yt-special-batch8-lr4-small / trainer_state.json
TheRains's picture
End of training
2229837
{
"best_metric": 43.84281175021075,
"best_model_checkpoint": "./output/small/yt-special-batch8-lr4-small/checkpoint-5000",
"epoch": 0.8588114050154586,
"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": 4.2000000000000004e-06,
"loss": 2.2413,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.2e-06,
"loss": 1.3995,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.42e-05,
"loss": 1.1991,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.9200000000000003e-05,
"loss": 1.0044,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 2.4200000000000002e-05,
"loss": 0.9937,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.9199999999999998e-05,
"loss": 1.061,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 3.4200000000000005e-05,
"loss": 0.9925,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 3.9200000000000004e-05,
"loss": 1.0398,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.4200000000000004e-05,
"loss": 1.2162,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 4.92e-05,
"loss": 1.1954,
"step": 250
},
{
"epoch": 0.05,
"learning_rate": 5.420000000000001e-05,
"loss": 1.2592,
"step": 275
},
{
"epoch": 0.05,
"learning_rate": 5.92e-05,
"loss": 1.2746,
"step": 300
},
{
"epoch": 0.06,
"learning_rate": 6.42e-05,
"loss": 1.2048,
"step": 325
},
{
"epoch": 0.06,
"learning_rate": 6.92e-05,
"loss": 1.3666,
"step": 350
},
{
"epoch": 0.06,
"learning_rate": 7.42e-05,
"loss": 1.5647,
"step": 375
},
{
"epoch": 0.07,
"learning_rate": 7.920000000000001e-05,
"loss": 1.3826,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 8.42e-05,
"loss": 1.3131,
"step": 425
},
{
"epoch": 0.08,
"learning_rate": 8.92e-05,
"loss": 1.5587,
"step": 450
},
{
"epoch": 0.08,
"learning_rate": 9.42e-05,
"loss": 1.6129,
"step": 475
},
{
"epoch": 0.09,
"learning_rate": 9.92e-05,
"loss": 1.6359,
"step": 500
},
{
"epoch": 0.09,
"learning_rate": 9.953333333333333e-05,
"loss": 1.5784,
"step": 525
},
{
"epoch": 0.09,
"learning_rate": 9.897777777777778e-05,
"loss": 1.702,
"step": 550
},
{
"epoch": 0.1,
"learning_rate": 9.842222222222223e-05,
"loss": 1.6738,
"step": 575
},
{
"epoch": 0.1,
"learning_rate": 9.786666666666667e-05,
"loss": 1.7857,
"step": 600
},
{
"epoch": 0.11,
"learning_rate": 9.731111111111112e-05,
"loss": 1.631,
"step": 625
},
{
"epoch": 0.11,
"learning_rate": 9.675555555555556e-05,
"loss": 1.6467,
"step": 650
},
{
"epoch": 0.12,
"learning_rate": 9.620000000000001e-05,
"loss": 1.6083,
"step": 675
},
{
"epoch": 0.12,
"learning_rate": 9.564444444444444e-05,
"loss": 1.5504,
"step": 700
},
{
"epoch": 0.12,
"learning_rate": 9.50888888888889e-05,
"loss": 1.6203,
"step": 725
},
{
"epoch": 0.13,
