whisper-base-common_voice0427 / trainer_state.json
liushaowei's picture
Upload 18 files
7fc16ac
{
"best_metric": 80.77868077868078,
"best_model_checkpoint": "./whisper-base-Chinese/checkpoint-12000",
"epoch": 5.649717514124294,
"global_step": 14000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 2.2e-07,
"loss": 2.1244,
"step": 25
},
{
"epoch": 0.02,
"learning_rate": 4.7000000000000005e-07,
"loss": 2.0967,
"step": 50
},
{
"epoch": 0.03,
"learning_rate": 7.2e-07,
"loss": 1.7948,
"step": 75
},
{
"epoch": 0.04,
"learning_rate": 9.7e-07,
"loss": 1.2271,
"step": 100
},
{
"epoch": 0.05,
"learning_rate": 1.2200000000000002e-06,
"loss": 0.9593,
"step": 125
},
{
"epoch": 0.06,
"learning_rate": 1.4700000000000001e-06,
"loss": 0.8795,
"step": 150
},
{
"epoch": 0.07,
"learning_rate": 1.72e-06,
"loss": 0.8847,
"step": 175
},
{
"epoch": 0.08,
"learning_rate": 1.97e-06,
"loss": 0.7976,
"step": 200
},
{
"epoch": 0.09,
"learning_rate": 2.21e-06,
"loss": 0.7272,
"step": 225
},
{
"epoch": 0.1,
"learning_rate": 2.46e-06,
"loss": 0.7368,
"step": 250
},
{
"epoch": 0.11,
"learning_rate": 2.7100000000000003e-06,
"loss": 0.7681,
"step": 275
},
{
"epoch": 0.12,
"learning_rate": 2.96e-06,
"loss": 0.7916,
"step": 300
},
{
"epoch": 0.13,
"learning_rate": 3.21e-06,
"loss": 0.8346,
"step": 325
},
{
"epoch": 0.14,
"learning_rate": 3.46e-06,
"loss": 0.7713,
"step": 350
},
{
"epoch": 0.15,
"learning_rate": 3.7100000000000005e-06,
"loss": 0.771,
"step": 375
},
{
"epoch": 0.16,
"learning_rate": 3.96e-06,
"loss": 0.6666,
"step": 400
},
{
"epoch": 0.17,
"learning_rate": 4.21e-06,
"loss": 0.7252,
"step": 425
},
{
"epoch": 0.18,
"learning_rate": 4.4600000000000005e-06,
"loss": 0.6224,
"step": 450
},
{
"epoch": 0.19,
"learning_rate": 4.71e-06,
"loss": 0.6665,
"step": 475
},
{
"epoch": 0.2,
"learning_rate": 4.960000000000001e-06,
"loss": 0.6073,
"step": 500
},
{
"epoch": 0.21,
"learning_rate": 5.210000000000001e-06,
"loss": 0.6667,
"step": 525
},
{
"epoch": 0.22,
"learning_rate": 5.460000000000001e-06,
"loss": 0.6848,
"step": 550
},
{
"epoch": 0.23,
"learning_rate": 5.71e-06,
"loss": 0.7169,
"step": 575
},
{
"epoch": 0.24,
"learning_rate": 5.9600000000000005e-06,
"loss": 0.6162,
"step": 600
},
{
"epoch": 0.25,
"learning_rate": 6.210000000000001e-06,
"loss": 0.6384,
"step": 625
},
{
"epoch": 0.26,
"learning_rate": 6.460000000000001e-06,
"loss": 0.6729,
"step": 650
},
{
"epoch": 0.27,
"learning_rate": 6.710000000000001e-06,
"loss": 0.761,
"step": 675
},
{
"epoch": 0.28,
"learning_rate": 6.96e-06,
"loss": 0.6227,
"step": 700
},
{
"epoch": 0.29,
"learning_rate": 7.2100000000000004e-06,
"loss": 0.6822,
"step": 725
},
{
"epoch": 0.3,
"learning_rate": 7.4600000000000006e-06,
"loss": 0.6466,
"step": 750
},
{
"epoch": 0.31,
"learning_rate": 7.71e-06,
"loss": 0.5998,
"step": 775
},
{
"epoch": 0.32,
"learning_rate": 7.960000000000002e-06,
"loss": 0.648,
"step": 800
},
{
"epoch": 0.33,
"learning_rate": 8.210000000000001e-06,
"loss": 0.5941,
"step": 825
},
{
"epoch": 0.34,
"learning_rate": 8.46e-06,
"loss": 0.6946,
"step": 850
},
{
"epoch": 0.35,
"learning_rate": 8.710000000000001e-06,
"loss": 0.632,
"step": 875
},
{
"epoch": 0.36,
"learning_rate": 8.96e-06,
"loss": 0.617,
"step": 900
},
{
"epoch": 0.37,
"learning_rate": 9.210000000000002e-06,
"loss": 0.5018,
"step": 925
},
{
"epoch": 0.38,
"learning_rate": 9.460000000000001e-06,
"loss": 0.6201,
"step": 950
},
{
"epoch": 0.39,
"learning_rate": 9.71e-06,
"loss": 0.6265,
"step": 975
},
{
"epoch": 0.4,
"learning_rate": 9.960000000000001e-06,
"loss": 0.564,
"step": 1000
},
{
"epoch": 0.41,
"learning_rate": 9.988947368421053e-06,
"loss": 0.6589,
"step": 1025
},
{
"epoch": 0.42,
"learning_rate": 9.975789473684211e-06,
"loss": 0.5549,
"step": 1050
},
{
"epoch": 0.43,
"learning_rate": 9.96263157894737e-06,
"loss": 0.5769,
"step": 1075
},
{
"epoch": 0.44,
"learning_rate": 9.949473684210526e-06,
"loss": 0.6697,
"step": 1100
},
{
"epoch": 0.45,
"learning_rate": 9.936315789473685e-06,
"loss": 0.6405,
"step": 1125
},
{
"epoch": 0.46,
"learning_rate": 9.923157894736844e-06,
"loss": 0.646,
"step": 1150
},
{
"epoch": 0.47,
"learning_rate": 9.91e-06,
"loss": 0.6236,
"step": 1175
},
{
"epoch": 0.48,
"learning_rate": 9.89684210526316e-06,
"loss": 0.6284,
"step": 1200
},
{
"epoch": 0.49,
"learning_rate": 9.883684210526317e-06,
"loss": 0.6444,
"step": 1225
},
{
"epoch": 0.5,
"learning_rate": 9.870526315789474e-06,
"loss": 0.6004,
"step": 1250
},
{
"epoch": 0.51,
"learning_rate": 9.857368421052632e-06,
"loss": 0.5655,
"step": 1275
},
{
"epoch": 0.52,
"learning_rate": 9.84421052631579e-06,
"loss": 0.5671,
"step": 1300
},
{
"epoch": 0.53,
"learning_rate": 9.831052631578948e-06,
"loss": 0.6141,
"step": 1325
},
{
"epoch": 0.54,
"learning_rate": 9.817894736842106e-06,
"loss": 0.5869,
"step": 1350
},
{
"epoch": 0.55,
"learning_rate": 9.804736842105263e-06,
"loss": 0.5613,
"step": 1375
},
{
"epoch": 0.56,
"learning_rate": 9.791578947368422e-06,
"loss": 0.5943,
"step": 1400
},
{
"epoch": 0.58,
"learning_rate": 9.77842105263158e-06,
"loss": 0.5147,
"step": 1425
},
{
"epoch": 0.59,
"learning_rate": 9.765263157894737e-06,
"loss": 0.6198,
"step": 1450
},
{
"epoch": 0.6,
"learning_rate": 9.752105263157897e-06,
"loss": 0.6331,
"step": 1475
},
{
"epoch": 0.61,
"learning_rate": 9.738947368421054e-06,
"loss": 0.5842,
"step": 1500
},
{
"epoch": 0.62,
"learning_rate": 9.725789473684212e-06,
"loss": 0.5856,
"step": 1525
},
{
"epoch": 0.63,
"learning_rate": 9.712631578947369e-06,
"loss": 0.5456,
"step": 1550
},
{
"epoch": 0.64,
"learning_rate": 9.699473684210528e-06,
"loss": 0.5578,
"step": 1575
},
{
"epoch": 0.65,
"learning_rate": 9.686315789473684e-06,
"loss": 0.5631,
"step": 1600
},
{
"epoch": 0.66,
"learning_rate": 9.673157894736843e-06,
"loss": 0.6264,
"step": 1625
},
{
"epoch": 0.67,
"learning_rate": 9.66e-06,
"loss": 0.597,
"step": 1650
},
{
"epoch": 0.68,
"learning_rate": 9.646842105263158e-06,
"loss": 0.5916,
"step": 1675
},
{
"epoch": 0.69,
"learning_rate": 9.633684210526316e-06,
"loss": 0.5901,
"step": 1700
},
{
"epoch": 0.7,
"learning_rate": 9.620526315789475e-06,
"loss": 0.5801,
"step": 1725
},
{
"epoch": 0.71,
"learning_rate": 9.607368421052632e-06,
"loss": 0.5317,
"step": 1750
},
{
"epoch": 0.72,
"learning_rate": 9.59421052631579e-06,
"loss": 0.5592,
"step": 1775
},
{
"epoch": 0.73,
"learning_rate": 9.581052631578947e-06,
"loss": 0.5614,
"step": 1800
},
{
"epoch": 0.74,
"learning_rate": 9.567894736842106e-06,
