TheRains's picture
End of training
7938f75
{
"best_metric": 43.84929641398094,
"best_model_checkpoint": "./output/small/yt-special-batch12-lr4-small-halfclean/checkpoint-5000",
"epoch": 2.0300446609825418,
"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": 1.7478,
"step": 25
},
{
"epoch": 0.02,
"learning_rate": 9.4e-06,
"loss": 1.0975,
"step": 50
},
{
"epoch": 0.03,
"learning_rate": 1.44e-05,
"loss": 0.7274,
"step": 75
},
{
"epoch": 0.04,
"learning_rate": 1.94e-05,
"loss": 0.7116,
"step": 100
},
{
"epoch": 0.05,
"learning_rate": 2.44e-05,
"loss": 0.825,
"step": 125
},
{
"epoch": 0.06,
"learning_rate": 2.94e-05,
"loss": 0.8863,
"step": 150
},
{
"epoch": 0.07,
"learning_rate": 3.4399999999999996e-05,
"loss": 0.7053,
"step": 175
},
{
"epoch": 0.08,
"learning_rate": 3.94e-05,
"loss": 0.7759,
"step": 200
},
{
"epoch": 0.09,
"learning_rate": 4.44e-05,
"loss": 0.7952,
"step": 225
},
{
"epoch": 0.1,
"learning_rate": 4.94e-05,
"loss": 0.8367,
"step": 250
},
{
"epoch": 0.11,
"learning_rate": 5.440000000000001e-05,
"loss": 0.8557,
"step": 275
},
{
"epoch": 0.12,
"learning_rate": 5.94e-05,
"loss": 0.934,
"step": 300
},
{
"epoch": 0.13,
"learning_rate": 6.440000000000001e-05,
"loss": 0.9576,
"step": 325
},
{
"epoch": 0.14,
"learning_rate": 6.939999999999999e-05,
"loss": 0.9859,
"step": 350
},
{
"epoch": 0.15,
"learning_rate": 7.44e-05,
"loss": 0.9099,
"step": 375
},
{
"epoch": 0.16,
"learning_rate": 7.94e-05,
"loss": 0.9983,
"step": 400
},
{
"epoch": 0.17,
"learning_rate": 8.44e-05,
"loss": 1.0015,
"step": 425
},
{
"epoch": 0.18,
"learning_rate": 8.94e-05,
"loss": 1.0961,
"step": 450
},
{
"epoch": 0.19,
"learning_rate": 9.44e-05,
"loss": 1.0579,
"step": 475
},
{
"epoch": 0.2,
"learning_rate": 9.94e-05,
"loss": 1.0551,
"step": 500
},
{
"epoch": 0.21,
"learning_rate": 9.95111111111111e-05,
"loss": 1.1646,
"step": 525
},
{
"epoch": 0.22,
"learning_rate": 9.895555555555556e-05,
"loss": 1.1698,
"step": 550
},
{
"epoch": 0.23,
"learning_rate": 9.84e-05,
"loss": 1.1542,
"step": 575
},
{
"epoch": 0.24,
"learning_rate": 9.784444444444444e-05,
"loss": 1.2021,
"step": 600
},
{
"epoch": 0.25,
"learning_rate": 9.72888888888889e-05,
"loss": 1.2163,
"step": 625
},
{
"epoch": 0.26,
"learning_rate": 9.673333333333334e-05,
"loss": 1.1018,
"step": 650
},
{
"epoch": 0.27,
"learning_rate": 9.617777777777778e-05,
"loss": 1.2459,
"step": 675
},
{
"epoch": 0.28,
"learning_rate": 9.562222222222222e-05,
"loss": 1.1877,
"step": 700
},
{
"epoch": 0.29,
"learning_rate": 9.506666666666667e-05,
"loss": 1.022,
"step": 725
},
{
"epoch": 0.3,
"learning_rate": 9.451111111111112e-05,
"loss": 1.0854,
"step": 750
},
{
"epoch": 0.31,
"learning_rate": 9.395555555555557e-05,
"loss": 1.0567,
"step": 775
