yt-special-batch8-small / trainer_state.json
TheRains's picture
End of training
f5d67d0
{
"best_metric": 39.1803384994488,
"best_model_checkpoint": "./output/small/yt-special-batch8-small/checkpoint-4000",
"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": 4.2000000000000006e-07,
"loss": 2.3615,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.200000000000001e-07,
"loss": 2.1267,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.42e-06,
"loss": 2.0558,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.9200000000000003e-06,
"loss": 1.5605,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 2.42e-06,
"loss": 1.2957,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.92e-06,
"loss": 1.1285,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 3.4200000000000007e-06,
"loss": 0.9258,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 3.920000000000001e-06,
"loss": 0.9519,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.42e-06,
"loss": 1.0465,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 4.92e-06,
"loss": 0.9585,
"step": 250
},
{
"epoch": 0.05,
"learning_rate": 5.420000000000001e-06,
"loss": 0.9696,
"step": 275
},
{
"epoch": 0.05,
"learning_rate": 5.92e-06,
"loss": 0.9763,
"step": 300
},
{
"epoch": 0.06,
"learning_rate": 6.42e-06,
"loss": 0.8259,
"step": 325
},
{
"epoch": 0.06,
"learning_rate": 6.92e-06,
"loss": 0.9723,
"step": 350
},
{
"epoch": 0.06,
"learning_rate": 7.420000000000001e-06,
"loss": 1.0433,
"step": 375
},
{
"epoch": 0.07,
"learning_rate": 7.92e-06,
"loss": 0.9234,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 8.42e-06,
"loss": 0.7719,
"step": 425
},
{
"epoch": 0.08,
"learning_rate": 8.920000000000001e-06,
"loss": 0.9647,
"step": 450
},
{
"epoch": 0.08,
"learning_rate": 9.42e-06,
"loss": 0.9199,
"step": 475
},
{
"epoch": 0.09,
"learning_rate": 9.920000000000002e-06,
"loss": 0.9555,
"step": 500
},
{
"epoch": 0.09,
"learning_rate": 9.953333333333333e-06,
"loss": 0.8905,
"step": 525
},
{
"epoch": 0.09,
"learning_rate": 9.89777777777778e-06,
"loss": 0.9262,
"step": 550
},
{
"epoch": 0.1,
"learning_rate": 9.842222222222223e-06,
"loss": 0.8921,
"step": 575
},
{
"epoch": 0.1,
"learning_rate": 9.786666666666667e-06,
"loss": 1.0335,
"step": 600
},
{
"epoch": 0.11,
"learning_rate": 9.731111111111113e-06,
"loss": 0.8921,
"step": 625
},
{
"epoch": 0.11,
"learning_rate": 9.675555555555555e-06,
"loss": 0.9255,
"step": 650
},
{
"epoch": 0.12,
"learning_rate": 9.620000000000001e-06,
"loss": 0.9188,
"step": 675
},
{
"epoch": 0.12,
"learning_rate": 9.564444444444445e-06,
"loss": 0.8102,
"step": 700
},
{
"epoch": 0.12,
"learning_rate": 9.508888888888889e-06,
"loss": 0.8651,
"step": 725
},
{
"epoch": 0.13,
"learning_rate": 9.453333333333335e-06,
