cv9-special-batch4-lr4-small / trainer_state.json
TheRains's picture
End of training
d14850f
{
"best_metric": 19.102829537612145,
"best_model_checkpoint": "./output/small/cv9-special-batch4-lr4-small/checkpoint-5000",
"epoch": 2.420135527589545,
"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": 4.4e-06,
"loss": 1.2416,
"step": 25
},
{
"epoch": 0.02,
"learning_rate": 9.4e-06,
"loss": 0.648,
"step": 50
},
{
"epoch": 0.04,
"learning_rate": 1.44e-05,
"loss": 0.5005,
"step": 75
},
{
"epoch": 0.05,
"learning_rate": 1.94e-05,
"loss": 0.4696,
"step": 100
},
{
"epoch": 0.06,
"learning_rate": 2.44e-05,
"loss": 0.4696,
"step": 125
},
{
"epoch": 0.07,
"learning_rate": 2.94e-05,
"loss": 0.4946,
"step": 150
},
{
"epoch": 0.08,
"learning_rate": 3.4399999999999996e-05,
"loss": 0.534,
"step": 175
},
{
"epoch": 0.1,
"learning_rate": 3.94e-05,
"loss": 0.5804,
"step": 200
},
{
"epoch": 0.11,
"learning_rate": 4.44e-05,
"loss": 0.5353,
"step": 225
},
{
"epoch": 0.12,
"learning_rate": 4.94e-05,
"loss": 0.7473,
"step": 250
},
{
"epoch": 0.13,
"learning_rate": 5.440000000000001e-05,
"loss": 0.6415,
"step": 275
},
{
"epoch": 0.15,
"learning_rate": 5.94e-05,
"loss": 0.6545,
"step": 300
},
{
"epoch": 0.16,
"learning_rate": 6.440000000000001e-05,
"loss": 0.733,
"step": 325
},
{
"epoch": 0.17,
"learning_rate": 6.939999999999999e-05,
"loss": 0.7825,
"step": 350
},
{
"epoch": 0.18,
"learning_rate": 7.44e-05,
"loss": 0.9102,
"step": 375
},
{
"epoch": 0.19,
"learning_rate": 7.94e-05,
"loss": 0.8767,
"step": 400
},
{
"epoch": 0.21,
"learning_rate": 8.42e-05,
"loss": 0.9197,
"step": 425
},
{
"epoch": 0.22,
"learning_rate": 8.92e-05,
"loss": 0.9347,
"step": 450
},
{
"epoch": 0.23,
"learning_rate": 9.42e-05,
"loss": 1.0261,
"step": 475
},
{
"epoch": 0.24,
"learning_rate": 9.92e-05,
"loss": 1.0067,
"step": 500
},
{
"epoch": 0.25,
"learning_rate": 9.953333333333333e-05,
"loss": 1.1174,
"step": 525
},
{
"epoch": 0.27,
"learning_rate": 9.897777777777778e-05,
"loss": 1.2204,
"step": 550
},
{
"epoch": 0.28,
"learning_rate": 9.842222222222223e-05,
"loss": 1.1001,
"step": 575
},
{
"epoch": 0.29,
"learning_rate": 9.786666666666667e-05,
"loss": 1.1297,
"step": 600
},
{
"epoch": 0.3,
"learning_rate": 9.731111111111112e-05,
"loss": 1.0853,
"step": 625
},
{
"epoch": 0.31,
"learning_rate": 9.677777777777778e-05,
"loss": 1.0081,
"step": 650
},
{
"epoch": 0.33,
"learning_rate": 9.622222222222222e-05,
"loss": 0.9878,
"step": 675
},
{
"epoch": 0.34,
"learning_rate": 9.566666666666667e-05,
"loss": 1.093,
"step": 700
},
{
"epoch": 0.35,
"learning_rate": 9.511111111111112e-05,
"loss": 1.2345,
"step": 725
},
{
"epoch": 0.36,
"learning_rate": 9.455555555555556e-05,
"loss": 1.1439,
"step": 750
},
{
"epoch": 0.38,
