test / trainer_state.json
wangpuupup's picture
Upload 18 files
654d627
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 9.0892,
"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": 5.000000000000001e-07,
"loss": 3.0757,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 1.0000000000000002e-06,
"loss": 2.884,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.5e-06,
"loss": 2.2098,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.3244,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.5e-06,
"loss": 0.7112,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.6626,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 3.5e-06,
"loss": 0.5772,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 4.000000000000001e-06,
"loss": 0.5288,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.5e-06,
"loss": 0.5758,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 5e-06,
"loss": 0.5197,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 5.500000000000001e-06,
"loss": 0.5935,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 6e-06,
"loss": 0.3028,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.4619,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 7e-06,
"loss": 0.3713,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 7.500000000000001e-06,
"loss": 0.3963,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 8.000000000000001e-06,
"loss": 0.4508,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 8.5e-06,
"loss": 0.4108,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 9e-06,
"loss": 0.5043,
"step": 450
},
{
"epoch": 0.1,
"learning_rate": 9.5e-06,
"loss": 0.2167,
"step": 475
},
{
"epoch": 0.1,
"learning_rate": 1e-05,
"loss": 0.3462,
"step": 500
},
{
"epoch": 0.1,
"eval_loss": 0.3720855116844177,
"eval_runtime": 314.77,
"eval_samples_per_second": 9.191,
"eval_steps_per_second": 0.289,
"eval_wer": 29.974628488582823,
"step": 500
},
{
"epoch": 1.0,
"learning_rate": 9.944444444444445e-06,
"loss": 0.2946,
"step": 525
},
{
"epoch": 1.01,
"learning_rate": 9.88888888888889e-06,
"loss": 0.3579,
"step": 550
},
{
"epoch": 1.01,
"learning_rate": 9.833333333333333e-06,
"loss": 0.4014,
"step": 575
},
{
"epoch": 1.02,
"learning_rate": 9.777777777777779e-06,
"loss": 0.3716,
"step": 600
},
{
"epoch": 1.02,
"learning_rate": 9.722222222222223e-06,
"loss": 0.3812,
"step": 625
},
{
"epoch": 1.03,
"learning_rate": 9.666666666666667e-06,
"loss": 0.2322,
"step": 650
},
{
"epoch": 1.03,
"learning_rate": 9.611111111111112e-06,
"loss": 0.2057,
"step": 675
},
{
"epoch": 1.04,
"learning_rate": 9.555555555555556e-06,
"loss": 0.2192,
"step": 700
},
{
"epoch": 1.04,
"learning_rate": 9.5e-06,
"loss": 0.2884,
"step": 725
},
{
"epoch": 1.05,
"learning_rate": 9.444444444444445e-06,
"loss": 0.3185,
"step": 750
},
{
"epoch": 1.05,
"learning_rate": 9.38888888888889e-06,
