TheRains's picture
End of training
65d1842
raw
history blame
25.8 kB
{
"best_metric": 28.224974200206397,
"best_model_checkpoint": "./output/small/yt-special-batch12-lr4-small-halfcleantest/checkpoint-5000",
"epoch": 1.2883277505797475,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.2000000000000004e-06,
"loss": 2.1789,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.2e-06,
"loss": 1.4684,
"step": 50
},
{
"epoch": 0.02,
"learning_rate": 1.42e-05,
"loss": 1.018,
"step": 75
},
{
"epoch": 0.03,
"learning_rate": 1.9200000000000003e-05,
"loss": 0.9873,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.4200000000000002e-05,
"loss": 0.917,
"step": 125
},
{
"epoch": 0.04,
"learning_rate": 2.9199999999999998e-05,
"loss": 1.0185,
"step": 150
},
{
"epoch": 0.05,
"learning_rate": 3.4200000000000005e-05,
"loss": 0.9819,
"step": 175
},
{
"epoch": 0.05,
"learning_rate": 3.9200000000000004e-05,
"loss": 1.1265,
"step": 200
},
{
"epoch": 0.06,
"learning_rate": 4.4200000000000004e-05,
"loss": 1.0499,
"step": 225
},
{
"epoch": 0.06,
"learning_rate": 4.92e-05,
"loss": 1.1659,
"step": 250
},
{
"epoch": 0.07,
"learning_rate": 5.420000000000001e-05,
"loss": 1.0453,
"step": 275
},
{
"epoch": 0.08,
"learning_rate": 5.92e-05,
"loss": 1.162,
"step": 300
},
{
"epoch": 0.08,
"learning_rate": 6.42e-05,
"loss": 1.1754,
"step": 325
},
{
"epoch": 0.09,
"learning_rate": 6.92e-05,
"loss": 1.2576,
"step": 350
},
{
"epoch": 0.1,
"learning_rate": 7.42e-05,
"loss": 1.2823,
"step": 375
},
{
"epoch": 0.1,
"learning_rate": 7.920000000000001e-05,
"loss": 1.4194,
"step": 400
},
{
"epoch": 0.11,
"learning_rate": 8.42e-05,
"loss": 1.3906,
"step": 425
},
{
"epoch": 0.12,
"learning_rate": 8.92e-05,
"loss": 1.3488,
"step": 450
},
{
"epoch": 0.12,
"learning_rate": 9.42e-05,
"loss": 1.3076,
"step": 475
},
{
"epoch": 0.13,
"learning_rate": 9.92e-05,
"loss": 1.4206,
"step": 500
},
{
"epoch": 0.14,
"learning_rate": 9.953333333333333e-05,
"loss": 1.3911,
"step": 525
},
{
"epoch": 0.14,
"learning_rate": 9.897777777777778e-05,
"loss": 1.405,
"step": 550
},
{
"epoch": 0.15,
"learning_rate": 9.842222222222223e-05,
"loss": 1.3885,
"step": 575
},
{
"epoch": 0.15,
"learning_rate": 9.786666666666667e-05,
"loss": 1.455,
"step": 600
},
{
"epoch": 0.16,
"learning_rate": 9.731111111111112e-05,
"loss": 1.479,
"step": 625
},
{
"epoch": 0.17,
"learning_rate": 9.675555555555556e-05,
"loss": 1.344,
"step": 650
},
{
"epoch": 0.17,
"learning_rate": 9.620000000000001e-05,
"loss": 1.4103,
"step": 675
},
{
"epoch": 0.18,
"learning_rate": 9.564444444444444e-05,
"loss": 1.5148,
"step": 700
},
{
"epoch": 0.19,
"learning_rate": 9.50888888888889e-05,
"loss": 1.446,
"step": 725
},
{
"epoch": 0.19,
"learning_rate": 9.453333333333335e-05,