"loss": 0.9106,
"step": 750
},
{
"epoch": 0.13,
"learning_rate": 9.397777777777779e-06,
"loss": 0.821,
"step": 775
},
{
"epoch": 0.14,
"learning_rate": 9.342222222222223e-06,
"loss": 0.8583,
"step": 800
},
{
"epoch": 0.14,
"learning_rate": 9.286666666666667e-06,
"loss": 0.7893,
"step": 825
},
{
"epoch": 0.15,
"learning_rate": 9.231111111111111e-06,
"loss": 0.824,
"step": 850
},
{
"epoch": 0.15,
"learning_rate": 9.175555555555557e-06,
"loss": 0.8617,
"step": 875
},
{
"epoch": 0.15,
"learning_rate": 9.12e-06,
"loss": 0.8933,
"step": 900
},
{
"epoch": 0.16,
"learning_rate": 9.064444444444447e-06,
"loss": 0.882,
"step": 925
},
{
"epoch": 0.16,
"learning_rate": 9.008888888888889e-06,
"loss": 0.7932,
"step": 950
},
{
"epoch": 0.17,
"learning_rate": 8.953333333333335e-06,
"loss": 0.8278,
"step": 975
},
{
"epoch": 0.17,
"learning_rate": 8.897777777777779e-06,
"loss": 0.8294,
"step": 1000
},
{
"epoch": 0.17,
"eval_loss": 0.861615777015686,
"eval_runtime": 460.5888,
"eval_samples_per_second": 5.321,
"eval_steps_per_second": 1.331,
"eval_wer": 61.325465274625515,
"step": 1000
},
{
"epoch": 0.18,
"learning_rate": 8.842222222222223e-06,
"loss": 0.9288,
"step": 1025
},
{
"epoch": 0.18,
"learning_rate": 8.786666666666668e-06,
"loss": 0.8081,
"step": 1050
},
{
"epoch": 0.18,
"learning_rate": 8.73111111111111e-06,
"loss": 0.9229,
"step": 1075
},
{
"epoch": 0.19,
"learning_rate": 8.675555555555556e-06,
"loss": 0.8032,
"step": 1100
},
{
"epoch": 0.19,
"learning_rate": 8.62e-06,
"loss": 0.8175,
"step": 1125
},
{
"epoch": 0.2,
"learning_rate": 8.564444444444445e-06,
"loss": 1.0077,
"step": 1150
},
{
"epoch": 0.2,
"learning_rate": 8.50888888888889e-06,
"loss": 0.9826,
"step": 1175
},
{
"epoch": 0.21,
"learning_rate": 8.453333333333334e-06,
"loss": 0.7871,
"step": 1200
},
{
"epoch": 0.21,
"learning_rate": 8.397777777777778e-06,
"loss": 0.8384,
"step": 1225
},
{
"epoch": 0.21,
"learning_rate": 8.342222222222222e-06,
"loss": 0.8991,
"step": 1250
},
{
"epoch": 0.22,
"learning_rate": 8.286666666666668e-06,
"loss": 0.7364,
"step": 1275
},
{
"epoch": 0.22,
"learning_rate": 8.231111111111112e-06,
"loss": 0.8427,
"step": 1300
},
{
"epoch": 0.23,
"learning_rate": 8.175555555555556e-06,
"loss": 0.8059,
"step": 1325
},
{
"epoch": 0.23,
"learning_rate": 8.120000000000002e-06,
"loss": 0.8373,
"step": 1350
},
{
"epoch": 0.24,
"learning_rate": 8.064444444444444e-06,
"loss": 0.825,
"step": 1375
},
{
"epoch": 0.24,
"learning_rate": 8.00888888888889e-06,
"loss": 0.7639,
"step": 1400
},
{
"epoch": 0.24,
"learning_rate": 7.953333333333334e-06,
"loss": 0.8784,
"step": 1425
},
{
"epoch": 0.25,
"learning_rate": 7.897777777777778e-06,
"loss": 0.7776,
"step": 1450
},
{
"epoch": 0.25,
"learning_rate": 7.842222222222224e-06,