},
{
"epoch": 0.32,
"learning_rate": 9.340000000000001e-05,
"loss": 1.0928,
"step": 800
},
{
"epoch": 0.33,
"learning_rate": 9.284444444444444e-05,
"loss": 1.049,
"step": 825
},
{
"epoch": 0.35,
"learning_rate": 9.22888888888889e-05,
"loss": 1.1207,
"step": 850
},
{
"epoch": 0.36,
"learning_rate": 9.173333333333333e-05,
"loss": 1.0236,
"step": 875
},
{
"epoch": 0.37,
"learning_rate": 9.117777777777778e-05,
"loss": 1.0685,
"step": 900
},
{
"epoch": 0.38,
"learning_rate": 9.062222222222223e-05,
"loss": 1.0391,
"step": 925
},
{
"epoch": 0.39,
"learning_rate": 9.006666666666667e-05,
"loss": 0.9874,
"step": 950
},
{
"epoch": 0.4,
"learning_rate": 8.95111111111111e-05,
"loss": 1.069,
"step": 975
},
{
"epoch": 0.41,
"learning_rate": 8.895555555555556e-05,
"loss": 0.9631,
"step": 1000
},
{
"epoch": 0.41,
"eval_loss": 1.3609400987625122,
"eval_runtime": 1797.368,
"eval_samples_per_second": 1.364,
"eval_steps_per_second": 0.228,
"eval_wer": 83.4770767135724,
"step": 1000
},
{
"epoch": 0.42,
"learning_rate": 8.840000000000001e-05,
"loss": 0.962,
"step": 1025
},
{
"epoch": 0.43,
"learning_rate": 8.784444444444446e-05,
"loss": 1.0816,
"step": 1050
},
{
"epoch": 0.44,
"learning_rate": 8.72888888888889e-05,
"loss": 1.0146,
"step": 1075
},
{
"epoch": 0.45,
"learning_rate": 8.673333333333333e-05,
"loss": 1.0391,
"step": 1100
},
{
"epoch": 0.46,
"learning_rate": 8.617777777777778e-05,
"loss": 1.0887,
"step": 1125
},
{
"epoch": 0.47,
"learning_rate": 8.562222222222222e-05,
"loss": 0.9843,
"step": 1150
},
{
"epoch": 0.48,
"learning_rate": 8.506666666666667e-05,
"loss": 0.955,
"step": 1175
},
{
"epoch": 0.49,
"learning_rate": 8.451111111111112e-05,
"loss": 1.0606,
"step": 1200
},
{
"epoch": 0.5,
"learning_rate": 8.395555555555557e-05,
"loss": 0.8814,
"step": 1225
},
{
"epoch": 0.51,
"learning_rate": 8.34e-05,
"loss": 0.9109,
"step": 1250
},
{
"epoch": 0.52,
"learning_rate": 8.284444444444444e-05,
"loss": 0.9574,
"step": 1275
},
{
"epoch": 0.53,
"learning_rate": 8.22888888888889e-05,
"loss": 0.9929,
"step": 1300
},
{
"epoch": 0.54,
"learning_rate": 8.173333333333335e-05,
"loss": 0.9634,
"step": 1325
},
{
"epoch": 0.55,
"learning_rate": 8.117777777777778e-05,
"loss": 0.9718,
"step": 1350
},
{
"epoch": 0.56,
"learning_rate": 8.062222222222222e-05,
"loss": 0.9303,
"step": 1375
},
{
"epoch": 0.57,
"learning_rate": 8.006666666666667e-05,
"loss": 0.8809,
"step": 1400
},
{
"epoch": 0.58,
"learning_rate": 7.951111111111111e-05,
"loss": 0.9295,
"step": 1425
},
{
"epoch": 0.59,
"learning_rate": 7.895555555555556e-05,
"loss": 0.9107,
"step": 1450
},
{
"epoch": 0.6,
"learning_rate": 7.840000000000001e-05,
"loss": 0.9706,
"step": 1475
},
{
"epoch": 0.61,
"learning_rate": 7.784444444444446e-05,