"learning_rate": 9.453333333333335e-05,
"loss": 1.7001,
"step": 750
},
{
"epoch": 0.13,
"learning_rate": 9.397777777777778e-05,
"loss": 1.5403,
"step": 775
},
{
"epoch": 0.14,
"learning_rate": 9.342222222222222e-05,
"loss": 1.4806,
"step": 800
},
{
"epoch": 0.14,
"learning_rate": 9.286666666666667e-05,
"loss": 1.5636,
"step": 825
},
{
"epoch": 0.15,
"learning_rate": 9.231111111111112e-05,
"loss": 1.4721,
"step": 850
},
{
"epoch": 0.15,
"learning_rate": 9.175555555555556e-05,
"loss": 1.5738,
"step": 875
},
{
"epoch": 0.15,
"learning_rate": 9.120000000000001e-05,
"loss": 1.5799,
"step": 900
},
{
"epoch": 0.16,
"learning_rate": 9.064444444444446e-05,
"loss": 1.6273,
"step": 925
},
{
"epoch": 0.16,
"learning_rate": 9.00888888888889e-05,
"loss": 1.44,
"step": 950
},
{
"epoch": 0.17,
"learning_rate": 8.953333333333333e-05,
"loss": 1.5182,
"step": 975
},
{
"epoch": 0.17,
"learning_rate": 8.897777777777778e-05,
"loss": 1.5084,
"step": 1000
},
{
"epoch": 0.17,
"eval_loss": 1.5453723669052124,
"eval_runtime": 963.9353,
"eval_samples_per_second": 2.543,
"eval_steps_per_second": 0.636,
"eval_wer": 207.42494001686015,
"step": 1000
},
{
"epoch": 0.18,
"learning_rate": 8.842222222222223e-05,
"loss": 1.6378,
"step": 1025
},
{
"epoch": 0.18,
"learning_rate": 8.786666666666667e-05,
"loss": 1.4759,
"step": 1050
},
{
"epoch": 0.18,
"learning_rate": 8.731111111111111e-05,
"loss": 1.569,
"step": 1075
},
{
"epoch": 0.19,
"learning_rate": 8.675555555555556e-05,
"loss": 1.492,
"step": 1100
},
{
"epoch": 0.19,
"learning_rate": 8.620000000000001e-05,
"loss": 1.4954,
"step": 1125
},
{
"epoch": 0.2,
"learning_rate": 8.564444444444445e-05,
"loss": 1.7086,
"step": 1150
},
{
"epoch": 0.2,
"learning_rate": 8.50888888888889e-05,
"loss": 1.6635,
"step": 1175
},
{
"epoch": 0.21,
"learning_rate": 8.453333333333335e-05,
"loss": 1.4504,
"step": 1200
},
{
"epoch": 0.21,
"learning_rate": 8.397777777777778e-05,
"loss": 1.4326,
"step": 1225
},
{
"epoch": 0.21,
"learning_rate": 8.342222222222222e-05,
"loss": 1.5646,
"step": 1250
},
{
"epoch": 0.22,
"learning_rate": 8.286666666666667e-05,
"loss": 1.331,
"step": 1275
},
{
"epoch": 0.22,
"learning_rate": 8.231111111111112e-05,
"loss": 1.449,
"step": 1300
},
{
"epoch": 0.23,
"learning_rate": 8.175555555555556e-05,
"loss": 1.3924,
"step": 1325
},
{
"epoch": 0.23,
"learning_rate": 8.120000000000001e-05,
"loss": 1.4227,
"step": 1350
},
{
"epoch": 0.24,
"learning_rate": 8.064444444444445e-05,
"loss": 1.4399,
"step": 1375
},
{
"epoch": 0.24,
"learning_rate": 8.00888888888889e-05,
"loss": 1.3774,
"step": 1400
},
{
"epoch": 0.24,
"learning_rate": 7.953333333333333e-05,
"loss": 1.4628,
"step": 1425
},
{
"epoch": 0.25,
"learning_rate": 7.897777777777778e-05,
"loss": 1.3502,