"learning_rate": 9.4e-05,
"loss": 1.0903,
"step": 775
},
{
"epoch": 0.39,
"learning_rate": 9.344444444444444e-05,
"loss": 1.1087,
"step": 800
},
{
"epoch": 0.4,
"learning_rate": 9.28888888888889e-05,
"loss": 1.0111,
"step": 825
},
{
"epoch": 0.41,
"learning_rate": 9.233333333333333e-05,
"loss": 1.0238,
"step": 850
},
{
"epoch": 0.42,
"learning_rate": 9.177777777777778e-05,
"loss": 1.0508,
"step": 875
},
{
"epoch": 0.44,
"learning_rate": 9.122222222222223e-05,
"loss": 1.0741,
"step": 900
},
{
"epoch": 0.45,
"learning_rate": 9.066666666666667e-05,
"loss": 0.983,
"step": 925
},
{
"epoch": 0.46,
"learning_rate": 9.011111111111111e-05,
"loss": 1.0221,
"step": 950
},
{
"epoch": 0.47,
"learning_rate": 8.955555555555556e-05,
"loss": 1.0972,
"step": 975
},
{
"epoch": 0.48,
"learning_rate": 8.900000000000001e-05,
"loss": 0.9789,
"step": 1000
},
{
"epoch": 0.48,
"eval_loss": 0.8521688580513,
"eval_runtime": 4248.9218,
"eval_samples_per_second": 0.852,
"eval_steps_per_second": 0.426,
"eval_wer": 41.74833218311479,
"step": 1000
},
{
"epoch": 0.5,
"learning_rate": 8.844444444444445e-05,
"loss": 1.0655,
"step": 1025
},
{
"epoch": 0.51,
"learning_rate": 8.78888888888889e-05,
"loss": 1.0255,
"step": 1050
},
{
"epoch": 0.52,
"learning_rate": 8.733333333333333e-05,
"loss": 0.929,
"step": 1075
},
{
"epoch": 0.53,
"learning_rate": 8.677777777777778e-05,
"loss": 0.9659,
"step": 1100
},
{
"epoch": 0.54,
"learning_rate": 8.622222222222222e-05,
"loss": 1.0644,
"step": 1125
},
{
"epoch": 0.56,
"learning_rate": 8.566666666666667e-05,
"loss": 0.9096,
"step": 1150
},
{
"epoch": 0.57,
"learning_rate": 8.511111111111112e-05,
"loss": 1.0427,
"step": 1175
},
{
"epoch": 0.58,
"learning_rate": 8.455555555555556e-05,
"loss": 0.9621,
"step": 1200
},
{
"epoch": 0.59,
"learning_rate": 8.4e-05,
"loss": 0.8934,
"step": 1225
},
{
"epoch": 0.61,
"learning_rate": 8.344444444444445e-05,
"loss": 1.0162,
"step": 1250
},
{
"epoch": 0.62,
"learning_rate": 8.28888888888889e-05,
"loss": 0.8607,
"step": 1275
},
{
"epoch": 0.63,
"learning_rate": 8.233333333333333e-05,
"loss": 0.8669,
"step": 1300
},
{
"epoch": 0.64,
"learning_rate": 8.177777777777778e-05,
"loss": 0.9272,
"step": 1325
},
{
"epoch": 0.65,
"learning_rate": 8.122222222222222e-05,
"loss": 1.0139,
"step": 1350
},
{
"epoch": 0.67,
"learning_rate": 8.066666666666667e-05,
"loss": 1.0201,
"step": 1375
},
{
"epoch": 0.68,
"learning_rate": 8.011111111111111e-05,
"loss": 0.8599,
"step": 1400
},
{
"epoch": 0.69,
"learning_rate": 7.955555555555556e-05,
"loss": 1.0144,
"step": 1425
},
{
"epoch": 0.7,
"learning_rate": 7.900000000000001e-05,
"loss": 1.052,
"step": 1450
},
{
"epoch": 0.71,
"learning_rate": 7.844444444444446e-05,
"loss": 0.8024,
"step": 1475
},
{
