whisper-large-v2-pt / trainer_state.json
jlondonobo's picture
End of training
6cb935c
{
"best_metric": 5.590020342630419,
"best_model_checkpoint": "./checkpoint-5000",
"epoch": 9.0226,
"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": 4.6000000000000004e-07,
"loss": 1.0639,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.400000000000001e-07,
"loss": 0.7438,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.44e-06,
"loss": 0.3249,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.94e-06,
"loss": 0.2418,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.42e-06,
"loss": 0.205,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.92e-06,
"loss": 0.1895,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 3.4200000000000007e-06,
"loss": 0.1833,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 3.920000000000001e-06,
"loss": 0.1505,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.42e-06,
"loss": 0.1774,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 4.92e-06,
"loss": 0.1608,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 5.420000000000001e-06,
"loss": 0.1627,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 5.92e-06,
"loss": 0.1642,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 6.42e-06,
"loss": 0.1766,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 6.92e-06,
"loss": 0.1437,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 7.420000000000001e-06,
"loss": 0.1627,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 7.92e-06,
"loss": 0.1545,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 8.42e-06,
"loss": 0.1397,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 8.920000000000001e-06,
"loss": 0.1523,
"step": 450
},
{
"epoch": 0.1,
"learning_rate": 9.42e-06,
"loss": 0.1657,
"step": 475
},
{
"epoch": 0.1,
"learning_rate": 9.920000000000002e-06,
"loss": 0.1614,
"step": 500
},
{
"epoch": 0.1,
"learning_rate": 9.953333333333333e-06,
"loss": 0.1647,
"step": 525
},
{
"epoch": 1.0,
"learning_rate": 9.89777777777778e-06,
"loss": 0.1549,
"step": 550
},
{
"epoch": 1.01,
"learning_rate": 9.842222222222223e-06,
"loss": 0.1632,
"step": 575
},
{
"epoch": 1.01,
"learning_rate": 9.786666666666667e-06,
"loss": 0.1515,
"step": 600
},
{
"epoch": 1.02,
"learning_rate": 9.731111111111113e-06,
"loss": 0.1428,
"step": 625
},
{
"epoch": 1.02,
"learning_rate": 9.675555555555555e-06,
"loss": 0.1325,
"step": 650
},
{
"epoch": 1.03,
"learning_rate": 9.620000000000001e-06,
"loss": 0.1207,
"step": 675
},
{
"epoch": 1.03,
"learning_rate": 9.564444444444445e-06,
"loss": 0.1298,
"step": 700
},
{
"epoch": 1.04,
"learning_rate": 9.508888888888889e-06,
"loss": 0.1149,
"step": 725
},
{
"epoch": 1.04,
"learning_rate": 9.453333333333335e-06,
"loss": 0.1051,
"step": 750
},
{
"epoch": 1.05,
"learning_rate": 9.397777777777779e-06,
"loss": 0.1174,
"step": 775
},
{
"epoch": 1.05,
"learning_rate": 9.342222222222223e-06,
"loss": 0.1014,
"step": 800
},
{
"epoch": 1.06,
"learning_rate": 9.286666666666667e-06,
"loss": 0.0988,
"step": 825
},
{
