate": 6.632631578947369e-05,
"loss": 0.0039,
"step": 3700
},
{
"epoch": 7.02,
"learning_rate": 6.606315789473684e-05,
"loss": 0.0046,
"step": 3725
},
{
"epoch": 7.02,
"learning_rate": 6.58e-05,
"loss": 0.0095,
"step": 3750
},
{
"epoch": 7.02,
"learning_rate": 6.553684210526316e-05,
"loss": 0.0071,
"step": 3775
},
{
"epoch": 7.03,
"learning_rate": 6.527368421052633e-05,
"loss": 0.0076,
"step": 3800
},
{
"epoch": 7.03,
"learning_rate": 6.501052631578948e-05,
"loss": 0.0056,
"step": 3825
},
{
"epoch": 7.03,
"learning_rate": 6.474736842105264e-05,
"loss": 0.0064,
"step": 3850
},
{
"epoch": 7.03,
"learning_rate": 6.44842105263158e-05,
"loss": 0.0028,
"step": 3875
},
{
"epoch": 7.04,
"learning_rate": 6.422105263157895e-05,
"loss": 0.005,
"step": 3900
},
{
"epoch": 7.04,
"learning_rate": 6.395789473684211e-05,
"loss": 0.0081,
"step": 3925
},
{
"epoch": 7.04,
"learning_rate": 6.369473684210527e-05,
"loss": 0.0052,
"step": 3950
},
{
"epoch": 7.04,
"learning_rate": 6.343157894736842e-05,
"loss": 0.0103,
"step": 3975
},
{
"epoch": 7.05,
"learning_rate": 6.316842105263158e-05,
"loss": 0.0096,
"step": 4000
},
{
"epoch": 7.05,
"eval_loss": 0.4389564096927643,
"eval_runtime": 312.6719,
"eval_samples_per_second": 9.253,
"eval_steps_per_second": 0.291,
"eval_wer": 23.30554548749547,
"step": 4000
},
{
"epoch": 7.05,
"learning_rate": 6.290526315789475e-05,
"loss": 0.006,
"step": 4025
},
{
"epoch": 8.0,
"learning_rate": 6.26421052631579e-05,
"loss": 0.0026,
"step": 4050
},
{
"epoch": 8.0,
"learning_rate": 6.237894736842106e-05,
"loss": 0.0133,
"step": 4075
},
{
"epoch": 8.01,
"learning_rate": 6.211578947368422e-05,
"loss": 0.0094,
"step": 4100
},
{
"epoch": 8.01,
"learning_rate": 6.185263157894737e-05,
"loss": 0.0098,
"step": 4125
},
{
"epoch": 8.01,
"learning_rate": 6.158947368421053e-05,
"loss": 0.007,
"step": 4150
},
{
"epoch": 8.01,
"learning_rate": 6.132631578947369e-05,
"loss": 0.0068,
"step": 4175
},
{
"epoch": 8.02,
"learning_rate": 6.106315789473684e-05,
"loss": 0.0034,
"step": 4200
},
{
"epoch": 8.02,
"learning_rate": 6.08e-05,
"loss": 0.0035,
"step": 4225
},
{
"epoch": 8.02,
"learning_rate": 6.053684210526316e-05,
"loss": 0.0055,
"step": 4250
},
{
"epoch": 8.02,
"learning_rate": 6.0273684210526314e-05,
"loss": 0.0046,
"step": 4275
},
{
"epoch": 8.03,
"learning_rate": 6.001052631578947e-05,
"loss": 0.0069,
"step": 4300
},
{
"epoch": 8.03,
"learning_rate": 5.9747368421052627e-05,
"loss": 0.0049,
"step": 4325
},
{
"epoch": 8.03,
"learning_rate": 5.94842105263158e-05,
"loss": 0.0054,
"step": 4350
},
{
"epoch": 8.03,
"learning_rate": 5.922105263157895e-05,
"loss": 0.0024,
"step": 4375
},
{
"epoch": 8.04,
"learning_rate": 5.895789473684211e-05,
"loss": 0.0063,
"step": 4400
},
{
"epoch": 8.04,
"learning_rate": 5.8694736842105266e-05,
"loss": 0.0059,
"step": 4425
},
{
"epoch": 8.04,
"learning_rate": 5.843157894736843e-05,
"loss": 0.0043,
"step": 4450
},
{
"epoch": 8.04,
"learning_rate": 5.8168421052631586e-05,
"loss": 0.0049,
"step": 4475
},
{
"epoch": 8.05,
"learning_rate": 5.790526315789474e-05,
"loss": 0.0054,
"step": 4500
},
{
"epoch": 8.05,
"eval_loss": 0.4440564513206482,
"eval_runtime": 317.6978,
"eval_samples_per_second": 9.106,
