whisper-small-ml / trainer_state.json
apzl's picture
folder structure
02cdc31
{
"best_metric": 37.29246487867178,
"best_model_checkpoint": "./checkpoint-3000",
"epoch": 92.5925925925926,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.46,
"learning_rate": 4.4e-07,
"loss": 2.4124,
"step": 25
},
{
"epoch": 0.93,
"learning_rate": 9.400000000000001e-07,
"loss": 2.0615,
"step": 50
},
{
"epoch": 1.39,
"learning_rate": 1.44e-06,
"loss": 1.676,
"step": 75
},
{
"epoch": 1.85,
"learning_rate": 1.94e-06,
"loss": 1.4242,
"step": 100
},
{
"epoch": 2.31,
"learning_rate": 2.4400000000000004e-06,
"loss": 1.2759,
"step": 125
},
{
"epoch": 2.78,
"learning_rate": 2.9400000000000002e-06,
"loss": 1.1563,
"step": 150
},
{
"epoch": 3.24,
"learning_rate": 3.44e-06,
"loss": 0.9762,
"step": 175
},
{
"epoch": 3.7,
"learning_rate": 3.94e-06,
"loss": 0.7968,
"step": 200
},
{
"epoch": 4.17,
"learning_rate": 4.440000000000001e-06,
"loss": 0.6836,
"step": 225
},
{
"epoch": 4.63,
"learning_rate": 4.94e-06,
"loss": 0.5482,
"step": 250
},
{
"epoch": 5.09,
"learning_rate": 5.4400000000000004e-06,
"loss": 0.4754,
"step": 275
},
{
"epoch": 5.56,
"learning_rate": 5.94e-06,
"loss": 0.3772,
"step": 300
},
{
"epoch": 6.02,
"learning_rate": 6.440000000000001e-06,
"loss": 0.3323,
"step": 325
},
{
"epoch": 6.48,
"learning_rate": 6.9400000000000005e-06,
"loss": 0.2471,
"step": 350
},
{
"epoch": 6.94,
"learning_rate": 7.440000000000001e-06,
"loss": 0.2263,
"step": 375
},
{
"epoch": 7.41,
"learning_rate": 7.94e-06,
"loss": 0.1282,
"step": 400
},
{
"epoch": 7.87,
"learning_rate": 8.44e-06,
"loss": 0.1105,
"step": 425
},
{
"epoch": 8.33,
"learning_rate": 8.94e-06,
"loss": 0.0807,
"step": 450
},
{
"epoch": 8.8,
"learning_rate": 9.440000000000001e-06,
"loss": 0.0694,
"step": 475
},
{
"epoch": 9.26,
"learning_rate": 9.940000000000001e-06,
"loss": 0.0722,
"step": 500
},
{
"epoch": 9.72,
"learning_rate": 9.951111111111111e-06,
"loss": 0.0625,
"step": 525
},
{
"epoch": 10.19,
"learning_rate": 9.895555555555557e-06,
"loss": 0.0508,
"step": 550
},
{
"epoch": 10.65,
"learning_rate": 9.84e-06,
"loss": 0.0428,
"step": 575
},
{
"epoch": 11.11,
"learning_rate": 9.784444444444445e-06,
"loss": 0.039,
"step": 600
},
{
"epoch": 11.57,
"learning_rate": 9.72888888888889e-06,
"loss": 0.0355,
"step": 625
},
{
"epoch": 12.04,
"learning_rate": 9.673333333333334e-06,
"loss": 0.0305,
"step": 650
},
{
"epoch": 12.5,
"learning_rate": 9.617777777777778e-06,
"loss": 0.0289,
"step": 675
},
{
"epoch": 12.96,
"learning_rate": 9.562222222222223e-06,
"loss": 0.0309,
"step": 700
},
{
"epoch": 13.43,
"learning_rate": 9.506666666666667e-06,
"loss": 0.0168,
"step": 725
},
{
"epoch": 13.89,