"loss": 1.3418,
"step": 750
},
{
"epoch": 0.2,
"learning_rate": 9.397777777777778e-05,
"loss": 1.5833,
"step": 775
},
{
"epoch": 0.21,
"learning_rate": 9.342222222222222e-05,
"loss": 1.4566,
"step": 800
},
{
"epoch": 0.21,
"learning_rate": 9.286666666666667e-05,
"loss": 1.3978,
"step": 825
},
{
"epoch": 0.22,
"learning_rate": 9.231111111111112e-05,
"loss": 1.3507,
"step": 850
},
{
"epoch": 0.23,
"learning_rate": 9.175555555555556e-05,
"loss": 1.3904,
"step": 875
},
{
"epoch": 0.23,
"learning_rate": 9.120000000000001e-05,
"loss": 1.333,
"step": 900
},
{
"epoch": 0.24,
"learning_rate": 9.064444444444446e-05,
"loss": 1.3419,
"step": 925
},
{
"epoch": 0.24,
"learning_rate": 9.00888888888889e-05,
"loss": 1.3867,
"step": 950
},
{
"epoch": 0.25,
"learning_rate": 8.953333333333333e-05,
"loss": 1.326,
"step": 975
},
{
"epoch": 0.26,
"learning_rate": 8.897777777777778e-05,
"loss": 1.363,
"step": 1000
},
{
"epoch": 0.26,
"eval_loss": 1.111358642578125,
"eval_runtime": 346.8465,
"eval_samples_per_second": 4.55,
"eval_steps_per_second": 0.758,
"eval_wer": 85.94771241830065,
"step": 1000
},
{
"epoch": 0.26,
"learning_rate": 8.842222222222223e-05,
"loss": 1.4236,
"step": 1025
},
{
"epoch": 0.27,
"learning_rate": 8.786666666666667e-05,
"loss": 1.3623,
"step": 1050
},
{
"epoch": 0.28,
"learning_rate": 8.731111111111111e-05,
"loss": 1.2477,
"step": 1075
},
{
"epoch": 0.28,
"learning_rate": 8.675555555555556e-05,
"loss": 1.3394,
"step": 1100
},
{
"epoch": 0.29,
"learning_rate": 8.620000000000001e-05,
"loss": 1.3311,
"step": 1125
},
{
"epoch": 0.3,
"learning_rate": 8.564444444444445e-05,
"loss": 1.234,
"step": 1150
},
{
"epoch": 0.3,
"learning_rate": 8.50888888888889e-05,
"loss": 1.2877,
"step": 1175
},
{
"epoch": 0.31,
"learning_rate": 8.453333333333335e-05,
"loss": 1.2526,
"step": 1200
},
{
"epoch": 0.32,
"learning_rate": 8.397777777777778e-05,
"loss": 1.2912,
"step": 1225
},
{
"epoch": 0.32,
"learning_rate": 8.342222222222222e-05,
"loss": 1.2711,
"step": 1250
},
{
"epoch": 0.33,
"learning_rate": 8.286666666666667e-05,
"loss": 1.3246,
"step": 1275
},
{
"epoch": 0.33,
"learning_rate": 8.231111111111112e-05,
"loss": 1.2938,
"step": 1300
},
{
"epoch": 0.34,
"learning_rate": 8.175555555555556e-05,
"loss": 1.2736,
"step": 1325
},
{
"epoch": 0.35,
"learning_rate": 8.120000000000001e-05,
"loss": 1.2443,
"step": 1350
},
{
"epoch": 0.35,
"learning_rate": 8.064444444444445e-05,
"loss": 1.2803,
"step": 1375
},
{
"epoch": 0.36,
"learning_rate": 8.00888888888889e-05,
"loss": 1.1925,
"step": 1400
},
{
"epoch": 0.37,
"learning_rate": 7.953333333333333e-05,
"loss": 1.1486,
"step": 1425
},
{
"epoch": 0.37,
"learning_rate": 7.897777777777778e-05,
"loss": 1.2941,
"step": 1450
},
{
"epoch": 0.38,