"eval_steps_per_second": 0.286,
"eval_wer": 23.704240666908298,
"step": 4500
},
{
"epoch": 8.05,
"learning_rate": 5.76421052631579e-05,
"loss": 0.0039,
"step": 4525
},
{
"epoch": 8.05,
"learning_rate": 5.7378947368421055e-05,
"loss": 0.0015,
"step": 4550
},
{
"epoch": 9.0,
"learning_rate": 5.711578947368421e-05,
"loss": 0.005,
"step": 4575
},
{
"epoch": 9.0,
"learning_rate": 5.685263157894737e-05,
"loss": 0.0065,
"step": 4600
},
{
"epoch": 9.01,
"learning_rate": 5.6589473684210525e-05,
"loss": 0.0071,
"step": 4625
},
{
"epoch": 9.01,
"learning_rate": 5.632631578947368e-05,
"loss": 0.0041,
"step": 4650
},
{
"epoch": 9.01,
"learning_rate": 5.606315789473684e-05,
"loss": 0.0073,
"step": 4675
},
{
"epoch": 9.01,
"learning_rate": 5.580000000000001e-05,
"loss": 0.0064,
"step": 4700
},
{
"epoch": 9.02,
"learning_rate": 5.5536842105263164e-05,
"loss": 0.0042,
"step": 4725
},
{
"epoch": 9.02,
"learning_rate": 5.527368421052632e-05,
"loss": 0.0037,
"step": 4750
},
{
"epoch": 9.02,
"learning_rate": 5.501052631578948e-05,
"loss": 0.0043,
"step": 4775
},
{
"epoch": 9.02,
"learning_rate": 5.4747368421052634e-05,
"loss": 0.0047,
"step": 4800
},
{
"epoch": 9.03,
"learning_rate": 5.448421052631579e-05,
"loss": 0.0052,
"step": 4825
},
{
"epoch": 9.03,
"learning_rate": 5.4221052631578954e-05,
"loss": 0.0045,
"step": 4850
},
{
"epoch": 9.03,
"learning_rate": 5.395789473684211e-05,
"loss": 0.0033,
"step": 4875
},
{
"epoch": 9.03,
"learning_rate": 5.3694736842105267e-05,
"loss": 0.0034,
"step": 4900
},
{
"epoch": 9.04,
"learning_rate": 5.343157894736842e-05,
"loss": 0.0036,
"step": 4925
},
{
"epoch": 9.04,
"learning_rate": 5.316842105263158e-05,
"loss": 0.0036,
"step": 4950
},
{
"epoch": 9.04,
"learning_rate": 5.2905263157894736e-05,
"loss": 0.0048,
"step": 4975
},
{
"epoch": 9.04,
"learning_rate": 5.264210526315789e-05,
"loss": 0.0032,
"step": 5000
},
{
"epoch": 9.04,
"eval_loss": 0.4493493139743805,
"eval_runtime": 313.6524,
"eval_samples_per_second": 9.224,
"eval_steps_per_second": 0.29,
"eval_wer": 23.26930047118521,
"step": 5000
},
{
"epoch": 9.05,
"learning_rate": 5.237894736842105e-05,
"loss": 0.0041,
"step": 5025
},
{
"epoch": 9.05,
"learning_rate": 5.211578947368422e-05,
"loss": 0.0014,
"step": 5050
},
{
"epoch": 10.0,
"learning_rate": 5.1852631578947376e-05,
"loss": 0.0024,
"step": 5075
},
{
"epoch": 10.0,
"learning_rate": 5.158947368421053e-05,
"loss": 0.0052,
"step": 5100
},
{
"epoch": 10.01,
"learning_rate": 5.132631578947369e-05,
"loss": 0.0052,
"step": 5125
},
{
"epoch": 10.01,
"learning_rate": 5.1063157894736845e-05,
"loss": 0.0058,
"step": 5150
},
{
"epoch": 10.01,
"learning_rate": 5.08e-05,
"loss": 0.0082,
"step": 5175
},
{
"epoch": 10.01,
"learning_rate": 5.053684210526316e-05,
"loss": 0.0054,
"step": 5200
},
{
"epoch": 10.02,
"learning_rate": 5.0273684210526314e-05,
"loss": 0.0017,
"step": 5225
},
{
"epoch": 10.02,
"learning_rate": 5.001052631578947e-05,
"loss": 0.0052,
"step": 5250
},
{
"epoch": 10.02,
"learning_rate": 4.9747368421052634e-05,
"loss": 0.0056,
"step": 5275
},
{
"epoch": 10.02,
"learning_rate": 4.948421052631579e-05,
"loss": 0.0046,
"step": 5300
},
{
"epoch": 10.03,
"learning_rate": 4.9221052631578954e-05,