"loss": 0.3511,
"step": 775
},
{
"epoch": 1.06,
"learning_rate": 9.333333333333334e-06,
"loss": 0.2279,
"step": 800
},
{
"epoch": 1.06,
"learning_rate": 9.277777777777778e-06,
"loss": 0.17,
"step": 825
},
{
"epoch": 1.07,
"learning_rate": 9.222222222222224e-06,
"loss": 0.1418,
"step": 850
},
{
"epoch": 1.07,
"learning_rate": 9.166666666666666e-06,
"loss": 0.2223,
"step": 875
},
{
"epoch": 1.08,
"learning_rate": 9.111111111111112e-06,
"loss": 0.2963,
"step": 900
},
{
"epoch": 1.08,
"learning_rate": 9.055555555555556e-06,
"loss": 0.2686,
"step": 925
},
{
"epoch": 1.09,
"learning_rate": 9e-06,
"loss": 0.3573,
"step": 950
},
{
"epoch": 1.09,
"learning_rate": 8.944444444444446e-06,
"loss": 0.1536,
"step": 975
},
{
"epoch": 1.1,
"learning_rate": 8.888888888888888e-06,
"loss": 0.1205,
"step": 1000
},
{
"epoch": 1.1,
"eval_loss": 0.3286285996437073,
"eval_runtime": 312.004,
"eval_samples_per_second": 9.272,
"eval_steps_per_second": 0.292,
"eval_wer": 26.422616890177604,
"step": 1000
},
{
"epoch": 2.0,
"learning_rate": 8.833333333333334e-06,
"loss": 0.1435,
"step": 1025
},
{
"epoch": 2.01,
"learning_rate": 8.777777777777778e-06,
"loss": 0.2308,
"step": 1050
},
{
"epoch": 2.01,
"learning_rate": 8.722222222222224e-06,
"loss": 0.2836,
"step": 1075
},
{
"epoch": 2.02,
"learning_rate": 8.666666666666668e-06,
"loss": 0.2702,
"step": 1100
},
{
"epoch": 2.02,
"learning_rate": 8.611111111111112e-06,
"loss": 0.3131,
"step": 1125
},
{
"epoch": 2.03,
"learning_rate": 8.555555555555556e-06,
"loss": 0.1122,
"step": 1150
},
{
"epoch": 2.03,
"learning_rate": 8.5e-06,
"loss": 0.0917,
"step": 1175
},
{
"epoch": 2.04,
"learning_rate": 8.444444444444446e-06,
"loss": 0.1198,
"step": 1200
},
{
"epoch": 2.04,
"learning_rate": 8.38888888888889e-06,
"loss": 0.1889,
"step": 1225
},
{
"epoch": 2.05,
"learning_rate": 8.333333333333334e-06,
"loss": 0.2362,
"step": 1250
},
{
"epoch": 2.05,
"learning_rate": 8.277777777777778e-06,
"loss": 0.2131,
"step": 1275
},
{
"epoch": 2.06,
"learning_rate": 8.222222222222222e-06,
"loss": 0.2183,
"step": 1300
},
{
"epoch": 2.06,
"learning_rate": 8.166666666666668e-06,
"loss": 0.0794,
"step": 1325
},
{
"epoch": 2.07,
"learning_rate": 8.111111111111112e-06,
"loss": 0.0543,
"step": 1350
},
{
"epoch": 2.07,
"learning_rate": 8.055555555555557e-06,
"loss": 0.1325,
"step": 1375
},
{
"epoch": 2.08,
"learning_rate": 8.000000000000001e-06,
"loss": 0.2008,
"step": 1400
},
{
"epoch": 2.08,
"learning_rate": 7.944444444444445e-06,
"loss": 0.2229,
"step": 1425
},
{
"epoch": 2.09,
"learning_rate": 7.88888888888889e-06,
"loss": 0.2455,
"step": 1450
},
{
"epoch": 2.09,
"learning_rate": 7.833333333333333e-06,
"loss": 0.159,
"step": 1475
},
{
"epoch": 2.1,
"learning_rate": 7.77777777777778e-06,
"loss": 0.0642,
"step": 1500
},
{
"epoch": 2.1,
"eval_loss": 0.31385740637779236,
"eval_runtime": 327.7385,
