yt-special-batch8 / trainer_state.json
TheRains's picture
End of training
6423534
raw
history blame
25.2 kB
{
"best_metric": 100.0,
"best_model_checkpoint": "./output/small/yt-special-batch8/checkpoint-1000",
"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": 2.3488,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.200000000000001e-07,
"loss": 1.8706,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.42e-06,
"loss": 1.5618,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.9200000000000003e-06,
"loss": 1.2087,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 2.42e-06,
"loss": 1.8115,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.92e-06,
"loss": 3.7546,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 3.4200000000000007e-06,
"loss": 4.7425,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 3.920000000000001e-06,
"loss": 11.5886,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.42e-06,
"loss": 16.6871,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 4.880000000000001e-06,
"loss": 30.0066,
"step": 250
},
{
"epoch": 0.05,
"learning_rate": 5.380000000000001e-06,
"loss": 40.1869,
"step": 275
},
{
"epoch": 0.05,
"learning_rate": 5.86e-06,
"loss": 38.9384,
"step": 300
},
{
"epoch": 0.06,
"learning_rate": 6.3200000000000005e-06,
"loss": 67.8038,
"step": 325
},
{
"epoch": 0.06,
"learning_rate": 6.800000000000001e-06,
"loss": 99.3353,
"step": 350
},
{
"epoch": 0.06,
"learning_rate": 7.280000000000001e-06,
"loss": 109.246,
"step": 375
},
{
"epoch": 0.07,
"learning_rate": 7.78e-06,
"loss": 127.3227,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 8.28e-06,
"loss": 133.3727,
"step": 425
},
{
"epoch": 0.08,
"learning_rate": 8.78e-06,
"loss": 139.7458,
"step": 450
},
{
"epoch": 0.08,
"learning_rate": 9.280000000000001e-06,
"loss": 146.024,
"step": 475
},
{
"epoch": 0.09,
"learning_rate": 9.760000000000001e-06,
"loss": 191.6763,
"step": 500
},
{
"epoch": 0.09,
"learning_rate": 9.97111111111111e-06,
"loss": 176.9206,
"step": 525
},
{
"epoch": 0.09,
"learning_rate": 9.917777777777779e-06,
"loss": 207.3772,
"step": 550
},
{
"epoch": 0.1,
"learning_rate": 9.862222222222223e-06,
"loss": 220.9518,
"step": 575
},
{
"epoch": 0.1,
"learning_rate": 9.806666666666667e-06,
"loss": 296.3531,
"step": 600
},
{
"epoch": 0.11,
"learning_rate": 9.751111111111113e-06,
"loss": 408.2827,
"step": 625
},
{
"epoch": 0.11,
"learning_rate": 9.695555555555555e-06,
"loss": 315.1208,
"step": 650
},
{
"epoch": 0.12,
"learning_rate": 9.640000000000001e-06,
"loss": 369.5574,
"step": 675
},
{
"epoch": 0.12,
"learning_rate": 9.584444444444445e-06,
"loss": 352.1215,
"step": 700
},
{
"epoch": 0.12,