"loss": 0.0049,
"step": 5325
},
{
"epoch": 10.03,
"learning_rate": 4.895789473684211e-05,
"loss": 0.0027,
"step": 5350
},
{
"epoch": 10.03,
"learning_rate": 4.869473684210527e-05,
"loss": 0.0043,
"step": 5375
},
{
"epoch": 10.03,
"learning_rate": 4.8431578947368423e-05,
"loss": 0.0013,
"step": 5400
},
{
"epoch": 10.04,
"learning_rate": 4.816842105263158e-05,
"loss": 0.0045,
"step": 5425
},
{
"epoch": 10.04,
"learning_rate": 4.7905263157894736e-05,
"loss": 0.0042,
"step": 5450
},
{
"epoch": 10.04,
"learning_rate": 4.76421052631579e-05,
"loss": 0.0023,
"step": 5475
},
{
"epoch": 10.04,
"learning_rate": 4.7378947368421056e-05,
"loss": 0.004,
"step": 5500
},
{
"epoch": 10.04,
"eval_loss": 0.45240676403045654,
"eval_runtime": 316.5904,
"eval_samples_per_second": 9.138,
"eval_steps_per_second": 0.287,
"eval_wer": 23.341790503805726,
"step": 5500
},
{
"epoch": 10.05,
"learning_rate": 4.711578947368421e-05,
"loss": 0.0034,
"step": 5525
},
{
"epoch": 10.05,
"learning_rate": 4.685263157894737e-05,
"loss": 0.0026,
"step": 5550
},
{
"epoch": 11.0,
"learning_rate": 4.6589473684210526e-05,
"loss": 0.0036,
"step": 5575
},
{
"epoch": 11.0,
"learning_rate": 4.632631578947368e-05,
"loss": 0.0038,
"step": 5600
},
{
"epoch": 11.01,
"learning_rate": 4.6063157894736845e-05,
"loss": 0.0058,
"step": 5625
},
{
"epoch": 11.01,
"learning_rate": 4.58e-05,
"loss": 0.0087,
"step": 5650
},
{
"epoch": 11.01,
"learning_rate": 4.553684210526316e-05,
"loss": 0.0038,
"step": 5675
},
{
"epoch": 11.01,
"learning_rate": 4.5273684210526315e-05,
"loss": 0.0054,
"step": 5700
},
{
"epoch": 11.02,
"learning_rate": 4.501052631578948e-05,
"loss": 0.002,
"step": 5725
},
{
"epoch": 11.02,
"learning_rate": 4.4747368421052635e-05,
"loss": 0.0034,
"step": 5750
},
{
"epoch": 11.02,
"learning_rate": 4.448421052631579e-05,
"loss": 0.0032,
"step": 5775
},
{
"epoch": 11.02,
"learning_rate": 4.4221052631578954e-05,
"loss": 0.0036,
"step": 5800
},
{
"epoch": 11.03,
"learning_rate": 4.395789473684211e-05,
"loss": 0.0034,
"step": 5825
},
{
"epoch": 11.03,
"learning_rate": 4.369473684210527e-05,
"loss": 0.0027,
"step": 5850
},
{
"epoch": 11.03,
"learning_rate": 4.3431578947368424e-05,
"loss": 0.0051,
"step": 5875
},
{
"epoch": 11.03,
"learning_rate": 4.316842105263158e-05,
"loss": 0.0012,
"step": 5900
},
{
"epoch": 11.04,
"learning_rate": 4.290526315789474e-05,
"loss": 0.0017,
"step": 5925
},
{
"epoch": 11.04,
"learning_rate": 4.2642105263157893e-05,
"loss": 0.0027,
"step": 5950
},
{
"epoch": 11.04,
"learning_rate": 4.237894736842106e-05,
"loss": 0.0038,
"step": 5975
},
{
"epoch": 11.04,
"learning_rate": 4.211578947368421e-05,
"loss": 0.0048,
"step": 6000
},
{
"epoch": 11.04,
"eval_loss": 0.44982999563217163,
"eval_runtime": 310.927,
"eval_samples_per_second": 9.304,
"eval_steps_per_second": 0.293,
"eval_wer": 23.72236317506343,
"step": 6000
},
{
"epoch": 11.05,
"learning_rate": 4.185263157894737e-05,
"loss": 0.0045,
"step": 6025
},
{
"epoch": 11.05,
"learning_rate": 4.1589473684210526e-05,
"loss": 0.0035,
"step": 6050
},
{
"epoch": 12.0,
"learning_rate": 4.132631578947368e-05,
"loss": 0.0019,
"step": 6075
},
{
"epoch": 12.0,
"learning_rate": 4.106315789473684e-05,
