output_dir / trainer_state.json
Kleber's picture
End of training
b2f1396
raw
history blame
25.7 kB
{
"best_metric": 43.75236083594792,
"best_model_checkpoint": "output_dir/checkpoint-5000",
"epoch": 0.19939384271813687,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.4e-07,
"loss": 3.8256,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 9.200000000000001e-07,
"loss": 3.5128,
"step": 50
},
{
"epoch": 0.0,
"learning_rate": 1.42e-06,
"loss": 2.8953,
"step": 75
},
{
"epoch": 0.0,
"learning_rate": 1.9200000000000003e-06,
"loss": 2.3779,
"step": 100
},
{
"epoch": 0.0,
"learning_rate": 2.42e-06,
"loss": 2.065,
"step": 125
},
{
"epoch": 0.01,
"learning_rate": 2.92e-06,
"loss": 1.8663,
"step": 150
},
{
"epoch": 0.01,
"learning_rate": 3.4200000000000007e-06,
"loss": 1.6813,
"step": 175
},
{
"epoch": 0.01,
"learning_rate": 3.920000000000001e-06,
"loss": 1.6136,
"step": 200
},
{
"epoch": 0.01,
"learning_rate": 4.42e-06,
"loss": 1.5066,
"step": 225
},
{
"epoch": 0.01,
"learning_rate": 4.92e-06,
"loss": 1.3971,
"step": 250
},
{
"epoch": 0.01,
"learning_rate": 5.420000000000001e-06,
"loss": 1.3437,
"step": 275
},
{
"epoch": 0.01,
"learning_rate": 5.92e-06,
"loss": 1.2227,
"step": 300
},
{
"epoch": 0.01,
"learning_rate": 6.42e-06,
"loss": 1.2204,
"step": 325
},
{
"epoch": 0.01,
"learning_rate": 6.92e-06,
"loss": 1.1546,
"step": 350
},
{
"epoch": 0.01,
"learning_rate": 7.420000000000001e-06,
"loss": 1.1521,
"step": 375
},
{
"epoch": 0.02,
"learning_rate": 7.92e-06,
"loss": 1.0848,
"step": 400
},
{
"epoch": 0.02,
"learning_rate": 8.42e-06,
"loss": 1.0626,
"step": 425
},
{
"epoch": 0.02,
"learning_rate": 8.920000000000001e-06,
"loss": 1.0112,
"step": 450
},
{
"epoch": 0.02,
"learning_rate": 9.42e-06,
"loss": 0.988,
"step": 475
},
{
"epoch": 0.02,
"learning_rate": 9.920000000000002e-06,
"loss": 0.9656,
"step": 500
},
{
"epoch": 0.02,
"learning_rate": 9.953333333333333e-06,
"loss": 0.9418,
"step": 525
},
{
"epoch": 0.02,
"learning_rate": 9.89777777777778e-06,
"loss": 0.9224,
"step": 550
},
{
"epoch": 0.02,
"learning_rate": 9.842222222222223e-06,
"loss": 0.9478,
"step": 575
},
{
"epoch": 0.02,
"learning_rate": 9.786666666666667e-06,
"loss": 0.8925,
"step": 600
},
{
"epoch": 0.02,
"learning_rate": 9.731111111111113e-06,
"loss": 0.8688,
"step": 625
},
{
"epoch": 0.03,
"learning_rate": 9.675555555555555e-06,
"loss": 0.8847,
"step": 650
},
{
"epoch": 0.03,
"learning_rate": 9.620000000000001e-06,
"loss": 0.8514,
"step": 675
},
{
"epoch": 0.03,
"learning_rate": 9.564444444444445e-06,
"loss": 0.8257,
"step": 700
},
{
"epoch": 0.03,
"learning_rate": 9.508888888888889e-06,
"loss": 0.817,
"step": 725
},
{
"epoch": 0.03,
"learning_rate": 9.453333333333335e-06,
"loss": 0.7702,
"step": 750