"step": 1450
},
{
"epoch": 0.25,
"learning_rate": 7.842222222222223e-05,
"loss": 1.4197,
"step": 1475
},
{
"epoch": 0.26,
"learning_rate": 7.786666666666667e-05,
"loss": 1.3844,
"step": 1500
},
{
"epoch": 0.26,
"learning_rate": 7.731111111111111e-05,
"loss": 1.5141,
"step": 1525
},
{
"epoch": 0.27,
"learning_rate": 7.675555555555556e-05,
"loss": 1.3164,
"step": 1550
},
{
"epoch": 0.27,
"learning_rate": 7.620000000000001e-05,
"loss": 1.3863,
"step": 1575
},
{
"epoch": 0.27,
"learning_rate": 7.564444444444445e-05,
"loss": 1.4064,
"step": 1600
},
{
"epoch": 0.28,
"learning_rate": 7.50888888888889e-05,
"loss": 1.2884,
"step": 1625
},
{
"epoch": 0.28,
"learning_rate": 7.453333333333333e-05,
"loss": 1.3053,
"step": 1650
},
{
"epoch": 0.29,
"learning_rate": 7.397777777777778e-05,
"loss": 1.3211,
"step": 1675
},
{
"epoch": 0.29,
"learning_rate": 7.342222222222222e-05,
"loss": 1.3747,
"step": 1700
},
{
"epoch": 0.3,
"learning_rate": 7.286666666666667e-05,
"loss": 1.2916,
"step": 1725
},
{
"epoch": 0.3,
"learning_rate": 7.231111111111112e-05,
"loss": 1.2992,
"step": 1750
},
{
"epoch": 0.3,
"learning_rate": 7.175555555555556e-05,
"loss": 1.3143,
"step": 1775
},
{
"epoch": 0.31,
"learning_rate": 7.12e-05,
"loss": 1.2348,
"step": 1800
},
{
"epoch": 0.31,
"learning_rate": 7.064444444444445e-05,
"loss": 1.3189,
"step": 1825
},
{
"epoch": 0.32,
"learning_rate": 7.00888888888889e-05,
"loss": 1.4008,
"step": 1850
},
{
"epoch": 0.32,
"learning_rate": 6.953333333333333e-05,
"loss": 1.2266,
"step": 1875
},
{
"epoch": 0.33,
"learning_rate": 6.897777777777778e-05,
"loss": 1.2858,
"step": 1900
},
{
"epoch": 0.33,
"learning_rate": 6.842222222222222e-05,
"loss": 1.3641,
"step": 1925
},
{
"epoch": 0.33,
"learning_rate": 6.786666666666667e-05,
"loss": 1.3612,
"step": 1950
},
{
"epoch": 0.34,
"learning_rate": 6.731111111111111e-05,
"loss": 1.2474,
"step": 1975
},
{
"epoch": 0.34,
"learning_rate": 6.675555555555556e-05,
"loss": 1.367,
"step": 2000
},
{
"epoch": 0.34,
"eval_loss": 1.2737663984298706,
"eval_runtime": 493.2068,
"eval_samples_per_second": 4.97,
"eval_steps_per_second": 1.243,
"eval_wer": 73.85383567862006,
"step": 2000
},
{
"epoch": 0.35,
"learning_rate": 6.620000000000001e-05,
"loss": 1.2534,
"step": 2025
},
{
"epoch": 0.35,
"learning_rate": 6.564444444444445e-05,
"loss": 1.3164,
"step": 2050
},
{
"epoch": 0.36,
"learning_rate": 6.508888888888888e-05,
"loss": 1.2474,
"step": 2075
},
{
"epoch": 0.36,
"learning_rate": 6.453333333333333e-05,
"loss": 1.1937,
"step": 2100
},
{
"epoch": 0.36,
"learning_rate": 6.397777777777778e-05,
"loss": 1.19,
"step": 2125
},
{
"epoch": 0.37,
"learning_rate": 6.342222222222222e-05,
"loss": 1.2326,
"step": 2150
},
{
"epoch": 0.37,
"learning_rate": 6.286666666666667e-05,