"loss": 0.6351,
"step": 1825
},
{
"epoch": 0.75,
"learning_rate": 9.554736842105264e-06,
"loss": 0.5655,
"step": 1850
},
{
"epoch": 0.76,
"learning_rate": 9.541578947368421e-06,
"loss": 0.5137,
"step": 1875
},
{
"epoch": 0.77,
"learning_rate": 9.52842105263158e-06,
"loss": 0.5644,
"step": 1900
},
{
"epoch": 0.78,
"learning_rate": 9.515263157894738e-06,
"loss": 0.4815,
"step": 1925
},
{
"epoch": 0.79,
"learning_rate": 9.502105263157896e-06,
"loss": 0.5776,
"step": 1950
},
{
"epoch": 0.8,
"learning_rate": 9.488947368421053e-06,
"loss": 0.575,
"step": 1975
},
{
"epoch": 0.81,
"learning_rate": 9.475789473684212e-06,
"loss": 0.5649,
"step": 2000
},
{
"epoch": 0.81,
"eval_loss": 0.4785773754119873,
"eval_runtime": 2365.5611,
"eval_samples_per_second": 4.473,
"eval_steps_per_second": 0.28,
"eval_wer": 85.7966357966358,
"step": 2000
},
{
"epoch": 0.82,
"learning_rate": 9.462631578947368e-06,
"loss": 0.5274,
"step": 2025
},
{
"epoch": 0.83,
"learning_rate": 9.449473684210527e-06,
"loss": 0.5418,
"step": 2050
},
{
"epoch": 0.84,
"learning_rate": 9.436315789473685e-06,
"loss": 0.5214,
"step": 2075
},
{
"epoch": 0.85,
"learning_rate": 9.423157894736842e-06,
"loss": 0.5318,
"step": 2100
},
{
"epoch": 0.86,
"learning_rate": 9.41e-06,
"loss": 0.5926,
"step": 2125
},
{
"epoch": 0.87,
"learning_rate": 9.396842105263159e-06,
"loss": 0.5911,
"step": 2150
},
{
"epoch": 0.88,
"learning_rate": 9.383684210526316e-06,
"loss": 0.5175,
"step": 2175
},
{
"epoch": 0.89,
"learning_rate": 9.370526315789474e-06,
"loss": 0.5669,
"step": 2200
},
{
"epoch": 0.9,
"learning_rate": 9.357368421052633e-06,
"loss": 0.5104,
"step": 2225
},
{
"epoch": 0.91,
"learning_rate": 9.34421052631579e-06,
"loss": 0.5228,
"step": 2250
},
{
"epoch": 0.92,
"learning_rate": 9.331052631578948e-06,
"loss": 0.5131,
"step": 2275
},
{
"epoch": 0.93,
"learning_rate": 9.317894736842105e-06,
"loss": 0.5767,
"step": 2300
},
{
"epoch": 0.94,
"learning_rate": 9.304736842105265e-06,
"loss": 0.5493,
"step": 2325
},
{
"epoch": 0.95,
"learning_rate": 9.291578947368422e-06,
"loss": 0.5046,
"step": 2350
},
{
"epoch": 0.96,
"learning_rate": 9.27842105263158e-06,
"loss": 0.5923,
"step": 2375
},
{
"epoch": 0.97,
"learning_rate": 9.265263157894737e-06,
"loss": 0.5278,
"step": 2400
},
{
"epoch": 0.98,
"learning_rate": 9.252105263157896e-06,
"loss": 0.5761,
"step": 2425
},
{
"epoch": 0.99,
"learning_rate": 9.238947368421052e-06,
"loss": 0.5782,
"step": 2450
},
{
"epoch": 1.0,
"learning_rate": 9.225789473684211e-06,
"loss": 0.5325,
"step": 2475
},
{
"epoch": 1.01,
"learning_rate": 9.21263157894737e-06,
"loss": 0.4361,
"step": 2500
},
{
"epoch": 1.02,
"learning_rate": 9.199473684210526e-06,
"loss": 0.4985,
"step": 2525
},
{
"epoch": 1.03,
"learning_rate": 9.186315789473685e-06,
"loss": 0.4505,
"step": 2550
},
{
"epoch": 1.04,
"learning_rate": 9.173157894736843e-06,
"loss": 0.459,
"step": 2575
},
{
"epoch": 1.05,
"learning_rate": 9.16e-06,
"loss": 0.4398,
"step": 2600
},
{
"epoch": 1.06,
"learning_rate": 9.146842105263158e-06,
"loss": 0.4314,
"step": 2625
},
{
"epoch": 1.07,
"learning_rate": 9.133684210526317e-06,
"loss": 0.4722,
"step": 2650
},
{
"epoch": 1.08,
"learning_rate": 9.120526315789475e-06,
"loss": 0.4472,
"step": 2675
},
{
"epoch": 1.09,
"learning_rate": 9.107368421052632e-06,
"loss": 0.423,
"step": 2700
},
{
"epoch": 1.1,
"learning_rate": 9.09421052631579e-06,
"loss": 0.4783,
"step": 2725
},
{
"epoch": 1.11,
"learning_rate": 9.081052631578949e-06,
"loss": 0.4566,
"step": 2750
},
{
"epoch": 1.12,
"learning_rate": 9.067894736842106e-06,
"loss": 0.469,
"step": 2775
},
{
"epoch": 1.13,
"learning_rate": 9.054736842105264e-06,
"loss": 0.507,
"step": 2800
},
{
"epoch": 1.14,
"learning_rate": 9.041578947368423e-06,
"loss": 0.4164,
"step": 2825
},
{
"epoch": 1.15,
"learning_rate": 9.02842105263158e-06,
"loss": 0.4619,
"step": 2850
},
{
"epoch": 1.16,
"learning_rate": 9.015263157894738e-06,
"loss": 0.4096,
"step": 2875
},
{
"epoch": 1.17,
"learning_rate": 9.002105263157895e-06,
"loss": 0.4047,
"step": 2900
},
{
"epoch": 1.18,
"learning_rate": 8.988947368421054e-06,
"loss": 0.4122,
"step": 2925
},
{
"epoch": 1.19,
"learning_rate": 8.97578947368421e-06,
"loss": 0.4237,
"step": 2950
},
{
"epoch": 1.2,
"learning_rate": 8.96263157894737e-06,
"loss": 0.4915,
"step": 2975
},
{
"epoch": 1.21,
"learning_rate": 8.949473684210527e-06,
"loss": 0.4522,
"step": 3000
},
{
"epoch": 1.22,
"learning_rate": 8.936315789473684e-06,
"loss": 0.475,
"step": 3025
},
{
"epoch": 1.23,
"learning_rate": 8.923157894736842e-06,
"loss": 0.421,
"step": 3050
},
{
"epoch": 1.24,
"learning_rate": 8.910000000000001e-06,
"loss": 0.4252,
"step": 3075
},
{
"epoch": 1.25,
"learning_rate": 8.896842105263159e-06,
"loss": 0.3984,
"step": 3100
},
{
"epoch": 1.26,
"learning_rate": 8.883684210526316e-06,
"loss": 0.4225,
"step": 3125
},
{
"epoch": 1.27,
"learning_rate": 8.870526315789474e-06,
"loss": 0.4671,
"step": 3150
},
{
"epoch": 1.28,
"learning_rate": 8.857368421052633e-06,
"loss": 0.513,
"step": 3175
},
{
"epoch": 1.29,
"learning_rate": 8.84421052631579e-06,
"loss": 0.4499,
"step": 3200
},
{
"epoch": 1.3,
"learning_rate": 8.831052631578948e-06,
"loss": 0.5166,
"step": 3225
},
{
"epoch": 1.31,
"learning_rate": 8.817894736842107e-06,
"loss": 0.4238,
"step": 3250
},
{
"epoch": 1.32,
"learning_rate": 8.804736842105264e-06,
"loss": 0.4494,
"step": 3275
},
{
"epoch": 1.33,
"learning_rate": 8.791578947368422e-06,
"loss": 0.4708,
"step": 3300
},
{
"epoch": 1.34,
"learning_rate": 8.77842105263158e-06,
"loss": 0.4304,
"step": 3325
},
{
"epoch": 1.35,
"learning_rate": 8.765263157894739e-06,
"loss": 0.441,
"step": 3350
},
{
"epoch": 1.36,
"learning_rate": 8.752105263157894e-06,
"loss": 0.5283,
"step": 3375
},
{
"epoch": 1.37,
"learning_rate": 8.738947368421053e-06,
"loss": 0.4366,
"step": 3400
},
{
"epoch": 1.38,
"learning_rate": 8.725789473684211e-06,
"loss": 0.4281,
"step": 3425
},
{
"epoch": 1.39,
"learning_rate": 8.712631578947368e-06,
"loss": 0.4667,
"step": 3450
},
{
"epoch": 1.4,
"learning_rate": 8.699473684210526e-06,
"loss": 0.4286,
"step": 3475
},
{
"epoch": 1.41,
"learning_rate": 8.686315789473685e-06,
"loss": 0.4623,
"step": 3500
},
{
"epoch": 1.42,
"learning_rate": 8.673157894736843e-06,
"loss": 0.4533,
"step": 3525
},
{
"epoch": 1.43,
"learning_rate": 8.66e-06,
"loss": 0.4646,
"step": 3550
},
{
"epoch": 1.44,
"learning_rate": 8.64684210526316e-06,
"loss": 0.4365,
"step": 3575