},
{
"epoch": 0.03,
"learning_rate": 9.397777777777779e-06,
"loss": 0.8117,
"step": 775
},
{
"epoch": 0.03,
"learning_rate": 9.342222222222223e-06,
"loss": 0.7728,
"step": 800
},
{
"epoch": 0.03,
"learning_rate": 9.286666666666667e-06,
"loss": 0.7943,
"step": 825
},
{
"epoch": 0.03,
"learning_rate": 9.231111111111111e-06,
"loss": 0.7576,
"step": 850
},
{
"epoch": 0.03,
"learning_rate": 9.175555555555557e-06,
"loss": 0.7636,
"step": 875
},
{
"epoch": 0.04,
"learning_rate": 9.12e-06,
"loss": 0.7577,
"step": 900
},
{
"epoch": 0.04,
"learning_rate": 9.064444444444447e-06,
"loss": 0.767,
"step": 925
},
{
"epoch": 0.04,
"learning_rate": 9.008888888888889e-06,
"loss": 0.7275,
"step": 950
},
{
"epoch": 0.04,
"learning_rate": 8.953333333333335e-06,
"loss": 0.7399,
"step": 975
},
{
"epoch": 0.04,
"learning_rate": 8.897777777777779e-06,
"loss": 0.7471,
"step": 1000
},
{
"epoch": 0.04,
"eval_loss": 0.9043671488761902,
"eval_runtime": 5186.5433,
"eval_samples_per_second": 3.126,
"eval_steps_per_second": 0.078,
"eval_wer": 59.29032307799046,
"step": 1000
},
{
"epoch": 0.04,
"learning_rate": 8.842222222222223e-06,
"loss": 0.7158,
"step": 1025
},
{
"epoch": 0.04,
"learning_rate": 8.786666666666668e-06,
"loss": 0.7059,
"step": 1050
},
{
"epoch": 0.04,
"learning_rate": 8.73111111111111e-06,
"loss": 0.7181,
"step": 1075
},
{
"epoch": 0.04,
"learning_rate": 8.675555555555556e-06,
"loss": 0.739,
"step": 1100
},
{
"epoch": 0.04,
"learning_rate": 8.62e-06,
"loss": 0.7154,
"step": 1125
},
{
"epoch": 0.05,
"learning_rate": 8.564444444444445e-06,
"loss": 0.6732,
"step": 1150
},
{
"epoch": 0.05,
"learning_rate": 8.50888888888889e-06,
"loss": 0.6945,
"step": 1175
},
{
"epoch": 0.05,
"learning_rate": 8.453333333333334e-06,
"loss": 0.6883,
"step": 1200
},
{
"epoch": 0.05,
"learning_rate": 8.397777777777778e-06,
"loss": 0.725,
"step": 1225
},
{
"epoch": 0.05,
"learning_rate": 8.342222222222222e-06,
"loss": 0.6912,
"step": 1250
},
{
"epoch": 0.05,
"learning_rate": 8.286666666666668e-06,
"loss": 0.6662,
"step": 1275
},
{
"epoch": 0.05,
"learning_rate": 8.231111111111112e-06,
"loss": 0.6781,
"step": 1300
},
{
"epoch": 0.05,
"learning_rate": 8.175555555555556e-06,
"loss": 0.6669,
"step": 1325
},
{
"epoch": 0.05,
"learning_rate": 8.120000000000002e-06,
"loss": 0.6523,
"step": 1350
},
{
"epoch": 0.05,
"learning_rate": 8.064444444444444e-06,
"loss": 0.6556,
"step": 1375
},
{
"epoch": 0.06,
"learning_rate": 8.00888888888889e-06,
"loss": 0.6521,
"step": 1400
},
{
"epoch": 0.06,
"learning_rate": 7.953333333333334e-06,
"loss": 0.6504,
"step": 1425
},
{
"epoch": 0.06,
"learning_rate": 7.897777777777778e-06,
"loss": 0.6477,
"step": 1450
},
{
"epoch": 0.06,
"learning_rate": 7.842222222222224e-06,
"loss": 0.6385,