"loss": 0.986,
"step": 1500
},
{
"epoch": 0.62,
"learning_rate": 7.728888888888888e-05,
"loss": 0.9268,
"step": 1525
},
{
"epoch": 0.63,
"learning_rate": 7.673333333333333e-05,
"loss": 0.9369,
"step": 1550
},
{
"epoch": 0.64,
"learning_rate": 7.617777777777778e-05,
"loss": 0.8768,
"step": 1575
},
{
"epoch": 0.65,
"learning_rate": 7.562222222222223e-05,
"loss": 0.8991,
"step": 1600
},
{
"epoch": 0.66,
"learning_rate": 7.506666666666667e-05,
"loss": 0.9176,
"step": 1625
},
{
"epoch": 0.67,
"learning_rate": 7.451111111111111e-05,
"loss": 0.8547,
"step": 1650
},
{
"epoch": 0.68,
"learning_rate": 7.395555555555556e-05,
"loss": 0.8989,
"step": 1675
},
{
"epoch": 0.69,
"learning_rate": 7.340000000000001e-05,
"loss": 0.9493,
"step": 1700
},
{
"epoch": 0.7,
"learning_rate": 7.284444444444445e-05,
"loss": 0.9159,
"step": 1725
},
{
"epoch": 0.71,
"learning_rate": 7.22888888888889e-05,
"loss": 0.8894,
"step": 1750
},
{
"epoch": 0.72,
"learning_rate": 7.173333333333335e-05,
"loss": 0.7883,
"step": 1775
},
{
"epoch": 0.73,
"learning_rate": 7.117777777777777e-05,
"loss": 0.8495,
"step": 1800
},
{
"epoch": 0.74,
"learning_rate": 7.062222222222222e-05,
"loss": 0.9049,
"step": 1825
},
{
"epoch": 0.75,
"learning_rate": 7.006666666666667e-05,
"loss": 0.8152,
"step": 1850
},
{
"epoch": 0.76,
"learning_rate": 6.951111111111112e-05,
"loss": 0.9479,
"step": 1875
},
{
"epoch": 0.77,
"learning_rate": 6.895555555555556e-05,
"loss": 0.8737,
"step": 1900
},
{
"epoch": 0.78,
"learning_rate": 6.840000000000001e-05,
"loss": 0.7462,
"step": 1925
},
{
"epoch": 0.79,
"learning_rate": 6.784444444444445e-05,
"loss": 0.8745,
"step": 1950
},
{
"epoch": 0.8,
"learning_rate": 6.72888888888889e-05,
"loss": 0.7957,
"step": 1975
},
{
"epoch": 0.81,
"learning_rate": 6.673333333333333e-05,
"loss": 0.7929,
"step": 2000
},
{
"epoch": 0.81,
"eval_loss": 1.1652641296386719,
"eval_runtime": 2010.9968,
"eval_samples_per_second": 1.219,
"eval_steps_per_second": 0.203,
"eval_wer": 89.64399195901692,
"step": 2000
},
{
"epoch": 0.82,
"learning_rate": 6.617777777777778e-05,
"loss": 0.8078,
"step": 2025
},
{
"epoch": 0.83,
"learning_rate": 6.562222222222223e-05,
"loss": 0.9222,
"step": 2050
},
{
"epoch": 0.84,
"learning_rate": 6.506666666666666e-05,
"loss": 0.847,
"step": 2075
},
{
"epoch": 0.85,
"learning_rate": 6.451111111111111e-05,
"loss": 0.7932,
"step": 2100
},
{
"epoch": 0.86,
"learning_rate": 6.395555555555556e-05,
"loss": 0.8526,
"step": 2125
},
{
"epoch": 0.87,
"learning_rate": 6.340000000000001e-05,
"loss": 0.8217,
"step": 2150
},
{
"epoch": 0.88,
"learning_rate": 6.284444444444445e-05,
"loss": 0.8245,
"step": 2175
},
{
"epoch": 0.89,
"learning_rate": 6.22888888888889e-05,
"loss": 0.8669,
"step": 2200