"learning_rate": 9.528888888888889e-06,
"loss": 406.8465,
"step": 725
},
{
"epoch": 0.13,
"learning_rate": 9.473333333333335e-06,
"loss": 385.1631,
"step": 750
},
{
"epoch": 0.13,
"learning_rate": 9.417777777777779e-06,
"loss": 382.0362,
"step": 775
},
{
"epoch": 0.14,
"learning_rate": 9.382222222222223e-06,
"loss": 417.6849,
"step": 800
},
{
"epoch": 0.14,
"learning_rate": 9.382222222222223e-06,
"loss": 0.0,
"step": 825
},
{
"epoch": 0.15,
"learning_rate": 9.382222222222223e-06,
"loss": 0.0,
"step": 850
},
{
"epoch": 0.15,
"learning_rate": 9.382222222222223e-06,
"loss": 0.0,
"step": 875
},
{
"epoch": 0.15,
"learning_rate": 9.382222222222223e-06,
"loss": 0.0,
"step": 900
},
{
"epoch": 0.16,
"learning_rate": 9.382222222222223e-06,
"loss": 0.0,
"step": 925
},
{
"epoch": 0.16,
"learning_rate": 9.368888888888889e-06,
"loss": 0.0,
"step": 950
},
{
"epoch": 0.17,
"learning_rate": 9.313333333333335e-06,
"loss": 0.0,
"step": 975
},
{
"epoch": 0.17,
"learning_rate": 9.257777777777779e-06,
"loss": 0.0,
"step": 1000
},
{
"epoch": 0.17,
"eval_loss": NaN,
"eval_runtime": 1471.3602,
"eval_samples_per_second": 1.665,
"eval_steps_per_second": 0.209,
"eval_wer": 100.0,
"step": 1000
},
{
"epoch": 0.18,
"learning_rate": 9.202222222222224e-06,
"loss": 0.0,
"step": 1025
},
{
"epoch": 0.18,
"learning_rate": 9.146666666666667e-06,
"loss": 0.0,
"step": 1050
},
{
"epoch": 0.18,
"learning_rate": 9.091111111111112e-06,
"loss": 0.0,
"step": 1075
},
{
"epoch": 0.19,
"learning_rate": 9.035555555555556e-06,
"loss": 0.0,
"step": 1100
},
{
"epoch": 0.19,
"learning_rate": 8.98e-06,
"loss": 0.0,
"step": 1125
},
{
"epoch": 0.2,
"learning_rate": 8.924444444444446e-06,
"loss": 0.0,
"step": 1150
},
{
"epoch": 0.2,
"learning_rate": 8.868888888888888e-06,
"loss": 0.0,
"step": 1175
},
{
"epoch": 0.21,
"learning_rate": 8.813333333333334e-06,
"loss": 0.0,
"step": 1200
},
{
"epoch": 0.21,
"learning_rate": 8.757777777777778e-06,
"loss": 0.0,
"step": 1225
},
{
"epoch": 0.21,
"learning_rate": 8.702222222222222e-06,
"loss": 0.0,
"step": 1250
},
{
"epoch": 0.22,
"learning_rate": 8.646666666666668e-06,
"loss": 0.0,
"step": 1275
},
{
"epoch": 0.22,
"learning_rate": 8.591111111111112e-06,
"loss": 0.0,
"step": 1300
},
{
"epoch": 0.23,
"learning_rate": 8.535555555555556e-06,
"loss": 0.0,
"step": 1325
},
{
"epoch": 0.23,
"learning_rate": 8.48e-06,
"loss": 0.0,
"step": 1350
},
{
"epoch": 0.24,
"learning_rate": 8.424444444444446e-06,
"loss": 0.0,
"step": 1375
},
{
"epoch": 0.24,
"learning_rate": 8.36888888888889e-06,
"loss": 0.0,
"step": 1400
},
{
"epoch": 0.24,
"learning_rate": 8.313333333333334e-06,
"loss": 0.0,
"step": 1425
},
{
"epoch": 0.25,