"loss": 0.004,
"step": 6100
},
{
"epoch": 12.01,
"learning_rate": 4.08e-05,
"loss": 0.0026,
"step": 6125
},
{
"epoch": 12.01,
"learning_rate": 4.053684210526316e-05,
"loss": 0.0036,
"step": 6150
},
{
"epoch": 12.01,
"learning_rate": 4.027368421052632e-05,
"loss": 0.0041,
"step": 6175
},
{
"epoch": 12.01,
"learning_rate": 4.001052631578948e-05,
"loss": 0.0031,
"step": 6200
},
{
"epoch": 12.02,
"learning_rate": 3.9747368421052635e-05,
"loss": 0.0018,
"step": 6225
},
{
"epoch": 12.02,
"learning_rate": 3.948421052631579e-05,
"loss": 0.002,
"step": 6250
},
{
"epoch": 12.02,
"learning_rate": 3.922105263157895e-05,
"loss": 0.0014,
"step": 6275
},
{
"epoch": 12.02,
"learning_rate": 3.895789473684211e-05,
"loss": 0.0015,
"step": 6300
},
{
"epoch": 12.03,
"learning_rate": 3.869473684210527e-05,
"loss": 0.0014,
"step": 6325
},
{
"epoch": 12.03,
"learning_rate": 3.8431578947368424e-05,
"loss": 0.0007,
"step": 6350
},
{
"epoch": 12.03,
"learning_rate": 3.816842105263158e-05,
"loss": 0.0015,
"step": 6375
},
{
"epoch": 12.03,
"learning_rate": 3.790526315789474e-05,
"loss": 0.0004,
"step": 6400
},
{
"epoch": 12.04,
"learning_rate": 3.7642105263157894e-05,
"loss": 0.0008,
"step": 6425
},
{
"epoch": 12.04,
"learning_rate": 3.737894736842105e-05,
"loss": 0.0011,
"step": 6450
},
{
"epoch": 12.04,
"learning_rate": 3.7115789473684214e-05,
"loss": 0.0006,
"step": 6475
},
{
"epoch": 12.04,
"learning_rate": 3.685263157894737e-05,
"loss": 0.001,
"step": 6500
},
{
"epoch": 12.04,
"eval_loss": 0.45768216252326965,
"eval_runtime": 315.5007,
"eval_samples_per_second": 9.17,
"eval_steps_per_second": 0.288,
"eval_wer": 22.888727799927512,
"step": 6500
},
{
"epoch": 12.05,
"learning_rate": 3.6589473684210527e-05,
"loss": 0.0021,
"step": 6525
},
{
"epoch": 12.05,
"learning_rate": 3.632631578947368e-05,
"loss": 0.0026,
"step": 6550
},
{
"epoch": 12.05,
"learning_rate": 3.6063157894736846e-05,
"loss": 0.0012,
"step": 6575
},
{
"epoch": 13.0,
"learning_rate": 3.58e-05,
"loss": 0.001,
"step": 6600
},
{
"epoch": 13.0,
"learning_rate": 3.553684210526316e-05,
"loss": 0.0005,
"step": 6625
},
{
"epoch": 13.01,
"learning_rate": 3.527368421052632e-05,
"loss": 0.0014,
"step": 6650
},
{
"epoch": 13.01,
"learning_rate": 3.501052631578948e-05,
"loss": 0.0012,
"step": 6675
},
{
"epoch": 13.01,
"learning_rate": 3.4747368421052636e-05,
"loss": 0.0008,
"step": 6700
},
{
"epoch": 13.01,
"learning_rate": 3.448421052631579e-05,
"loss": 0.0009,
"step": 6725
},
{
"epoch": 13.02,
"learning_rate": 3.422105263157895e-05,
"loss": 0.0006,
"step": 6750
},
{
"epoch": 13.02,
"learning_rate": 3.3957894736842105e-05,
"loss": 0.0008,
"step": 6775
},
{
"epoch": 13.02,
"learning_rate": 3.369473684210526e-05,
"loss": 0.0014,
"step": 6800
},
{
"epoch": 13.02,
"learning_rate": 3.3431578947368425e-05,
"loss": 0.0002,
"step": 6825
},
{
"epoch": 13.03,
"learning_rate": 3.316842105263158e-05,
"loss": 0.0001,
"step": 6850
},
{
"epoch": 13.03,
"learning_rate": 3.290526315789474e-05,
"loss": 0.0011,
"step": 6875
},
{
"epoch": 13.03,
"learning_rate": 3.2642105263157894e-05,
"loss": 0.0005,
"step": 6900
},
{
"epoch": 13.03,
"learning_rate": 3.237894736842105e-05,
"loss": 0.0016,
"step": 6925