"epoch": 1.06,
"learning_rate": 9.231111111111111e-06,
"loss": 0.0986,
"step": 850
},
{
"epoch": 1.07,
"learning_rate": 9.175555555555557e-06,
"loss": 0.0915,
"step": 875
},
{
"epoch": 1.07,
"learning_rate": 9.12e-06,
"loss": 0.0892,
"step": 900
},
{
"epoch": 1.08,
"learning_rate": 9.064444444444447e-06,
"loss": 0.0829,
"step": 925
},
{
"epoch": 1.08,
"learning_rate": 9.008888888888889e-06,
"loss": 0.0872,
"step": 950
},
{
"epoch": 1.09,
"learning_rate": 8.953333333333335e-06,
"loss": 0.0814,
"step": 975
},
{
"epoch": 1.09,
"learning_rate": 8.900000000000001e-06,
"loss": 0.0828,
"step": 1000
},
{
"epoch": 1.09,
"eval_loss": 0.18684269487857819,
"eval_runtime": 8136.8393,
"eval_samples_per_second": 1.068,
"eval_steps_per_second": 0.134,
"eval_wer": 6.7786351944318355,
"step": 1000
},
{
"epoch": 1.1,
"learning_rate": 8.844444444444445e-06,
"loss": 0.0821,
"step": 1025
},
{
"epoch": 1.1,
"learning_rate": 8.788888888888891e-06,
"loss": 0.0785,
"step": 1050
},
{
"epoch": 1.11,
"learning_rate": 8.733333333333333e-06,
"loss": 0.0741,
"step": 1075
},
{
"epoch": 2.0,
"learning_rate": 8.677777777777779e-06,
"loss": 0.0778,
"step": 1100
},
{
"epoch": 2.01,
"learning_rate": 8.622222222222223e-06,
"loss": 0.0809,
"step": 1125
},
{
"epoch": 2.01,
"learning_rate": 8.566666666666667e-06,
"loss": 0.0721,
"step": 1150
},
{
"epoch": 2.02,
"learning_rate": 8.511111111111113e-06,
"loss": 0.0669,
"step": 1175
},
{
"epoch": 2.02,
"learning_rate": 8.455555555555555e-06,
"loss": 0.072,
"step": 1200
},
{
"epoch": 2.03,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0618,
"step": 1225
},
{
"epoch": 2.03,
"learning_rate": 8.344444444444445e-06,
"loss": 0.064,
"step": 1250
},
{
"epoch": 2.04,
"learning_rate": 8.288888888888889e-06,
"loss": 0.058,
"step": 1275
},
{
"epoch": 2.04,
"learning_rate": 8.233333333333335e-06,
"loss": 0.0599,
"step": 1300
},
{
"epoch": 2.05,
"learning_rate": 8.177777777777779e-06,
"loss": 0.0595,
"step": 1325
},
{
"epoch": 2.05,
"learning_rate": 8.122222222222223e-06,
"loss": 0.0578,
"step": 1350
},
{
"epoch": 2.06,
"learning_rate": 8.066666666666667e-06,
"loss": 0.058,
"step": 1375
},
{
"epoch": 2.06,
"learning_rate": 8.011111111111113e-06,
"loss": 0.0558,
"step": 1400
},
{
"epoch": 2.07,
"learning_rate": 7.955555555555557e-06,
"loss": 0.0502,
"step": 1425
},
{
"epoch": 2.07,
"learning_rate": 7.9e-06,
"loss": 0.0509,
"step": 1450
},
{
"epoch": 2.08,
"learning_rate": 7.844444444444446e-06,
"loss": 0.044,
"step": 1475
},
{
"epoch": 2.08,
"learning_rate": 7.788888888888889e-06,
"loss": 0.0454,
"step": 1500
},
{
"epoch": 2.09,
"learning_rate": 7.733333333333334e-06,
"loss": 0.0485,
"step": 1525
},
{
"epoch": 2.09,
"learning_rate": 7.677777777777778e-06,
"loss": 0.047,
"step": 1550
},
{
"epoch": 2.1,
"learning_rate": 7.622222222222223e-06,
"loss": 0.0506,
"step": 1575
},
{
"epoch": 2.1,
"learning_rate": 7.566666666666667e-06,
"loss": 0.0437,
"step": 1600
},
{
"epoch": 2.11,
"learning_rate": 7.511111111111111e-06,
"loss": 0.0394,
"step": 1625
},
{
"epoch": 3.0,