"learning_rate": 7.842222222222223e-05,
"loss": 1.2756,
"step": 1475
},
{
"epoch": 0.39,
"learning_rate": 7.786666666666667e-05,
"loss": 1.1799,
"step": 1500
},
{
"epoch": 0.39,
"learning_rate": 7.731111111111111e-05,
"loss": 1.1272,
"step": 1525
},
{
"epoch": 0.4,
"learning_rate": 7.675555555555556e-05,
"loss": 1.2084,
"step": 1550
},
{
"epoch": 0.41,
"learning_rate": 7.620000000000001e-05,
"loss": 1.3061,
"step": 1575
},
{
"epoch": 0.41,
"learning_rate": 7.564444444444445e-05,
"loss": 1.1156,
"step": 1600
},
{
"epoch": 0.42,
"learning_rate": 7.50888888888889e-05,
"loss": 1.209,
"step": 1625
},
{
"epoch": 0.43,
"learning_rate": 7.453333333333333e-05,
"loss": 1.148,
"step": 1650
},
{
"epoch": 0.43,
"learning_rate": 7.397777777777778e-05,
"loss": 1.1437,
"step": 1675
},
{
"epoch": 0.44,
"learning_rate": 7.342222222222222e-05,
"loss": 1.2291,
"step": 1700
},
{
"epoch": 0.44,
"learning_rate": 7.286666666666667e-05,
"loss": 1.1929,
"step": 1725
},
{
"epoch": 0.45,
"learning_rate": 7.231111111111112e-05,
"loss": 1.104,
"step": 1750
},
{
"epoch": 0.46,
"learning_rate": 7.175555555555556e-05,
"loss": 1.2001,
"step": 1775
},
{
"epoch": 0.46,
"learning_rate": 7.12e-05,
"loss": 1.2435,
"step": 1800
},
{
"epoch": 0.47,
"learning_rate": 7.064444444444445e-05,
"loss": 1.1789,
"step": 1825
},
{
"epoch": 0.48,
"learning_rate": 7.00888888888889e-05,
"loss": 1.265,
"step": 1850
},
{
"epoch": 0.48,
"learning_rate": 6.953333333333333e-05,
"loss": 1.0453,
"step": 1875
},
{
"epoch": 0.49,
"learning_rate": 6.897777777777778e-05,
"loss": 1.1507,
"step": 1900
},
{
"epoch": 0.5,
"learning_rate": 6.842222222222222e-05,
"loss": 1.1322,
"step": 1925
},
{
"epoch": 0.5,
"learning_rate": 6.786666666666667e-05,
"loss": 1.1704,
"step": 1950
},
{
"epoch": 0.51,
"learning_rate": 6.731111111111111e-05,
"loss": 1.2049,
"step": 1975
},
{
"epoch": 0.52,
"learning_rate": 6.675555555555556e-05,
"loss": 1.0764,
"step": 2000
},
{
"epoch": 0.52,
"eval_loss": 0.8777403235435486,
"eval_runtime": 277.2991,
"eval_samples_per_second": 5.691,
"eval_steps_per_second": 0.948,
"eval_wer": 63.47609219126247,
"step": 2000
},
{
"epoch": 0.52,
"learning_rate": 6.620000000000001e-05,
"loss": 1.0643,
"step": 2025
},
{
"epoch": 0.53,
"learning_rate": 6.564444444444445e-05,
"loss": 0.9919,
"step": 2050
},
{
"epoch": 0.53,
"learning_rate": 6.508888888888888e-05,
"loss": 1.1142,
"step": 2075
},
{
"epoch": 0.54,
"learning_rate": 6.453333333333333e-05,
"loss": 1.1743,
"step": 2100
},
{
"epoch": 0.55,
"learning_rate": 6.397777777777778e-05,
"loss": 1.0676,
"step": 2125
},
{
"epoch": 0.55,
"learning_rate": 6.342222222222222e-05,
"loss": 1.0886,
"step": 2150
},
{
"epoch": 0.56,
"learning_rate": 6.286666666666667e-05,
"loss": 1.0819,
"step": 2175
},