},
{
"epoch": 0.9,
"learning_rate": 6.173333333333333e-05,
"loss": 0.8484,
"step": 2225
},
{
"epoch": 0.91,
"learning_rate": 6.117777777777778e-05,
"loss": 0.8192,
"step": 2250
},
{
"epoch": 0.92,
"learning_rate": 6.062222222222222e-05,
"loss": 0.8295,
"step": 2275
},
{
"epoch": 0.93,
"learning_rate": 6.006666666666667e-05,
"loss": 0.7852,
"step": 2300
},
{
"epoch": 0.94,
"learning_rate": 5.9511111111111116e-05,
"loss": 0.745,
"step": 2325
},
{
"epoch": 0.95,
"learning_rate": 5.8955555555555566e-05,
"loss": 0.8653,
"step": 2350
},
{
"epoch": 0.96,
"learning_rate": 5.8399999999999997e-05,
"loss": 0.7092,
"step": 2375
},
{
"epoch": 0.97,
"learning_rate": 5.784444444444445e-05,
"loss": 0.7092,
"step": 2400
},
{
"epoch": 0.98,
"learning_rate": 5.728888888888889e-05,
"loss": 0.7167,
"step": 2425
},
{
"epoch": 0.99,
"learning_rate": 5.673333333333334e-05,
"loss": 0.764,
"step": 2450
},
{
"epoch": 1.0,
"learning_rate": 5.6177777777777785e-05,
"loss": 0.5982,
"step": 2475
},
{
"epoch": 1.02,
"learning_rate": 5.562222222222222e-05,
"loss": 0.4638,
"step": 2500
},
{
"epoch": 1.03,
"learning_rate": 5.5066666666666666e-05,
"loss": 0.4064,
"step": 2525
},
{
"epoch": 1.04,
"learning_rate": 5.451111111111111e-05,
"loss": 0.4044,
"step": 2550
},
{
"epoch": 1.05,
"learning_rate": 5.395555555555556e-05,
"loss": 0.428,
"step": 2575
},
{
"epoch": 1.06,
"learning_rate": 5.3400000000000004e-05,
"loss": 0.4443,
"step": 2600
},
{
"epoch": 1.07,
"learning_rate": 5.2844444444444454e-05,
"loss": 0.4222,
"step": 2625
},
{
"epoch": 1.08,
"learning_rate": 5.2288888888888884e-05,
"loss": 0.3849,
"step": 2650
},
{
"epoch": 1.09,
"learning_rate": 5.1733333333333335e-05,
"loss": 0.4201,
"step": 2675
},
{
"epoch": 1.1,
"learning_rate": 5.117777777777778e-05,
"loss": 0.4051,
"step": 2700
},
{
"epoch": 1.11,
"learning_rate": 5.062222222222223e-05,
"loss": 0.4121,
"step": 2725
},
{
"epoch": 1.12,
"learning_rate": 5.006666666666667e-05,
"loss": 0.3973,
"step": 2750
},
{
"epoch": 1.13,
"learning_rate": 4.951111111111112e-05,
"loss": 0.3674,
"step": 2775
},
{
"epoch": 1.14,
"learning_rate": 4.895555555555556e-05,
"loss": 0.4242,
"step": 2800
},
{
"epoch": 1.15,
"learning_rate": 4.8400000000000004e-05,
"loss": 0.3909,
"step": 2825
},
{
"epoch": 1.16,
"learning_rate": 4.784444444444445e-05,
"loss": 0.4518,
"step": 2850
},
{
"epoch": 1.17,
"learning_rate": 4.728888888888889e-05,
"loss": 0.4253,
"step": 2875
},
{
"epoch": 1.18,
"learning_rate": 4.6733333333333335e-05,
"loss": 0.4116,
"step": 2900
},
{
"epoch": 1.19,
"learning_rate": 4.617777777777778e-05,
"loss": 0.4096,
"step": 2925
},
{
"epoch": 1.2,
"learning_rate": 4.562222222222222e-05,
"loss": 0.4302,
"step": 2950
},
{
"epoch": 1.21,