},
{
"epoch": 13.04,
"learning_rate": 3.211578947368421e-05,
"loss": 0.0002,
"step": 6950
},
{
"epoch": 13.04,
"learning_rate": 3.185263157894737e-05,
"loss": 0.0003,
"step": 6975
},
{
"epoch": 13.04,
"learning_rate": 3.158947368421053e-05,
"loss": 0.0002,
"step": 7000
},
{
"epoch": 13.04,
"eval_loss": 0.45770931243896484,
"eval_runtime": 310.8089,
"eval_samples_per_second": 9.308,
"eval_steps_per_second": 0.293,
"eval_wer": 22.09133744110185,
"step": 7000
},
{
"epoch": 13.04,
"learning_rate": 3.132631578947369e-05,
"loss": 0.0004,
"step": 7025
},
{
"epoch": 13.05,
"learning_rate": 3.106315789473685e-05,
"loss": 0.0005,
"step": 7050
},
{
"epoch": 13.05,
"learning_rate": 3.08e-05,
"loss": 0.0004,
"step": 7075
},
{
"epoch": 14.0,
"learning_rate": 3.053684210526316e-05,
"loss": 0.0005,
"step": 7100
},
{
"epoch": 14.0,
"learning_rate": 3.0273684210526316e-05,
"loss": 0.0004,
"step": 7125
},
{
"epoch": 14.01,
"learning_rate": 3.0010526315789476e-05,
"loss": 0.0006,
"step": 7150
},
{
"epoch": 14.01,
"learning_rate": 2.9747368421052636e-05,
"loss": 0.0002,
"step": 7175
},
{
"epoch": 14.01,
"learning_rate": 2.9484210526315792e-05,
"loss": 0.0003,
"step": 7200
},
{
"epoch": 14.01,
"learning_rate": 2.922105263157895e-05,
"loss": 0.0002,
"step": 7225
},
{
"epoch": 14.02,
"learning_rate": 2.8957894736842105e-05,
"loss": 0.0004,
"step": 7250
},
{
"epoch": 14.02,
"learning_rate": 2.8694736842105262e-05,
"loss": 0.0002,
"step": 7275
},
{
"epoch": 14.02,
"learning_rate": 2.8431578947368422e-05,
"loss": 0.0001,
"step": 7300
},
{
"epoch": 14.02,
"learning_rate": 2.8168421052631582e-05,
"loss": 0.0001,
"step": 7325
},
{
"epoch": 14.03,
"learning_rate": 2.7905263157894738e-05,
"loss": 0.0001,
"step": 7350
},
{
"epoch": 14.03,
"learning_rate": 2.7642105263157898e-05,
"loss": 0.0003,
"step": 7375
},
{
"epoch": 14.03,
"learning_rate": 2.7378947368421055e-05,
"loss": 0.0001,
"step": 7400
},
{
"epoch": 14.03,
"learning_rate": 2.711578947368421e-05,
"loss": 0.0003,
"step": 7425
},
{
"epoch": 14.04,
"learning_rate": 2.6852631578947368e-05,
"loss": 0.0002,
"step": 7450
},
{
"epoch": 14.04,
"learning_rate": 2.6589473684210524e-05,
"loss": 0.0001,
"step": 7475
},
{
"epoch": 14.04,
"learning_rate": 2.6326315789473687e-05,
"loss": 0.0001,
"step": 7500
},
{
"epoch": 14.04,
"eval_loss": 0.4616090655326843,
"eval_runtime": 310.6542,
"eval_samples_per_second": 9.313,
"eval_steps_per_second": 0.293,
"eval_wer": 22.127582457412107,
"step": 7500
},
{
"epoch": 14.04,
"learning_rate": 2.6063157894736844e-05,
"loss": 0.0001,
"step": 7525
},
{
"epoch": 14.05,
"learning_rate": 2.58e-05,
"loss": 0.0005,
"step": 7550
},
{
"epoch": 14.05,
"learning_rate": 2.553684210526316e-05,
"loss": 0.0002,
"step": 7575
},
{
"epoch": 15.0,
"learning_rate": 2.5273684210526317e-05,
"loss": 0.0002,
"step": 7600
},
{
"epoch": 15.0,
"learning_rate": 2.5010526315789473e-05,
"loss": 0.0002,
"step": 7625
},
{
"epoch": 15.01,
"learning_rate": 2.4747368421052633e-05,
"loss": 0.0003,
"step": 7650
},
{
"epoch": 15.01,
"learning_rate": 2.448421052631579e-05,
"loss": 0.0001,
"step": 7675
},
{
"epoch": 15.01,
"learning_rate": 2.422105263157895e-05,
"loss": 0.0004,
"step": 7700
},
{