"eval_samples_per_second": 8.827,
"eval_steps_per_second": 0.278,
"eval_wer": 25.26277636824937,
"step": 1500
},
{
"epoch": 3.0,
"learning_rate": 7.722222222222223e-06,
"loss": 0.0678,
"step": 1525
},
{
"epoch": 3.01,
"learning_rate": 7.666666666666667e-06,
"loss": 0.1429,
"step": 1550
},
{
"epoch": 3.01,
"learning_rate": 7.611111111111111e-06,
"loss": 0.2157,
"step": 1575
},
{
"epoch": 3.02,
"learning_rate": 7.555555555555556e-06,
"loss": 0.2085,
"step": 1600
},
{
"epoch": 3.02,
"learning_rate": 7.500000000000001e-06,
"loss": 0.2483,
"step": 1625
},
{
"epoch": 3.03,
"learning_rate": 7.444444444444445e-06,
"loss": 0.1015,
"step": 1650
},
{
"epoch": 3.03,
"learning_rate": 7.38888888888889e-06,
"loss": 0.0447,
"step": 1675
},
{
"epoch": 3.04,
"learning_rate": 7.333333333333333e-06,
"loss": 0.0576,
"step": 1700
},
{
"epoch": 3.04,
"learning_rate": 7.277777777777778e-06,
"loss": 0.1164,
"step": 1725
},
{
"epoch": 3.05,
"learning_rate": 7.222222222222223e-06,
"loss": 0.1618,
"step": 1750
},
{
"epoch": 3.05,
"learning_rate": 7.166666666666667e-06,
"loss": 0.1469,
"step": 1775
},
{
"epoch": 3.06,
"learning_rate": 7.111111111111112e-06,
"loss": 0.1904,
"step": 1800
},
{
"epoch": 3.06,
"learning_rate": 7.055555555555557e-06,
"loss": 0.0464,
"step": 1825
},
{
"epoch": 3.07,
"learning_rate": 7e-06,
"loss": 0.0178,
"step": 1850
},
{
"epoch": 3.07,
"learning_rate": 6.944444444444445e-06,
"loss": 0.0773,
"step": 1875
},
{
"epoch": 3.08,
"learning_rate": 6.88888888888889e-06,
"loss": 0.1279,
"step": 1900
},
{
"epoch": 3.08,
"learning_rate": 6.833333333333334e-06,
"loss": 0.1693,
"step": 1925
},
{
"epoch": 3.09,
"learning_rate": 6.777777777777779e-06,
"loss": 0.1585,
"step": 1950
},
{
"epoch": 3.09,
"learning_rate": 6.7222222222222235e-06,
"loss": 0.1513,
"step": 1975
},
{
"epoch": 3.1,
"learning_rate": 6.666666666666667e-06,
"loss": 0.0443,
"step": 2000
},
{
"epoch": 3.1,
"eval_loss": 0.31686556339263916,
"eval_runtime": 330.2543,
"eval_samples_per_second": 8.76,
"eval_steps_per_second": 0.276,
"eval_wer": 25.045306270387822,
"step": 2000
},
{
"epoch": 4.0,
"learning_rate": 6.6111111111111115e-06,
"loss": 0.0176,
"step": 2025
},
{
"epoch": 4.01,
"learning_rate": 6.555555555555556e-06,
"loss": 0.1027,
"step": 2050
},
{
"epoch": 4.01,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.1393,
"step": 2075
},
{
"epoch": 4.02,
"learning_rate": 6.444444444444445e-06,
"loss": 0.1621,
"step": 2100
},
{
"epoch": 4.02,
"learning_rate": 6.3888888888888885e-06,
"loss": 0.1748,
"step": 2125
},
{
"epoch": 4.03,
"learning_rate": 6.333333333333333e-06,
"loss": 0.1044,
"step": 2150
},
{
"epoch": 4.03,
"learning_rate": 6.277777777777778e-06,
"loss": 0.0312,
"step": 2175
},
{
"epoch": 4.04,
"learning_rate": 6.222222222222223e-06,
"loss": 0.0265,
"step": 2200
},
{
"epoch": 4.04,
"learning_rate": 6.166666666666667e-06,