"learning_rate": 4.5066666666666667e-05,
"loss": 0.4254,
"step": 2975
},
{
"epoch": 1.22,
"learning_rate": 4.451111111111112e-05,
"loss": 0.4189,
"step": 3000
},
{
"epoch": 1.22,
"eval_loss": 1.0419443845748901,
"eval_runtime": 1762.9691,
"eval_samples_per_second": 1.39,
"eval_steps_per_second": 0.232,
"eval_wer": 61.64969846313468,
"step": 3000
},
{
"epoch": 1.23,
"learning_rate": 4.3955555555555554e-05,
"loss": 0.3847,
"step": 3025
},
{
"epoch": 1.24,
"learning_rate": 4.3400000000000005e-05,
"loss": 0.3829,
"step": 3050
},
{
"epoch": 1.25,
"learning_rate": 4.284444444444445e-05,
"loss": 0.3793,
"step": 3075
},
{
"epoch": 1.26,
"learning_rate": 4.228888888888889e-05,
"loss": 0.3805,
"step": 3100
},
{
"epoch": 1.27,
"learning_rate": 4.1733333333333336e-05,
"loss": 0.4464,
"step": 3125
},
{
"epoch": 1.28,
"learning_rate": 4.117777777777778e-05,
"loss": 0.3997,
"step": 3150
},
{
"epoch": 1.29,
"learning_rate": 4.062222222222222e-05,
"loss": 0.408,
"step": 3175
},
{
"epoch": 1.3,
"learning_rate": 4.006666666666667e-05,
"loss": 0.4427,
"step": 3200
},
{
"epoch": 1.31,
"learning_rate": 3.9533333333333337e-05,
"loss": 0.4223,
"step": 3225
},
{
"epoch": 1.32,
"learning_rate": 3.897777777777778e-05,
"loss": 0.4477,
"step": 3250
},
{
"epoch": 1.33,
"learning_rate": 3.8422222222222224e-05,
"loss": 0.3458,
"step": 3275
},
{
"epoch": 1.34,
"learning_rate": 3.786666666666667e-05,
"loss": 0.3754,
"step": 3300
},
{
"epoch": 1.35,
"learning_rate": 3.731111111111111e-05,
"loss": 0.3914,
"step": 3325
},
{
"epoch": 1.36,
"learning_rate": 3.675555555555556e-05,
"loss": 0.4206,
"step": 3350
},
{
"epoch": 1.37,
"learning_rate": 3.62e-05,
"loss": 0.4399,
"step": 3375
},
{
"epoch": 1.38,
"learning_rate": 3.564444444444445e-05,
"loss": 0.3971,
"step": 3400
},
{
"epoch": 1.39,
"learning_rate": 3.5088888888888886e-05,
"loss": 0.3811,
"step": 3425
},
{
"epoch": 1.4,
"learning_rate": 3.453333333333334e-05,
"loss": 0.3807,
"step": 3450
},
{
"epoch": 1.41,
"learning_rate": 3.397777777777778e-05,
"loss": 0.3495,
"step": 3475
},
{
"epoch": 1.42,
"learning_rate": 3.3422222222222224e-05,
"loss": 0.3401,
"step": 3500
},
{
"epoch": 1.43,
"learning_rate": 3.286666666666667e-05,
"loss": 0.3621,
"step": 3525
},
{
"epoch": 1.44,
"learning_rate": 3.231111111111111e-05,
"loss": 0.4035,
"step": 3550
},
{
"epoch": 1.45,
"learning_rate": 3.1755555555555556e-05,
"loss": 0.3977,
"step": 3575
},
{
"epoch": 1.46,
"learning_rate": 3.12e-05,
"loss": 0.3543,
"step": 3600
},
{
"epoch": 1.47,
"learning_rate": 3.064444444444445e-05,
"loss": 0.356,
"step": 3625
},
{
"epoch": 1.48,
"learning_rate": 3.008888888888889e-05,
"loss": 0.3299,
"step": 3650
},
{
"epoch": 1.49,
"learning_rate": 2.9533333333333334e-05,
"loss": 0.3907,