"step": 1475
},
{
"epoch": 0.06,
"learning_rate": 7.786666666666666e-06,
"loss": 0.6352,
"step": 1500
},
{
"epoch": 0.06,
"learning_rate": 7.731111111111112e-06,
"loss": 0.6638,
"step": 1525
},
{
"epoch": 0.06,
"learning_rate": 7.675555555555556e-06,
"loss": 0.6207,
"step": 1550
},
{
"epoch": 0.06,
"learning_rate": 7.620000000000001e-06,
"loss": 0.6453,
"step": 1575
},
{
"epoch": 0.06,
"learning_rate": 7.564444444444446e-06,
"loss": 0.5993,
"step": 1600
},
{
"epoch": 0.06,
"learning_rate": 7.50888888888889e-06,
"loss": 0.6559,
"step": 1625
},
{
"epoch": 0.07,
"learning_rate": 7.453333333333334e-06,
"loss": 0.6495,
"step": 1650
},
{
"epoch": 0.07,
"learning_rate": 7.3977777777777786e-06,
"loss": 0.5972,
"step": 1675
},
{
"epoch": 0.07,
"learning_rate": 7.342222222222223e-06,
"loss": 0.6116,
"step": 1700
},
{
"epoch": 0.07,
"learning_rate": 7.2866666666666675e-06,
"loss": 0.62,
"step": 1725
},
{
"epoch": 0.07,
"learning_rate": 7.231111111111112e-06,
"loss": 0.6004,
"step": 1750
},
{
"epoch": 0.07,
"learning_rate": 7.1755555555555556e-06,
"loss": 0.6225,
"step": 1775
},
{
"epoch": 0.07,
"learning_rate": 7.1200000000000004e-06,
"loss": 0.5935,
"step": 1800
},
{
"epoch": 0.07,
"learning_rate": 7.0644444444444445e-06,
"loss": 0.6184,
"step": 1825
},
{
"epoch": 0.07,
"learning_rate": 7.008888888888889e-06,
"loss": 0.602,
"step": 1850
},
{
"epoch": 0.07,
"learning_rate": 6.953333333333334e-06,
"loss": 0.6025,
"step": 1875
},
{
"epoch": 0.08,
"learning_rate": 6.897777777777779e-06,
"loss": 0.5939,
"step": 1900
},
{
"epoch": 0.08,
"learning_rate": 6.842222222222222e-06,
"loss": 0.6152,
"step": 1925
},
{
"epoch": 0.08,
"learning_rate": 6.786666666666667e-06,
"loss": 0.5773,
"step": 1950
},
{
"epoch": 0.08,
"learning_rate": 6.731111111111111e-06,
"loss": 0.5991,
"step": 1975
},
{
"epoch": 0.08,
"learning_rate": 6.675555555555556e-06,
"loss": 0.5987,
"step": 2000
},
{
"epoch": 0.08,
"eval_loss": 0.7523181438446045,
"eval_runtime": 5459.3591,
"eval_samples_per_second": 2.97,
"eval_steps_per_second": 0.074,
"eval_wer": 52.0231882270411,
"step": 2000
},
{
"epoch": 0.08,
"learning_rate": 6.620000000000001e-06,
"loss": 0.5844,
"step": 2025
},
{
"epoch": 0.08,
"learning_rate": 6.564444444444446e-06,
"loss": 0.5849,
"step": 2050
},
{
"epoch": 0.08,
"learning_rate": 6.508888888888889e-06,
"loss": 0.591,
"step": 2075
},
{
"epoch": 0.08,
"learning_rate": 6.453333333333334e-06,
"loss": 0.5725,
"step": 2100
},
{
"epoch": 0.08,
"learning_rate": 6.397777777777778e-06,
"loss": 0.5989,
"step": 2125
},
{
"epoch": 0.09,
"learning_rate": 6.342222222222223e-06,
"loss": 0.5848,
"step": 2150
},
{
"epoch": 0.09,
"learning_rate": 6.286666666666668e-06,
"loss": 0.5574,
"step": 2175
},
{
"epoch": 0.09,