"epoch": 15.01,
"learning_rate": 2.3957894736842106e-05,
"loss": 0.0008,
"step": 7725
},
{
"epoch": 15.02,
"learning_rate": 2.3694736842105262e-05,
"loss": 0.0002,
"step": 7750
},
{
"epoch": 15.02,
"learning_rate": 2.3431578947368422e-05,
"loss": 0.0001,
"step": 7775
},
{
"epoch": 15.02,
"learning_rate": 2.3168421052631582e-05,
"loss": 0.0001,
"step": 7800
},
{
"epoch": 15.02,
"learning_rate": 2.290526315789474e-05,
"loss": 0.0001,
"step": 7825
},
{
"epoch": 15.03,
"learning_rate": 2.2642105263157895e-05,
"loss": 0.0001,
"step": 7850
},
{
"epoch": 15.03,
"learning_rate": 2.2378947368421055e-05,
"loss": 0.0003,
"step": 7875
},
{
"epoch": 15.03,
"learning_rate": 2.211578947368421e-05,
"loss": 0.0001,
"step": 7900
},
{
"epoch": 15.03,
"learning_rate": 2.1852631578947368e-05,
"loss": 0.0001,
"step": 7925
},
{
"epoch": 15.04,
"learning_rate": 2.1589473684210525e-05,
"loss": 0.0002,
"step": 7950
},
{
"epoch": 15.04,
"learning_rate": 2.1326315789473684e-05,
"loss": 0.0001,
"step": 7975
},
{
"epoch": 15.04,
"learning_rate": 2.1063157894736844e-05,
"loss": 0.0001,
"step": 8000
},
{
"epoch": 15.04,
"eval_loss": 0.46391692757606506,
"eval_runtime": 317.4946,
"eval_samples_per_second": 9.112,
"eval_steps_per_second": 0.287,
"eval_wer": 22.272562522653136,
"step": 8000
},
{
"epoch": 15.04,
"learning_rate": 2.08e-05,
"loss": 0.0001,
"step": 8025
},
{
"epoch": 15.05,
"learning_rate": 2.053684210526316e-05,
"loss": 0.0001,
"step": 8050
},
{
"epoch": 15.05,
"learning_rate": 2.0273684210526317e-05,
"loss": 0.0001,
"step": 8075
},
{
"epoch": 16.0,
"learning_rate": 2.0010526315789474e-05,
"loss": 0.0003,
"step": 8100
},
{
"epoch": 16.0,
"learning_rate": 1.9747368421052633e-05,
"loss": 0.0001,
"step": 8125
},
{
"epoch": 16.01,
"learning_rate": 1.948421052631579e-05,
"loss": 0.0001,
"step": 8150
},
{
"epoch": 16.01,
"learning_rate": 1.9221052631578946e-05,
"loss": 0.0001,
"step": 8175
},
{
"epoch": 16.01,
"learning_rate": 1.8957894736842106e-05,
"loss": 0.0001,
"step": 8200
},
{
"epoch": 16.01,
"learning_rate": 1.8694736842105266e-05,
"loss": 0.0002,
"step": 8225
},
{
"epoch": 16.02,
"learning_rate": 1.8431578947368423e-05,
"loss": 0.0001,
"step": 8250
},
{
"epoch": 16.02,
"learning_rate": 1.816842105263158e-05,
"loss": 0.0003,
"step": 8275
},
{
"epoch": 16.02,
"learning_rate": 1.790526315789474e-05,
"loss": 0.0,
"step": 8300
},
{
"epoch": 16.02,
"learning_rate": 1.7642105263157896e-05,
"loss": 0.0,
"step": 8325
},
{
"epoch": 16.03,
"learning_rate": 1.7378947368421052e-05,
"loss": 0.0,
"step": 8350
},
{
"epoch": 16.03,
"learning_rate": 1.711578947368421e-05,
"loss": 0.0001,
"step": 8375
},
{
"epoch": 16.03,
"learning_rate": 1.685263157894737e-05,
"loss": 0.0,
"step": 8400
},
{
"epoch": 16.03,
"learning_rate": 1.658947368421053e-05,
"loss": 0.0001,
"step": 8425
},
{
"epoch": 16.04,
"learning_rate": 1.6326315789473685e-05,
"loss": 0.0001,
"step": 8450
},
{
"epoch": 16.04,
"learning_rate": 1.6063157894736845e-05,
"loss": 0.0001,
"step": 8475
},
{
"epoch": 16.04,
"learning_rate": 1.58e-05,
"loss": 0.0001,
"step": 8500
},
{
"epoch": 16.04,
"eval_loss": 0.46621617674827576,
"eval_runtime": 312.5044,
"eval_samples_per_second": 9.257,