"step": 3675
},
{
"epoch": 1.5,
"learning_rate": 2.897777777777778e-05,
"loss": 0.3948,
"step": 3700
},
{
"epoch": 1.51,
"learning_rate": 2.842222222222222e-05,
"loss": 0.3801,
"step": 3725
},
{
"epoch": 1.52,
"learning_rate": 2.786666666666667e-05,
"loss": 0.344,
"step": 3750
},
{
"epoch": 1.53,
"learning_rate": 2.7311111111111116e-05,
"loss": 0.3511,
"step": 3775
},
{
"epoch": 1.54,
"learning_rate": 2.6755555555555556e-05,
"loss": 0.322,
"step": 3800
},
{
"epoch": 1.55,
"learning_rate": 2.6200000000000003e-05,
"loss": 0.3605,
"step": 3825
},
{
"epoch": 1.56,
"learning_rate": 2.5644444444444444e-05,
"loss": 0.3082,
"step": 3850
},
{
"epoch": 1.57,
"learning_rate": 2.508888888888889e-05,
"loss": 0.3407,
"step": 3875
},
{
"epoch": 1.58,
"learning_rate": 2.4533333333333334e-05,
"loss": 0.3504,
"step": 3900
},
{
"epoch": 1.59,
"learning_rate": 2.3977777777777778e-05,
"loss": 0.3835,
"step": 3925
},
{
"epoch": 1.6,
"learning_rate": 2.3422222222222222e-05,
"loss": 0.3127,
"step": 3950
},
{
"epoch": 1.61,
"learning_rate": 2.2866666666666666e-05,
"loss": 0.3362,
"step": 3975
},
{
"epoch": 1.62,
"learning_rate": 2.2311111111111113e-05,
"loss": 0.3195,
"step": 4000
},
{
"epoch": 1.62,
"eval_loss": 0.9384303689002991,
"eval_runtime": 1634.7148,
"eval_samples_per_second": 1.499,
"eval_steps_per_second": 0.25,
"eval_wer": 45.4380390376759,
"step": 4000
},
{
"epoch": 1.63,
"learning_rate": 2.1755555555555557e-05,
"loss": 0.3324,
"step": 4025
},
{
"epoch": 1.64,
"learning_rate": 2.12e-05,
"loss": 0.3142,
"step": 4050
},
{
"epoch": 1.65,
"learning_rate": 2.0644444444444447e-05,
"loss": 0.3164,
"step": 4075
},
{
"epoch": 1.66,
"learning_rate": 2.008888888888889e-05,
"loss": 0.365,
"step": 4100
},
{
"epoch": 1.67,
"learning_rate": 1.9533333333333335e-05,
"loss": 0.326,
"step": 4125
},
{
"epoch": 1.68,
"learning_rate": 1.897777777777778e-05,
"loss": 0.3771,
"step": 4150
},
{
"epoch": 1.7,
"learning_rate": 1.8422222222222222e-05,
"loss": 0.3725,
"step": 4175
},
{
"epoch": 1.71,
"learning_rate": 1.7866666666666666e-05,
"loss": 0.3122,
"step": 4200
},
{
"epoch": 1.72,
"learning_rate": 1.731111111111111e-05,
"loss": 0.3656,
"step": 4225
},
{
"epoch": 1.73,
"learning_rate": 1.6755555555555557e-05,
"loss": 0.33,
"step": 4250
},
{
"epoch": 1.74,
"learning_rate": 1.62e-05,
"loss": 0.3264,
"step": 4275
},
{
"epoch": 1.75,
"learning_rate": 1.5644444444444444e-05,
"loss": 0.2959,
"step": 4300
},
{
"epoch": 1.76,
"learning_rate": 1.5088888888888888e-05,
"loss": 0.2978,
"step": 4325
},
{
"epoch": 1.77,
"learning_rate": 1.4533333333333335e-05,
"loss": 0.3247,
"step": 4350
},
{
"epoch": 1.78,
"learning_rate": 1.3977777777777779e-05,
"loss": 0.3401,
"step": 4375
},
{
"epoch": 1.79,