},
{
"epoch": 1.45,
"learning_rate": 8.633684210526317e-06,
"loss": 0.4762,
"step": 3600
},
{
"epoch": 1.46,
"learning_rate": 8.620526315789474e-06,
"loss": 0.5173,
"step": 3625
},
{
"epoch": 1.47,
"learning_rate": 8.607368421052632e-06,
"loss": 0.4368,
"step": 3650
},
{
"epoch": 1.48,
"learning_rate": 8.594210526315791e-06,
"loss": 0.4165,
"step": 3675
},
{
"epoch": 1.49,
"learning_rate": 8.581052631578948e-06,
"loss": 0.416,
"step": 3700
},
{
"epoch": 1.5,
"learning_rate": 8.567894736842106e-06,
"loss": 0.4247,
"step": 3725
},
{
"epoch": 1.51,
"learning_rate": 8.554736842105263e-06,
"loss": 0.4047,
"step": 3750
},
{
"epoch": 1.52,
"learning_rate": 8.541578947368423e-06,
"loss": 0.4254,
"step": 3775
},
{
"epoch": 1.53,
"learning_rate": 8.528421052631578e-06,
"loss": 0.4822,
"step": 3800
},
{
"epoch": 1.54,
"learning_rate": 8.515263157894738e-06,
"loss": 0.3774,
"step": 3825
},
{
"epoch": 1.55,
"learning_rate": 8.502105263157897e-06,
"loss": 0.4308,
"step": 3850
},
{
"epoch": 1.56,
"learning_rate": 8.488947368421052e-06,
"loss": 0.4439,
"step": 3875
},
{
"epoch": 1.57,
"learning_rate": 8.475789473684212e-06,
"loss": 0.5038,
"step": 3900
},
{
"epoch": 1.58,
"learning_rate": 8.46263157894737e-06,
"loss": 0.4451,
"step": 3925
},
{
"epoch": 1.59,
"learning_rate": 8.449473684210527e-06,
"loss": 0.4103,
"step": 3950
},
{
"epoch": 1.6,
"learning_rate": 8.436315789473684e-06,
"loss": 0.3975,
"step": 3975
},
{
"epoch": 1.61,
"learning_rate": 8.423157894736843e-06,
"loss": 0.4097,
"step": 4000
},
{
"epoch": 1.61,
"eval_loss": 0.4367630183696747,
"eval_runtime": 2455.6092,
"eval_samples_per_second": 4.309,
"eval_steps_per_second": 0.27,
"eval_wer": 83.44358344358345,
"step": 4000
},
{
"epoch": 1.62,
"learning_rate": 8.41e-06,
"loss": 0.4376,
"step": 4025
},
{
"epoch": 1.63,
"learning_rate": 8.396842105263158e-06,
"loss": 0.4917,
"step": 4050
},
{
"epoch": 1.64,
"learning_rate": 8.383684210526316e-06,
"loss": 0.4294,
"step": 4075
},
{
"epoch": 1.65,
"learning_rate": 8.370526315789475e-06,
"loss": 0.5441,
"step": 4100
},
{
"epoch": 1.66,
"learning_rate": 8.357368421052632e-06,
"loss": 0.4649,
"step": 4125
},
{
"epoch": 1.67,
"learning_rate": 8.34421052631579e-06,
"loss": 0.4271,
"step": 4150
},
{
"epoch": 1.68,
"learning_rate": 8.331052631578947e-06,
"loss": 0.4814,
"step": 4175
},
{
"epoch": 1.69,
"learning_rate": 8.317894736842107e-06,
"loss": 0.4261,
"step": 4200
},
{
"epoch": 1.71,
"learning_rate": 8.304736842105264e-06,
"loss": 0.4557,
"step": 4225
},
{
"epoch": 1.72,
"learning_rate": 8.291578947368422e-06,
"loss": 0.4564,
"step": 4250
},
{
"epoch": 1.73,
"learning_rate": 8.27842105263158e-06,
"loss": 0.3951,
"step": 4275
},
{
"epoch": 1.74,
"learning_rate": 8.265263157894737e-06,
"loss": 0.4213,
"step": 4300
},
{
"epoch": 1.75,
"learning_rate": 8.252105263157896e-06,
"loss": 0.4574,
"step": 4325
},
{
"epoch": 1.76,
"learning_rate": 8.238947368421053e-06,
"loss": 0.5026,
"step": 4350
},
{
"epoch": 1.77,
"learning_rate": 8.22578947368421e-06,
"loss": 0.4842,
"step": 4375
},
{
"epoch": 1.78,
"learning_rate": 8.212631578947368e-06,
"loss": 0.4095,
"step": 4400
},
{
"epoch": 1.79,
"learning_rate": 8.199473684210527e-06,
"loss": 0.3993,
"step": 4425
},
{
"epoch": 1.8,
"learning_rate": 8.186315789473685e-06,
"loss": 0.4402,
"step": 4450
},
{
"epoch": 1.81,
"learning_rate": 8.173157894736842e-06,
"loss": 0.3791,
"step": 4475
},
{
"epoch": 1.82,
"learning_rate": 8.16e-06,
"loss": 0.4047,
"step": 4500
},
{
"epoch": 1.83,
"learning_rate": 8.146842105263159e-06,
"loss": 0.4431,
"step": 4525
},
{
"epoch": 1.84,
"learning_rate": 8.133684210526316e-06,
"loss": 0.4196,
"step": 4550
},
{
"epoch": 1.85,
"learning_rate": 8.120526315789474e-06,
"loss": 0.3802,
"step": 4575
},
{
"epoch": 1.86,
"learning_rate": 8.107368421052633e-06,
"loss": 0.4199,
"step": 4600
},
{
"epoch": 1.87,
"learning_rate": 8.09421052631579e-06,
"loss": 0.4356,
"step": 4625
},
{
"epoch": 1.88,
"learning_rate": 8.081052631578948e-06,
"loss": 0.4633,
"step": 4650
},
{
"epoch": 1.89,
"learning_rate": 8.067894736842106e-06,
"loss": 0.4335,
"step": 4675
},
{
"epoch": 1.9,
"learning_rate": 8.054736842105265e-06,
"loss": 0.4721,
"step": 4700
},
{
"epoch": 1.91,
"learning_rate": 8.042105263157896e-06,
"loss": 0.4326,
"step": 4725
},
{
"epoch": 1.92,
"learning_rate": 8.028947368421054e-06,
"loss": 0.4377,
"step": 4750
},
{
"epoch": 1.93,
"learning_rate": 8.015789473684211e-06,
"loss": 0.375,
"step": 4775
},
{
"epoch": 1.94,
"learning_rate": 8.002631578947368e-06,
"loss": 0.4189,
"step": 4800
},
{
"epoch": 1.95,
"learning_rate": 7.989473684210528e-06,
"loss": 0.4016,
"step": 4825
},
{
"epoch": 1.96,
"learning_rate": 7.976315789473685e-06,
"loss": 0.4372,
"step": 4850
},
{
"epoch": 1.97,
"learning_rate": 7.963157894736843e-06,
"loss": 0.4668,
"step": 4875
},
{
"epoch": 1.98,
"learning_rate": 7.950000000000002e-06,
"loss": 0.414,
"step": 4900
},
{
"epoch": 1.99,
"learning_rate": 7.936842105263158e-06,
"loss": 0.4023,
"step": 4925
},
{
"epoch": 2.0,
"learning_rate": 7.923684210526317e-06,
"loss": 0.4011,
"step": 4950
},
{
"epoch": 2.01,
"learning_rate": 7.910526315789474e-06,
"loss": 0.3677,
"step": 4975
},
{
"epoch": 2.02,
"learning_rate": 7.897368421052632e-06,
"loss": 0.3262,
"step": 5000
},
{
"epoch": 2.03,
"learning_rate": 7.88421052631579e-06,
"loss": 0.3211,
"step": 5025
},
{
"epoch": 2.04,
"learning_rate": 7.871052631578948e-06,
"loss": 0.3237,
"step": 5050
},
{
"epoch": 2.05,
"learning_rate": 7.857894736842106e-06,
"loss": 0.3344,
"step": 5075
},
{
"epoch": 2.06,
"learning_rate": 7.844736842105263e-06,
"loss": 0.304,
"step": 5100
},
{
"epoch": 2.07,
"learning_rate": 7.831578947368421e-06,
"loss": 0.3276,
"step": 5125
},
{
"epoch": 2.08,
"learning_rate": 7.81842105263158e-06,
"loss": 0.3361,
"step": 5150
},
{
"epoch": 2.09,
"learning_rate": 7.805263157894738e-06,
"loss": 0.3259,
"step": 5175
},
{
"epoch": 2.1,
"learning_rate": 7.792105263157895e-06,
"loss": 0.373,
"step": 5200
},
{
"epoch": 2.11,
"learning_rate": 7.778947368421054e-06,
"loss": 0.3513,
"step": 5225
},
{
"epoch": 2.12,
"learning_rate": 7.765789473684212e-06,
"loss": 0.3154,
"step": 5250
},
{
"epoch": 2.13,
"learning_rate": 7.75263157894737e-06,
"loss": 0.3178,
"step": 5275
},
{
"epoch": 2.14,
"learning_rate": 7.739473684210527e-06,
"loss": 0.3195,
"step": 5300
},
{
"epoch": 2.15,
"learning_rate": 7.726315789473686e-06,
"loss": 0.4051,
"step": 5325
},
{
"epoch": 2.16,