"learning_rate": 8.25777777777778e-06,
"loss": 0.0,
"step": 1450
},
{
"epoch": 0.25,
"learning_rate": 8.202222222222222e-06,
"loss": 0.0,
"step": 1475
},
{
"epoch": 0.26,
"learning_rate": 8.146666666666668e-06,
"loss": 0.0,
"step": 1500
},
{
"epoch": 0.26,
"learning_rate": 8.091111111111112e-06,
"loss": 0.0,
"step": 1525
},
{
"epoch": 0.27,
"learning_rate": 8.035555555555556e-06,
"loss": 0.0,
"step": 1550
},
{
"epoch": 0.27,
"learning_rate": 7.980000000000002e-06,
"loss": 0.0,
"step": 1575
},
{
"epoch": 0.27,
"learning_rate": 7.924444444444444e-06,
"loss": 0.0,
"step": 1600
},
{
"epoch": 0.28,
"learning_rate": 7.86888888888889e-06,
"loss": 0.0,
"step": 1625
},
{
"epoch": 0.28,
"learning_rate": 7.813333333333334e-06,
"loss": 0.0,
"step": 1650
},
{
"epoch": 0.29,
"learning_rate": 7.75777777777778e-06,
"loss": 0.0,
"step": 1675
},
{
"epoch": 0.29,
"learning_rate": 7.702222222222223e-06,
"loss": 0.0,
"step": 1700
},
{
"epoch": 0.3,
"learning_rate": 7.646666666666667e-06,
"loss": 0.0,
"step": 1725
},
{
"epoch": 0.3,
"learning_rate": 7.5911111111111115e-06,
"loss": 0.0,
"step": 1750
},
{
"epoch": 0.3,
"learning_rate": 7.535555555555556e-06,
"loss": 0.0,
"step": 1775
},
{
"epoch": 0.31,
"learning_rate": 7.48e-06,
"loss": 0.0,
"step": 1800
},
{
"epoch": 0.31,
"learning_rate": 7.424444444444445e-06,
"loss": 0.0,
"step": 1825
},
{
"epoch": 0.32,
"learning_rate": 7.36888888888889e-06,
"loss": 0.0,
"step": 1850
},
{
"epoch": 0.32,
"learning_rate": 7.313333333333333e-06,
"loss": 0.0,
"step": 1875
},
{
"epoch": 0.33,
"learning_rate": 7.257777777777778e-06,
"loss": 0.0,
"step": 1900
},
{
"epoch": 0.33,
"learning_rate": 7.202222222222222e-06,
"loss": 0.0,
"step": 1925
},
{
"epoch": 0.33,
"learning_rate": 7.146666666666667e-06,
"loss": 0.0,
"step": 1950
},
{
"epoch": 0.34,
"learning_rate": 7.091111111111112e-06,
"loss": 0.0,
"step": 1975
},
{
"epoch": 0.34,
"learning_rate": 7.035555555555557e-06,
"loss": 0.0,
"step": 2000
},
{
"epoch": 0.34,
"eval_loss": NaN,
"eval_runtime": 1641.9943,
"eval_samples_per_second": 1.492,
"eval_steps_per_second": 0.187,
"eval_wer": 100.0,
"step": 2000
},
{
"epoch": 0.35,
"learning_rate": 6.98e-06,
"loss": 0.0,
"step": 2025
},
{
"epoch": 0.35,
"learning_rate": 6.924444444444445e-06,
"loss": 0.0,
"step": 2050
},
{
"epoch": 0.36,
"learning_rate": 6.868888888888889e-06,
"loss": 0.0,
"step": 2075
},
{
"epoch": 0.36,
"learning_rate": 6.813333333333334e-06,
"loss": 0.0,
"step": 2100
},
{
"epoch": 0.36,
"learning_rate": 6.757777777777779e-06,
"loss": 0.0,
"step": 2125
},
{
"epoch": 0.37,
"learning_rate": 6.702222222222224e-06,
"loss": 0.0,
"step": 2150
},
{
"epoch": 0.37,