"learning_rate": 7.455555555555556e-06,
"loss": 0.0467,
"step": 1650
},
{
"epoch": 3.01,
"learning_rate": 7.4e-06,
"loss": 0.0379,
"step": 1675
},
{
"epoch": 3.01,
"learning_rate": 7.344444444444445e-06,
"loss": 0.0379,
"step": 1700
},
{
"epoch": 3.02,
"learning_rate": 7.28888888888889e-06,
"loss": 0.037,
"step": 1725
},
{
"epoch": 3.02,
"learning_rate": 7.233333333333334e-06,
"loss": 0.0376,
"step": 1750
},
{
"epoch": 3.03,
"learning_rate": 7.177777777777778e-06,
"loss": 0.0326,
"step": 1775
},
{
"epoch": 3.03,
"learning_rate": 7.122222222222222e-06,
"loss": 0.0389,
"step": 1800
},
{
"epoch": 3.04,
"learning_rate": 7.066666666666667e-06,
"loss": 0.0373,
"step": 1825
},
{
"epoch": 3.04,
"learning_rate": 7.011111111111112e-06,
"loss": 0.0295,
"step": 1850
},
{
"epoch": 3.05,
"learning_rate": 6.955555555555557e-06,
"loss": 0.0334,
"step": 1875
},
{
"epoch": 3.05,
"learning_rate": 6.9e-06,
"loss": 0.0309,
"step": 1900
},
{
"epoch": 3.06,
"learning_rate": 6.844444444444445e-06,
"loss": 0.0306,
"step": 1925
},
{
"epoch": 3.06,
"learning_rate": 6.788888888888889e-06,
"loss": 0.0279,
"step": 1950
},
{
"epoch": 3.07,
"learning_rate": 6.733333333333334e-06,
"loss": 0.0284,
"step": 1975
},
{
"epoch": 3.07,
"learning_rate": 6.677777777777779e-06,
"loss": 0.0241,
"step": 2000
},
{
"epoch": 3.07,
"eval_loss": 0.20566053688526154,
"eval_runtime": 8076.5511,
"eval_samples_per_second": 1.076,
"eval_steps_per_second": 0.135,
"eval_wer": 6.109513962441788,
"step": 2000
},
{
"epoch": 3.08,
"learning_rate": 6.6222222222222236e-06,
"loss": 0.0239,
"step": 2025
},
{
"epoch": 3.08,
"learning_rate": 6.566666666666667e-06,
"loss": 0.0238,
"step": 2050
},
{
"epoch": 3.09,
"learning_rate": 6.511111111111112e-06,
"loss": 0.023,
"step": 2075
},
{
"epoch": 3.09,
"learning_rate": 6.455555555555556e-06,
"loss": 0.0269,
"step": 2100
},
{
"epoch": 3.1,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.0235,
"step": 2125
},
{
"epoch": 3.1,
"learning_rate": 6.3444444444444454e-06,
"loss": 0.0224,
"step": 2150
},
{
"epoch": 4.0,
"learning_rate": 6.28888888888889e-06,
"loss": 0.0182,
"step": 2175
},
{
"epoch": 4.01,
"learning_rate": 6.2333333333333335e-06,
"loss": 0.0194,
"step": 2200
},
{
"epoch": 4.01,
"learning_rate": 6.177777777777778e-06,
"loss": 0.0211,
"step": 2225
},
{
"epoch": 4.02,
"learning_rate": 6.1222222222222224e-06,
"loss": 0.0189,
"step": 2250
},
{
"epoch": 4.02,
"learning_rate": 6.066666666666667e-06,
"loss": 0.0195,
"step": 2275
},
{
"epoch": 4.03,
"learning_rate": 6.011111111111112e-06,
"loss": 0.0207,
"step": 2300
},
{
"epoch": 4.03,
"learning_rate": 5.955555555555555e-06,
"loss": 0.0196,
"step": 2325
},
{
"epoch": 4.04,
"learning_rate": 5.9e-06,
"loss": 0.019,
"step": 2350
},
{
"epoch": 4.04,
"learning_rate": 5.844444444444445e-06,
"loss": 0.0162,
"step": 2375
},
{
"epoch": 4.05,
"learning_rate": 5.788888888888889e-06,
"loss": 0.0172,
"step": 2400
},
{
"epoch": 4.05,
"learning_rate": 5.733333333333334e-06,
"loss": 0.0167,
"step": 2425
},
{
"epoch": 4.06,
"learning_rate": 5.677777777777779e-06,