{
"epoch": 0.57,
"learning_rate": 6.231111111111111e-05,
"loss": 0.9847,
"step": 2200
},
{
"epoch": 0.57,
"learning_rate": 6.175555555555556e-05,
"loss": 1.0215,
"step": 2225
},
{
"epoch": 0.58,
"learning_rate": 6.12e-05,
"loss": 1.046,
"step": 2250
},
{
"epoch": 0.59,
"learning_rate": 6.064444444444445e-05,
"loss": 1.0571,
"step": 2275
},
{
"epoch": 0.59,
"learning_rate": 6.00888888888889e-05,
"loss": 1.084,
"step": 2300
},
{
"epoch": 0.6,
"learning_rate": 5.953333333333334e-05,
"loss": 1.1521,
"step": 2325
},
{
"epoch": 0.61,
"learning_rate": 5.897777777777778e-05,
"loss": 1.0532,
"step": 2350
},
{
"epoch": 0.61,
"learning_rate": 5.842222222222222e-05,
"loss": 0.9993,
"step": 2375
},
{
"epoch": 0.62,
"learning_rate": 5.7866666666666666e-05,
"loss": 1.0385,
"step": 2400
},
{
"epoch": 0.62,
"learning_rate": 5.7311111111111117e-05,
"loss": 1.0749,
"step": 2425
},
{
"epoch": 0.63,
"learning_rate": 5.675555555555556e-05,
"loss": 1.083,
"step": 2450
},
{
"epoch": 0.64,
"learning_rate": 5.620000000000001e-05,
"loss": 1.0904,
"step": 2475
},
{
"epoch": 0.64,
"learning_rate": 5.564444444444444e-05,
"loss": 1.0137,
"step": 2500
},
{
"epoch": 0.65,
"learning_rate": 5.508888888888889e-05,
"loss": 0.9908,
"step": 2525
},
{
"epoch": 0.66,
"learning_rate": 5.4533333333333335e-05,
"loss": 1.0105,
"step": 2550
},
{
"epoch": 0.66,
"learning_rate": 5.3977777777777786e-05,
"loss": 0.9957,
"step": 2575
},
{
"epoch": 0.67,
"learning_rate": 5.342222222222223e-05,
"loss": 1.0214,
"step": 2600
},
{
"epoch": 0.68,
"learning_rate": 5.2866666666666666e-05,
"loss": 0.9245,
"step": 2625
},
{
"epoch": 0.68,
"learning_rate": 5.231111111111111e-05,
"loss": 0.9816,
"step": 2650
},
{
"epoch": 0.69,
"learning_rate": 5.1755555555555554e-05,
"loss": 0.9857,
"step": 2675
},
{
"epoch": 0.7,
"learning_rate": 5.1200000000000004e-05,
"loss": 1.0385,
"step": 2700
},
{
"epoch": 0.7,
"learning_rate": 5.064444444444445e-05,
"loss": 0.9415,
"step": 2725
},
{
"epoch": 0.71,
"learning_rate": 5.00888888888889e-05,
"loss": 0.9923,
"step": 2750
},
{
"epoch": 0.72,
"learning_rate": 4.9533333333333336e-05,
"loss": 0.9511,
"step": 2775
},
{
"epoch": 0.72,
"learning_rate": 4.897777777777778e-05,
"loss": 1.011,
"step": 2800
},
{
"epoch": 0.73,
"learning_rate": 4.842222222222222e-05,
"loss": 0.9509,
"step": 2825
},
{
"epoch": 0.73,
"learning_rate": 4.7866666666666674e-05,
"loss": 1.0017,
"step": 2850
},
{
"epoch": 0.74,
"learning_rate": 4.731111111111111e-05,
"loss": 0.9228,
"step": 2875
},
{
"epoch": 0.75,
"learning_rate": 4.675555555555556e-05,
"loss": 0.9577,
"step": 2900
},
{
"epoch": 0.75,
"learning_rate": 4.6200000000000005e-05,
"loss": 0.9699,
"step": 2925
},
{
"epoch": 0.76,
"learning_rate": 4.564444444444444e-05,