"learning_rate": 1.3422222222222223e-05,
"loss": 0.3142,
"step": 4400
},
{
"epoch": 1.8,
"learning_rate": 1.2866666666666668e-05,
"loss": 0.3089,
"step": 4425
},
{
"epoch": 1.81,
"learning_rate": 1.2311111111111112e-05,
"loss": 0.2932,
"step": 4450
},
{
"epoch": 1.82,
"learning_rate": 1.1755555555555556e-05,
"loss": 0.3154,
"step": 4475
},
{
"epoch": 1.83,
"learning_rate": 1.1200000000000001e-05,
"loss": 0.348,
"step": 4500
},
{
"epoch": 1.84,
"learning_rate": 1.0644444444444445e-05,
"loss": 0.3508,
"step": 4525
},
{
"epoch": 1.85,
"learning_rate": 1.0088888888888889e-05,
"loss": 0.3336,
"step": 4550
},
{
"epoch": 1.86,
"learning_rate": 9.533333333333334e-06,
"loss": 0.2896,
"step": 4575
},
{
"epoch": 1.87,
"learning_rate": 8.977777777777778e-06,
"loss": 0.3015,
"step": 4600
},
{
"epoch": 1.88,
"learning_rate": 8.422222222222223e-06,
"loss": 0.2998,
"step": 4625
},
{
"epoch": 1.89,
"learning_rate": 7.866666666666667e-06,
"loss": 0.3126,
"step": 4650
},
{
"epoch": 1.9,
"learning_rate": 7.311111111111112e-06,
"loss": 0.2938,
"step": 4675
},
{
"epoch": 1.91,
"learning_rate": 6.755555555555555e-06,
"loss": 0.3071,
"step": 4700
},
{
"epoch": 1.92,
"learning_rate": 6.2e-06,
"loss": 0.3095,
"step": 4725
},
{
"epoch": 1.93,
"learning_rate": 5.6444444444444445e-06,
"loss": 0.3427,
"step": 4750
},
{
"epoch": 1.94,
"learning_rate": 5.088888888888889e-06,
"loss": 0.3112,
"step": 4775
},
{
"epoch": 1.95,
"learning_rate": 4.533333333333334e-06,
"loss": 0.3198,
"step": 4800
},
{
"epoch": 1.96,
"learning_rate": 3.977777777777778e-06,
"loss": 0.2896,
"step": 4825
},
{
"epoch": 1.97,
"learning_rate": 3.4222222222222224e-06,
"loss": 0.2687,
"step": 4850
},
{
"epoch": 1.98,
"learning_rate": 2.8666666666666666e-06,
"loss": 0.2945,
"step": 4875
},
{
"epoch": 1.99,
"learning_rate": 2.311111111111111e-06,
"loss": 0.2803,
"step": 4900
},
{
"epoch": 2.0,
"learning_rate": 1.7555555555555558e-06,
"loss": 0.2775,
"step": 4925
},
{
"epoch": 2.01,
"learning_rate": 1.2000000000000002e-06,
"loss": 0.1358,
"step": 4950
},
{
"epoch": 2.02,
"learning_rate": 6.444444444444444e-07,
"loss": 0.1088,
"step": 4975
},
{
"epoch": 2.03,
"learning_rate": 8.88888888888889e-08,
"loss": 0.1179,
"step": 5000
},
{
"epoch": 2.03,
"eval_loss": 0.861615777015686,
"eval_runtime": 1635.5901,
"eval_samples_per_second": 1.499,
"eval_steps_per_second": 0.25,
"eval_wer": 43.84929641398094,
"step": 5000
},
{
"epoch": 2.03,
"step": 5000,
"total_flos": 1.731108382359552e+19,
"train_loss": 0.6445532434463501,
"train_runtime": 46427.0848,
"train_samples_per_second": 1.292,
"train_steps_per_second": 0.108
}
],
"max_steps": 5000,
"num_train_epochs": 3,
"total_flos": 1.731108382359552e+19,
"trial_name": null,
"trial_params": null
}