"epoch": 0.73,
"learning_rate": 7.788888888888888e-05,
"loss": 0.921,
"step": 1500
},
{
"epoch": 0.74,
"learning_rate": 7.733333333333333e-05,
"loss": 0.9408,
"step": 1525
},
{
"epoch": 0.75,
"learning_rate": 7.677777777777778e-05,
"loss": 0.9626,
"step": 1550
},
{
"epoch": 0.76,
"learning_rate": 7.622222222222223e-05,
"loss": 0.819,
"step": 1575
},
{
"epoch": 0.77,
"learning_rate": 7.566666666666667e-05,
"loss": 0.8516,
"step": 1600
},
{
"epoch": 0.79,
"learning_rate": 7.511111111111111e-05,
"loss": 0.8963,
"step": 1625
},
{
"epoch": 0.8,
"learning_rate": 7.455555555555556e-05,
"loss": 0.8619,
"step": 1650
},
{
"epoch": 0.81,
"learning_rate": 7.4e-05,
"loss": 0.9302,
"step": 1675
},
{
"epoch": 0.82,
"learning_rate": 7.344444444444445e-05,
"loss": 0.9227,
"step": 1700
},
{
"epoch": 0.83,
"learning_rate": 7.28888888888889e-05,
"loss": 0.8421,
"step": 1725
},
{
"epoch": 0.85,
"learning_rate": 7.233333333333335e-05,
"loss": 0.8214,
"step": 1750
},
{
"epoch": 0.86,
"learning_rate": 7.177777777777777e-05,
"loss": 0.8977,
"step": 1775
},
{
"epoch": 0.87,
"learning_rate": 7.122222222222222e-05,
"loss": 0.9047,
"step": 1800
},
{
"epoch": 0.88,
"learning_rate": 7.066666666666667e-05,
"loss": 0.9844,
"step": 1825
},
{
"epoch": 0.9,
"learning_rate": 7.011111111111112e-05,
"loss": 0.8347,
"step": 1850
},
{
"epoch": 0.91,
"learning_rate": 6.955555555555556e-05,
"loss": 0.8313,
"step": 1875
},
{
"epoch": 0.92,
"learning_rate": 6.9e-05,
"loss": 0.8901,
"step": 1900
},
{
"epoch": 0.93,
"learning_rate": 6.844444444444445e-05,
"loss": 0.7889,
"step": 1925
},
{
"epoch": 0.94,
"learning_rate": 6.788888888888888e-05,
"loss": 0.9418,
"step": 1950
},
{
"epoch": 0.96,
"learning_rate": 6.733333333333333e-05,
"loss": 0.7617,
"step": 1975
},
{
"epoch": 0.97,
"learning_rate": 6.677777777777779e-05,
"loss": 0.7245,
"step": 2000
},
{
"epoch": 0.97,
"eval_loss": 0.7121276259422302,
"eval_runtime": 3080.8111,
"eval_samples_per_second": 1.176,
"eval_steps_per_second": 0.588,
"eval_wer": 36.81619507706464,
"step": 2000
},
{
"epoch": 0.98,
"learning_rate": 6.622222222222224e-05,
"loss": 1.0453,
"step": 2025
},
{
"epoch": 0.99,
"learning_rate": 6.566666666666666e-05,
"loss": 0.9144,
"step": 2050
},
{
"epoch": 1.0,
"learning_rate": 6.511111111111111e-05,
"loss": 0.7348,
"step": 2075
},
{
"epoch": 1.02,
"learning_rate": 6.455555555555556e-05,
"loss": 0.4077,
"step": 2100
},
{
"epoch": 1.03,
"learning_rate": 6.400000000000001e-05,
"loss": 0.3705,
"step": 2125
},
{
"epoch": 1.04,
"learning_rate": 6.344444444444445e-05,
"loss": 0.3388,
"step": 2150
},
{
"epoch": 1.05,
"learning_rate": 6.28888888888889e-05,
"loss": 0.3734,
"step": 2175
},
{
"epoch": 1.06,
"learning_rate": 6.233333333333334e-05,