"loss": 0.0166,
"step": 2450
},
{
"epoch": 4.06,
"learning_rate": 5.622222222222222e-06,
"loss": 0.014,
"step": 2475
},
{
"epoch": 4.07,
"learning_rate": 5.566666666666667e-06,
"loss": 0.0139,
"step": 2500
},
{
"epoch": 4.07,
"learning_rate": 5.511111111111112e-06,
"loss": 0.0154,
"step": 2525
},
{
"epoch": 4.08,
"learning_rate": 5.455555555555556e-06,
"loss": 0.0109,
"step": 2550
},
{
"epoch": 4.08,
"learning_rate": 5.400000000000001e-06,
"loss": 0.0121,
"step": 2575
},
{
"epoch": 4.09,
"learning_rate": 5.344444444444446e-06,
"loss": 0.0115,
"step": 2600
},
{
"epoch": 4.09,
"learning_rate": 5.288888888888889e-06,
"loss": 0.0128,
"step": 2625
},
{
"epoch": 4.1,
"learning_rate": 5.233333333333334e-06,
"loss": 0.0127,
"step": 2650
},
{
"epoch": 4.1,
"learning_rate": 5.177777777777779e-06,
"loss": 0.0114,
"step": 2675
},
{
"epoch": 4.11,
"learning_rate": 5.122222222222223e-06,
"loss": 0.011,
"step": 2700
},
{
"epoch": 5.0,
"learning_rate": 5.0666666666666676e-06,
"loss": 0.0095,
"step": 2725
},
{
"epoch": 5.01,
"learning_rate": 5.011111111111111e-06,
"loss": 0.0098,
"step": 2750
},
{
"epoch": 5.01,
"learning_rate": 4.9555555555555565e-06,
"loss": 0.0105,
"step": 2775
},
{
"epoch": 5.02,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.0078,
"step": 2800
},
{
"epoch": 5.02,
"learning_rate": 4.8444444444444446e-06,
"loss": 0.0126,
"step": 2825
},
{
"epoch": 5.03,
"learning_rate": 4.7888888888888894e-06,
"loss": 0.0088,
"step": 2850
},
{
"epoch": 5.03,
"learning_rate": 4.7333333333333335e-06,
"loss": 0.0088,
"step": 2875
},
{
"epoch": 5.04,
"learning_rate": 4.677777777777778e-06,
"loss": 0.0069,
"step": 2900
},
{
"epoch": 5.04,
"learning_rate": 4.622222222222222e-06,
"loss": 0.012,
"step": 2925
},
{
"epoch": 5.05,
"learning_rate": 4.566666666666667e-06,
"loss": 0.0106,
"step": 2950
},
{
"epoch": 5.05,
"learning_rate": 4.511111111111111e-06,
"loss": 0.0065,
"step": 2975
},
{
"epoch": 5.06,
"learning_rate": 4.455555555555555e-06,
"loss": 0.0084,
"step": 3000
},
{
"epoch": 5.06,
"eval_loss": 0.23669926822185516,
"eval_runtime": 8039.9929,
"eval_samples_per_second": 1.081,
"eval_steps_per_second": 0.135,
"eval_wer": 6.028815924412838,
"step": 3000
},
{
"epoch": 5.06,
"learning_rate": 4.4e-06,
"loss": 0.0068,
"step": 3025
},
{
"epoch": 5.07,
"learning_rate": 4.344444444444445e-06,
"loss": 0.0087,
"step": 3050
},
{
"epoch": 5.07,
"learning_rate": 4.288888888888889e-06,
"loss": 0.0058,
"step": 3075
},
{
"epoch": 5.08,
"learning_rate": 4.233333333333334e-06,
"loss": 0.0053,
"step": 3100
},
{
"epoch": 5.08,
"learning_rate": 4.177777777777778e-06,
"loss": 0.0049,
"step": 3125
},
{
"epoch": 5.09,
"learning_rate": 4.122222222222222e-06,
"loss": 0.0059,
"step": 3150
},
{
"epoch": 5.09,
"learning_rate": 4.066666666666667e-06,
"loss": 0.0068,
"step": 3175
},
{
"epoch": 5.1,
"learning_rate": 4.011111111111111e-06,
"loss": 0.0079,
"step": 3200
},
{
"epoch": 5.1,
"learning_rate": 3.955555555555556e-06,
"loss": 0.0075,
"step": 3225
},
{
"epoch": 5.11,
"learning_rate": 3.900000000000001e-06,