"learning_rate": 6.646666666666667e-06,
"loss": 0.0,
"step": 2175
},
{
"epoch": 0.38,
"learning_rate": 6.591111111111112e-06,
"loss": 0.0,
"step": 2200
},
{
"epoch": 0.38,
"learning_rate": 6.535555555555556e-06,
"loss": 0.0,
"step": 2225
},
{
"epoch": 0.39,
"learning_rate": 6.480000000000001e-06,
"loss": 0.0,
"step": 2250
},
{
"epoch": 0.39,
"learning_rate": 6.4244444444444455e-06,
"loss": 0.0,
"step": 2275
},
{
"epoch": 0.4,
"learning_rate": 6.368888888888889e-06,
"loss": 0.0,
"step": 2300
},
{
"epoch": 0.4,
"learning_rate": 6.313333333333334e-06,
"loss": 0.0,
"step": 2325
},
{
"epoch": 0.4,
"learning_rate": 6.2577777777777785e-06,
"loss": 0.0,
"step": 2350
},
{
"epoch": 0.41,
"learning_rate": 6.2022222222222225e-06,
"loss": 0.0,
"step": 2375
},
{
"epoch": 0.41,
"learning_rate": 6.146666666666667e-06,
"loss": 0.0,
"step": 2400
},
{
"epoch": 0.42,
"learning_rate": 6.091111111111112e-06,
"loss": 0.0,
"step": 2425
},
{
"epoch": 0.42,
"learning_rate": 6.0355555555555555e-06,
"loss": 0.0,
"step": 2450
},
{
"epoch": 0.43,
"learning_rate": 5.98e-06,
"loss": 0.0,
"step": 2475
},
{
"epoch": 0.43,
"learning_rate": 5.924444444444445e-06,
"loss": 0.0,
"step": 2500
},
{
"epoch": 0.43,
"learning_rate": 5.868888888888889e-06,
"loss": 0.0,
"step": 2525
},
{
"epoch": 0.44,
"learning_rate": 5.813333333333334e-06,
"loss": 0.0,
"step": 2550
},
{
"epoch": 0.44,
"learning_rate": 5.757777777777779e-06,
"loss": 0.0,
"step": 2575
},
{
"epoch": 0.45,
"learning_rate": 5.702222222222222e-06,
"loss": 0.0,
"step": 2600
},
{
"epoch": 0.45,
"learning_rate": 5.646666666666667e-06,
"loss": 0.0,
"step": 2625
},
{
"epoch": 0.46,
"learning_rate": 5.591111111111112e-06,
"loss": 0.0,
"step": 2650
},
{
"epoch": 0.46,
"learning_rate": 5.535555555555556e-06,
"loss": 0.0,
"step": 2675
},
{
"epoch": 0.46,
"learning_rate": 5.480000000000001e-06,
"loss": 0.0,
"step": 2700
},
{
"epoch": 0.47,
"learning_rate": 5.424444444444444e-06,
"loss": 0.0,
"step": 2725
},
{
"epoch": 0.47,
"learning_rate": 5.368888888888889e-06,
"loss": 0.0,
"step": 2750
},
{
"epoch": 0.48,
"learning_rate": 5.313333333333334e-06,
"loss": 0.0,
"step": 2775
},
{
"epoch": 0.48,
"learning_rate": 5.257777777777779e-06,
"loss": 0.0,
"step": 2800
},
{
"epoch": 0.49,
"learning_rate": 5.202222222222223e-06,
"loss": 0.0,
"step": 2825
},
{
"epoch": 0.49,
"learning_rate": 5.146666666666668e-06,
"loss": 0.0,
"step": 2850
},
{
"epoch": 0.49,
"learning_rate": 5.091111111111111e-06,
"loss": 0.0,
"step": 2875
},
{
"epoch": 0.5,
"learning_rate": 5.035555555555556e-06,
"loss": 0.0,
"step": 2900
},
{
"epoch": 0.5,
"learning_rate": 4.980000000000001e-06,
"loss": 0.0,
"step": 2925
},
{