"loss": 0.8619,
"step": 1475
},
{
"epoch": 0.26,
"learning_rate": 7.786666666666666e-06,
"loss": 0.791,
"step": 1500
},
{
"epoch": 0.26,
"learning_rate": 7.731111111111112e-06,
"loss": 0.9299,
"step": 1525
},
{
"epoch": 0.27,
"learning_rate": 7.675555555555556e-06,
"loss": 0.7594,
"step": 1550
},
{
"epoch": 0.27,
"learning_rate": 7.620000000000001e-06,
"loss": 0.853,
"step": 1575
},
{
"epoch": 0.27,
"learning_rate": 7.564444444444446e-06,
"loss": 0.8122,
"step": 1600
},
{
"epoch": 0.28,
"learning_rate": 7.50888888888889e-06,
"loss": 0.7526,
"step": 1625
},
{
"epoch": 0.28,
"learning_rate": 7.453333333333334e-06,
"loss": 0.824,
"step": 1650
},
{
"epoch": 0.29,
"learning_rate": 7.3977777777777786e-06,
"loss": 0.8211,
"step": 1675
},
{
"epoch": 0.29,
"learning_rate": 7.342222222222223e-06,
"loss": 0.7614,
"step": 1700
},
{
"epoch": 0.3,
"learning_rate": 7.2866666666666675e-06,
"loss": 0.7568,
"step": 1725
},
{
"epoch": 0.3,
"learning_rate": 7.231111111111112e-06,
"loss": 0.7584,
"step": 1750
},
{
"epoch": 0.3,
"learning_rate": 7.1755555555555556e-06,
"loss": 0.8613,
"step": 1775
},
{
"epoch": 0.31,
"learning_rate": 7.1200000000000004e-06,
"loss": 0.7136,
"step": 1800
},
{
"epoch": 0.31,
"learning_rate": 7.0644444444444445e-06,
"loss": 0.8231,
"step": 1825
},
{
"epoch": 0.32,
"learning_rate": 7.008888888888889e-06,
"loss": 0.8276,
"step": 1850
},
{
"epoch": 0.32,
"learning_rate": 6.953333333333334e-06,
"loss": 0.7536,
"step": 1875
},
{
"epoch": 0.33,
"learning_rate": 6.897777777777779e-06,
"loss": 0.7945,
"step": 1900
},
{
"epoch": 0.33,
"learning_rate": 6.842222222222222e-06,
"loss": 0.8086,
"step": 1925
},
{
"epoch": 0.33,
"learning_rate": 6.786666666666667e-06,
"loss": 0.8741,
"step": 1950
},
{
"epoch": 0.34,
"learning_rate": 6.731111111111111e-06,
"loss": 0.7293,
"step": 1975
},
{
"epoch": 0.34,
"learning_rate": 6.675555555555556e-06,
"loss": 0.8198,
"step": 2000
},
{
"epoch": 0.34,
"eval_loss": 0.7873214483261108,
"eval_runtime": 416.0464,
"eval_samples_per_second": 5.891,
"eval_steps_per_second": 1.473,
"eval_wer": 48.19402114000389,
"step": 2000
},
{
"epoch": 0.35,
"learning_rate": 6.620000000000001e-06,
"loss": 0.7576,
"step": 2025
},
{
"epoch": 0.35,
"learning_rate": 6.564444444444446e-06,
"loss": 0.8485,
"step": 2050
},
{
"epoch": 0.36,
"learning_rate": 6.508888888888889e-06,
"loss": 0.799,
"step": 2075
},
{
"epoch": 0.36,
"learning_rate": 6.453333333333334e-06,
"loss": 0.7521,
"step": 2100
},
{
"epoch": 0.36,
"learning_rate": 6.397777777777778e-06,
"loss": 0.7132,
"step": 2125
},
{
"epoch": 0.37,
"learning_rate": 6.342222222222223e-06,
"loss": 0.8021,
"step": 2150
},
{
"epoch": 0.37,
"learning_rate": 6.286666666666668e-06,
"loss": 0.7834,
"step": 2175
},