"loss": 0.4216,
"step": 2200
},
{
"epoch": 1.08,
"learning_rate": 6.177777777777779e-05,
"loss": 0.4496,
"step": 2225
},
{
"epoch": 1.09,
"learning_rate": 6.122222222222222e-05,
"loss": 0.3623,
"step": 2250
},
{
"epoch": 1.1,
"learning_rate": 6.066666666666667e-05,
"loss": 0.3613,
"step": 2275
},
{
"epoch": 1.11,
"learning_rate": 6.011111111111112e-05,
"loss": 0.4565,
"step": 2300
},
{
"epoch": 1.13,
"learning_rate": 5.9555555555555554e-05,
"loss": 0.3732,
"step": 2325
},
{
"epoch": 1.14,
"learning_rate": 5.9e-05,
"loss": 0.3877,
"step": 2350
},
{
"epoch": 1.15,
"learning_rate": 5.844444444444445e-05,
"loss": 0.3241,
"step": 2375
},
{
"epoch": 1.16,
"learning_rate": 5.788888888888889e-05,
"loss": 0.388,
"step": 2400
},
{
"epoch": 1.17,
"learning_rate": 5.7333333333333336e-05,
"loss": 0.4183,
"step": 2425
},
{
"epoch": 1.19,
"learning_rate": 5.6777777777777786e-05,
"loss": 0.3712,
"step": 2450
},
{
"epoch": 1.2,
"learning_rate": 5.622222222222222e-05,
"loss": 0.3725,
"step": 2475
},
{
"epoch": 1.21,
"learning_rate": 5.566666666666667e-05,
"loss": 0.3257,
"step": 2500
},
{
"epoch": 1.22,
"learning_rate": 5.511111111111111e-05,
"loss": 0.38,
"step": 2525
},
{
"epoch": 1.23,
"learning_rate": 5.455555555555556e-05,
"loss": 0.3076,
"step": 2550
},
{
"epoch": 1.25,
"learning_rate": 5.4000000000000005e-05,
"loss": 0.3722,
"step": 2575
},
{
"epoch": 1.26,
"learning_rate": 5.3444444444444455e-05,
"loss": 0.3794,
"step": 2600
},
{
"epoch": 1.27,
"learning_rate": 5.2888888888888885e-05,
"loss": 0.4177,
"step": 2625
},
{
"epoch": 1.28,
"learning_rate": 5.2333333333333336e-05,
"loss": 0.3532,
"step": 2650
},
{
"epoch": 1.29,
"learning_rate": 5.177777777777778e-05,
"loss": 0.312,
"step": 2675
},
{
"epoch": 1.31,
"learning_rate": 5.122222222222223e-05,
"loss": 0.3835,
"step": 2700
},
{
"epoch": 1.32,
"learning_rate": 5.0666666666666674e-05,
"loss": 0.3245,
"step": 2725
},
{
"epoch": 1.33,
"learning_rate": 5.011111111111111e-05,
"loss": 0.362,
"step": 2750
},
{
"epoch": 1.34,
"learning_rate": 4.955555555555556e-05,
"loss": 0.3939,
"step": 2775
},
{
"epoch": 1.36,
"learning_rate": 4.9e-05,
"loss": 0.3364,
"step": 2800
},
{
"epoch": 1.37,
"learning_rate": 4.844444444444445e-05,
"loss": 0.3421,
"step": 2825
},
{
"epoch": 1.38,
"learning_rate": 4.7888888888888886e-05,
"loss": 0.4141,
"step": 2850
},
{
"epoch": 1.39,
"learning_rate": 4.7333333333333336e-05,
"loss": 0.3327,
"step": 2875
},
{
"epoch": 1.4,
"learning_rate": 4.677777777777778e-05,
"loss": 0.3997,
"step": 2900
},
{
"epoch": 1.42,
"learning_rate": 4.6222222222222224e-05,
"loss": 0.2963,
"step": 2925
},
{
"epoch": 1.43,
"learning_rate": 4.566666666666667e-05,
"loss": 0.3121,
"step": 2950
},
{
"epoch": 1.44,
"learning_rate": 4.511111111111112e-05,