"loss": 0.0631,
"step": 2225
},
{
"epoch": 4.05,
"learning_rate": 6.111111111111112e-06,
"loss": 0.1107,
"step": 2250
},
{
"epoch": 4.05,
"learning_rate": 6.055555555555555e-06,
"loss": 0.093,
"step": 2275
},
{
"epoch": 4.06,
"learning_rate": 6e-06,
"loss": 0.1339,
"step": 2300
},
{
"epoch": 4.06,
"learning_rate": 5.944444444444445e-06,
"loss": 0.0413,
"step": 2325
},
{
"epoch": 4.07,
"learning_rate": 5.88888888888889e-06,
"loss": 0.012,
"step": 2350
},
{
"epoch": 4.07,
"learning_rate": 5.833333333333334e-06,
"loss": 0.0367,
"step": 2375
},
{
"epoch": 4.08,
"learning_rate": 5.777777777777778e-06,
"loss": 0.0755,
"step": 2400
},
{
"epoch": 4.08,
"learning_rate": 5.722222222222222e-06,
"loss": 0.1216,
"step": 2425
},
{
"epoch": 4.09,
"learning_rate": 5.666666666666667e-06,
"loss": 0.0972,
"step": 2450
},
{
"epoch": 4.09,
"learning_rate": 5.611111111111112e-06,
"loss": 0.1324,
"step": 2475
},
{
"epoch": 4.1,
"learning_rate": 5.555555555555557e-06,
"loss": 0.0356,
"step": 2500
},
{
"epoch": 4.1,
"eval_loss": 0.322619765996933,
"eval_runtime": 312.2275,
"eval_samples_per_second": 9.266,
"eval_steps_per_second": 0.291,
"eval_wer": 25.06342877854295,
"step": 2500
},
{
"epoch": 4.1,
"learning_rate": 5.500000000000001e-06,
"loss": 0.0096,
"step": 2525
},
{
"epoch": 5.0,
"learning_rate": 5.444444444444445e-06,
"loss": 0.0589,
"step": 2550
},
{
"epoch": 5.01,
"learning_rate": 5.388888888888889e-06,
"loss": 0.0871,
"step": 2575
},
{
"epoch": 5.01,
"learning_rate": 5.333333333333334e-06,
"loss": 0.1181,
"step": 2600
},
{
"epoch": 5.02,
"learning_rate": 5.2777777777777785e-06,
"loss": 0.106,
"step": 2625
},
{
"epoch": 5.02,
"learning_rate": 5.2222222222222226e-06,
"loss": 0.1025,
"step": 2650
},
{
"epoch": 5.03,
"learning_rate": 5.1666666666666675e-06,
"loss": 0.0294,
"step": 2675
},
{
"epoch": 5.03,
"learning_rate": 5.1111111111111115e-06,
"loss": 0.0117,
"step": 2700
},
{
"epoch": 5.04,
"learning_rate": 5.0555555555555555e-06,
"loss": 0.0298,
"step": 2725
},
{
"epoch": 5.04,
"learning_rate": 5e-06,
"loss": 0.0558,
"step": 2750
},
{
"epoch": 5.05,
"learning_rate": 4.944444444444445e-06,
"loss": 0.0727,
"step": 2775
},
{
"epoch": 5.05,
"learning_rate": 4.888888888888889e-06,
"loss": 0.0781,
"step": 2800
},
{
"epoch": 5.06,
"learning_rate": 4.833333333333333e-06,
"loss": 0.0446,
"step": 2825
},
{
"epoch": 5.06,
"learning_rate": 4.777777777777778e-06,
"loss": 0.0104,
"step": 2850
},
{
"epoch": 5.07,
"learning_rate": 4.722222222222222e-06,
"loss": 0.0161,
"step": 2875
},
{
"epoch": 5.07,
"learning_rate": 4.666666666666667e-06,
"loss": 0.0472,
"step": 2900
},
{
"epoch": 5.08,
"learning_rate": 4.611111111111112e-06,
"loss": 0.0736,
"step": 2925
},
{
"epoch": 5.08,
"learning_rate": 4.555555555555556e-06,
"loss": 0.0654,
"step": 2950
},
{
"epoch": 5.09,
"learning_rate": 4.5e-06,