{
"epoch": 0.38,
"learning_rate": 6.231111111111111e-06,
"loss": 0.7665,
"step": 2200
},
{
"epoch": 0.38,
"learning_rate": 6.175555555555556e-06,
"loss": 0.7646,
"step": 2225
},
{
"epoch": 0.39,
"learning_rate": 6.120000000000001e-06,
"loss": 0.7053,
"step": 2250
},
{
"epoch": 0.39,
"learning_rate": 6.064444444444445e-06,
"loss": 0.664,
"step": 2275
},
{
"epoch": 0.4,
"learning_rate": 6.00888888888889e-06,
"loss": 0.8094,
"step": 2300
},
{
"epoch": 0.4,
"learning_rate": 5.9533333333333345e-06,
"loss": 0.7934,
"step": 2325
},
{
"epoch": 0.4,
"learning_rate": 5.897777777777778e-06,
"loss": 0.8695,
"step": 2350
},
{
"epoch": 0.41,
"learning_rate": 5.8422222222222226e-06,
"loss": 0.822,
"step": 2375
},
{
"epoch": 0.41,
"learning_rate": 5.7866666666666674e-06,
"loss": 0.6992,
"step": 2400
},
{
"epoch": 0.42,
"learning_rate": 5.7311111111111115e-06,
"loss": 0.757,
"step": 2425
},
{
"epoch": 0.42,
"learning_rate": 5.675555555555556e-06,
"loss": 0.7336,
"step": 2450
},
{
"epoch": 0.43,
"learning_rate": 5.620000000000001e-06,
"loss": 0.7451,
"step": 2475
},
{
"epoch": 0.43,
"learning_rate": 5.5644444444444444e-06,
"loss": 0.776,
"step": 2500
},
{
"epoch": 0.43,
"learning_rate": 5.508888888888889e-06,
"loss": 0.7442,
"step": 2525
},
{
"epoch": 0.44,
"learning_rate": 5.453333333333334e-06,
"loss": 0.7801,
"step": 2550
},
{
"epoch": 0.44,
"learning_rate": 5.397777777777778e-06,
"loss": 0.7308,
"step": 2575
},
{
"epoch": 0.45,
"learning_rate": 5.342222222222223e-06,
"loss": 0.7105,
"step": 2600
},
{
"epoch": 0.45,
"learning_rate": 5.286666666666666e-06,
"loss": 0.7256,
"step": 2625
},
{
"epoch": 0.46,
"learning_rate": 5.231111111111111e-06,
"loss": 0.8504,
"step": 2650
},
{
"epoch": 0.46,
"learning_rate": 5.175555555555556e-06,
"loss": 0.7455,
"step": 2675
},
{
"epoch": 0.46,
"learning_rate": 5.12e-06,
"loss": 0.7802,
"step": 2700
},
{
"epoch": 0.47,
"learning_rate": 5.064444444444445e-06,
"loss": 0.7823,
"step": 2725
},
{
"epoch": 0.47,
"learning_rate": 5.00888888888889e-06,
"loss": 0.7633,
"step": 2750
},
{
"epoch": 0.48,
"learning_rate": 4.953333333333334e-06,
"loss": 0.8035,
"step": 2775
},
{
"epoch": 0.48,
"learning_rate": 4.897777777777778e-06,
"loss": 0.6477,
"step": 2800
},
{
"epoch": 0.49,
"learning_rate": 4.842222222222223e-06,
"loss": 0.7061,
"step": 2825
},
{
"epoch": 0.49,
"learning_rate": 4.786666666666667e-06,
"loss": 0.7505,
"step": 2850
},
{
"epoch": 0.49,
"learning_rate": 4.731111111111112e-06,
"loss": 0.8425,
"step": 2875
},
{
"epoch": 0.5,
"learning_rate": 4.675555555555556e-06,
"loss": 0.6431,
"step": 2900
},
{
"epoch": 0.5,
"learning_rate": 4.620000000000001e-06,
"loss": 0.8037,
"step": 2925
},
{
"epoch": 0.51,
"learning_rate": 4.564444444444445e-06,
"loss": 0.7983,