"learning_rate": 9.451111111111112e-06,
"loss": 0.0183,
"step": 750
},
{
"epoch": 14.35,
"learning_rate": 9.395555555555556e-06,
"loss": 0.0273,
"step": 775
},
{
"epoch": 14.81,
"learning_rate": 9.340000000000002e-06,
"loss": 0.0194,
"step": 800
},
{
"epoch": 15.28,
"learning_rate": 9.284444444444444e-06,
"loss": 0.0178,
"step": 825
},
{
"epoch": 15.74,
"learning_rate": 9.22888888888889e-06,
"loss": 0.0147,
"step": 850
},
{
"epoch": 16.2,
"learning_rate": 9.173333333333334e-06,
"loss": 0.0136,
"step": 875
},
{
"epoch": 16.67,
"learning_rate": 9.117777777777778e-06,
"loss": 0.014,
"step": 900
},
{
"epoch": 17.13,
"learning_rate": 9.062222222222224e-06,
"loss": 0.014,
"step": 925
},
{
"epoch": 17.59,
"learning_rate": 9.006666666666666e-06,
"loss": 0.0126,
"step": 950
},
{
"epoch": 18.06,
"learning_rate": 8.951111111111112e-06,
"loss": 0.0124,
"step": 975
},
{
"epoch": 18.52,
"learning_rate": 8.895555555555556e-06,
"loss": 0.0102,
"step": 1000
},
{
"epoch": 18.52,
"eval_loss": 0.548794686794281,
"eval_runtime": 479.252,
"eval_samples_per_second": 0.234,
"eval_steps_per_second": 0.029,
"eval_wer": 42.017879948914434,
"step": 1000
},
{
"epoch": 18.98,
"learning_rate": 8.84e-06,
"loss": 0.0138,
"step": 1025
},
{
"epoch": 19.44,
"learning_rate": 8.784444444444446e-06,
"loss": 0.0084,
"step": 1050
},
{
"epoch": 19.91,
"learning_rate": 8.72888888888889e-06,
"loss": 0.0094,
"step": 1075
},
{
"epoch": 20.37,
"learning_rate": 8.673333333333334e-06,
"loss": 0.0086,
"step": 1100
},
{
"epoch": 20.83,
"learning_rate": 8.617777777777778e-06,
"loss": 0.0078,
"step": 1125
},
{
"epoch": 21.3,
"learning_rate": 8.562222222222224e-06,
"loss": 0.0063,
"step": 1150
},
{
"epoch": 21.76,
"learning_rate": 8.506666666666668e-06,
"loss": 0.0067,
"step": 1175
},
{
"epoch": 22.22,
"learning_rate": 8.451111111111112e-06,
"loss": 0.0054,
"step": 1200
},
{
"epoch": 22.69,
"learning_rate": 8.395555555555557e-06,
"loss": 0.0068,
"step": 1225
},
{
"epoch": 23.15,
"learning_rate": 8.34e-06,
"loss": 0.0073,
"step": 1250
},
{
"epoch": 23.61,
"learning_rate": 8.284444444444446e-06,
"loss": 0.0056,
"step": 1275
},
{
"epoch": 24.07,
"learning_rate": 8.22888888888889e-06,
"loss": 0.0094,
"step": 1300
},
{
"epoch": 24.54,
"learning_rate": 8.173333333333334e-06,
"loss": 0.0047,
"step": 1325
},
{
"epoch": 25.0,
"learning_rate": 8.11777777777778e-06,
"loss": 0.0073,
"step": 1350
},
{
"epoch": 25.46,
"learning_rate": 8.062222222222222e-06,
"loss": 0.0052,
"step": 1375
},
{
"epoch": 25.93,
"learning_rate": 8.006666666666667e-06,
"loss": 0.0042,
"step": 1400
},
{
"epoch": 26.39,
"learning_rate": 7.951111111111111e-06,
"loss": 0.002,
"step": 1425
},
{
"epoch": 26.85,