"loss": 0.1007,
"step": 2975
},
{
"epoch": 5.09,
"learning_rate": 4.444444444444444e-06,
"loss": 0.0321,
"step": 3000
},
{
"epoch": 5.09,
"eval_loss": 0.32373714447021484,
"eval_runtime": 307.9215,
"eval_samples_per_second": 9.395,
"eval_steps_per_second": 0.296,
"eval_wer": 24.719101123595504,
"step": 3000
},
{
"epoch": 5.1,
"learning_rate": 4.388888888888889e-06,
"loss": 0.0088,
"step": 3025
},
{
"epoch": 6.0,
"learning_rate": 4.333333333333334e-06,
"loss": 0.0323,
"step": 3050
},
{
"epoch": 6.01,
"learning_rate": 4.277777777777778e-06,
"loss": 0.0523,
"step": 3075
},
{
"epoch": 6.01,
"learning_rate": 4.222222222222223e-06,
"loss": 0.0798,
"step": 3100
},
{
"epoch": 6.02,
"learning_rate": 4.166666666666667e-06,
"loss": 0.0746,
"step": 3125
},
{
"epoch": 6.02,
"learning_rate": 4.111111111111111e-06,
"loss": 0.0819,
"step": 3150
},
{
"epoch": 6.03,
"learning_rate": 4.055555555555556e-06,
"loss": 0.0265,
"step": 3175
},
{
"epoch": 6.03,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0065,
"step": 3200
},
{
"epoch": 6.04,
"learning_rate": 3.944444444444445e-06,
"loss": 0.0157,
"step": 3225
},
{
"epoch": 6.04,
"learning_rate": 3.88888888888889e-06,
"loss": 0.0303,
"step": 3250
},
{
"epoch": 6.05,
"learning_rate": 3.833333333333334e-06,
"loss": 0.0464,
"step": 3275
},
{
"epoch": 6.05,
"learning_rate": 3.777777777777778e-06,
"loss": 0.0404,
"step": 3300
},
{
"epoch": 6.06,
"learning_rate": 3.7222222222222225e-06,
"loss": 0.0411,
"step": 3325
},
{
"epoch": 6.06,
"learning_rate": 3.6666666666666666e-06,
"loss": 0.0095,
"step": 3350
},
{
"epoch": 6.07,
"learning_rate": 3.6111111111111115e-06,
"loss": 0.0061,
"step": 3375
},
{
"epoch": 6.07,
"learning_rate": 3.555555555555556e-06,
"loss": 0.0265,
"step": 3400
},
{
"epoch": 6.08,
"learning_rate": 3.5e-06,
"loss": 0.0433,
"step": 3425
},
{
"epoch": 6.08,
"learning_rate": 3.444444444444445e-06,
"loss": 0.0495,
"step": 3450
},
{
"epoch": 6.09,
"learning_rate": 3.3888888888888893e-06,
"loss": 0.0637,
"step": 3475
},
{
"epoch": 6.09,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.0317,
"step": 3500
},
{
"epoch": 6.09,
"eval_loss": 0.3322317600250244,
"eval_runtime": 327.7773,
"eval_samples_per_second": 8.826,
"eval_steps_per_second": 0.278,
"eval_wer": 24.77346864806089,
"step": 3500
},
{
"epoch": 6.1,
"learning_rate": 3.277777777777778e-06,
"loss": 0.0094,
"step": 3525
},
{
"epoch": 7.0,
"learning_rate": 3.2222222222222227e-06,
"loss": 0.0148,
"step": 3550
},
{
"epoch": 7.01,
"learning_rate": 3.1666666666666667e-06,
"loss": 0.0339,
"step": 3575
},
{
"epoch": 7.01,
"learning_rate": 3.1111111111111116e-06,
"loss": 0.0567,
"step": 3600
},
{
"epoch": 7.02,
"learning_rate": 3.055555555555556e-06,
"loss": 0.0513,
"step": 3625
},
{
"epoch": 7.02,
"learning_rate": 3e-06,
"loss": 0.0621,
"step": 3650
},
{
"epoch": 7.03,