"loss": 0.9944,
"step": 2950
},
{
"epoch": 0.77,
"learning_rate": 4.508888888888889e-05,
"loss": 0.9538,
"step": 2975
},
{
"epoch": 0.77,
"learning_rate": 4.4533333333333336e-05,
"loss": 0.9416,
"step": 3000
},
{
"epoch": 0.77,
"eval_loss": 0.6950076818466187,
"eval_runtime": 241.5796,
"eval_samples_per_second": 6.532,
"eval_steps_per_second": 1.089,
"eval_wer": 43.395252837977296,
"step": 3000
},
{
"epoch": 0.78,
"learning_rate": 4.397777777777778e-05,
"loss": 0.8654,
"step": 3025
},
{
"epoch": 0.79,
"learning_rate": 4.3422222222222224e-05,
"loss": 0.9921,
"step": 3050
},
{
"epoch": 0.79,
"learning_rate": 4.286666666666667e-05,
"loss": 0.9004,
"step": 3075
},
{
"epoch": 0.8,
"learning_rate": 4.231111111111111e-05,
"loss": 0.9512,
"step": 3100
},
{
"epoch": 0.81,
"learning_rate": 4.175555555555556e-05,
"loss": 0.923,
"step": 3125
},
{
"epoch": 0.81,
"learning_rate": 4.12e-05,
"loss": 0.9511,
"step": 3150
},
{
"epoch": 0.82,
"learning_rate": 4.064444444444445e-05,
"loss": 0.7917,
"step": 3175
},
{
"epoch": 0.82,
"learning_rate": 4.008888888888889e-05,
"loss": 0.9599,
"step": 3200
},
{
"epoch": 0.83,
"learning_rate": 3.9533333333333337e-05,
"loss": 0.8576,
"step": 3225
},
{
"epoch": 0.84,
"learning_rate": 3.897777777777778e-05,
"loss": 0.8192,
"step": 3250
},
{
"epoch": 0.84,
"learning_rate": 3.8422222222222224e-05,
"loss": 0.8875,
"step": 3275
},
{
"epoch": 0.85,
"learning_rate": 3.786666666666667e-05,
"loss": 0.8694,
"step": 3300
},
{
"epoch": 0.86,
"learning_rate": 3.731111111111111e-05,
"loss": 0.7971,
"step": 3325
},
{
"epoch": 0.86,
"learning_rate": 3.675555555555556e-05,
"loss": 0.8654,
"step": 3350
},
{
"epoch": 0.87,
"learning_rate": 3.62e-05,
"loss": 0.9021,
"step": 3375
},
{
"epoch": 0.88,
"learning_rate": 3.564444444444445e-05,
"loss": 0.9026,
"step": 3400
},
{
"epoch": 0.88,
"learning_rate": 3.5088888888888886e-05,
"loss": 0.8727,
"step": 3425
},
{
"epoch": 0.89,
"learning_rate": 3.453333333333334e-05,
"loss": 0.8259,
"step": 3450
},
{
"epoch": 0.9,
"learning_rate": 3.397777777777778e-05,
"loss": 0.8467,
"step": 3475
},
{
"epoch": 0.9,
"learning_rate": 3.3422222222222224e-05,
"loss": 0.8096,
"step": 3500
},
{
"epoch": 0.91,
"learning_rate": 3.286666666666667e-05,
"loss": 0.8784,
"step": 3525
},
{
"epoch": 0.91,
"learning_rate": 3.231111111111111e-05,
"loss": 0.8185,
"step": 3550
},
{
"epoch": 0.92,
"learning_rate": 3.1755555555555556e-05,
"loss": 0.856,
"step": 3575
},
{
"epoch": 0.93,
"learning_rate": 3.12e-05,
"loss": 0.8323,
"step": 3600
},
{
"epoch": 0.93,
"learning_rate": 3.064444444444445e-05,
"loss": 0.8538,
"step": 3625
},
{
"epoch": 0.94,
"learning_rate": 3.008888888888889e-05,
"loss": 0.8714,
"step": 3650
},
{
"epoch": 0.95,
"learning_rate": 2.9533333333333334e-05,