"learning_rate": 7.895555555555557e-06,
"loss": 0.0048,
"step": 1450
},
{
"epoch": 27.31,
"learning_rate": 7.840000000000001e-06,
"loss": 0.0045,
"step": 1475
},
{
"epoch": 27.78,
"learning_rate": 7.784444444444445e-06,
"loss": 0.0042,
"step": 1500
},
{
"epoch": 28.24,
"learning_rate": 7.72888888888889e-06,
"loss": 0.0042,
"step": 1525
},
{
"epoch": 28.7,
"learning_rate": 7.673333333333333e-06,
"loss": 0.0036,
"step": 1550
},
{
"epoch": 29.17,
"learning_rate": 7.617777777777778e-06,
"loss": 0.0047,
"step": 1575
},
{
"epoch": 29.63,
"learning_rate": 7.562222222222223e-06,
"loss": 0.0062,
"step": 1600
},
{
"epoch": 30.09,
"learning_rate": 7.506666666666668e-06,
"loss": 0.0047,
"step": 1625
},
{
"epoch": 30.56,
"learning_rate": 7.451111111111111e-06,
"loss": 0.0032,
"step": 1650
},
{
"epoch": 31.02,
"learning_rate": 7.395555555555556e-06,
"loss": 0.0042,
"step": 1675
},
{
"epoch": 31.48,
"learning_rate": 7.340000000000001e-06,
"loss": 0.0021,
"step": 1700
},
{
"epoch": 31.94,
"learning_rate": 7.284444444444445e-06,
"loss": 0.0021,
"step": 1725
},
{
"epoch": 32.41,
"learning_rate": 7.22888888888889e-06,
"loss": 0.0017,
"step": 1750
},
{
"epoch": 32.87,
"learning_rate": 7.173333333333335e-06,
"loss": 0.0015,
"step": 1775
},
{
"epoch": 33.33,
"learning_rate": 7.117777777777778e-06,
"loss": 0.0024,
"step": 1800
},
{
"epoch": 33.8,
"learning_rate": 7.062222222222223e-06,
"loss": 0.0009,
"step": 1825
},
{
"epoch": 34.26,
"learning_rate": 7.006666666666667e-06,
"loss": 0.0023,
"step": 1850
},
{
"epoch": 34.72,
"learning_rate": 6.951111111111112e-06,
"loss": 0.0007,
"step": 1875
},
{
"epoch": 35.19,
"learning_rate": 6.8955555555555565e-06,
"loss": 0.0009,
"step": 1900
},
{
"epoch": 35.65,
"learning_rate": 6.8400000000000014e-06,
"loss": 0.0006,
"step": 1925
},
{
"epoch": 36.11,
"learning_rate": 6.784444444444445e-06,
"loss": 0.0017,
"step": 1950
},
{
"epoch": 36.57,
"learning_rate": 6.7288888888888895e-06,
"loss": 0.0002,
"step": 1975
},
{
"epoch": 37.04,
"learning_rate": 6.6733333333333335e-06,
"loss": 0.0009,
"step": 2000
},
{
"epoch": 37.04,
"eval_loss": 0.615847110748291,
"eval_runtime": 504.1418,
"eval_samples_per_second": 0.222,
"eval_steps_per_second": 0.028,
"eval_wer": 37.867177522349934,
"step": 2000
},
{
"epoch": 37.5,
"learning_rate": 6.617777777777778e-06,
"loss": 0.0002,
"step": 2025
},
{
"epoch": 37.96,
"learning_rate": 6.562222222222223e-06,
"loss": 0.0001,
"step": 2050
},
{
"epoch": 38.43,
"learning_rate": 6.5066666666666665e-06,
"loss": 0.0001,
"step": 2075
},
{
"epoch": 38.89,
"learning_rate": 6.451111111111111e-06,
"loss": 0.0001,
"step": 2100
},
{
"epoch": 39.35,
"learning_rate": 6.395555555555556e-06,
"loss": 0.0,