"epoch": 0.51,
"learning_rate": 4.924444444444445e-06,
"loss": 0.0,
"step": 2950
},
{
"epoch": 0.51,
"learning_rate": 4.8688888888888895e-06,
"loss": 0.0,
"step": 2975
},
{
"epoch": 0.52,
"learning_rate": 4.8133333333333336e-06,
"loss": 0.0,
"step": 3000
},
{
"epoch": 0.52,
"eval_loss": NaN,
"eval_runtime": 1478.728,
"eval_samples_per_second": 1.657,
"eval_steps_per_second": 0.208,
"eval_wer": 100.0,
"step": 3000
},
{
"epoch": 0.52,
"learning_rate": 4.7577777777777784e-06,
"loss": 0.0,
"step": 3025
},
{
"epoch": 0.52,
"learning_rate": 4.7022222222222225e-06,
"loss": 0.0,
"step": 3050
},
{
"epoch": 0.53,
"learning_rate": 4.646666666666667e-06,
"loss": 0.0,
"step": 3075
},
{
"epoch": 0.53,
"learning_rate": 4.591111111111111e-06,
"loss": 0.0,
"step": 3100
},
{
"epoch": 0.54,
"learning_rate": 4.535555555555555e-06,
"loss": 0.0,
"step": 3125
},
{
"epoch": 0.54,
"learning_rate": 4.48e-06,
"loss": 0.0,
"step": 3150
},
{
"epoch": 0.55,
"learning_rate": 4.424444444444444e-06,
"loss": 0.0,
"step": 3175
},
{
"epoch": 0.55,
"learning_rate": 4.368888888888889e-06,
"loss": 0.0,
"step": 3200
},
{
"epoch": 0.55,
"learning_rate": 4.313333333333334e-06,
"loss": 0.0,
"step": 3225
},
{
"epoch": 0.56,
"learning_rate": 4.257777777777778e-06,
"loss": 0.0,
"step": 3250
},
{
"epoch": 0.56,
"learning_rate": 4.202222222222222e-06,
"loss": 0.0,
"step": 3275
},
{
"epoch": 0.57,
"learning_rate": 4.146666666666667e-06,
"loss": 0.0,
"step": 3300
},
{
"epoch": 0.57,
"learning_rate": 4.091111111111111e-06,
"loss": 0.0,
"step": 3325
},
{
"epoch": 0.58,
"learning_rate": 4.035555555555556e-06,
"loss": 0.0,
"step": 3350
},
{
"epoch": 0.58,
"learning_rate": 3.980000000000001e-06,
"loss": 0.0,
"step": 3375
},
{
"epoch": 0.58,
"learning_rate": 3.924444444444445e-06,
"loss": 0.0,
"step": 3400
},
{
"epoch": 0.59,
"learning_rate": 3.868888888888889e-06,
"loss": 0.0,
"step": 3425
},
{
"epoch": 0.59,
"learning_rate": 3.813333333333334e-06,
"loss": 0.0,
"step": 3450
},
{
"epoch": 0.6,
"learning_rate": 3.757777777777778e-06,
"loss": 0.0,
"step": 3475
},
{
"epoch": 0.6,
"learning_rate": 3.7022222222222227e-06,
"loss": 0.0,
"step": 3500
},
{
"epoch": 0.61,
"learning_rate": 3.6466666666666668e-06,
"loss": 0.0,
"step": 3525
},
{
"epoch": 0.61,
"learning_rate": 3.5911111111111112e-06,
"loss": 0.0,
"step": 3550
},
{
"epoch": 0.61,
"learning_rate": 3.535555555555556e-06,
"loss": 0.0,
"step": 3575
},
{
"epoch": 0.62,
"learning_rate": 3.48e-06,
"loss": 0.0,
"step": 3600
},
{
"epoch": 0.62,
"learning_rate": 3.4244444444444446e-06,
"loss": 0.0,
"step": 3625
},
{
"epoch": 0.63,
"learning_rate": 3.3688888888888895e-06,
"loss": 0.0,
"step": 3650
},
{