"loss": 0.8262,
"step": 3675
},
{
"epoch": 0.95,
"learning_rate": 2.897777777777778e-05,
"loss": 0.6751,
"step": 3700
},
{
"epoch": 0.96,
"learning_rate": 2.842222222222222e-05,
"loss": 0.8182,
"step": 3725
},
{
"epoch": 0.97,
"learning_rate": 2.786666666666667e-05,
"loss": 0.8517,
"step": 3750
},
{
"epoch": 0.97,
"learning_rate": 2.7311111111111116e-05,
"loss": 0.7702,
"step": 3775
},
{
"epoch": 0.98,
"learning_rate": 2.6755555555555556e-05,
"loss": 0.753,
"step": 3800
},
{
"epoch": 0.99,
"learning_rate": 2.6200000000000003e-05,
"loss": 0.8402,
"step": 3825
},
{
"epoch": 0.99,
"learning_rate": 2.5644444444444444e-05,
"loss": 0.7936,
"step": 3850
},
{
"epoch": 1.0,
"learning_rate": 2.508888888888889e-05,
"loss": 0.8543,
"step": 3875
},
{
"epoch": 1.0,
"learning_rate": 2.4533333333333334e-05,
"loss": 0.5651,
"step": 3900
},
{
"epoch": 1.01,
"learning_rate": 2.3977777777777778e-05,
"loss": 0.4491,
"step": 3925
},
{
"epoch": 1.02,
"learning_rate": 2.3422222222222222e-05,
"loss": 0.3744,
"step": 3950
},
{
"epoch": 1.02,
"learning_rate": 2.2866666666666666e-05,
"loss": 0.4442,
"step": 3975
},
{
"epoch": 1.03,
"learning_rate": 2.2311111111111113e-05,
"loss": 0.4797,
"step": 4000
},
{
"epoch": 1.03,
"eval_loss": 0.5948242545127869,
"eval_runtime": 224.621,
"eval_samples_per_second": 7.025,
"eval_steps_per_second": 1.171,
"eval_wer": 34.786721706226345,
"step": 4000
},
{
"epoch": 1.04,
"learning_rate": 2.1755555555555557e-05,
"loss": 0.423,
"step": 4025
},
{
"epoch": 1.04,
"learning_rate": 2.12e-05,
"loss": 0.4616,
"step": 4050
},
{
"epoch": 1.05,
"learning_rate": 2.0644444444444447e-05,
"loss": 0.431,
"step": 4075
},
{
"epoch": 1.06,
"learning_rate": 2.008888888888889e-05,
"loss": 0.4234,
"step": 4100
},
{
"epoch": 1.06,
"learning_rate": 1.9533333333333335e-05,
"loss": 0.4697,
"step": 4125
},
{
"epoch": 1.07,
"learning_rate": 1.897777777777778e-05,
"loss": 0.4951,
"step": 4150
},
{
"epoch": 1.08,
"learning_rate": 1.8422222222222222e-05,
"loss": 0.4445,
"step": 4175
},
{
"epoch": 1.08,
"learning_rate": 1.7866666666666666e-05,
"loss": 0.4076,
"step": 4200
},
{
"epoch": 1.09,
"learning_rate": 1.731111111111111e-05,
"loss": 0.3996,
"step": 4225
},
{
"epoch": 1.1,
"learning_rate": 1.6755555555555557e-05,
"loss": 0.4251,
"step": 4250
},
{
"epoch": 1.1,
"learning_rate": 1.62e-05,
"loss": 0.3814,
"step": 4275
},
{
"epoch": 1.11,
"learning_rate": 1.5644444444444444e-05,
"loss": 0.3839,
"step": 4300
},
{
"epoch": 1.11,
"learning_rate": 1.5088888888888888e-05,
"loss": 0.4362,
"step": 4325
},
{
"epoch": 1.12,
"learning_rate": 1.4533333333333335e-05,
"loss": 0.395,
"step": 4350
},
{
"epoch": 1.13,
"learning_rate": 1.3977777777777779e-05,
"loss": 0.3756,
"step": 4375
},
{