"learning_rate": 7.713157894736842e-06,
"loss": 0.3527,
"step": 5350
},
{
"epoch": 2.17,
"learning_rate": 7.7e-06,
"loss": 0.2731,
"step": 5375
},
{
"epoch": 2.18,
"learning_rate": 7.686842105263158e-06,
"loss": 0.3218,
"step": 5400
},
{
"epoch": 2.19,
"learning_rate": 7.673684210526316e-06,
"loss": 0.3277,
"step": 5425
},
{
"epoch": 2.2,
"learning_rate": 7.660526315789473e-06,
"loss": 0.3669,
"step": 5450
},
{
"epoch": 2.21,
"learning_rate": 7.647368421052632e-06,
"loss": 0.3595,
"step": 5475
},
{
"epoch": 2.22,
"learning_rate": 7.63421052631579e-06,
"loss": 0.3525,
"step": 5500
},
{
"epoch": 2.23,
"learning_rate": 7.621052631578948e-06,
"loss": 0.3428,
"step": 5525
},
{
"epoch": 2.24,
"learning_rate": 7.607894736842107e-06,
"loss": 0.361,
"step": 5550
},
{
"epoch": 2.25,
"learning_rate": 7.594736842105263e-06,
"loss": 0.3812,
"step": 5575
},
{
"epoch": 2.26,
"learning_rate": 7.581578947368422e-06,
"loss": 0.3576,
"step": 5600
},
{
"epoch": 2.27,
"learning_rate": 7.568421052631579e-06,
"loss": 0.3477,
"step": 5625
},
{
"epoch": 2.28,
"learning_rate": 7.555263157894737e-06,
"loss": 0.3456,
"step": 5650
},
{
"epoch": 2.29,
"learning_rate": 7.542105263157895e-06,
"loss": 0.407,
"step": 5675
},
{
"epoch": 2.3,
"learning_rate": 7.528947368421053e-06,
"loss": 0.3203,
"step": 5700
},
{
"epoch": 2.31,
"learning_rate": 7.515789473684211e-06,
"loss": 0.3543,
"step": 5725
},
{
"epoch": 2.32,
"learning_rate": 7.502631578947369e-06,
"loss": 0.3663,
"step": 5750
},
{
"epoch": 2.33,
"learning_rate": 7.4894736842105265e-06,
"loss": 0.3229,
"step": 5775
},
{
"epoch": 2.34,
"learning_rate": 7.476315789473685e-06,
"loss": 0.3527,
"step": 5800
},
{
"epoch": 2.35,
"learning_rate": 7.463157894736843e-06,
"loss": 0.3663,
"step": 5825
},
{
"epoch": 2.36,
"learning_rate": 7.450000000000001e-06,
"loss": 0.3202,
"step": 5850
},
{
"epoch": 2.37,
"learning_rate": 7.436842105263159e-06,
"loss": 0.3239,
"step": 5875
},
{
"epoch": 2.38,
"learning_rate": 7.4236842105263165e-06,
"loss": 0.3481,
"step": 5900
},
{
"epoch": 2.39,
"learning_rate": 7.410526315789475e-06,
"loss": 0.4125,
"step": 5925
},
{
"epoch": 2.4,
"learning_rate": 7.397368421052632e-06,
"loss": 0.3412,
"step": 5950
},
{
"epoch": 2.41,
"learning_rate": 7.384210526315791e-06,
"loss": 0.3212,
"step": 5975
},
{
"epoch": 2.42,
"learning_rate": 7.371052631578947e-06,
"loss": 0.4217,
"step": 6000
},
{
"epoch": 2.42,
"eval_loss": 0.4236811399459839,
"eval_runtime": 2468.8217,
"eval_samples_per_second": 4.286,
"eval_steps_per_second": 0.268,
"eval_wer": 81.98828198828198,
"step": 6000
},
{
"epoch": 2.43,
"learning_rate": 7.3578947368421065e-06,
"loss": 0.3245,
"step": 6025
},
{
"epoch": 2.44,
"learning_rate": 7.344736842105263e-06,
"loss": 0.3201,
"step": 6050
},
{
"epoch": 2.45,
"learning_rate": 7.3315789473684214e-06,
"loss": 0.3448,
"step": 6075
},
{
"epoch": 2.46,
"learning_rate": 7.318421052631579e-06,
"loss": 0.3403,
"step": 6100
},
{
"epoch": 2.47,
"learning_rate": 7.305263157894737e-06,
"loss": 0.3708,
"step": 6125
},
{
"epoch": 2.48,
"learning_rate": 7.292105263157895e-06,
"loss": 0.325,
"step": 6150
},
{
"epoch": 2.49,
"learning_rate": 7.278947368421053e-06,
"loss": 0.3329,
"step": 6175
},
{
"epoch": 2.5,
"learning_rate": 7.265789473684211e-06,
"loss": 0.3146,
"step": 6200
},
{
"epoch": 2.51,
"learning_rate": 7.252631578947369e-06,
"loss": 0.3314,
"step": 6225
},
{
"epoch": 2.52,
"learning_rate": 7.239473684210527e-06,
"loss": 0.3497,
"step": 6250
},
{
"epoch": 2.53,
"learning_rate": 7.226315789473685e-06,
"loss": 0.3507,
"step": 6275
},
{
"epoch": 2.54,
"learning_rate": 7.213157894736843e-06,
"loss": 0.3504,
"step": 6300
},
{
"epoch": 2.55,
"learning_rate": 7.2000000000000005e-06,
"loss": 0.3633,
"step": 6325
},
{
"epoch": 2.56,
"learning_rate": 7.186842105263159e-06,
"loss": 0.3498,
"step": 6350
},
{
"epoch": 2.57,
"learning_rate": 7.17421052631579e-06,
"loss": 0.33,
"step": 6375
},
{
"epoch": 2.58,
"learning_rate": 7.161052631578948e-06,
"loss": 0.2975,
"step": 6400
},
{
"epoch": 2.59,
"learning_rate": 7.147894736842106e-06,
"loss": 0.3353,
"step": 6425
},
{
"epoch": 2.6,
"learning_rate": 7.134736842105264e-06,
"loss": 0.3518,
"step": 6450
},
{
"epoch": 2.61,
"learning_rate": 7.121578947368422e-06,
"loss": 0.3593,
"step": 6475
},
{
"epoch": 2.62,
"learning_rate": 7.10842105263158e-06,
"loss": 0.345,
"step": 6500
},
{
"epoch": 2.63,
"learning_rate": 7.0952631578947376e-06,
"loss": 0.303,
"step": 6525
},
{
"epoch": 2.64,
"learning_rate": 7.082105263157896e-06,
"loss": 0.3235,
"step": 6550
},
{
"epoch": 2.65,
"learning_rate": 7.068947368421053e-06,
"loss": 0.312,
"step": 6575
},
{
"epoch": 2.66,
"learning_rate": 7.055789473684212e-06,
"loss": 0.3263,
"step": 6600
},
{
"epoch": 2.67,
"learning_rate": 7.042631578947368e-06,
"loss": 0.3387,
"step": 6625
},
{
"epoch": 2.68,
"learning_rate": 7.0294736842105275e-06,
"loss": 0.3045,
"step": 6650
},
{
"epoch": 2.69,
"learning_rate": 7.016315789473684e-06,
"loss": 0.3826,
"step": 6675
},
{
"epoch": 2.7,
"learning_rate": 7.0031578947368425e-06,
"loss": 0.3605,
"step": 6700
},
{
"epoch": 2.71,
"learning_rate": 6.99e-06,
"loss": 0.3156,
"step": 6725
},
{
"epoch": 2.72,
"learning_rate": 6.976842105263158e-06,
"loss": 0.3727,
"step": 6750
},
{
"epoch": 2.73,
"learning_rate": 6.963684210526316e-06,
"loss": 0.3281,
"step": 6775
},
{
"epoch": 2.74,
"learning_rate": 6.950526315789474e-06,
"loss": 0.3222,
"step": 6800
},
{
"epoch": 2.75,
"learning_rate": 6.9373684210526325e-06,
"loss": 0.4355,
"step": 6825
},
{
"epoch": 2.76,
"learning_rate": 6.92421052631579e-06,
"loss": 0.3451,
"step": 6850
},
{
"epoch": 2.77,
"learning_rate": 6.911052631578948e-06,
"loss": 0.3344,
"step": 6875
},
{
"epoch": 2.78,
"learning_rate": 6.897894736842106e-06,
"loss": 0.3711,
"step": 6900
},
{
"epoch": 2.79,
"learning_rate": 6.884736842105264e-06,
"loss": 0.3621,
"step": 6925
},
{
"epoch": 2.8,
"learning_rate": 6.871578947368422e-06,
"loss": 0.3344,
"step": 6950
},
{
"epoch": 2.81,
"learning_rate": 6.85842105263158e-06,
"loss": 0.2669,
"step": 6975
},
{
"epoch": 2.82,
"learning_rate": 6.845263157894737e-06,
"loss": 0.3905,
"step": 7000
},
{
"epoch": 2.83,
"learning_rate": 6.832105263157896e-06,
"loss": 0.3882,
"step": 7025
},
{
"epoch": 2.85,
"learning_rate": 6.818947368421052e-06,
"loss": 0.3703,
"step": 7050
},
{
"epoch": 2.86,
"learning_rate": 6.8057894736842116e-06,
"loss": 0.3917,
"step": 7075
},
{
"epoch": 2.87,