"loss": 1.2876,
"step": 2175
},
{
"epoch": 0.38,
"learning_rate": 6.231111111111111e-05,
"loss": 1.2383,
"step": 2200
},
{
"epoch": 0.38,
"learning_rate": 6.175555555555556e-05,
"loss": 1.2132,
"step": 2225
},
{
"epoch": 0.39,
"learning_rate": 6.12e-05,
"loss": 1.2148,
"step": 2250
},
{
"epoch": 0.39,
"learning_rate": 6.064444444444445e-05,
"loss": 1.0691,
"step": 2275
},
{
"epoch": 0.4,
"learning_rate": 6.00888888888889e-05,
"loss": 1.2139,
"step": 2300
},
{
"epoch": 0.4,
"learning_rate": 5.953333333333334e-05,
"loss": 1.2247,
"step": 2325
},
{
"epoch": 0.4,
"learning_rate": 5.897777777777778e-05,
"loss": 1.3401,
"step": 2350
},
{
"epoch": 0.41,
"learning_rate": 5.842222222222222e-05,
"loss": 1.2732,
"step": 2375
},
{
"epoch": 0.41,
"learning_rate": 5.7866666666666666e-05,
"loss": 1.1014,
"step": 2400
},
{
"epoch": 0.42,
"learning_rate": 5.7311111111111117e-05,
"loss": 1.188,
"step": 2425
},
{
"epoch": 0.42,
"learning_rate": 5.675555555555556e-05,
"loss": 1.2357,
"step": 2450
},
{
"epoch": 0.43,
"learning_rate": 5.620000000000001e-05,
"loss": 1.1116,
"step": 2475
},
{
"epoch": 0.43,
"learning_rate": 5.564444444444444e-05,
"loss": 1.1697,
"step": 2500
},
{
"epoch": 0.43,
"learning_rate": 5.508888888888889e-05,
"loss": 1.1183,
"step": 2525
},
{
"epoch": 0.44,
"learning_rate": 5.4533333333333335e-05,
"loss": 1.2608,
"step": 2550
},
{
"epoch": 0.44,
"learning_rate": 5.3977777777777786e-05,
"loss": 1.1675,
"step": 2575
},
{
"epoch": 0.45,
"learning_rate": 5.342222222222223e-05,
"loss": 1.1062,
"step": 2600
},
{
"epoch": 0.45,
"learning_rate": 5.2866666666666666e-05,
"loss": 1.1097,
"step": 2625
},
{
"epoch": 0.46,
"learning_rate": 5.231111111111111e-05,
"loss": 1.2606,
"step": 2650
},
{
"epoch": 0.46,
"learning_rate": 5.1755555555555554e-05,
"loss": 1.1913,
"step": 2675
},
{
"epoch": 0.46,
"learning_rate": 5.1200000000000004e-05,
"loss": 1.1217,
"step": 2700
},
{
"epoch": 0.47,
"learning_rate": 5.064444444444445e-05,
"loss": 1.172,
"step": 2725
},
{
"epoch": 0.47,
"learning_rate": 5.00888888888889e-05,
"loss": 1.1292,
"step": 2750
},
{
"epoch": 0.48,
"learning_rate": 4.9533333333333336e-05,
"loss": 1.1528,
"step": 2775
},
{
"epoch": 0.48,
"learning_rate": 4.897777777777778e-05,
"loss": 0.9916,
"step": 2800
},
{
"epoch": 0.49,
"learning_rate": 4.842222222222222e-05,
"loss": 1.1021,
"step": 2825
},
{
"epoch": 0.49,
"learning_rate": 4.7866666666666674e-05,
"loss": 1.128,
"step": 2850
},
{
"epoch": 0.49,
"learning_rate": 4.731111111111111e-05,
"loss": 1.2314,
"step": 2875
},
{
"epoch": 0.5,
"learning_rate": 4.675555555555556e-05,
"loss": 0.9696,
"step": 2900
},
{
"epoch": 0.5,
"learning_rate": 4.6200000000000005e-05,
"loss": 1.1669,
"step": 2925
},
{
"epoch": 0.51,