"step": 2950
},
{
"epoch": 0.51,
"learning_rate": 4.50888888888889e-06,
"loss": 0.7182,
"step": 2975
},
{
"epoch": 0.52,
"learning_rate": 4.453333333333334e-06,
"loss": 0.7103,
"step": 3000
},
{
"epoch": 0.52,
"eval_loss": 0.7308079600334167,
"eval_runtime": 475.46,
"eval_samples_per_second": 5.155,
"eval_steps_per_second": 1.289,
"eval_wer": 45.379677063744246,
"step": 3000
},
{
"epoch": 0.52,
"learning_rate": 4.397777777777778e-06,
"loss": 0.7418,
"step": 3025
},
{
"epoch": 0.52,
"learning_rate": 4.3422222222222225e-06,
"loss": 0.7133,
"step": 3050
},
{
"epoch": 0.53,
"learning_rate": 4.2866666666666666e-06,
"loss": 0.5783,
"step": 3075
},
{
"epoch": 0.53,
"learning_rate": 4.2311111111111114e-06,
"loss": 0.7437,
"step": 3100
},
{
"epoch": 0.54,
"learning_rate": 4.175555555555556e-06,
"loss": 0.7718,
"step": 3125
},
{
"epoch": 0.54,
"learning_rate": 4.12e-06,
"loss": 0.7528,
"step": 3150
},
{
"epoch": 0.55,
"learning_rate": 4.064444444444444e-06,
"loss": 0.759,
"step": 3175
},
{
"epoch": 0.55,
"learning_rate": 4.008888888888889e-06,
"loss": 0.7942,
"step": 3200
},
{
"epoch": 0.55,
"learning_rate": 3.953333333333333e-06,
"loss": 0.7287,
"step": 3225
},
{
"epoch": 0.56,
"learning_rate": 3.897777777777778e-06,
"loss": 0.6905,
"step": 3250
},
{
"epoch": 0.56,
"learning_rate": 3.842222222222223e-06,
"loss": 0.7392,
"step": 3275
},
{
"epoch": 0.57,
"learning_rate": 3.7866666666666667e-06,
"loss": 0.6416,
"step": 3300
},
{
"epoch": 0.57,
"learning_rate": 3.7311111111111116e-06,
"loss": 0.6988,
"step": 3325
},
{
"epoch": 0.58,
"learning_rate": 3.675555555555556e-06,
"loss": 0.661,
"step": 3350
},
{
"epoch": 0.58,
"learning_rate": 3.62e-06,
"loss": 0.6951,
"step": 3375
},
{
"epoch": 0.58,
"learning_rate": 3.564444444444445e-06,
"loss": 0.6569,
"step": 3400
},
{
"epoch": 0.59,
"learning_rate": 3.508888888888889e-06,
"loss": 0.7821,
"step": 3425
},
{
"epoch": 0.59,
"learning_rate": 3.4533333333333334e-06,
"loss": 0.7501,
"step": 3450
},
{
"epoch": 0.6,
"learning_rate": 3.3977777777777783e-06,
"loss": 0.8415,
"step": 3475
},
{
"epoch": 0.6,
"learning_rate": 3.3422222222222224e-06,
"loss": 0.6945,
"step": 3500
},
{
"epoch": 0.61,
"learning_rate": 3.286666666666667e-06,
"loss": 0.7447,
"step": 3525
},
{
"epoch": 0.61,
"learning_rate": 3.2311111111111117e-06,
"loss": 0.5961,
"step": 3550
},
{
"epoch": 0.61,
"learning_rate": 3.1755555555555557e-06,
"loss": 0.7029,
"step": 3575
},
{
"epoch": 0.62,
"learning_rate": 3.12e-06,
"loss": 0.723,
"step": 3600
},
{
"epoch": 0.62,
"learning_rate": 3.064444444444445e-06,
"loss": 0.7213,
"step": 3625
},
{
"epoch": 0.63,
"learning_rate": 3.008888888888889e-06,
"loss": 0.6929,
"step": 3650
},
{
"epoch": 0.63,
"learning_rate": 2.9533333333333336e-06,