"learning_rate": 6.792631578947368e-06,
"loss": 0.3513,
"step": 7100
},
{
"epoch": 2.88,
"learning_rate": 6.7794736842105265e-06,
"loss": 0.3085,
"step": 7125
},
{
"epoch": 2.89,
"learning_rate": 6.766315789473685e-06,
"loss": 0.401,
"step": 7150
},
{
"epoch": 2.9,
"learning_rate": 6.753157894736842e-06,
"loss": 0.3057,
"step": 7175
},
{
"epoch": 2.91,
"learning_rate": 6.740000000000001e-06,
"loss": 0.3422,
"step": 7200
},
{
"epoch": 2.92,
"learning_rate": 6.726842105263158e-06,
"loss": 0.2756,
"step": 7225
},
{
"epoch": 2.93,
"learning_rate": 6.7136842105263165e-06,
"loss": 0.3096,
"step": 7250
},
{
"epoch": 2.94,
"learning_rate": 6.700526315789474e-06,
"loss": 0.316,
"step": 7275
},
{
"epoch": 2.95,
"learning_rate": 6.687368421052632e-06,
"loss": 0.3057,
"step": 7300
},
{
"epoch": 2.96,
"learning_rate": 6.67421052631579e-06,
"loss": 0.3559,
"step": 7325
},
{
"epoch": 2.97,
"learning_rate": 6.661052631578948e-06,
"loss": 0.3157,
"step": 7350
},
{
"epoch": 2.98,
"learning_rate": 6.647894736842106e-06,
"loss": 0.3663,
"step": 7375
},
{
"epoch": 2.99,
"learning_rate": 6.634736842105264e-06,
"loss": 0.3204,
"step": 7400
},
{
"epoch": 3.0,
"learning_rate": 6.6215789473684215e-06,
"loss": 0.329,
"step": 7425
},
{
"epoch": 3.01,
"learning_rate": 6.60842105263158e-06,
"loss": 0.2707,
"step": 7450
},
{
"epoch": 3.02,
"learning_rate": 6.595263157894738e-06,
"loss": 0.2595,
"step": 7475
},
{
"epoch": 3.03,
"learning_rate": 6.582105263157896e-06,
"loss": 0.2542,
"step": 7500
},
{
"epoch": 3.04,
"learning_rate": 6.568947368421054e-06,
"loss": 0.3032,
"step": 7525
},
{
"epoch": 3.05,
"learning_rate": 6.5557894736842106e-06,
"loss": 0.2394,
"step": 7550
},
{
"epoch": 3.06,
"learning_rate": 6.54263157894737e-06,
"loss": 0.2639,
"step": 7575
},
{
"epoch": 3.07,
"learning_rate": 6.529473684210526e-06,
"loss": 0.267,
"step": 7600
},
{
"epoch": 3.08,
"learning_rate": 6.516315789473685e-06,
"loss": 0.2483,
"step": 7625
},
{
"epoch": 3.09,
"learning_rate": 6.503157894736842e-06,
"loss": 0.2586,
"step": 7650
},
{
"epoch": 3.1,
"learning_rate": 6.4900000000000005e-06,
"loss": 0.2974,
"step": 7675
},
{
"epoch": 3.11,
"learning_rate": 6.476842105263158e-06,
"loss": 0.2334,
"step": 7700
},
{
"epoch": 3.12,
"learning_rate": 6.463684210526316e-06,
"loss": 0.2346,
"step": 7725
},
{
"epoch": 3.13,
"learning_rate": 6.450526315789474e-06,
"loss": 0.2362,
"step": 7750
},
{
"epoch": 3.14,
"learning_rate": 6.437368421052632e-06,
"loss": 0.2634,
"step": 7775
},
{
"epoch": 3.15,
"learning_rate": 6.42421052631579e-06,
"loss": 0.2799,
"step": 7800
},
{
"epoch": 3.16,
"learning_rate": 6.411052631578948e-06,
"loss": 0.2796,
"step": 7825
},
{
"epoch": 3.17,
"learning_rate": 6.397894736842106e-06,
"loss": 0.2421,
"step": 7850
},
{
"epoch": 3.18,
"learning_rate": 6.384736842105264e-06,
"loss": 0.2624,
"step": 7875
},
{
"epoch": 3.19,
"learning_rate": 6.371578947368422e-06,
"loss": 0.2594,
"step": 7900
},
{
"epoch": 3.2,
"learning_rate": 6.35842105263158e-06,
"loss": 0.234,
"step": 7925
},
{
"epoch": 3.21,
"learning_rate": 6.345263157894738e-06,
"loss": 0.3122,
"step": 7950
},
{
"epoch": 3.22,
"learning_rate": 6.332105263157895e-06,
"loss": 0.3375,
"step": 7975
},
{
"epoch": 3.23,
"learning_rate": 6.318947368421054e-06,
"loss": 0.3233,
"step": 8000
},
{
"epoch": 3.23,
"eval_loss": 0.4236887991428375,
"eval_runtime": 2462.9792,
"eval_samples_per_second": 4.296,
"eval_steps_per_second": 0.269,
"eval_wer": 81.42128142128142,
"step": 8000
},
{
"epoch": 3.24,
"learning_rate": 6.3057894736842104e-06,
"loss": 0.3036,
"step": 8025
},
{
"epoch": 3.25,
"learning_rate": 6.292631578947369e-06,
"loss": 0.2769,
"step": 8050
},
{
"epoch": 3.26,
"learning_rate": 6.279473684210526e-06,
"loss": 0.2669,
"step": 8075
},
{
"epoch": 3.27,
"learning_rate": 6.266315789473685e-06,
"loss": 0.2534,
"step": 8100
},
{
"epoch": 3.28,
"learning_rate": 6.253157894736842e-06,
"loss": 0.2368,
"step": 8125
},
{
"epoch": 3.29,
"learning_rate": 6.24e-06,
"loss": 0.2883,
"step": 8150
},
{
"epoch": 3.3,
"learning_rate": 6.226842105263159e-06,
"loss": 0.2737,
"step": 8175
},
{
"epoch": 3.31,
"learning_rate": 6.213684210526316e-06,
"loss": 0.2353,
"step": 8200
},
{
"epoch": 3.32,
"learning_rate": 6.2005263157894745e-06,
"loss": 0.3043,
"step": 8225
},
{
"epoch": 3.33,
"learning_rate": 6.187368421052632e-06,
"loss": 0.2836,
"step": 8250
},
{
"epoch": 3.34,
"learning_rate": 6.17421052631579e-06,
"loss": 0.2994,
"step": 8275
},
{
"epoch": 3.35,
"learning_rate": 6.161052631578948e-06,
"loss": 0.242,
"step": 8300
},
{
"epoch": 3.36,
"learning_rate": 6.147894736842106e-06,
"loss": 0.2759,
"step": 8325
},
{
"epoch": 3.37,
"learning_rate": 6.134736842105264e-06,
"loss": 0.2133,
"step": 8350
},
{
"epoch": 3.38,
"learning_rate": 6.121578947368422e-06,
"loss": 0.2784,
"step": 8375
},
{
"epoch": 3.39,
"learning_rate": 6.108421052631579e-06,
"loss": 0.2288,
"step": 8400
},
{
"epoch": 3.4,
"learning_rate": 6.095263157894738e-06,
"loss": 0.2712,
"step": 8425
},
{
"epoch": 3.41,
"learning_rate": 6.0821052631578945e-06,
"loss": 0.2372,
"step": 8450
},
{
"epoch": 3.42,
"learning_rate": 6.068947368421053e-06,
"loss": 0.2514,
"step": 8475
},
{
"epoch": 3.43,
"learning_rate": 6.055789473684212e-06,
"loss": 0.2825,
"step": 8500
},
{
"epoch": 3.44,
"learning_rate": 6.042631578947369e-06,
"loss": 0.2537,
"step": 8525
},
{
"epoch": 3.45,
"learning_rate": 6.029473684210527e-06,
"loss": 0.2464,
"step": 8550
},
{
"epoch": 3.46,
"learning_rate": 6.0163157894736844e-06,
"loss": 0.2583,
"step": 8575
},
{
"epoch": 3.47,
"learning_rate": 6.003157894736843e-06,
"loss": 0.2925,
"step": 8600
},
{
"epoch": 3.48,
"learning_rate": 5.99e-06,
"loss": 0.2858,
"step": 8625
},
{
"epoch": 3.49,
"learning_rate": 5.976842105263159e-06,
"loss": 0.2566,
"step": 8650
},
{
"epoch": 3.5,
"learning_rate": 5.963684210526316e-06,
"loss": 0.2389,
"step": 8675
},
{
"epoch": 3.51,
"learning_rate": 5.950526315789474e-06,
"loss": 0.3029,
"step": 8700
},
{
"epoch": 3.52,
"learning_rate": 5.937368421052632e-06,
"loss": 0.2667,
"step": 8725
},
{
"epoch": 3.53,
"learning_rate": 5.92421052631579e-06,
"loss": 0.2945,
"step": 8750
},
{
"epoch": 3.54,
"learning_rate": 5.911052631578948e-06,
"loss": 0.2727,
"step": 8775
},
{
"epoch": 3.55,
"learning_rate": 5.897894736842106e-06,
"loss": 0.2834,
"step": 8800
},
{
"epoch": 3.56,
"learning_rate": 5.884736842105264e-06,
"loss": 0.3156,
"step": 8825
},
{
"epoch": 3.57,