"learning_rate": 6.231111111111111e-06,
"loss": 0.5885,
"step": 2200
},
{
"epoch": 0.09,
"learning_rate": 6.175555555555556e-06,
"loss": 0.5412,
"step": 2225
},
{
"epoch": 0.09,
"learning_rate": 6.120000000000001e-06,
"loss": 0.5748,
"step": 2250
},
{
"epoch": 0.09,
"learning_rate": 6.064444444444445e-06,
"loss": 0.5809,
"step": 2275
},
{
"epoch": 0.09,
"learning_rate": 6.00888888888889e-06,
"loss": 0.5775,
"step": 2300
},
{
"epoch": 0.09,
"learning_rate": 5.9533333333333345e-06,
"loss": 0.5722,
"step": 2325
},
{
"epoch": 0.09,
"learning_rate": 5.897777777777778e-06,
"loss": 0.5922,
"step": 2350
},
{
"epoch": 0.09,
"learning_rate": 5.8422222222222226e-06,
"loss": 0.5839,
"step": 2375
},
{
"epoch": 0.1,
"learning_rate": 5.7866666666666674e-06,
"loss": 0.5619,
"step": 2400
},
{
"epoch": 0.1,
"learning_rate": 5.7311111111111115e-06,
"loss": 0.5533,
"step": 2425
},
{
"epoch": 0.1,
"learning_rate": 5.675555555555556e-06,
"loss": 0.5389,
"step": 2450
},
{
"epoch": 0.1,
"learning_rate": 5.620000000000001e-06,
"loss": 0.5516,
"step": 2475
},
{
"epoch": 0.1,
"learning_rate": 5.5644444444444444e-06,
"loss": 0.556,
"step": 2500
},
{
"epoch": 0.1,
"learning_rate": 5.508888888888889e-06,
"loss": 0.5455,
"step": 2525
},
{
"epoch": 0.1,
"learning_rate": 5.453333333333334e-06,
"loss": 0.5473,
"step": 2550
},
{
"epoch": 0.1,
"learning_rate": 5.397777777777778e-06,
"loss": 0.5358,
"step": 2575
},
{
"epoch": 0.1,
"learning_rate": 5.342222222222223e-06,
"loss": 0.5525,
"step": 2600
},
{
"epoch": 0.1,
"learning_rate": 5.286666666666666e-06,
"loss": 0.5777,
"step": 2625
},
{
"epoch": 0.11,
"learning_rate": 5.231111111111111e-06,
"loss": 0.5523,
"step": 2650
},
{
"epoch": 0.11,
"learning_rate": 5.175555555555556e-06,
"loss": 0.5621,
"step": 2675
},
{
"epoch": 0.11,
"learning_rate": 5.12e-06,
"loss": 0.521,
"step": 2700
},
{
"epoch": 0.11,
"learning_rate": 5.064444444444445e-06,
"loss": 0.542,
"step": 2725
},
{
"epoch": 0.11,
"learning_rate": 5.00888888888889e-06,
"loss": 0.5566,
"step": 2750
},
{
"epoch": 0.11,
"learning_rate": 4.953333333333334e-06,
"loss": 0.5657,
"step": 2775
},
{
"epoch": 0.11,
"learning_rate": 4.897777777777778e-06,
"loss": 0.5579,
"step": 2800
},
{
"epoch": 0.11,
"learning_rate": 4.842222222222223e-06,
"loss": 0.5528,
"step": 2825
},
{
"epoch": 0.11,
"learning_rate": 4.786666666666667e-06,
"loss": 0.5554,
"step": 2850
},
{
"epoch": 0.11,
"learning_rate": 4.731111111111112e-06,
"loss": 0.556,
"step": 2875
},
{
"epoch": 0.12,
"learning_rate": 4.675555555555556e-06,
"loss": 0.5335,
"step": 2900
},
{
"epoch": 0.12,
"learning_rate": 4.620000000000001e-06,
"loss": 0.5579,
"step": 2925
},
{
"epoch": 0.12,
"learning_rate": 4.564444444444445e-06,
"loss": 0.5436,
"step": 2950