"epoch": 0.63,
"learning_rate": 3.3133333333333335e-06,
"loss": 0.0,
"step": 3675
},
{
"epoch": 0.64,
"learning_rate": 3.257777777777778e-06,
"loss": 0.0,
"step": 3700
},
{
"epoch": 0.64,
"learning_rate": 3.202222222222223e-06,
"loss": 0.0,
"step": 3725
},
{
"epoch": 0.64,
"learning_rate": 3.146666666666667e-06,
"loss": 0.0,
"step": 3750
},
{
"epoch": 0.65,
"learning_rate": 3.0911111111111114e-06,
"loss": 0.0,
"step": 3775
},
{
"epoch": 0.65,
"learning_rate": 3.0355555555555562e-06,
"loss": 0.0,
"step": 3800
},
{
"epoch": 0.66,
"learning_rate": 2.9800000000000003e-06,
"loss": 0.0,
"step": 3825
},
{
"epoch": 0.66,
"learning_rate": 2.9244444444444447e-06,
"loss": 0.0,
"step": 3850
},
{
"epoch": 0.67,
"learning_rate": 2.868888888888889e-06,
"loss": 0.0,
"step": 3875
},
{
"epoch": 0.67,
"learning_rate": 2.8133333333333336e-06,
"loss": 0.0,
"step": 3900
},
{
"epoch": 0.67,
"learning_rate": 2.757777777777778e-06,
"loss": 0.0,
"step": 3925
},
{
"epoch": 0.68,
"learning_rate": 2.702222222222222e-06,
"loss": 0.0,
"step": 3950
},
{
"epoch": 0.68,
"learning_rate": 2.646666666666667e-06,
"loss": 0.0,
"step": 3975
},
{
"epoch": 0.69,
"learning_rate": 2.5911111111111115e-06,
"loss": 0.0,
"step": 4000
},
{
"epoch": 0.69,
"eval_loss": NaN,
"eval_runtime": 1441.4308,
"eval_samples_per_second": 1.7,
"eval_steps_per_second": 0.213,
"eval_wer": 100.0,
"step": 4000
},
{
"epoch": 0.69,
"learning_rate": 2.5355555555555555e-06,
"loss": 0.0,
"step": 4025
},
{
"epoch": 0.7,
"learning_rate": 2.4800000000000004e-06,
"loss": 0.0,
"step": 4050
},
{
"epoch": 0.7,
"learning_rate": 2.4244444444444444e-06,
"loss": 0.0,
"step": 4075
},
{
"epoch": 0.7,
"learning_rate": 2.3688888888888893e-06,
"loss": 0.0,
"step": 4100
},
{
"epoch": 0.71,
"learning_rate": 2.3133333333333333e-06,
"loss": 0.0,
"step": 4125
},
{
"epoch": 0.71,
"learning_rate": 2.257777777777778e-06,
"loss": 0.0,
"step": 4150
},
{
"epoch": 0.72,
"learning_rate": 2.2022222222222227e-06,
"loss": 0.0,
"step": 4175
},
{
"epoch": 0.72,
"learning_rate": 2.1466666666666667e-06,
"loss": 0.0,
"step": 4200
},
{
"epoch": 0.73,
"learning_rate": 2.091111111111111e-06,
"loss": 0.0,
"step": 4225
},
{
"epoch": 0.73,
"learning_rate": 2.0355555555555556e-06,
"loss": 0.0,
"step": 4250
},
{
"epoch": 0.73,
"learning_rate": 1.98e-06,
"loss": 0.0,
"step": 4275
},
{
"epoch": 0.74,
"learning_rate": 1.9244444444444446e-06,
"loss": 0.0,
"step": 4300
},
{
"epoch": 0.74,
"learning_rate": 1.868888888888889e-06,
"loss": 0.0,
"step": 4325
},
{
"epoch": 0.75,
"learning_rate": 1.8133333333333337e-06,
"loss": 0.0,
"step": 4350
},
{
"epoch": 0.75,
"learning_rate": 1.757777777777778e-06,
"loss": 0.0,