"loss": 0.3688,
"step": 2975
},
{
"epoch": 1.45,
"learning_rate": 4.4555555555555555e-05,
"loss": 0.3335,
"step": 3000
},
{
"epoch": 1.45,
"eval_loss": 0.5499204993247986,
"eval_runtime": 3433.5578,
"eval_samples_per_second": 1.055,
"eval_steps_per_second": 0.527,
"eval_wer": 28.71405567057741,
"step": 3000
},
{
"epoch": 1.46,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.3282,
"step": 3025
},
{
"epoch": 1.48,
"learning_rate": 4.344444444444445e-05,
"loss": 0.3278,
"step": 3050
},
{
"epoch": 1.49,
"learning_rate": 4.2888888888888886e-05,
"loss": 0.3028,
"step": 3075
},
{
"epoch": 1.5,
"learning_rate": 4.233333333333334e-05,
"loss": 0.2976,
"step": 3100
},
{
"epoch": 1.51,
"learning_rate": 4.177777777777778e-05,
"loss": 0.3663,
"step": 3125
},
{
"epoch": 1.52,
"learning_rate": 4.1222222222222224e-05,
"loss": 0.4239,
"step": 3150
},
{
"epoch": 1.54,
"learning_rate": 4.066666666666667e-05,
"loss": 0.3031,
"step": 3175
},
{
"epoch": 1.55,
"learning_rate": 4.011111111111111e-05,
"loss": 0.2766,
"step": 3200
},
{
"epoch": 1.56,
"learning_rate": 3.9555555555555556e-05,
"loss": 0.2827,
"step": 3225
},
{
"epoch": 1.57,
"learning_rate": 3.9000000000000006e-05,
"loss": 0.3636,
"step": 3250
},
{
"epoch": 1.59,
"learning_rate": 3.844444444444444e-05,
"loss": 0.2879,
"step": 3275
},
{
"epoch": 1.6,
"learning_rate": 3.7888888888888894e-05,
"loss": 0.2919,
"step": 3300
},
{
"epoch": 1.61,
"learning_rate": 3.733333333333334e-05,
"loss": 0.2733,
"step": 3325
},
{
"epoch": 1.62,
"learning_rate": 3.677777777777778e-05,
"loss": 0.3007,
"step": 3350
},
{
"epoch": 1.63,
"learning_rate": 3.6222222222222225e-05,
"loss": 0.2882,
"step": 3375
},
{
"epoch": 1.65,
"learning_rate": 3.566666666666667e-05,
"loss": 0.3166,
"step": 3400
},
{
"epoch": 1.66,
"learning_rate": 3.511111111111111e-05,
"loss": 0.3175,
"step": 3425
},
{
"epoch": 1.67,
"learning_rate": 3.4555555555555556e-05,
"loss": 0.3146,
"step": 3450
},
{
"epoch": 1.68,
"learning_rate": 3.4000000000000007e-05,
"loss": 0.305,
"step": 3475
},
{
"epoch": 1.69,
"learning_rate": 3.3444444444444443e-05,
"loss": 0.2458,
"step": 3500
},
{
"epoch": 1.71,
"learning_rate": 3.2888888888888894e-05,
"loss": 0.3051,
"step": 3525
},
{
"epoch": 1.72,
"learning_rate": 3.233333333333333e-05,
"loss": 0.2955,
"step": 3550
},
{
"epoch": 1.73,
"learning_rate": 3.177777777777778e-05,
"loss": 0.2435,
"step": 3575
},
{
"epoch": 1.74,
"learning_rate": 3.1222222222222225e-05,
"loss": 0.2842,
"step": 3600
},
{
"epoch": 1.75,
"learning_rate": 3.066666666666667e-05,
"loss": 0.3142,
"step": 3625
},
{
"epoch": 1.77,
"learning_rate": 3.0111111111111113e-05,
"loss": 0.3038,
"step": 3650
},
{
"epoch": 1.78,
"learning_rate": 2.955555555555556e-05,