"learning_rate": 4.564444444444444e-05,
"loss": 1.1663,
"step": 2950
},
{
"epoch": 0.51,
"learning_rate": 4.508888888888889e-05,
"loss": 1.0362,
"step": 2975
},
{
"epoch": 0.52,
"learning_rate": 4.4533333333333336e-05,
"loss": 1.0699,
"step": 3000
},
{
"epoch": 0.52,
"eval_loss": 1.0418812036514282,
"eval_runtime": 596.1185,
"eval_samples_per_second": 4.112,
"eval_steps_per_second": 1.028,
"eval_wer": 100.37611049867064,
"step": 3000
},
{
"epoch": 0.52,
"learning_rate": 4.397777777777778e-05,
"loss": 1.0995,
"step": 3025
},
{
"epoch": 0.52,
"learning_rate": 4.3422222222222224e-05,
"loss": 1.0575,
"step": 3050
},
{
"epoch": 0.53,
"learning_rate": 4.286666666666667e-05,
"loss": 0.8525,
"step": 3075
},
{
"epoch": 0.53,
"learning_rate": 4.231111111111111e-05,
"loss": 1.0964,
"step": 3100
},
{
"epoch": 0.54,
"learning_rate": 4.175555555555556e-05,
"loss": 1.1041,
"step": 3125
},
{
"epoch": 0.54,
"learning_rate": 4.12e-05,
"loss": 1.0856,
"step": 3150
},
{
"epoch": 0.55,
"learning_rate": 4.064444444444445e-05,
"loss": 1.0335,
"step": 3175
},
{
"epoch": 0.55,
"learning_rate": 4.008888888888889e-05,
"loss": 1.1112,
"step": 3200
},
{
"epoch": 0.55,
"learning_rate": 3.9533333333333337e-05,
"loss": 1.0152,
"step": 3225
},
{
"epoch": 0.56,
"learning_rate": 3.897777777777778e-05,
"loss": 0.983,
"step": 3250
},
{
"epoch": 0.56,
"learning_rate": 3.8422222222222224e-05,
"loss": 1.0262,
"step": 3275
},
{
"epoch": 0.57,
"learning_rate": 3.786666666666667e-05,
"loss": 0.9556,
"step": 3300
},
{
"epoch": 0.57,
"learning_rate": 3.731111111111111e-05,
"loss": 0.999,
"step": 3325
},
{
"epoch": 0.58,
"learning_rate": 3.675555555555556e-05,
"loss": 0.9377,
"step": 3350
},
{
"epoch": 0.58,
"learning_rate": 3.62e-05,
"loss": 0.9901,
"step": 3375
},
{
"epoch": 0.58,
"learning_rate": 3.564444444444445e-05,
"loss": 0.9327,
"step": 3400
},
{
"epoch": 0.59,
"learning_rate": 3.5088888888888886e-05,
"loss": 1.0878,
"step": 3425
},
{
"epoch": 0.59,
"learning_rate": 3.453333333333334e-05,
"loss": 0.9694,
"step": 3450
},
{
"epoch": 0.6,
"learning_rate": 3.397777777777778e-05,
"loss": 1.1698,
"step": 3475
},
{
"epoch": 0.6,
"learning_rate": 3.3422222222222224e-05,
"loss": 0.9757,
"step": 3500
},
{
"epoch": 0.61,
"learning_rate": 3.286666666666667e-05,
"loss": 1.0011,
"step": 3525
},
{
"epoch": 0.61,
"learning_rate": 3.231111111111111e-05,
"loss": 0.8511,
"step": 3550
},
{
"epoch": 0.61,
"learning_rate": 3.1755555555555556e-05,
"loss": 0.9943,
"step": 3575
},
{
"epoch": 0.62,
"learning_rate": 3.12e-05,
"loss": 0.9637,
"step": 3600
},
{
"epoch": 0.62,
"learning_rate": 3.064444444444445e-05,
"loss": 0.982,
"step": 3625
},
{
"epoch": 0.63,
"learning_rate": 3.008888888888889e-05,
"loss": 0.9277,
"step": 3650
},
{
"epoch": 0.63,