"learning_rate": 2.944444444444445e-06,
"loss": 0.0234,
"step": 3675
},
{
"epoch": 7.03,
"learning_rate": 2.888888888888889e-06,
"loss": 0.0065,
"step": 3700
},
{
"epoch": 7.04,
"learning_rate": 2.8333333333333335e-06,
"loss": 0.0089,
"step": 3725
},
{
"epoch": 7.04,
"learning_rate": 2.7777777777777783e-06,
"loss": 0.0174,
"step": 3750
},
{
"epoch": 7.05,
"learning_rate": 2.7222222222222224e-06,
"loss": 0.0279,
"step": 3775
},
{
"epoch": 7.05,
"learning_rate": 2.666666666666667e-06,
"loss": 0.0251,
"step": 3800
},
{
"epoch": 7.06,
"learning_rate": 2.6111111111111113e-06,
"loss": 0.0315,
"step": 3825
},
{
"epoch": 7.06,
"learning_rate": 2.5555555555555557e-06,
"loss": 0.0077,
"step": 3850
},
{
"epoch": 7.07,
"learning_rate": 2.5e-06,
"loss": 0.0036,
"step": 3875
},
{
"epoch": 7.07,
"learning_rate": 2.4444444444444447e-06,
"loss": 0.0153,
"step": 3900
},
{
"epoch": 7.08,
"learning_rate": 2.388888888888889e-06,
"loss": 0.0285,
"step": 3925
},
{
"epoch": 7.08,
"learning_rate": 2.3333333333333336e-06,
"loss": 0.0366,
"step": 3950
},
{
"epoch": 7.09,
"learning_rate": 2.277777777777778e-06,
"loss": 0.0375,
"step": 3975
},
{
"epoch": 7.09,
"learning_rate": 2.222222222222222e-06,
"loss": 0.0347,
"step": 4000
},
{
"epoch": 7.09,
"eval_loss": 0.3355422019958496,
"eval_runtime": 314.4963,
"eval_samples_per_second": 9.199,
"eval_steps_per_second": 0.289,
"eval_wer": 25.06342877854295,
"step": 4000
},
{
"epoch": 7.1,
"learning_rate": 2.166666666666667e-06,
"loss": 0.0084,
"step": 4025
},
{
"epoch": 8.0,
"learning_rate": 2.1111111111111114e-06,
"loss": 0.0055,
"step": 4050
},
{
"epoch": 8.01,
"learning_rate": 2.0555555555555555e-06,
"loss": 0.0252,
"step": 4075
},
{
"epoch": 8.01,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.0366,
"step": 4100
},
{
"epoch": 8.02,
"learning_rate": 1.944444444444445e-06,
"loss": 0.0421,
"step": 4125
},
{
"epoch": 8.02,
"learning_rate": 1.888888888888889e-06,
"loss": 0.0465,
"step": 4150
},
{
"epoch": 8.03,
"learning_rate": 1.8333333333333333e-06,
"loss": 0.0252,
"step": 4175
},
{
"epoch": 8.03,
"learning_rate": 1.777777777777778e-06,
"loss": 0.0075,
"step": 4200
},
{
"epoch": 8.04,
"learning_rate": 1.7222222222222224e-06,
"loss": 0.0049,
"step": 4225
},
{
"epoch": 8.04,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.0119,
"step": 4250
},
{
"epoch": 8.05,
"learning_rate": 1.6111111111111113e-06,
"loss": 0.0193,
"step": 4275
},
{
"epoch": 8.05,
"learning_rate": 1.5555555555555558e-06,
"loss": 0.0169,
"step": 4300
},
{
"epoch": 8.06,
"learning_rate": 1.5e-06,
"loss": 0.0246,
"step": 4325
},
{
"epoch": 8.06,
"learning_rate": 1.4444444444444445e-06,
"loss": 0.0083,
"step": 4350
},
{
"epoch": 8.07,
"learning_rate": 1.3888888888888892e-06,
"loss": 0.0035,
"step": 4375
},
{
"epoch": 8.07,
"learning_rate": 1.3333333333333334e-06,
"loss": 0.0095,
"step": 4400
},
{