"loss": 0.2343,
"step": 3675
},
{
"epoch": 1.79,
"learning_rate": 2.9e-05,
"loss": 0.2375,
"step": 3700
},
{
"epoch": 1.8,
"learning_rate": 2.8444444444444447e-05,
"loss": 0.2849,
"step": 3725
},
{
"epoch": 1.82,
"learning_rate": 2.788888888888889e-05,
"loss": 0.2585,
"step": 3750
},
{
"epoch": 1.83,
"learning_rate": 2.733333333333333e-05,
"loss": 0.2881,
"step": 3775
},
{
"epoch": 1.84,
"learning_rate": 2.677777777777778e-05,
"loss": 0.2755,
"step": 3800
},
{
"epoch": 1.85,
"learning_rate": 2.6222222222222226e-05,
"loss": 0.2618,
"step": 3825
},
{
"epoch": 1.86,
"learning_rate": 2.5666666666666666e-05,
"loss": 0.2275,
"step": 3850
},
{
"epoch": 1.88,
"learning_rate": 2.5111111111111113e-05,
"loss": 0.3041,
"step": 3875
},
{
"epoch": 1.89,
"learning_rate": 2.4555555555555557e-05,
"loss": 0.2588,
"step": 3900
},
{
"epoch": 1.9,
"learning_rate": 2.4e-05,
"loss": 0.199,
"step": 3925
},
{
"epoch": 1.91,
"learning_rate": 2.3444444444444448e-05,
"loss": 0.2782,
"step": 3950
},
{
"epoch": 1.92,
"learning_rate": 2.288888888888889e-05,
"loss": 0.2275,
"step": 3975
},
{
"epoch": 1.94,
"learning_rate": 2.2333333333333335e-05,
"loss": 0.2575,
"step": 4000
},
{
"epoch": 1.94,
"eval_loss": 0.4435468912124634,
"eval_runtime": 3466.8416,
"eval_samples_per_second": 1.045,
"eval_steps_per_second": 0.522,
"eval_wer": 22.093397745571657,
"step": 4000
},
{
"epoch": 1.95,
"learning_rate": 2.177777777777778e-05,
"loss": 0.2514,
"step": 4025
},
{
"epoch": 1.96,
"learning_rate": 2.1222222222222223e-05,
"loss": 0.2417,
"step": 4050
},
{
"epoch": 1.97,
"learning_rate": 2.0666666666666666e-05,
"loss": 0.254,
"step": 4075
},
{
"epoch": 1.98,
"learning_rate": 2.011111111111111e-05,
"loss": 0.2734,
"step": 4100
},
{
"epoch": 2.0,
"learning_rate": 1.9555555555555557e-05,
"loss": 0.2374,
"step": 4125
},
{
"epoch": 2.01,
"learning_rate": 1.9e-05,
"loss": 0.1387,
"step": 4150
},
{
"epoch": 2.02,
"learning_rate": 1.8444444444444445e-05,
"loss": 0.0846,
"step": 4175
},
{
"epoch": 2.03,
"learning_rate": 1.788888888888889e-05,
"loss": 0.0897,
"step": 4200
},
{
"epoch": 2.05,
"learning_rate": 1.7333333333333336e-05,
"loss": 0.0716,
"step": 4225
},
{
"epoch": 2.06,
"learning_rate": 1.677777777777778e-05,
"loss": 0.0628,
"step": 4250
},
{
"epoch": 2.07,
"learning_rate": 1.6222222222222223e-05,
"loss": 0.0779,
"step": 4275
},
{
"epoch": 2.08,
"learning_rate": 1.5666666666666667e-05,
"loss": 0.054,
"step": 4300
},
{
"epoch": 2.09,
"learning_rate": 1.5111111111111112e-05,
"loss": 0.0495,
"step": 4325
},
{
"epoch": 2.11,
"learning_rate": 1.4555555555555556e-05,
"loss": 0.0579,
"step": 4350
},
{
"epoch": 2.12,
"learning_rate": 1.4000000000000001e-05,
"loss": 0.0864,
"step": 4375
},
{
"epoch": 2.13,