"learning_rate": 2.9533333333333334e-05,
"loss": 1.0569,
"step": 3675
},
{
"epoch": 0.64,
"learning_rate": 2.897777777777778e-05,
"loss": 1.0322,
"step": 3700
},
{
"epoch": 0.64,
"learning_rate": 2.842222222222222e-05,
"loss": 0.8123,
"step": 3725
},
{
"epoch": 0.64,
"learning_rate": 2.786666666666667e-05,
"loss": 1.0553,
"step": 3750
},
{
"epoch": 0.65,
"learning_rate": 2.7311111111111116e-05,
"loss": 0.947,
"step": 3775
},
{
"epoch": 0.65,
"learning_rate": 2.6755555555555556e-05,
"loss": 0.8815,
"step": 3800
},
{
"epoch": 0.66,
"learning_rate": 2.6200000000000003e-05,
"loss": 0.9577,
"step": 3825
},
{
"epoch": 0.66,
"learning_rate": 2.5644444444444444e-05,
"loss": 0.9257,
"step": 3850
},
{
"epoch": 0.67,
"learning_rate": 2.508888888888889e-05,
"loss": 0.9539,
"step": 3875
},
{
"epoch": 0.67,
"learning_rate": 2.4533333333333334e-05,
"loss": 0.9067,
"step": 3900
},
{
"epoch": 0.67,
"learning_rate": 2.3977777777777778e-05,
"loss": 0.8073,
"step": 3925
},
{
"epoch": 0.68,
"learning_rate": 2.3422222222222222e-05,
"loss": 0.9439,
"step": 3950
},
{
"epoch": 0.68,
"learning_rate": 2.2866666666666666e-05,
"loss": 0.8694,
"step": 3975
},
{
"epoch": 0.69,
"learning_rate": 2.2311111111111113e-05,
"loss": 0.8606,
"step": 4000
},
{
"epoch": 0.69,
"eval_loss": 0.8900753855705261,
"eval_runtime": 449.9864,
"eval_samples_per_second": 5.447,
"eval_steps_per_second": 1.362,
"eval_wer": 49.41313792879839,
"step": 4000
},
{
"epoch": 0.69,
"learning_rate": 2.1755555555555557e-05,
"loss": 0.945,
"step": 4025
},
{
"epoch": 0.7,
"learning_rate": 2.12e-05,
"loss": 0.9358,
"step": 4050
},
{
"epoch": 0.7,
"learning_rate": 2.0666666666666666e-05,
"loss": 0.8622,
"step": 4075
},
{
"epoch": 0.7,
"learning_rate": 2.011111111111111e-05,
"loss": 0.8693,
"step": 4100
},
{
"epoch": 0.71,
"learning_rate": 1.9555555555555557e-05,
"loss": 0.8911,
"step": 4125
},
{
"epoch": 0.71,
"learning_rate": 1.9e-05,
"loss": 0.8337,
"step": 4150
},
{
"epoch": 0.72,
"learning_rate": 1.8444444444444445e-05,
"loss": 0.95,
"step": 4175
},
{
"epoch": 0.72,
"learning_rate": 1.788888888888889e-05,
"loss": 0.8776,
"step": 4200
},
{
"epoch": 0.73,
"learning_rate": 1.7333333333333336e-05,
"loss": 0.9119,
"step": 4225
},
{
"epoch": 0.73,
"learning_rate": 1.677777777777778e-05,
"loss": 0.8364,
"step": 4250
},
{
"epoch": 0.73,
"learning_rate": 1.6222222222222223e-05,
"loss": 0.911,
"step": 4275
},
{
"epoch": 0.74,
"learning_rate": 1.5666666666666667e-05,
"loss": 0.7708,
"step": 4300
},
{
"epoch": 0.74,
"learning_rate": 1.5111111111111112e-05,
"loss": 0.7506,
"step": 4325
},
{
"epoch": 0.75,
"learning_rate": 1.4555555555555556e-05,
"loss": 0.9162,
"step": 4350
},
{
"epoch": 0.75,
"learning_rate": 1.4000000000000001e-05,
"loss": 0.7746,