"loss": 0.7402,
"step": 3675
},
{
"epoch": 0.64,
"learning_rate": 2.8977777777777785e-06,
"loss": 0.755,
"step": 3700
},
{
"epoch": 0.64,
"learning_rate": 2.8422222222222225e-06,
"loss": 0.6064,
"step": 3725
},
{
"epoch": 0.64,
"learning_rate": 2.786666666666667e-06,
"loss": 0.7812,
"step": 3750
},
{
"epoch": 0.65,
"learning_rate": 2.7311111111111114e-06,
"loss": 0.7057,
"step": 3775
},
{
"epoch": 0.65,
"learning_rate": 2.675555555555556e-06,
"loss": 0.6693,
"step": 3800
},
{
"epoch": 0.66,
"learning_rate": 2.6200000000000003e-06,
"loss": 0.6943,
"step": 3825
},
{
"epoch": 0.66,
"learning_rate": 2.5644444444444444e-06,
"loss": 0.7031,
"step": 3850
},
{
"epoch": 0.67,
"learning_rate": 2.5088888888888892e-06,
"loss": 0.7577,
"step": 3875
},
{
"epoch": 0.67,
"learning_rate": 2.4533333333333333e-06,
"loss": 0.6929,
"step": 3900
},
{
"epoch": 0.67,
"learning_rate": 2.397777777777778e-06,
"loss": 0.5809,
"step": 3925
},
{
"epoch": 0.68,
"learning_rate": 2.342222222222222e-06,
"loss": 0.7553,
"step": 3950
},
{
"epoch": 0.68,
"learning_rate": 2.2866666666666667e-06,
"loss": 0.6366,
"step": 3975
},
{
"epoch": 0.69,
"learning_rate": 2.2311111111111115e-06,
"loss": 0.6905,
"step": 4000
},
{
"epoch": 0.69,
"eval_loss": 0.7024663090705872,
"eval_runtime": 395.1039,
"eval_samples_per_second": 6.203,
"eval_steps_per_second": 1.551,
"eval_wer": 39.1803384994488,
"step": 4000
},
{
"epoch": 0.69,
"learning_rate": 2.1755555555555556e-06,
"loss": 0.7029,
"step": 4025
},
{
"epoch": 0.7,
"learning_rate": 2.12e-06,
"loss": 0.7601,
"step": 4050
},
{
"epoch": 0.7,
"learning_rate": 2.064444444444445e-06,
"loss": 0.6789,
"step": 4075
},
{
"epoch": 0.7,
"learning_rate": 2.008888888888889e-06,
"loss": 0.6326,
"step": 4100
},
{
"epoch": 0.71,
"learning_rate": 1.9533333333333334e-06,
"loss": 0.6689,
"step": 4125
},
{
"epoch": 0.71,
"learning_rate": 1.8977777777777779e-06,
"loss": 0.6985,
"step": 4150
},
{
"epoch": 0.72,
"learning_rate": 1.8422222222222225e-06,
"loss": 0.7512,
"step": 4175
},
{
"epoch": 0.72,
"learning_rate": 1.7866666666666668e-06,
"loss": 0.674,
"step": 4200
},
{
"epoch": 0.73,
"learning_rate": 1.7311111111111112e-06,
"loss": 0.7391,
"step": 4225
},
{
"epoch": 0.73,
"learning_rate": 1.675555555555556e-06,
"loss": 0.69,
"step": 4250
},
{
"epoch": 0.73,
"learning_rate": 1.6222222222222223e-06,
"loss": 0.7855,
"step": 4275
},
{
"epoch": 0.74,
"learning_rate": 1.566666666666667e-06,
"loss": 0.6303,
"step": 4300
},
{
"epoch": 0.74,
"learning_rate": 1.5111111111111112e-06,
"loss": 0.6136,
"step": 4325
},
{
"epoch": 0.75,
"learning_rate": 1.4555555555555557e-06,
"loss": 0.7719,
"step": 4350
},
{
"epoch": 0.75,
"learning_rate": 1.4000000000000001e-06,
"loss": 0.6262,
"step": 4375
},
{