"step": 2125
},
{
"epoch": 39.81,
"learning_rate": 6.34e-06,
"loss": 0.0,
"step": 2150
},
{
"epoch": 40.28,
"learning_rate": 6.284444444444445e-06,
"loss": 0.0,
"step": 2175
},
{
"epoch": 40.74,
"learning_rate": 6.22888888888889e-06,
"loss": 0.0,
"step": 2200
},
{
"epoch": 41.2,
"learning_rate": 6.173333333333333e-06,
"loss": 0.0,
"step": 2225
},
{
"epoch": 41.67,
"learning_rate": 6.117777777777778e-06,
"loss": 0.0,
"step": 2250
},
{
"epoch": 42.13,
"learning_rate": 6.062222222222223e-06,
"loss": 0.0,
"step": 2275
},
{
"epoch": 42.59,
"learning_rate": 6.006666666666667e-06,
"loss": 0.0,
"step": 2300
},
{
"epoch": 43.06,
"learning_rate": 5.951111111111112e-06,
"loss": 0.0,
"step": 2325
},
{
"epoch": 43.52,
"learning_rate": 5.895555555555557e-06,
"loss": 0.0,
"step": 2350
},
{
"epoch": 43.98,
"learning_rate": 5.84e-06,
"loss": 0.0,
"step": 2375
},
{
"epoch": 44.44,
"learning_rate": 5.784444444444445e-06,
"loss": 0.0,
"step": 2400
},
{
"epoch": 44.91,
"learning_rate": 5.72888888888889e-06,
"loss": 0.0,
"step": 2425
},
{
"epoch": 45.37,
"learning_rate": 5.673333333333334e-06,
"loss": 0.0,
"step": 2450
},
{
"epoch": 45.83,
"learning_rate": 5.617777777777779e-06,
"loss": 0.0,
"step": 2475
},
{
"epoch": 46.3,
"learning_rate": 5.562222222222222e-06,
"loss": 0.0,
"step": 2500
},
{
"epoch": 46.76,
"learning_rate": 5.506666666666667e-06,
"loss": 0.0,
"step": 2525
},
{
"epoch": 47.22,
"learning_rate": 5.451111111111112e-06,
"loss": 0.0,
"step": 2550
},
{
"epoch": 47.69,
"learning_rate": 5.3955555555555565e-06,
"loss": 0.0,
"step": 2575
},
{
"epoch": 48.15,
"learning_rate": 5.3400000000000005e-06,
"loss": 0.0,
"step": 2600
},
{
"epoch": 48.61,
"learning_rate": 5.2844444444444454e-06,
"loss": 0.0,
"step": 2625
},
{
"epoch": 49.07,
"learning_rate": 5.228888888888889e-06,
"loss": 0.0,
"step": 2650
},
{
"epoch": 49.54,
"learning_rate": 5.1733333333333335e-06,
"loss": 0.0,
"step": 2675
},
{
"epoch": 50.0,
"learning_rate": 5.117777777777778e-06,
"loss": 0.0,
"step": 2700
},
{
"epoch": 50.46,
"learning_rate": 5.062222222222222e-06,
"loss": 0.0,
"step": 2725
},
{
"epoch": 50.93,
"learning_rate": 5.006666666666667e-06,
"loss": 0.0,
"step": 2750
},
{
"epoch": 51.39,
"learning_rate": 4.951111111111111e-06,
"loss": 0.0,
"step": 2775
},
{
"epoch": 51.85,
"learning_rate": 4.895555555555556e-06,
"loss": 0.0,
"step": 2800
},
{
"epoch": 52.31,
"learning_rate": 4.84e-06,
"loss": 0.0,
"step": 2825
},
{
"epoch": 52.78,
"learning_rate": 4.784444444444445e-06,
"loss": 0.0,
"step": 2850
},
{
"epoch": 53.24,
"learning_rate": 4.728888888888889e-06,
"loss": 0.0,
"step": 2875
},
{
"epoch": 53.7,
"learning_rate": 4.673333333333333e-06,
"loss": 0.0,
"step": 2900