"learning_rate": 5.871578947368422e-06,
"loss": 0.2616,
"step": 8850
},
{
"epoch": 3.58,
"learning_rate": 5.85842105263158e-06,
"loss": 0.3207,
"step": 8875
},
{
"epoch": 3.59,
"learning_rate": 5.845263157894737e-06,
"loss": 0.2761,
"step": 8900
},
{
"epoch": 3.6,
"learning_rate": 5.832105263157896e-06,
"loss": 0.2306,
"step": 8925
},
{
"epoch": 3.61,
"learning_rate": 5.818947368421053e-06,
"loss": 0.2425,
"step": 8950
},
{
"epoch": 3.62,
"learning_rate": 5.805789473684211e-06,
"loss": 0.2644,
"step": 8975
},
{
"epoch": 3.63,
"learning_rate": 5.7926315789473685e-06,
"loss": 0.2573,
"step": 9000
},
{
"epoch": 3.64,
"learning_rate": 5.779473684210527e-06,
"loss": 0.3367,
"step": 9025
},
{
"epoch": 3.65,
"learning_rate": 5.766315789473684e-06,
"loss": 0.2718,
"step": 9050
},
{
"epoch": 3.66,
"learning_rate": 5.753157894736843e-06,
"loss": 0.2629,
"step": 9075
},
{
"epoch": 3.67,
"learning_rate": 5.74e-06,
"loss": 0.2871,
"step": 9100
},
{
"epoch": 3.68,
"learning_rate": 5.7268421052631584e-06,
"loss": 0.2655,
"step": 9125
},
{
"epoch": 3.69,
"learning_rate": 5.713684210526316e-06,
"loss": 0.2442,
"step": 9150
},
{
"epoch": 3.7,
"learning_rate": 5.700526315789474e-06,
"loss": 0.3244,
"step": 9175
},
{
"epoch": 3.71,
"learning_rate": 5.687368421052633e-06,
"loss": 0.2898,
"step": 9200
},
{
"epoch": 3.72,
"learning_rate": 5.67421052631579e-06,
"loss": 0.3365,
"step": 9225
},
{
"epoch": 3.73,
"learning_rate": 5.661052631578948e-06,
"loss": 0.2929,
"step": 9250
},
{
"epoch": 3.74,
"learning_rate": 5.64842105263158e-06,
"loss": 0.2152,
"step": 9275
},
{
"epoch": 3.75,
"learning_rate": 5.635263157894737e-06,
"loss": 0.3137,
"step": 9300
},
{
"epoch": 3.76,
"learning_rate": 5.6221052631578955e-06,
"loss": 0.2491,
"step": 9325
},
{
"epoch": 3.77,
"learning_rate": 5.608947368421053e-06,
"loss": 0.2962,
"step": 9350
},
{
"epoch": 3.78,
"learning_rate": 5.595789473684211e-06,
"loss": 0.2777,
"step": 9375
},
{
"epoch": 3.79,
"learning_rate": 5.582631578947369e-06,
"loss": 0.2967,
"step": 9400
},
{
"epoch": 3.8,
"learning_rate": 5.569473684210527e-06,
"loss": 0.2816,
"step": 9425
},
{
"epoch": 3.81,
"learning_rate": 5.5563157894736854e-06,
"loss": 0.2792,
"step": 9450
},
{
"epoch": 3.82,
"learning_rate": 5.543157894736843e-06,
"loss": 0.3122,
"step": 9475
},
{
"epoch": 3.83,
"learning_rate": 5.530000000000001e-06,
"loss": 0.2693,
"step": 9500
},
{
"epoch": 3.84,
"learning_rate": 5.516842105263158e-06,
"loss": 0.2775,
"step": 9525
},
{
"epoch": 3.85,
"learning_rate": 5.503684210526316e-06,
"loss": 0.2548,
"step": 9550
},
{
"epoch": 3.86,
"learning_rate": 5.490526315789474e-06,
"loss": 0.2586,
"step": 9575
},
{
"epoch": 3.87,
"learning_rate": 5.477368421052632e-06,
"loss": 0.2463,
"step": 9600
},
{
"epoch": 3.88,
"learning_rate": 5.4642105263157895e-06,
"loss": 0.2656,
"step": 9625
},
{
"epoch": 3.89,
"learning_rate": 5.451052631578948e-06,
"loss": 0.2784,
"step": 9650
},
{
"epoch": 3.9,
"learning_rate": 5.437894736842105e-06,
"loss": 0.2934,
"step": 9675
},
{
"epoch": 3.91,
"learning_rate": 5.424736842105264e-06,
"loss": 0.2493,
"step": 9700
},
{
"epoch": 3.92,
"learning_rate": 5.411578947368421e-06,
"loss": 0.2646,
"step": 9725
},
{
"epoch": 3.93,
"learning_rate": 5.3984210526315795e-06,
"loss": 0.2225,
"step": 9750
},
{
"epoch": 3.94,
"learning_rate": 5.385263157894737e-06,
"loss": 0.2899,
"step": 9775
},
{
"epoch": 3.95,
"learning_rate": 5.372105263157895e-06,
"loss": 0.2976,
"step": 9800
},
{
"epoch": 3.96,
"learning_rate": 5.358947368421054e-06,
"loss": 0.2258,
"step": 9825
},
{
"epoch": 3.97,
"learning_rate": 5.345789473684211e-06,
"loss": 0.268,
"step": 9850
},
{
"epoch": 3.99,
"learning_rate": 5.3326315789473695e-06,
"loss": 0.293,
"step": 9875
},
{
"epoch": 4.0,
"learning_rate": 5.319473684210527e-06,
"loss": 0.2605,
"step": 9900
},
{
"epoch": 4.01,
"learning_rate": 5.306315789473685e-06,
"loss": 0.2278,
"step": 9925
},
{
"epoch": 4.02,
"learning_rate": 5.293157894736842e-06,
"loss": 0.1947,
"step": 9950
},
{
"epoch": 4.03,
"learning_rate": 5.28e-06,
"loss": 0.2423,
"step": 9975
},
{
"epoch": 4.04,
"learning_rate": 5.266842105263158e-06,
"loss": 0.2427,
"step": 10000
},
{
"epoch": 4.04,
"eval_loss": 0.4252447485923767,
"eval_runtime": 2509.0387,
"eval_samples_per_second": 4.217,
"eval_steps_per_second": 0.264,
"eval_wer": 81.2039312039312,
"step": 10000
},
{
"epoch": 4.05,
"learning_rate": 5.253684210526316e-06,
"loss": 0.2614,
"step": 10025
},
{
"epoch": 4.06,
"learning_rate": 5.2405263157894736e-06,
"loss": 0.1787,
"step": 10050
},
{
"epoch": 4.07,
"learning_rate": 5.227368421052632e-06,
"loss": 0.1866,
"step": 10075
},
{
"epoch": 4.08,
"learning_rate": 5.214210526315789e-06,
"loss": 0.1783,
"step": 10100
},
{
"epoch": 4.09,
"learning_rate": 5.201052631578948e-06,
"loss": 0.2145,
"step": 10125
},
{
"epoch": 4.1,
"learning_rate": 5.187894736842106e-06,
"loss": 0.1948,
"step": 10150
},
{
"epoch": 4.11,
"learning_rate": 5.1747368421052635e-06,
"loss": 0.2457,
"step": 10175
},
{
"epoch": 4.12,
"learning_rate": 5.161578947368422e-06,
"loss": 0.2081,
"step": 10200
},
{
"epoch": 4.13,
"learning_rate": 5.148421052631579e-06,
"loss": 0.2211,
"step": 10225
},
{
"epoch": 4.14,
"learning_rate": 5.135263157894738e-06,
"loss": 0.2095,
"step": 10250
},
{
"epoch": 4.15,
"learning_rate": 5.122105263157895e-06,
"loss": 0.2311,
"step": 10275
},
{
"epoch": 4.16,
"learning_rate": 5.1089473684210535e-06,
"loss": 0.2148,
"step": 10300
},
{
"epoch": 4.17,
"learning_rate": 5.095789473684211e-06,
"loss": 0.1881,
"step": 10325
},
{
"epoch": 4.18,
"learning_rate": 5.082631578947369e-06,
"loss": 0.1991,
"step": 10350
},
{
"epoch": 4.19,
"learning_rate": 5.069473684210526e-06,
"loss": 0.2676,
"step": 10375
},
{
"epoch": 4.2,
"learning_rate": 5.056315789473685e-06,
"loss": 0.1964,
"step": 10400
},
{
"epoch": 4.21,
"learning_rate": 5.043157894736842e-06,
"loss": 0.2195,
"step": 10425
},
{
"epoch": 4.22,
"learning_rate": 5.03e-06,
"loss": 0.1729,
"step": 10450
},
{
"epoch": 4.23,
"learning_rate": 5.0168421052631585e-06,
"loss": 0.22,
"step": 10475
},
{
"epoch": 4.24,
"learning_rate": 5.003684210526316e-06,
"loss": 0.1634,
"step": 10500
},
{
"epoch": 4.25,
"learning_rate": 4.990526315789474e-06,
"loss": 0.2266,
"step": 10525
},
{
"epoch": 4.26,
"learning_rate": 4.977368421052632e-06,
"loss": 0.2059,
"step": 10550
},
{
"epoch": 4.27,
"learning_rate": 4.96421052631579e-06,
"loss": 0.1848,
"step": 10575
},
{
"epoch": 4.28,