"eval_steps_per_second": 0.291,
"eval_wer": 22.109459949256976,
"step": 8500
},
{
"epoch": 16.04,
"learning_rate": 1.5536842105263158e-05,
"loss": 0.0001,
"step": 8525
},
{
"epoch": 16.05,
"learning_rate": 1.5273684210526318e-05,
"loss": 0.0001,
"step": 8550
},
{
"epoch": 16.05,
"learning_rate": 1.5010526315789474e-05,
"loss": 0.0001,
"step": 8575
},
{
"epoch": 16.05,
"learning_rate": 1.4747368421052632e-05,
"loss": 0.0001,
"step": 8600
},
{
"epoch": 17.0,
"learning_rate": 1.4484210526315789e-05,
"loss": 0.0001,
"step": 8625
},
{
"epoch": 17.0,
"learning_rate": 1.4221052631578949e-05,
"loss": 0.0001,
"step": 8650
},
{
"epoch": 17.01,
"learning_rate": 1.3957894736842105e-05,
"loss": 0.0001,
"step": 8675
},
{
"epoch": 17.01,
"learning_rate": 1.3694736842105263e-05,
"loss": 0.0001,
"step": 8700
},
{
"epoch": 17.01,
"learning_rate": 1.3431578947368423e-05,
"loss": 0.0001,
"step": 8725
},
{
"epoch": 17.01,
"learning_rate": 1.316842105263158e-05,
"loss": 0.0001,
"step": 8750
},
{
"epoch": 17.02,
"learning_rate": 1.2905263157894736e-05,
"loss": 0.0001,
"step": 8775
},
{
"epoch": 17.02,
"learning_rate": 1.2642105263157894e-05,
"loss": 0.0,
"step": 8800
},
{
"epoch": 17.02,
"learning_rate": 1.2378947368421054e-05,
"loss": 0.0,
"step": 8825
},
{
"epoch": 17.02,
"learning_rate": 1.211578947368421e-05,
"loss": 0.0,
"step": 8850
},
{
"epoch": 17.03,
"learning_rate": 1.1852631578947369e-05,
"loss": 0.0,
"step": 8875
},
{
"epoch": 17.03,
"learning_rate": 1.1589473684210527e-05,
"loss": 0.0,
"step": 8900
},
{
"epoch": 17.03,
"learning_rate": 1.1326315789473685e-05,
"loss": 0.0001,
"step": 8925
},
{
"epoch": 17.03,
"learning_rate": 1.1063157894736842e-05,
"loss": 0.0001,
"step": 8950
},
{
"epoch": 17.04,
"learning_rate": 1.08e-05,
"loss": 0.0001,
"step": 8975
},
{
"epoch": 17.04,
"learning_rate": 1.0536842105263158e-05,
"loss": 0.0001,
"step": 9000
},
{
"epoch": 17.04,
"eval_loss": 0.4683971703052521,
"eval_runtime": 317.1275,
"eval_samples_per_second": 9.123,
"eval_steps_per_second": 0.287,
"eval_wer": 22.145704965567234,
"step": 9000
},
{
"epoch": 17.04,
"learning_rate": 1.0273684210526316e-05,
"loss": 0.0001,
"step": 9025
},
{
"epoch": 17.04,
"learning_rate": 1.0010526315789474e-05,
"loss": 0.0001,
"step": 9050
},
{
"epoch": 17.05,
"learning_rate": 9.747368421052631e-06,
"loss": 0.0001,
"step": 9075
},
{
"epoch": 17.05,
"learning_rate": 9.48421052631579e-06,
"loss": 0.0001,
"step": 9100
},
{
"epoch": 18.0,
"learning_rate": 9.221052631578949e-06,
"loss": 0.0001,
"step": 9125
},
{
"epoch": 18.0,
"learning_rate": 8.957894736842106e-06,
"loss": 0.0001,
"step": 9150
},
{
"epoch": 18.01,
"learning_rate": 8.694736842105264e-06,
"loss": 0.0001,
"step": 9175
},
{
"epoch": 18.01,
"learning_rate": 8.43157894736842e-06,
"loss": 0.0001,
"step": 9200
},
{
"epoch": 18.01,
"learning_rate": 8.16842105263158e-06,
"loss": 0.0001,
"step": 9225
},
{
"epoch": 18.01,
"learning_rate": 7.905263157894738e-06,
"loss": 0.0001,
"step": 9250
},
{
"epoch": 18.02,
"learning_rate": 7.642105263157895e-06,
"loss": 0.0001,
"step": 9275
},
{
"epoch": 18.02,
"learning_rate": 7.378947368421053e-06,
"loss": 0.0,
"step": 9300
},
{
"epoch": 18.02,
"learning_rate": 7.11578947368421e-06,