},
{
"epoch": 54.17,
"learning_rate": 4.617777777777778e-06,
"loss": 0.0,
"step": 2925
},
{
"epoch": 54.63,
"learning_rate": 4.562222222222222e-06,
"loss": 0.0,
"step": 2950
},
{
"epoch": 55.09,
"learning_rate": 4.506666666666667e-06,
"loss": 0.0,
"step": 2975
},
{
"epoch": 55.56,
"learning_rate": 4.451111111111112e-06,
"loss": 0.0,
"step": 3000
},
{
"epoch": 55.56,
"eval_loss": 0.6553739905357361,
"eval_runtime": 504.5772,
"eval_samples_per_second": 0.222,
"eval_steps_per_second": 0.028,
"eval_wer": 37.29246487867178,
"step": 3000
},
{
"epoch": 56.02,
"learning_rate": 4.395555555555556e-06,
"loss": 0.0,
"step": 3025
},
{
"epoch": 56.48,
"learning_rate": 4.34e-06,
"loss": 0.0,
"step": 3050
},
{
"epoch": 56.94,
"learning_rate": 4.284444444444445e-06,
"loss": 0.0,
"step": 3075
},
{
"epoch": 57.41,
"learning_rate": 4.228888888888889e-06,
"loss": 0.0,
"step": 3100
},
{
"epoch": 57.87,
"learning_rate": 4.173333333333334e-06,
"loss": 0.0,
"step": 3125
},
{
"epoch": 58.33,
"learning_rate": 4.117777777777779e-06,
"loss": 0.0,
"step": 3150
},
{
"epoch": 58.8,
"learning_rate": 4.062222222222223e-06,
"loss": 0.0,
"step": 3175
},
{
"epoch": 59.26,
"learning_rate": 4.006666666666667e-06,
"loss": 0.0,
"step": 3200
},
{
"epoch": 59.72,
"learning_rate": 3.951111111111112e-06,
"loss": 0.0,
"step": 3225
},
{
"epoch": 60.19,
"learning_rate": 3.895555555555556e-06,
"loss": 0.0,
"step": 3250
},
{
"epoch": 60.65,
"learning_rate": 3.8400000000000005e-06,
"loss": 0.0,
"step": 3275
},
{
"epoch": 61.11,
"learning_rate": 3.784444444444445e-06,
"loss": 0.0,
"step": 3300
},
{
"epoch": 61.57,
"learning_rate": 3.728888888888889e-06,
"loss": 0.0,
"step": 3325
},
{
"epoch": 62.04,
"learning_rate": 3.673333333333334e-06,
"loss": 0.0,
"step": 3350
},
{
"epoch": 62.5,
"learning_rate": 3.617777777777778e-06,
"loss": 0.0,
"step": 3375
},
{
"epoch": 62.96,
"learning_rate": 3.5622222222222224e-06,
"loss": 0.0,
"step": 3400
},
{
"epoch": 63.43,
"learning_rate": 3.5066666666666673e-06,
"loss": 0.0,
"step": 3425
},
{
"epoch": 63.89,
"learning_rate": 3.4511111111111113e-06,
"loss": 0.0,
"step": 3450
},
{
"epoch": 64.35,
"learning_rate": 3.3955555555555558e-06,
"loss": 0.0,
"step": 3475
},
{
"epoch": 64.81,
"learning_rate": 3.3400000000000006e-06,
"loss": 0.0,
"step": 3500
},
{
"epoch": 65.28,
"learning_rate": 3.2844444444444447e-06,
"loss": 0.0,
"step": 3525
},
{
"epoch": 65.74,
"learning_rate": 3.228888888888889e-06,
"loss": 0.0,
"step": 3550
},
{
"epoch": 66.2,
"learning_rate": 3.173333333333334e-06,
"loss": 0.0,
"step": 3575
},
{
"epoch": 66.67,
"learning_rate": 3.117777777777778e-06,
"loss": 0.0,
"step": 3600
},
{
"epoch": 67.13,
"learning_rate": 3.0622222222222225e-06,