"epoch": 1.13,
"learning_rate": 1.3422222222222223e-05,
"loss": 0.432,
"step": 4400
},
{
"epoch": 1.14,
"learning_rate": 1.2888888888888889e-05,
"loss": 0.4587,
"step": 4425
},
{
"epoch": 1.15,
"learning_rate": 1.2333333333333334e-05,
"loss": 0.3781,
"step": 4450
},
{
"epoch": 1.15,
"learning_rate": 1.1777777777777778e-05,
"loss": 0.3629,
"step": 4475
},
{
"epoch": 1.16,
"learning_rate": 1.1222222222222224e-05,
"loss": 0.3563,
"step": 4500
},
{
"epoch": 1.17,
"learning_rate": 1.0666666666666667e-05,
"loss": 0.365,
"step": 4525
},
{
"epoch": 1.17,
"learning_rate": 1.0111111111111111e-05,
"loss": 0.4544,
"step": 4550
},
{
"epoch": 1.18,
"learning_rate": 9.555555555555556e-06,
"loss": 0.3495,
"step": 4575
},
{
"epoch": 1.19,
"learning_rate": 9e-06,
"loss": 0.4056,
"step": 4600
},
{
"epoch": 1.19,
"learning_rate": 8.444444444444446e-06,
"loss": 0.4023,
"step": 4625
},
{
"epoch": 1.2,
"learning_rate": 7.88888888888889e-06,
"loss": 0.396,
"step": 4650
},
{
"epoch": 1.2,
"learning_rate": 7.333333333333334e-06,
"loss": 0.3691,
"step": 4675
},
{
"epoch": 1.21,
"learning_rate": 6.777777777777779e-06,
"loss": 0.4036,
"step": 4700
},
{
"epoch": 1.22,
"learning_rate": 6.222222222222222e-06,
"loss": 0.4391,
"step": 4725
},
{
"epoch": 1.22,
"learning_rate": 5.666666666666667e-06,
"loss": 0.3607,
"step": 4750
},
{
"epoch": 1.23,
"learning_rate": 5.1111111111111115e-06,
"loss": 0.3998,
"step": 4775
},
{
"epoch": 1.24,
"learning_rate": 4.555555555555556e-06,
"loss": 0.3795,
"step": 4800
},
{
"epoch": 1.24,
"learning_rate": 4.000000000000001e-06,
"loss": 0.3987,
"step": 4825
},
{
"epoch": 1.25,
"learning_rate": 3.4444444444444444e-06,
"loss": 0.3567,
"step": 4850
},
{
"epoch": 1.26,
"learning_rate": 2.888888888888889e-06,
"loss": 0.3642,
"step": 4875
},
{
"epoch": 1.26,
"learning_rate": 2.3333333333333336e-06,
"loss": 0.3979,
"step": 4900
},
{
"epoch": 1.27,
"learning_rate": 1.777777777777778e-06,
"loss": 0.389,
"step": 4925
},
{
"epoch": 1.28,
"learning_rate": 1.2222222222222223e-06,
"loss": 0.3782,
"step": 4950
},
{
"epoch": 1.28,
"learning_rate": 6.666666666666667e-07,
"loss": 0.3333,
"step": 4975
},
{
"epoch": 1.29,
"learning_rate": 1.1111111111111112e-07,
"loss": 0.4025,
"step": 5000
},
{
"epoch": 1.29,
"eval_loss": 0.5436723828315735,
"eval_runtime": 216.4987,
"eval_samples_per_second": 7.289,
"eval_steps_per_second": 1.215,
"eval_wer": 28.224974200206397,
"step": 5000
},
{
"epoch": 1.29,
"step": 5000,
"total_flos": 1.731454684839936e+19,
"train_loss": 0.9546309856414795,
"train_runtime": 8040.6439,
"train_samples_per_second": 7.462,
"train_steps_per_second": 0.622
}
],
"max_steps": 5000,
"num_train_epochs": 2,
"total_flos": 1.731454684839936e+19,
"trial_name": null,
"trial_params": null
}