"epoch": 8.08,
"learning_rate": 1.2777777777777779e-06,
"loss": 0.0196,
"step": 4425
},
{
"epoch": 8.08,
"learning_rate": 1.2222222222222223e-06,
"loss": 0.0269,
"step": 4450
},
{
"epoch": 8.09,
"learning_rate": 1.1666666666666668e-06,
"loss": 0.0255,
"step": 4475
},
{
"epoch": 8.09,
"learning_rate": 1.111111111111111e-06,
"loss": 0.0357,
"step": 4500
},
{
"epoch": 8.09,
"eval_loss": 0.33636125922203064,
"eval_runtime": 316.5763,
"eval_samples_per_second": 9.138,
"eval_steps_per_second": 0.287,
"eval_wer": 25.22653135193911,
"step": 4500
},
{
"epoch": 8.1,
"learning_rate": 1.0555555555555557e-06,
"loss": 0.0088,
"step": 4525
},
{
"epoch": 8.1,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.0032,
"step": 4550
},
{
"epoch": 9.0,
"learning_rate": 9.444444444444445e-07,
"loss": 0.0169,
"step": 4575
},
{
"epoch": 9.01,
"learning_rate": 8.88888888888889e-07,
"loss": 0.0328,
"step": 4600
},
{
"epoch": 9.01,
"learning_rate": 8.333333333333333e-07,
"loss": 0.0384,
"step": 4625
},
{
"epoch": 9.02,
"learning_rate": 7.777777777777779e-07,
"loss": 0.0361,
"step": 4650
},
{
"epoch": 9.02,
"learning_rate": 7.222222222222222e-07,
"loss": 0.0315,
"step": 4675
},
{
"epoch": 9.03,
"learning_rate": 6.666666666666667e-07,
"loss": 0.0091,
"step": 4700
},
{
"epoch": 9.03,
"learning_rate": 6.111111111111112e-07,
"loss": 0.0038,
"step": 4725
},
{
"epoch": 9.04,
"learning_rate": 5.555555555555555e-07,
"loss": 0.0076,
"step": 4750
},
{
"epoch": 9.04,
"learning_rate": 5.000000000000001e-07,
"loss": 0.0182,
"step": 4775
},
{
"epoch": 9.05,
"learning_rate": 4.444444444444445e-07,
"loss": 0.0178,
"step": 4800
},
{
"epoch": 9.05,
"learning_rate": 3.8888888888888895e-07,
"loss": 0.0185,
"step": 4825
},
{
"epoch": 9.06,
"learning_rate": 3.3333333333333335e-07,
"loss": 0.0119,
"step": 4850
},
{
"epoch": 9.06,
"learning_rate": 2.7777777777777776e-07,
"loss": 0.0046,
"step": 4875
},
{
"epoch": 9.07,
"learning_rate": 2.2222222222222224e-07,
"loss": 0.0071,
"step": 4900
},
{
"epoch": 9.07,
"learning_rate": 1.6666666666666668e-07,
"loss": 0.0143,
"step": 4925
},
{
"epoch": 9.08,
"learning_rate": 1.1111111111111112e-07,
"loss": 0.0289,
"step": 4950
},
{
"epoch": 9.08,
"learning_rate": 5.555555555555556e-08,
"loss": 0.0231,
"step": 4975
},
{
"epoch": 9.09,
"learning_rate": 0.0,
"loss": 0.0318,
"step": 5000
},
{
"epoch": 9.09,
"eval_loss": 0.33143386244773865,
"eval_runtime": 315.5878,
"eval_samples_per_second": 9.167,
"eval_steps_per_second": 0.288,
"eval_wer": 25.08155128669808,
"step": 5000
},
{
"epoch": 9.09,
"step": 5000,
"total_flos": 9.231875814776832e+19,
"train_loss": 0.1675409890934825,
"train_runtime": 22134.9725,
"train_samples_per_second": 14.457,
"train_steps_per_second": 0.226
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 9.231875814776832e+19,
"trial_name": null,
"trial_params": null
}