"loss": 0.0,
"step": 3625
},
{
"epoch": 67.59,
"learning_rate": 3.0066666666666674e-06,
"loss": 0.0,
"step": 3650
},
{
"epoch": 68.06,
"learning_rate": 2.9511111111111114e-06,
"loss": 0.0,
"step": 3675
},
{
"epoch": 68.52,
"learning_rate": 2.895555555555556e-06,
"loss": 0.0,
"step": 3700
},
{
"epoch": 68.98,
"learning_rate": 2.84e-06,
"loss": 0.0,
"step": 3725
},
{
"epoch": 69.44,
"learning_rate": 2.784444444444445e-06,
"loss": 0.0,
"step": 3750
},
{
"epoch": 69.91,
"learning_rate": 2.7288888888888893e-06,
"loss": 0.0,
"step": 3775
},
{
"epoch": 70.37,
"learning_rate": 2.6733333333333333e-06,
"loss": 0.0,
"step": 3800
},
{
"epoch": 70.83,
"learning_rate": 2.617777777777778e-06,
"loss": 0.0,
"step": 3825
},
{
"epoch": 71.3,
"learning_rate": 2.5622222222222226e-06,
"loss": 0.0,
"step": 3850
},
{
"epoch": 71.76,
"learning_rate": 2.5066666666666667e-06,
"loss": 0.0,
"step": 3875
},
{
"epoch": 72.22,
"learning_rate": 2.451111111111111e-06,
"loss": 0.0,
"step": 3900
},
{
"epoch": 72.69,
"learning_rate": 2.3955555555555556e-06,
"loss": 0.0,
"step": 3925
},
{
"epoch": 73.15,
"learning_rate": 2.3400000000000005e-06,
"loss": 0.0,
"step": 3950
},
{
"epoch": 73.61,
"learning_rate": 2.2844444444444445e-06,
"loss": 0.0,
"step": 3975
},
{
"epoch": 74.07,
"learning_rate": 2.228888888888889e-06,
"loss": 0.0,
"step": 4000
},
{
"epoch": 74.07,
"eval_loss": 0.6719385981559753,
"eval_runtime": 508.516,
"eval_samples_per_second": 0.22,
"eval_steps_per_second": 0.028,
"eval_wer": 37.29246487867178,
"step": 4000
},
{
"epoch": 74.54,
"learning_rate": 2.1733333333333334e-06,
"loss": 0.0,
"step": 4025
},
{
"epoch": 75.0,
"learning_rate": 2.117777777777778e-06,
"loss": 0.0,
"step": 4050
},
{
"epoch": 75.46,
"learning_rate": 2.0622222222222223e-06,
"loss": 0.0,
"step": 4075
},
{
"epoch": 75.93,
"learning_rate": 2.006666666666667e-06,
"loss": 0.0,
"step": 4100
},
{
"epoch": 76.39,
"learning_rate": 1.9511111111111113e-06,
"loss": 0.0,
"step": 4125
},
{
"epoch": 76.85,
"learning_rate": 1.8955555555555557e-06,
"loss": 0.0,
"step": 4150
},
{
"epoch": 77.31,
"learning_rate": 1.8400000000000002e-06,
"loss": 0.0,
"step": 4175
},
{
"epoch": 77.78,
"learning_rate": 1.7844444444444444e-06,
"loss": 0.0,
"step": 4200
},
{
"epoch": 78.24,
"learning_rate": 1.728888888888889e-06,
"loss": 0.0,
"step": 4225
},
{
"epoch": 78.7,
"learning_rate": 1.6733333333333335e-06,
"loss": 0.0,
"step": 4250
},
{
"epoch": 79.17,
"learning_rate": 1.6177777777777778e-06,
"loss": 0.0,
"step": 4275
},
{
"epoch": 79.63,
"learning_rate": 1.5622222222222225e-06,
"loss": 0.0,
"step": 4300
},
{
"epoch": 80.09,
"learning_rate": 1.506666666666667e-06,
"loss": 0.0,
"step": 4325