},
{
"epoch": 0.12,
"learning_rate": 4.50888888888889e-06,
"loss": 0.5249,
"step": 2975
},
{
"epoch": 0.12,
"learning_rate": 4.453333333333334e-06,
"loss": 0.5168,
"step": 3000
},
{
"epoch": 0.12,
"eval_loss": 0.6889816522598267,
"eval_runtime": 4843.9803,
"eval_samples_per_second": 3.347,
"eval_steps_per_second": 0.084,
"eval_wer": 47.76096391486344,
"step": 3000
},
{
"epoch": 0.12,
"learning_rate": 4.397777777777778e-06,
"loss": 0.5189,
"step": 3025
},
{
"epoch": 0.12,
"learning_rate": 4.3422222222222225e-06,
"loss": 0.5599,
"step": 3050
},
{
"epoch": 0.12,
"learning_rate": 4.2866666666666666e-06,
"loss": 0.5367,
"step": 3075
},
{
"epoch": 0.12,
"learning_rate": 4.2311111111111114e-06,
"loss": 0.528,
"step": 3100
},
{
"epoch": 0.12,
"learning_rate": 4.175555555555556e-06,
"loss": 0.4945,
"step": 3125
},
{
"epoch": 0.13,
"learning_rate": 4.12e-06,
"loss": 0.5145,
"step": 3150
},
{
"epoch": 0.13,
"learning_rate": 4.064444444444444e-06,
"loss": 0.5346,
"step": 3175
},
{
"epoch": 0.13,
"learning_rate": 4.008888888888889e-06,
"loss": 0.5075,
"step": 3200
},
{
"epoch": 0.13,
"learning_rate": 3.953333333333333e-06,
"loss": 0.5444,
"step": 3225
},
{
"epoch": 0.13,
"learning_rate": 3.897777777777778e-06,
"loss": 0.4993,
"step": 3250
},
{
"epoch": 0.13,
"learning_rate": 3.842222222222223e-06,
"loss": 0.5234,
"step": 3275
},
{
"epoch": 0.13,
"learning_rate": 3.7866666666666667e-06,
"loss": 0.5473,
"step": 3300
},
{
"epoch": 0.13,
"learning_rate": 3.7311111111111116e-06,
"loss": 0.5193,
"step": 3325
},
{
"epoch": 0.13,
"learning_rate": 3.675555555555556e-06,
"loss": 0.5115,
"step": 3350
},
{
"epoch": 0.13,
"learning_rate": 3.62e-06,
"loss": 0.5361,
"step": 3375
},
{
"epoch": 0.14,
"learning_rate": 3.564444444444445e-06,
"loss": 0.5508,
"step": 3400
},
{
"epoch": 0.14,
"learning_rate": 3.508888888888889e-06,
"loss": 0.5012,
"step": 3425
},
{
"epoch": 0.14,
"learning_rate": 3.4533333333333334e-06,
"loss": 0.5409,
"step": 3450
},
{
"epoch": 0.14,
"learning_rate": 3.3977777777777783e-06,
"loss": 0.5133,
"step": 3475
},
{
"epoch": 0.14,
"learning_rate": 3.3422222222222224e-06,
"loss": 0.5283,
"step": 3500
},
{
"epoch": 0.14,
"learning_rate": 3.286666666666667e-06,
"loss": 0.528,
"step": 3525
},
{
"epoch": 0.14,
"learning_rate": 3.2311111111111117e-06,
"loss": 0.5424,
"step": 3550
},
{
"epoch": 0.14,
"learning_rate": 3.1755555555555557e-06,
"loss": 0.5166,
"step": 3575
},
{
"epoch": 0.14,
"learning_rate": 3.12e-06,
"loss": 0.5046,
"step": 3600
},
{
"epoch": 0.14,
"learning_rate": 3.064444444444445e-06,
"loss": 0.5139,
"step": 3625
},
{
"epoch": 0.15,
"learning_rate": 3.008888888888889e-06,
"loss": 0.5262,
"step": 3650
},
{
"epoch": 0.15,
"learning_rate": 2.9533333333333336e-06,