"step": 4375
},
{
"epoch": 0.76,
"learning_rate": 1.7022222222222224e-06,
"loss": 0.0,
"step": 4400
},
{
"epoch": 0.76,
"learning_rate": 1.6466666666666666e-06,
"loss": 0.0,
"step": 4425
},
{
"epoch": 0.76,
"learning_rate": 1.5911111111111113e-06,
"loss": 0.0,
"step": 4450
},
{
"epoch": 0.77,
"learning_rate": 1.5355555555555558e-06,
"loss": 0.0,
"step": 4475
},
{
"epoch": 0.77,
"learning_rate": 1.48e-06,
"loss": 0.0,
"step": 4500
},
{
"epoch": 0.78,
"learning_rate": 1.4244444444444447e-06,
"loss": 0.0,
"step": 4525
},
{
"epoch": 0.78,
"learning_rate": 1.3688888888888891e-06,
"loss": 0.0,
"step": 4550
},
{
"epoch": 0.79,
"learning_rate": 1.3133333333333334e-06,
"loss": 0.0,
"step": 4575
},
{
"epoch": 0.79,
"learning_rate": 1.2577777777777779e-06,
"loss": 0.0,
"step": 4600
},
{
"epoch": 0.79,
"learning_rate": 1.2022222222222223e-06,
"loss": 0.0,
"step": 4625
},
{
"epoch": 0.8,
"learning_rate": 1.1466666666666668e-06,
"loss": 0.0,
"step": 4650
},
{
"epoch": 0.8,
"learning_rate": 1.0911111111111112e-06,
"loss": 0.0,
"step": 4675
},
{
"epoch": 0.81,
"learning_rate": 1.0355555555555557e-06,
"loss": 0.0,
"step": 4700
},
{
"epoch": 0.81,
"learning_rate": 9.800000000000001e-07,
"loss": 0.0,
"step": 4725
},
{
"epoch": 0.82,
"learning_rate": 9.244444444444445e-07,
"loss": 0.0,
"step": 4750
},
{
"epoch": 0.82,
"learning_rate": 8.68888888888889e-07,
"loss": 0.0,
"step": 4775
},
{
"epoch": 0.82,
"learning_rate": 8.133333333333333e-07,
"loss": 0.0,
"step": 4800
},
{
"epoch": 0.83,
"learning_rate": 7.577777777777779e-07,
"loss": 0.0,
"step": 4825
},
{
"epoch": 0.83,
"learning_rate": 7.022222222222223e-07,
"loss": 0.0,
"step": 4850
},
{
"epoch": 0.84,
"learning_rate": 6.466666666666667e-07,
"loss": 0.0,
"step": 4875
},
{
"epoch": 0.84,
"learning_rate": 5.911111111111111e-07,
"loss": 0.0,
"step": 4900
},
{
"epoch": 0.85,
"learning_rate": 5.355555555555556e-07,
"loss": 0.0,
"step": 4925
},
{
"epoch": 0.85,
"learning_rate": 4.800000000000001e-07,
"loss": 0.0,
"step": 4950
},
{
"epoch": 0.85,
"learning_rate": 4.2444444444444447e-07,
"loss": 0.0,
"step": 4975
},
{
"epoch": 0.86,
"learning_rate": 3.6888888888888893e-07,
"loss": 0.0,
"step": 5000
},
{
"epoch": 0.86,
"eval_loss": NaN,
"eval_runtime": 1432.9984,
"eval_samples_per_second": 1.71,
"eval_steps_per_second": 0.214,
"eval_wer": 100.0,
"step": 5000
},
{
"epoch": 0.86,
"step": 5000,
"total_flos": 1.15434160128e+19,
"train_loss": 25.538242671966554,
"train_runtime": 15810.8062,
"train_samples_per_second": 2.53,
"train_steps_per_second": 0.316
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 1.15434160128e+19,
"trial_name": null,
"trial_params": null
}