"learning_rate": 4.951052631578948e-06,
"loss": 0.2126,
"step": 10600
},
{
"epoch": 4.29,
"learning_rate": 4.937894736842105e-06,
"loss": 0.2196,
"step": 10625
},
{
"epoch": 4.3,
"learning_rate": 4.924736842105263e-06,
"loss": 0.2757,
"step": 10650
},
{
"epoch": 4.31,
"learning_rate": 4.911578947368422e-06,
"loss": 0.2349,
"step": 10675
},
{
"epoch": 4.32,
"learning_rate": 4.898421052631579e-06,
"loss": 0.1847,
"step": 10700
},
{
"epoch": 4.33,
"learning_rate": 4.8852631578947375e-06,
"loss": 0.1931,
"step": 10725
},
{
"epoch": 4.34,
"learning_rate": 4.872105263157895e-06,
"loss": 0.2502,
"step": 10750
},
{
"epoch": 4.35,
"learning_rate": 4.858947368421053e-06,
"loss": 0.2491,
"step": 10775
},
{
"epoch": 4.36,
"learning_rate": 4.845789473684211e-06,
"loss": 0.2052,
"step": 10800
},
{
"epoch": 4.37,
"learning_rate": 4.832631578947369e-06,
"loss": 0.2249,
"step": 10825
},
{
"epoch": 4.38,
"learning_rate": 4.819473684210527e-06,
"loss": 0.2375,
"step": 10850
},
{
"epoch": 4.39,
"learning_rate": 4.806315789473684e-06,
"loss": 0.1814,
"step": 10875
},
{
"epoch": 4.4,
"learning_rate": 4.7931578947368425e-06,
"loss": 0.1998,
"step": 10900
},
{
"epoch": 4.41,
"learning_rate": 4.78e-06,
"loss": 0.2292,
"step": 10925
},
{
"epoch": 4.42,
"learning_rate": 4.766842105263158e-06,
"loss": 0.1699,
"step": 10950
},
{
"epoch": 4.43,
"learning_rate": 4.753684210526316e-06,
"loss": 0.2363,
"step": 10975
},
{
"epoch": 4.44,
"learning_rate": 4.740526315789474e-06,
"loss": 0.2432,
"step": 11000
},
{
"epoch": 4.45,
"learning_rate": 4.7273684210526325e-06,
"loss": 0.2329,
"step": 11025
},
{
"epoch": 4.46,
"learning_rate": 4.71421052631579e-06,
"loss": 0.2425,
"step": 11050
},
{
"epoch": 4.47,
"learning_rate": 4.701052631578948e-06,
"loss": 0.2366,
"step": 11075
},
{
"epoch": 4.48,
"learning_rate": 4.687894736842106e-06,
"loss": 0.1882,
"step": 11100
},
{
"epoch": 4.49,
"learning_rate": 4.674736842105263e-06,
"loss": 0.2082,
"step": 11125
},
{
"epoch": 4.5,
"learning_rate": 4.661578947368422e-06,
"loss": 0.1908,
"step": 11150
},
{
"epoch": 4.51,
"learning_rate": 4.648421052631579e-06,
"loss": 0.2407,
"step": 11175
},
{
"epoch": 4.52,
"learning_rate": 4.635263157894737e-06,
"loss": 0.2344,
"step": 11200
},
{
"epoch": 4.53,
"learning_rate": 4.622105263157895e-06,
"loss": 0.2027,
"step": 11225
},
{
"epoch": 4.54,
"learning_rate": 4.608947368421053e-06,
"loss": 0.2054,
"step": 11250
},
{
"epoch": 4.55,
"learning_rate": 4.595789473684211e-06,
"loss": 0.2328,
"step": 11275
},
{
"epoch": 4.56,
"learning_rate": 4.582631578947368e-06,
"loss": 0.1991,
"step": 11300
},
{
"epoch": 4.57,
"learning_rate": 4.569473684210527e-06,
"loss": 0.2174,
"step": 11325
},
{
"epoch": 4.58,
"learning_rate": 4.556315789473685e-06,
"loss": 0.2299,
"step": 11350
},
{
"epoch": 4.59,
"learning_rate": 4.543157894736842e-06,
"loss": 0.2337,
"step": 11375
},
{
"epoch": 4.6,
"learning_rate": 4.530000000000001e-06,
"loss": 0.2049,
"step": 11400
},
{
"epoch": 4.61,
"learning_rate": 4.516842105263158e-06,
"loss": 0.1766,
"step": 11425
},
{
"epoch": 4.62,
"learning_rate": 4.5036842105263165e-06,
"loss": 0.2309,
"step": 11450
},
{
"epoch": 4.63,
"learning_rate": 4.490526315789474e-06,
"loss": 0.2206,
"step": 11475
},
{
"epoch": 4.64,
"learning_rate": 4.477368421052632e-06,
"loss": 0.2116,
"step": 11500
},
{
"epoch": 4.65,
"learning_rate": 4.46421052631579e-06,
"loss": 0.1833,
"step": 11525
},
{
"epoch": 4.66,
"learning_rate": 4.451052631578947e-06,
"loss": 0.204,
"step": 11550
},
{
"epoch": 4.67,
"learning_rate": 4.437894736842106e-06,
"loss": 0.203,
"step": 11575
},
{
"epoch": 4.68,
"learning_rate": 4.424736842105263e-06,
"loss": 0.2191,
"step": 11600
},
{
"epoch": 4.69,
"learning_rate": 4.4115789473684214e-06,
"loss": 0.238,
"step": 11625
},
{
"epoch": 4.7,
"learning_rate": 4.398421052631579e-06,
"loss": 0.2561,
"step": 11650
},
{
"epoch": 4.71,
"learning_rate": 4.385263157894737e-06,
"loss": 0.2249,
"step": 11675
},
{
"epoch": 4.72,
"learning_rate": 4.372105263157896e-06,
"loss": 0.2355,
"step": 11700
},
{
"epoch": 4.73,
"learning_rate": 4.358947368421053e-06,
"loss": 0.2124,
"step": 11725
},
{
"epoch": 4.74,
"learning_rate": 4.345789473684211e-06,
"loss": 0.2428,
"step": 11750
},
{
"epoch": 4.75,
"learning_rate": 4.332631578947369e-06,
"loss": 0.2097,
"step": 11775
},
{
"epoch": 4.76,
"learning_rate": 4.319473684210526e-06,
"loss": 0.2173,
"step": 11800
},
{
"epoch": 4.77,
"learning_rate": 4.306315789473685e-06,
"loss": 0.1947,
"step": 11825
},
{
"epoch": 4.78,
"learning_rate": 4.293157894736842e-06,
"loss": 0.1949,
"step": 11850
},
{
"epoch": 4.79,
"learning_rate": 4.2805263157894734e-06,
"loss": 0.2461,
"step": 11875
},
{
"epoch": 4.8,
"learning_rate": 4.267368421052632e-06,
"loss": 0.2219,
"step": 11900
},
{
"epoch": 4.81,
"learning_rate": 4.254210526315789e-06,
"loss": 0.2434,
"step": 11925
},
{
"epoch": 4.82,
"learning_rate": 4.241052631578948e-06,
"loss": 0.2148,
"step": 11950
},
{
"epoch": 4.83,
"learning_rate": 4.227894736842106e-06,
"loss": 0.2691,
"step": 11975
},
{
"epoch": 4.84,
"learning_rate": 4.214736842105263e-06,
"loss": 0.2354,
"step": 12000
},
{
"epoch": 4.84,
"eval_loss": 0.428153395652771,
"eval_runtime": 2466.9052,
"eval_samples_per_second": 4.289,
"eval_steps_per_second": 0.268,
"eval_wer": 80.77868077868078,
"step": 12000
},
{
"epoch": 4.85,
"learning_rate": 4.201578947368422e-06,
"loss": 0.2402,
"step": 12025
},
{
"epoch": 4.86,
"learning_rate": 4.188421052631579e-06,
"loss": 0.2179,
"step": 12050
},
{
"epoch": 4.87,
"learning_rate": 4.1752631578947376e-06,
"loss": 0.2082,
"step": 12075
},
{
"epoch": 4.88,
"learning_rate": 4.162105263157895e-06,
"loss": 0.2015,
"step": 12100
},
{
"epoch": 4.89,
"learning_rate": 4.1489473684210525e-06,
"loss": 0.2152,
"step": 12125
},
{
"epoch": 4.9,
"learning_rate": 4.135789473684211e-06,
"loss": 0.2462,
"step": 12150
},
{
"epoch": 4.91,
"learning_rate": 4.122631578947368e-06,
"loss": 0.2099,
"step": 12175
},
{
"epoch": 4.92,
"learning_rate": 4.109473684210527e-06,
"loss": 0.2106,
"step": 12200
},
{
"epoch": 4.93,
"learning_rate": 4.096315789473684e-06,
"loss": 0.1814,
"step": 12225
},
{
"epoch": 4.94,
"learning_rate": 4.0831578947368425e-06,
"loss": 0.2406,
"step": 12250
},
{
"epoch": 4.95,
"learning_rate": 4.07e-06,
"loss": 0.187,
"step": 12275
},
{
"epoch": 4.96,
"learning_rate": 4.056842105263158e-06,
"loss": 0.2205,
"step": 12300
},
{
"epoch": 4.97,
"learning_rate": 4.043684210526317e-06,