"loss": 0.5232,
"step": 3675
},
{
"epoch": 0.15,
"learning_rate": 2.8977777777777785e-06,
"loss": 0.4996,
"step": 3700
},
{
"epoch": 0.15,
"learning_rate": 2.8422222222222225e-06,
"loss": 0.5143,
"step": 3725
},
{
"epoch": 0.15,
"learning_rate": 2.786666666666667e-06,
"loss": 0.5055,
"step": 3750
},
{
"epoch": 0.15,
"learning_rate": 2.7311111111111114e-06,
"loss": 0.4845,
"step": 3775
},
{
"epoch": 0.15,
"learning_rate": 2.675555555555556e-06,
"loss": 0.5279,
"step": 3800
},
{
"epoch": 0.15,
"learning_rate": 2.6200000000000003e-06,
"loss": 0.4984,
"step": 3825
},
{
"epoch": 0.15,
"learning_rate": 2.5644444444444444e-06,
"loss": 0.5428,
"step": 3850
},
{
"epoch": 0.15,
"learning_rate": 2.5088888888888892e-06,
"loss": 0.485,
"step": 3875
},
{
"epoch": 0.16,
"learning_rate": 2.4533333333333333e-06,
"loss": 0.5173,
"step": 3900
},
{
"epoch": 0.16,
"learning_rate": 2.397777777777778e-06,
"loss": 0.5162,
"step": 3925
},
{
"epoch": 0.16,
"learning_rate": 2.342222222222222e-06,
"loss": 0.4972,
"step": 3950
},
{
"epoch": 0.16,
"learning_rate": 2.2866666666666667e-06,
"loss": 0.5033,
"step": 3975
},
{
"epoch": 0.16,
"learning_rate": 2.2311111111111115e-06,
"loss": 0.5082,
"step": 4000
},
{
"epoch": 0.16,
"eval_loss": 0.6595572233200073,
"eval_runtime": 4924.8872,
"eval_samples_per_second": 3.292,
"eval_steps_per_second": 0.082,
"eval_wer": 45.40128429475567,
"step": 4000
},
{
"epoch": 0.16,
"learning_rate": 2.1755555555555556e-06,
"loss": 0.4878,
"step": 4025
},
{
"epoch": 0.16,
"learning_rate": 2.12e-06,
"loss": 0.5226,
"step": 4050
},
{
"epoch": 0.16,
"learning_rate": 2.064444444444445e-06,
"loss": 0.4863,
"step": 4075
},
{
"epoch": 0.16,
"learning_rate": 2.008888888888889e-06,
"loss": 0.5097,
"step": 4100
},
{
"epoch": 0.16,
"learning_rate": 1.9533333333333334e-06,
"loss": 0.5113,
"step": 4125
},
{
"epoch": 0.17,
"learning_rate": 1.8977777777777779e-06,
"loss": 0.5124,
"step": 4150
},
{
"epoch": 0.17,
"learning_rate": 1.8422222222222225e-06,
"loss": 0.5198,
"step": 4175
},
{
"epoch": 0.17,
"learning_rate": 1.7866666666666668e-06,
"loss": 0.5018,
"step": 4200
},
{
"epoch": 0.17,
"learning_rate": 1.7311111111111112e-06,
"loss": 0.4917,
"step": 4225
},
{
"epoch": 0.17,
"learning_rate": 1.675555555555556e-06,
"loss": 0.4952,
"step": 4250
},
{
"epoch": 0.17,
"learning_rate": 1.6200000000000002e-06,
"loss": 0.5068,
"step": 4275
},
{
"epoch": 0.17,
"learning_rate": 1.5644444444444446e-06,
"loss": 0.4884,
"step": 4300
},
{
"epoch": 0.17,
"learning_rate": 1.5088888888888889e-06,
"loss": 0.4925,
"step": 4325
},
{
"epoch": 0.17,
"learning_rate": 1.4533333333333335e-06,
"loss": 0.4859,
"step": 4350
},
{
"epoch": 0.17,
"learning_rate": 1.397777777777778e-06,
"loss": 0.4889,
"step": 4375
},