"loss": 0.0,
"step": 9325
},
{
"epoch": 18.02,
"learning_rate": 6.8526315789473685e-06,
"loss": 0.0,
"step": 9350
},
{
"epoch": 18.03,
"learning_rate": 6.5894736842105275e-06,
"loss": 0.0,
"step": 9375
},
{
"epoch": 18.03,
"learning_rate": 6.326315789473684e-06,
"loss": 0.0,
"step": 9400
},
{
"epoch": 18.03,
"learning_rate": 6.063157894736842e-06,
"loss": 0.0001,
"step": 9425
},
{
"epoch": 18.03,
"learning_rate": 5.8e-06,
"loss": 0.0001,
"step": 9450
},
{
"epoch": 18.04,
"learning_rate": 5.5368421052631586e-06,
"loss": 0.0001,
"step": 9475
},
{
"epoch": 18.04,
"learning_rate": 5.273684210526316e-06,
"loss": 0.0001,
"step": 9500
},
{
"epoch": 18.04,
"eval_loss": 0.46967026591300964,
"eval_runtime": 319.0877,
"eval_samples_per_second": 9.066,
"eval_steps_per_second": 0.285,
"eval_wer": 22.145704965567234,
"step": 9500
},
{
"epoch": 18.04,
"learning_rate": 5.010526315789474e-06,
"loss": 0.0001,
"step": 9525
},
{
"epoch": 18.04,
"learning_rate": 4.747368421052631e-06,
"loss": 0.0001,
"step": 9550
},
{
"epoch": 18.05,
"learning_rate": 4.48421052631579e-06,
"loss": 0.0001,
"step": 9575
},
{
"epoch": 18.05,
"learning_rate": 4.221052631578948e-06,
"loss": 0.0001,
"step": 9600
},
{
"epoch": 19.0,
"learning_rate": 3.957894736842106e-06,
"loss": 0.0001,
"step": 9625
},
{
"epoch": 19.0,
"learning_rate": 3.6947368421052633e-06,
"loss": 0.0001,
"step": 9650
},
{
"epoch": 19.01,
"learning_rate": 3.431578947368421e-06,
"loss": 0.0001,
"step": 9675
},
{
"epoch": 19.01,
"learning_rate": 3.168421052631579e-06,
"loss": 0.0001,
"step": 9700
},
{
"epoch": 19.01,
"learning_rate": 2.905263157894737e-06,
"loss": 0.0001,
"step": 9725
},
{
"epoch": 19.01,
"learning_rate": 2.6421052631578948e-06,
"loss": 0.0001,
"step": 9750
},
{
"epoch": 19.02,
"learning_rate": 2.378947368421053e-06,
"loss": 0.0001,
"step": 9775
},
{
"epoch": 19.02,
"learning_rate": 2.1157894736842107e-06,
"loss": 0.0,
"step": 9800
},
{
"epoch": 19.02,
"learning_rate": 1.8526315789473687e-06,
"loss": 0.0,
"step": 9825
},
{
"epoch": 19.02,
"learning_rate": 1.5894736842105265e-06,
"loss": 0.0,
"step": 9850
},
{
"epoch": 19.03,
"learning_rate": 1.3263157894736842e-06,
"loss": 0.0,
"step": 9875
},
{
"epoch": 19.03,
"learning_rate": 1.0631578947368422e-06,
"loss": 0.0,
"step": 9900
},
{
"epoch": 19.03,
"learning_rate": 8.000000000000001e-07,
"loss": 0.0,
"step": 9925
},
{
"epoch": 19.03,
"learning_rate": 5.368421052631579e-07,
"loss": 0.0001,
"step": 9950
},
{
"epoch": 19.04,
"learning_rate": 2.736842105263158e-07,
"loss": 0.0001,
"step": 9975
},
{
"epoch": 19.04,
"learning_rate": 1.0526315789473686e-08,
"loss": 0.0001,
"step": 10000
},
{
"epoch": 19.04,
"eval_loss": 0.47016531229019165,
"eval_runtime": 311.2623,
"eval_samples_per_second": 9.294,
"eval_steps_per_second": 0.292,
"eval_wer": 22.163827473722364,
"step": 10000
},
{
"epoch": 19.04,
"step": 10000,
"total_flos": 1.8463434185613312e+20,
"train_loss": 0.043854061097936936,
"train_runtime": 43314.2602,
"train_samples_per_second": 14.776,
"train_steps_per_second": 0.231
}
],
"max_steps": 10000,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.8463434185613312e+20,
"trial_name": null,
"trial_params": null
}