"epoch": 0.76,
"learning_rate": 1.3444444444444446e-06,
"loss": 0.8025,
"step": 4400
},
{
"epoch": 0.76,
"learning_rate": 1.288888888888889e-06,
"loss": 0.7488,
"step": 4425
},
{
"epoch": 0.76,
"learning_rate": 1.2333333333333335e-06,
"loss": 0.6676,
"step": 4450
},
{
"epoch": 0.77,
"learning_rate": 1.1777777777777778e-06,
"loss": 0.7918,
"step": 4475
},
{
"epoch": 0.77,
"learning_rate": 1.1222222222222222e-06,
"loss": 0.6667,
"step": 4500
},
{
"epoch": 0.78,
"learning_rate": 1.066666666666667e-06,
"loss": 0.6284,
"step": 4525
},
{
"epoch": 0.78,
"learning_rate": 1.0111111111111111e-06,
"loss": 0.6707,
"step": 4550
},
{
"epoch": 0.79,
"learning_rate": 9.555555555555556e-07,
"loss": 0.7195,
"step": 4575
},
{
"epoch": 0.79,
"learning_rate": 9.000000000000001e-07,
"loss": 0.6856,
"step": 4600
},
{
"epoch": 0.79,
"learning_rate": 8.444444444444445e-07,
"loss": 0.6795,
"step": 4625
},
{
"epoch": 0.8,
"learning_rate": 7.888888888888889e-07,
"loss": 0.7202,
"step": 4650
},
{
"epoch": 0.8,
"learning_rate": 7.333333333333334e-07,
"loss": 0.7359,
"step": 4675
},
{
"epoch": 0.81,
"learning_rate": 6.777777777777779e-07,
"loss": 0.7316,
"step": 4700
},
{
"epoch": 0.81,
"learning_rate": 6.222222222222223e-07,
"loss": 0.6835,
"step": 4725
},
{
"epoch": 0.82,
"learning_rate": 5.666666666666667e-07,
"loss": 0.6065,
"step": 4750
},
{
"epoch": 0.82,
"learning_rate": 5.111111111111112e-07,
"loss": 0.7881,
"step": 4775
},
{
"epoch": 0.82,
"learning_rate": 4.5555555555555563e-07,
"loss": 0.6415,
"step": 4800
},
{
"epoch": 0.83,
"learning_rate": 4.0000000000000003e-07,
"loss": 0.6293,
"step": 4825
},
{
"epoch": 0.83,
"learning_rate": 3.444444444444445e-07,
"loss": 0.6411,
"step": 4850
},
{
"epoch": 0.84,
"learning_rate": 2.888888888888889e-07,
"loss": 0.6106,
"step": 4875
},
{
"epoch": 0.84,
"learning_rate": 2.3333333333333336e-07,
"loss": 0.691,
"step": 4900
},
{
"epoch": 0.85,
"learning_rate": 1.777777777777778e-07,
"loss": 0.6845,
"step": 4925
},
{
"epoch": 0.85,
"learning_rate": 1.2222222222222225e-07,
"loss": 0.6526,
"step": 4950
},
{
"epoch": 0.85,
"learning_rate": 6.666666666666668e-08,
"loss": 0.5854,
"step": 4975
},
{
"epoch": 0.86,
"learning_rate": 1.1111111111111112e-08,
"loss": 0.6467,
"step": 5000
},
{
"epoch": 0.86,
"eval_loss": 0.684894859790802,
"eval_runtime": 398.7356,
"eval_samples_per_second": 6.147,
"eval_steps_per_second": 1.537,
"eval_wer": 41.71584203359056,
"step": 5000
},
{
"epoch": 0.86,
"step": 5000,
"total_flos": 1.15434160128e+19,
"train_loss": 0.8022269817352294,
"train_runtime": 7257.0629,
"train_samples_per_second": 5.512,
"train_steps_per_second": 0.689
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 1.15434160128e+19,
"trial_name": null,
"trial_params": null
}