"step": 4375
},
{
"epoch": 0.76,
"learning_rate": 1.3444444444444445e-05,
"loss": 0.9284,
"step": 4400
},
{
"epoch": 0.76,
"learning_rate": 1.2888888888888889e-05,
"loss": 0.8859,
"step": 4425
},
{
"epoch": 0.76,
"learning_rate": 1.2333333333333334e-05,
"loss": 0.8251,
"step": 4450
},
{
"epoch": 0.77,
"learning_rate": 1.1777777777777778e-05,
"loss": 0.93,
"step": 4475
},
{
"epoch": 0.77,
"learning_rate": 1.1222222222222224e-05,
"loss": 0.8064,
"step": 4500
},
{
"epoch": 0.78,
"learning_rate": 1.0666666666666667e-05,
"loss": 0.7628,
"step": 4525
},
{
"epoch": 0.78,
"learning_rate": 1.0111111111111111e-05,
"loss": 0.7912,
"step": 4550
},
{
"epoch": 0.79,
"learning_rate": 9.555555555555556e-06,
"loss": 0.8796,
"step": 4575
},
{
"epoch": 0.79,
"learning_rate": 9e-06,
"loss": 0.8324,
"step": 4600
},
{
"epoch": 0.79,
"learning_rate": 8.444444444444446e-06,
"loss": 0.8019,
"step": 4625
},
{
"epoch": 0.8,
"learning_rate": 7.88888888888889e-06,
"loss": 0.8745,
"step": 4650
},
{
"epoch": 0.8,
"learning_rate": 7.333333333333334e-06,
"loss": 0.8631,
"step": 4675
},
{
"epoch": 0.81,
"learning_rate": 6.777777777777779e-06,
"loss": 0.8496,
"step": 4700
},
{
"epoch": 0.81,
"learning_rate": 6.222222222222222e-06,
"loss": 0.7834,
"step": 4725
},
{
"epoch": 0.82,
"learning_rate": 5.666666666666667e-06,
"loss": 0.6906,
"step": 4750
},
{
"epoch": 0.82,
"learning_rate": 5.1111111111111115e-06,
"loss": 0.8864,
"step": 4775
},
{
"epoch": 0.82,
"learning_rate": 4.555555555555556e-06,
"loss": 0.725,
"step": 4800
},
{
"epoch": 0.83,
"learning_rate": 4.000000000000001e-06,
"loss": 0.7104,
"step": 4825
},
{
"epoch": 0.83,
"learning_rate": 3.4444444444444444e-06,
"loss": 0.7191,
"step": 4850
},
{
"epoch": 0.84,
"learning_rate": 2.888888888888889e-06,
"loss": 0.7249,
"step": 4875
},
{
"epoch": 0.84,
"learning_rate": 2.3333333333333336e-06,
"loss": 0.768,
"step": 4900
},
{
"epoch": 0.85,
"learning_rate": 1.777777777777778e-06,
"loss": 0.8022,
"step": 4925
},
{
"epoch": 0.85,
"learning_rate": 1.2222222222222223e-06,
"loss": 0.7637,
"step": 4950
},
{
"epoch": 0.85,
"learning_rate": 6.666666666666667e-07,
"loss": 0.7204,
"step": 4975
},
{
"epoch": 0.86,
"learning_rate": 1.1111111111111112e-07,
"loss": 0.7381,
"step": 5000
},
{
"epoch": 0.86,
"eval_loss": 0.7810165286064148,
"eval_runtime": 441.0966,
"eval_samples_per_second": 5.557,
"eval_steps_per_second": 1.39,
"eval_wer": 43.84281175021075,
"step": 5000
},
{
"epoch": 0.86,
"step": 5000,
"total_flos": 1.15434160128e+19,
"train_loss": 1.1686982261657715,
"train_runtime": 8241.3689,
"train_samples_per_second": 4.854,
"train_steps_per_second": 0.607
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 1.15434160128e+19,
"trial_name": null,
"trial_params": null
}