},
{
"epoch": 80.56,
"learning_rate": 1.4511111111111112e-06,
"loss": 0.0,
"step": 4350
},
{
"epoch": 81.02,
"learning_rate": 1.3955555555555556e-06,
"loss": 0.0,
"step": 4375
},
{
"epoch": 81.48,
"learning_rate": 1.34e-06,
"loss": 0.0,
"step": 4400
},
{
"epoch": 81.94,
"learning_rate": 1.2844444444444445e-06,
"loss": 0.0,
"step": 4425
},
{
"epoch": 82.41,
"learning_rate": 1.228888888888889e-06,
"loss": 0.0,
"step": 4450
},
{
"epoch": 82.87,
"learning_rate": 1.1733333333333335e-06,
"loss": 0.0,
"step": 4475
},
{
"epoch": 83.33,
"learning_rate": 1.117777777777778e-06,
"loss": 0.0,
"step": 4500
},
{
"epoch": 83.8,
"learning_rate": 1.0622222222222222e-06,
"loss": 0.0,
"step": 4525
},
{
"epoch": 84.26,
"learning_rate": 1.0066666666666668e-06,
"loss": 0.0,
"step": 4550
},
{
"epoch": 84.72,
"learning_rate": 9.511111111111111e-07,
"loss": 0.0,
"step": 4575
},
{
"epoch": 85.19,
"learning_rate": 8.955555555555557e-07,
"loss": 0.0,
"step": 4600
},
{
"epoch": 85.65,
"learning_rate": 8.400000000000001e-07,
"loss": 0.0,
"step": 4625
},
{
"epoch": 86.11,
"learning_rate": 7.844444444444445e-07,
"loss": 0.0,
"step": 4650
},
{
"epoch": 86.57,
"learning_rate": 7.28888888888889e-07,
"loss": 0.0,
"step": 4675
},
{
"epoch": 87.04,
"learning_rate": 6.733333333333334e-07,
"loss": 0.0,
"step": 4700
},
{
"epoch": 87.5,
"learning_rate": 6.177777777777778e-07,
"loss": 0.0,
"step": 4725
},
{
"epoch": 87.96,
"learning_rate": 5.622222222222223e-07,
"loss": 0.0,
"step": 4750
},
{
"epoch": 88.43,
"learning_rate": 5.066666666666667e-07,
"loss": 0.0,
"step": 4775
},
{
"epoch": 88.89,
"learning_rate": 4.511111111111111e-07,
"loss": 0.0,
"step": 4800
},
{
"epoch": 89.35,
"learning_rate": 3.9555555555555557e-07,
"loss": 0.0,
"step": 4825
},
{
"epoch": 89.81,
"learning_rate": 3.4000000000000003e-07,
"loss": 0.0,
"step": 4850
},
{
"epoch": 90.28,
"learning_rate": 2.844444444444445e-07,
"loss": 0.0,
"step": 4875
},
{
"epoch": 90.74,
"learning_rate": 2.2888888888888892e-07,
"loss": 0.0,
"step": 4900
},
{
"epoch": 91.2,
"learning_rate": 1.7333333333333335e-07,
"loss": 0.0,
"step": 4925
},
{
"epoch": 91.67,
"learning_rate": 1.1777777777777778e-07,
"loss": 0.0,
"step": 4950
},
{
"epoch": 92.13,
"learning_rate": 6.222222222222223e-08,
"loss": 0.0,
"step": 4975
},
{
"epoch": 92.59,
"learning_rate": 6.666666666666667e-09,
"loss": 0.0,
"step": 5000
},
{
"epoch": 92.59,
"eval_loss": 0.6782652735710144,
"eval_runtime": 503.7576,
"eval_samples_per_second": 0.222,
"eval_steps_per_second": 0.028,
"eval_wer": 37.29246487867178,
"step": 5000
}
],
"max_steps": 5000,
"num_train_epochs": 93,
"total_flos": 1.149031629914112e+19,
"trial_name": null,
"trial_params": null
}