"loss": 0.1837,
"step": 12325
},
{
"epoch": 4.98,
"learning_rate": 4.030526315789474e-06,
"loss": 0.2276,
"step": 12350
},
{
"epoch": 4.99,
"learning_rate": 4.017368421052632e-06,
"loss": 0.2214,
"step": 12375
},
{
"epoch": 5.0,
"learning_rate": 4.00421052631579e-06,
"loss": 0.2049,
"step": 12400
},
{
"epoch": 5.01,
"learning_rate": 3.9910526315789475e-06,
"loss": 0.1361,
"step": 12425
},
{
"epoch": 5.02,
"learning_rate": 3.977894736842106e-06,
"loss": 0.1333,
"step": 12450
},
{
"epoch": 5.03,
"learning_rate": 3.964736842105263e-06,
"loss": 0.1165,
"step": 12475
},
{
"epoch": 5.04,
"learning_rate": 3.951578947368422e-06,
"loss": 0.1938,
"step": 12500
},
{
"epoch": 5.05,
"learning_rate": 3.938421052631579e-06,
"loss": 0.1865,
"step": 12525
},
{
"epoch": 5.06,
"learning_rate": 3.9252631578947366e-06,
"loss": 0.1818,
"step": 12550
},
{
"epoch": 5.07,
"learning_rate": 3.912105263157895e-06,
"loss": 0.1339,
"step": 12575
},
{
"epoch": 5.08,
"learning_rate": 3.898947368421052e-06,
"loss": 0.1774,
"step": 12600
},
{
"epoch": 5.09,
"learning_rate": 3.885789473684211e-06,
"loss": 0.1914,
"step": 12625
},
{
"epoch": 5.1,
"learning_rate": 3.872631578947369e-06,
"loss": 0.145,
"step": 12650
},
{
"epoch": 5.12,
"learning_rate": 3.8594736842105265e-06,
"loss": 0.1834,
"step": 12675
},
{
"epoch": 5.13,
"learning_rate": 3.846315789473685e-06,
"loss": 0.192,
"step": 12700
},
{
"epoch": 5.14,
"learning_rate": 3.833157894736842e-06,
"loss": 0.1634,
"step": 12725
},
{
"epoch": 5.15,
"learning_rate": 3.820000000000001e-06,
"loss": 0.1492,
"step": 12750
},
{
"epoch": 5.16,
"learning_rate": 3.806842105263158e-06,
"loss": 0.1767,
"step": 12775
},
{
"epoch": 5.17,
"learning_rate": 3.793684210526316e-06,
"loss": 0.1718,
"step": 12800
},
{
"epoch": 5.18,
"learning_rate": 3.780526315789474e-06,
"loss": 0.1809,
"step": 12825
},
{
"epoch": 5.19,
"learning_rate": 3.767368421052632e-06,
"loss": 0.1596,
"step": 12850
},
{
"epoch": 5.2,
"learning_rate": 3.75421052631579e-06,
"loss": 0.2012,
"step": 12875
},
{
"epoch": 5.21,
"learning_rate": 3.7410526315789473e-06,
"loss": 0.1775,
"step": 12900
},
{
"epoch": 5.22,
"learning_rate": 3.7278947368421052e-06,
"loss": 0.1629,
"step": 12925
},
{
"epoch": 5.23,
"learning_rate": 3.714736842105263e-06,
"loss": 0.1666,
"step": 12950
},
{
"epoch": 5.24,
"learning_rate": 3.7015789473684215e-06,
"loss": 0.1658,
"step": 12975
},
{
"epoch": 5.25,
"learning_rate": 3.6884210526315794e-06,
"loss": 0.1775,
"step": 13000
},
{
"epoch": 5.26,
"learning_rate": 3.6752631578947373e-06,
"loss": 0.165,
"step": 13025
},
{
"epoch": 5.27,
"learning_rate": 3.662105263157895e-06,
"loss": 0.2025,
"step": 13050
},
{
"epoch": 5.28,
"learning_rate": 3.648947368421053e-06,
"loss": 0.1429,
"step": 13075
},
{
"epoch": 5.29,
"learning_rate": 3.635789473684211e-06,
"loss": 0.1881,
"step": 13100
},
{
"epoch": 5.3,
"learning_rate": 3.622631578947369e-06,
"loss": 0.1547,
"step": 13125
},
{
"epoch": 5.31,
"learning_rate": 3.6094736842105264e-06,
"loss": 0.1598,
"step": 13150
},
{
"epoch": 5.32,
"learning_rate": 3.5963157894736843e-06,
"loss": 0.1785,
"step": 13175
},
{
"epoch": 5.33,
"learning_rate": 3.5831578947368422e-06,
"loss": 0.1553,
"step": 13200
},
{
"epoch": 5.34,
"learning_rate": 3.57e-06,
"loss": 0.1592,
"step": 13225
},
{
"epoch": 5.35,
"learning_rate": 3.556842105263158e-06,
"loss": 0.1806,
"step": 13250
},
{
"epoch": 5.36,
"learning_rate": 3.543684210526316e-06,
"loss": 0.1694,
"step": 13275
},
{
"epoch": 5.37,
"learning_rate": 3.5305263157894743e-06,
"loss": 0.195,
"step": 13300
},
{
"epoch": 5.38,
"learning_rate": 3.517368421052632e-06,
"loss": 0.1885,
"step": 13325
},
{
"epoch": 5.39,
"learning_rate": 3.50421052631579e-06,
"loss": 0.1745,
"step": 13350
},
{
"epoch": 5.4,
"learning_rate": 3.491052631578948e-06,
"loss": 0.1266,
"step": 13375
},
{
"epoch": 5.41,
"learning_rate": 3.4778947368421055e-06,
"loss": 0.1676,
"step": 13400
},
{
"epoch": 5.42,
"learning_rate": 3.4647368421052634e-06,
"loss": 0.2647,
"step": 13425
},
{
"epoch": 5.43,
"learning_rate": 3.4515789473684213e-06,
"loss": 0.1765,
"step": 13450
},
{
"epoch": 5.44,
"learning_rate": 3.4384210526315792e-06,
"loss": 0.2035,
"step": 13475
},
{
"epoch": 5.45,
"learning_rate": 3.425263157894737e-06,
"loss": 0.1631,
"step": 13500
},
{
"epoch": 5.46,
"learning_rate": 3.412105263157895e-06,
"loss": 0.1818,
"step": 13525
},
{
"epoch": 5.47,
"learning_rate": 3.398947368421053e-06,
"loss": 0.1654,
"step": 13550
},
{
"epoch": 5.48,
"learning_rate": 3.3857894736842104e-06,
"loss": 0.1794,
"step": 13575
},
{
"epoch": 5.49,
"learning_rate": 3.3726315789473683e-06,
"loss": 0.2104,
"step": 13600
},
{
"epoch": 5.5,
"learning_rate": 3.3594736842105262e-06,
"loss": 0.1636,
"step": 13625
},
{
"epoch": 5.51,
"learning_rate": 3.3463157894736846e-06,
"loss": 0.1844,
"step": 13650
},
{
"epoch": 5.52,
"learning_rate": 3.3331578947368425e-06,
"loss": 0.2168,
"step": 13675
},
{
"epoch": 5.53,
"learning_rate": 3.3200000000000004e-06,
"loss": 0.156,
"step": 13700
},
{
"epoch": 5.54,
"learning_rate": 3.3068421052631583e-06,
"loss": 0.1995,
"step": 13725
},
{
"epoch": 5.55,
"learning_rate": 3.2936842105263162e-06,
"loss": 0.1682,
"step": 13750
},
{
"epoch": 5.56,
"learning_rate": 3.280526315789474e-06,
"loss": 0.1614,
"step": 13775
},
{
"epoch": 5.57,
"learning_rate": 3.267368421052632e-06,
"loss": 0.1796,
"step": 13800
},
{
"epoch": 5.58,
"learning_rate": 3.2542105263157895e-06,
"loss": 0.1699,
"step": 13825
},
{
"epoch": 5.59,
"learning_rate": 3.2410526315789474e-06,
"loss": 0.179,
"step": 13850
},
{
"epoch": 5.6,
"learning_rate": 3.2278947368421053e-06,
"loss": 0.163,
"step": 13875
},
{
"epoch": 5.61,
"learning_rate": 3.2147368421052633e-06,
"loss": 0.1874,
"step": 13900
},
{
"epoch": 5.62,
"learning_rate": 3.201578947368421e-06,
"loss": 0.1712,
"step": 13925
},
{
"epoch": 5.63,
"learning_rate": 3.188421052631579e-06,
"loss": 0.1586,
"step": 13950
},
{
"epoch": 5.64,
"learning_rate": 3.175263157894737e-06,
"loss": 0.1495,
"step": 13975
},
{
"epoch": 5.65,
"learning_rate": 3.1621052631578953e-06,
"loss": 0.2202,
"step": 14000
},
{
"epoch": 5.65,
"eval_loss": 0.4407210648059845,
"eval_runtime": 2507.3297,
"eval_samples_per_second": 4.22,
"eval_steps_per_second": 0.264,
"eval_wer": 81.43073143073143,
"step": 14000
}
],
"max_steps": 20000,
"num_train_epochs": 9,
"total_flos": 1.45250718400512e+19,
"trial_name": null,
"trial_params": null
}