"learning_rate": 1.3444444444444445e-05,
"loss": 0.0634,
"step": 4400
},
{
"epoch": 2.14,
"learning_rate": 1.2888888888888889e-05,
"loss": 0.0834,
"step": 4425
},
{
"epoch": 2.15,
"learning_rate": 1.2333333333333334e-05,
"loss": 0.1128,
"step": 4450
},
{
"epoch": 2.17,
"learning_rate": 1.1777777777777778e-05,
"loss": 0.0793,
"step": 4475
},
{
"epoch": 2.18,
"learning_rate": 1.1222222222222224e-05,
"loss": 0.068,
"step": 4500
},
{
"epoch": 2.19,
"learning_rate": 1.0666666666666667e-05,
"loss": 0.0765,
"step": 4525
},
{
"epoch": 2.2,
"learning_rate": 1.0111111111111111e-05,
"loss": 0.0593,
"step": 4550
},
{
"epoch": 2.21,
"learning_rate": 9.555555555555556e-06,
"loss": 0.0625,
"step": 4575
},
{
"epoch": 2.23,
"learning_rate": 9e-06,
"loss": 0.0597,
"step": 4600
},
{
"epoch": 2.24,
"learning_rate": 8.444444444444446e-06,
"loss": 0.0728,
"step": 4625
},
{
"epoch": 2.25,
"learning_rate": 7.88888888888889e-06,
"loss": 0.0505,
"step": 4650
},
{
"epoch": 2.26,
"learning_rate": 7.333333333333334e-06,
"loss": 0.081,
"step": 4675
},
{
"epoch": 2.27,
"learning_rate": 6.777777777777779e-06,
"loss": 0.0639,
"step": 4700
},
{
"epoch": 2.29,
"learning_rate": 6.222222222222222e-06,
"loss": 0.0849,
"step": 4725
},
{
"epoch": 2.3,
"learning_rate": 5.666666666666667e-06,
"loss": 0.0741,
"step": 4750
},
{
"epoch": 2.31,
"learning_rate": 5.1111111111111115e-06,
"loss": 0.0578,
"step": 4775
},
{
"epoch": 2.32,
"learning_rate": 4.555555555555556e-06,
"loss": 0.0649,
"step": 4800
},
{
"epoch": 2.34,
"learning_rate": 4.000000000000001e-06,
"loss": 0.0713,
"step": 4825
},
{
"epoch": 2.35,
"learning_rate": 3.4444444444444444e-06,
"loss": 0.0605,
"step": 4850
},
{
"epoch": 2.36,
"learning_rate": 2.888888888888889e-06,
"loss": 0.0542,
"step": 4875
},
{
"epoch": 2.37,
"learning_rate": 2.3333333333333336e-06,
"loss": 0.057,
"step": 4900
},
{
"epoch": 2.38,
"learning_rate": 1.777777777777778e-06,
"loss": 0.0549,
"step": 4925
},
{
"epoch": 2.4,
"learning_rate": 1.2222222222222223e-06,
"loss": 0.0526,
"step": 4950
},
{
"epoch": 2.41,
"learning_rate": 6.666666666666667e-07,
"loss": 0.0569,
"step": 4975
},
{
"epoch": 2.42,
"learning_rate": 1.1111111111111112e-07,
"loss": 0.0593,
"step": 5000
},
{
"epoch": 2.42,
"eval_loss": 0.4076603949069977,
"eval_runtime": 3471.3095,
"eval_samples_per_second": 1.043,
"eval_steps_per_second": 0.522,
"eval_wer": 19.102829537612145,
"step": 5000
},
{
"epoch": 2.42,
"step": 5000,
"total_flos": 5.76997649399808e+18,
"train_loss": 0.5223588705062866,
"train_runtime": 34909.596,
"train_samples_per_second": 0.573,
"train_steps_per_second": 0.143
}
],
"max_steps": 5000,
"num_train_epochs": 3,
"total_flos": 5.76997649399808e+18,
"trial_name": null,
"trial_params": null
}