{
"epoch": 0.18,
"learning_rate": 1.3422222222222222e-06,
"loss": 0.493,
"step": 4400
},
{
"epoch": 0.18,
"learning_rate": 1.286666666666667e-06,
"loss": 0.4818,
"step": 4425
},
{
"epoch": 0.18,
"learning_rate": 1.2311111111111112e-06,
"loss": 0.4829,
"step": 4450
},
{
"epoch": 0.18,
"learning_rate": 1.1755555555555556e-06,
"loss": 0.503,
"step": 4475
},
{
"epoch": 0.18,
"learning_rate": 1.12e-06,
"loss": 0.4801,
"step": 4500
},
{
"epoch": 0.18,
"learning_rate": 1.0644444444444445e-06,
"loss": 0.5011,
"step": 4525
},
{
"epoch": 0.18,
"learning_rate": 1.008888888888889e-06,
"loss": 0.503,
"step": 4550
},
{
"epoch": 0.18,
"learning_rate": 9.533333333333335e-07,
"loss": 0.5109,
"step": 4575
},
{
"epoch": 0.18,
"learning_rate": 8.977777777777778e-07,
"loss": 0.4963,
"step": 4600
},
{
"epoch": 0.18,
"learning_rate": 8.422222222222224e-07,
"loss": 0.5007,
"step": 4625
},
{
"epoch": 0.19,
"learning_rate": 7.866666666666667e-07,
"loss": 0.4924,
"step": 4650
},
{
"epoch": 0.19,
"learning_rate": 7.311111111111112e-07,
"loss": 0.4968,
"step": 4675
},
{
"epoch": 0.19,
"learning_rate": 6.755555555555555e-07,
"loss": 0.5075,
"step": 4700
},
{
"epoch": 0.19,
"learning_rate": 6.200000000000001e-07,
"loss": 0.515,
"step": 4725
},
{
"epoch": 0.19,
"learning_rate": 5.644444444444445e-07,
"loss": 0.4967,
"step": 4750
},
{
"epoch": 0.19,
"learning_rate": 5.088888888888889e-07,
"loss": 0.4789,
"step": 4775
},
{
"epoch": 0.19,
"learning_rate": 4.533333333333334e-07,
"loss": 0.4948,
"step": 4800
},
{
"epoch": 0.19,
"learning_rate": 3.9777777777777783e-07,
"loss": 0.4988,
"step": 4825
},
{
"epoch": 0.19,
"learning_rate": 3.422222222222223e-07,
"loss": 0.513,
"step": 4850
},
{
"epoch": 0.19,
"learning_rate": 2.866666666666667e-07,
"loss": 0.5294,
"step": 4875
},
{
"epoch": 0.2,
"learning_rate": 2.3111111111111112e-07,
"loss": 0.4793,
"step": 4900
},
{
"epoch": 0.2,
"learning_rate": 1.7555555555555558e-07,
"loss": 0.4868,
"step": 4925
},
{
"epoch": 0.2,
"learning_rate": 1.2000000000000002e-07,
"loss": 0.5204,
"step": 4950
},
{
"epoch": 0.2,
"learning_rate": 6.444444444444445e-08,
"loss": 0.4798,
"step": 4975
},
{
"epoch": 0.2,
"learning_rate": 8.88888888888889e-09,
"loss": 0.4748,
"step": 5000
},
{
"epoch": 0.2,
"eval_loss": 0.6424447894096375,
"eval_runtime": 4804.4238,
"eval_samples_per_second": 3.375,
"eval_steps_per_second": 0.085,
"eval_wer": 43.75236083594792,
"step": 5000
},
{
"epoch": 0.2,
"step": 5000,
"total_flos": 5.771708306565366e+19,
"train_loss": 0.6964670913696289,
"train_runtime": 39393.4491,
"train_samples_per_second": 5.077,
"train_steps_per_second": 0.127
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 5.771708306565366e+19,
"trial_name": null,
"trial_params": null
}