"loss": 0.0056,
"step": 3250
},
{
"epoch": 6.0,
"learning_rate": 3.844444444444445e-06,
"loss": 0.0052,
"step": 3275
},
{
"epoch": 6.01,
"learning_rate": 3.7888888888888893e-06,
"loss": 0.0049,
"step": 3300
},
{
"epoch": 6.01,
"learning_rate": 3.7333333333333337e-06,
"loss": 0.0052,
"step": 3325
},
{
"epoch": 6.02,
"learning_rate": 3.6777777777777778e-06,
"loss": 0.004,
"step": 3350
},
{
"epoch": 6.02,
"learning_rate": 3.6222222222222226e-06,
"loss": 0.0047,
"step": 3375
},
{
"epoch": 6.03,
"learning_rate": 3.566666666666667e-06,
"loss": 0.0044,
"step": 3400
},
{
"epoch": 6.03,
"learning_rate": 3.511111111111111e-06,
"loss": 0.0053,
"step": 3425
},
{
"epoch": 6.04,
"learning_rate": 3.455555555555556e-06,
"loss": 0.0052,
"step": 3450
},
{
"epoch": 6.04,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.0038,
"step": 3475
},
{
"epoch": 6.05,
"learning_rate": 3.3444444444444445e-06,
"loss": 0.0031,
"step": 3500
},
{
"epoch": 6.05,
"learning_rate": 3.2888888888888894e-06,
"loss": 0.0029,
"step": 3525
},
{
"epoch": 6.06,
"learning_rate": 3.2333333333333334e-06,
"loss": 0.004,
"step": 3550
},
{
"epoch": 6.06,
"learning_rate": 3.177777777777778e-06,
"loss": 0.0046,
"step": 3575
},
{
"epoch": 6.07,
"learning_rate": 3.1222222222222228e-06,
"loss": 0.0028,
"step": 3600
},
{
"epoch": 6.07,
"learning_rate": 3.066666666666667e-06,
"loss": 0.0024,
"step": 3625
},
{
"epoch": 6.08,
"learning_rate": 3.0111111111111113e-06,
"loss": 0.0043,
"step": 3650
},
{
"epoch": 6.08,
"learning_rate": 2.955555555555556e-06,
"loss": 0.0021,
"step": 3675
},
{
"epoch": 6.09,
"learning_rate": 2.9e-06,
"loss": 0.003,
"step": 3700
},
{
"epoch": 6.09,
"learning_rate": 2.8444444444444446e-06,
"loss": 0.0028,
"step": 3725
},
{
"epoch": 6.1,
"learning_rate": 2.788888888888889e-06,
"loss": 0.0033,
"step": 3750
},
{
"epoch": 6.1,
"learning_rate": 2.7333333333333336e-06,
"loss": 0.0035,
"step": 3775
},
{
"epoch": 6.11,
"learning_rate": 2.677777777777778e-06,
"loss": 0.0025,
"step": 3800
},
{
"epoch": 7.0,
"learning_rate": 2.6222222222222225e-06,
"loss": 0.0025,
"step": 3825
},
{
"epoch": 7.01,
"learning_rate": 2.566666666666667e-06,
"loss": 0.0019,
"step": 3850
},
{
"epoch": 7.01,
"learning_rate": 2.5111111111111114e-06,
"loss": 0.0021,
"step": 3875
},
{
"epoch": 7.02,
"learning_rate": 2.455555555555556e-06,
"loss": 0.002,
"step": 3900
},
{
"epoch": 7.02,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.0028,
"step": 3925
},
{
"epoch": 7.03,
"learning_rate": 2.3444444444444448e-06,
"loss": 0.0027,
"step": 3950
},
{
"epoch": 7.03,
"learning_rate": 2.2888888888888892e-06,
"loss": 0.0023,
"step": 3975
},
{
"epoch": 7.04,
"learning_rate": 2.2333333333333333e-06,
"loss": 0.0015,
"step": 4000
},
{
"epoch": 7.04,
"eval_loss": 0.24688950181007385,
"eval_runtime": 8026.8581,
"eval_samples_per_second": 1.083,
"eval_steps_per_second": 0.135,
"eval_wer": 5.709386190548242,
"step": 4000
},
{
"epoch": 7.04,
"learning_rate": 2.1777777777777777e-06,
"loss": 0.0017,
"step": 4025
},
{
"epoch": 7.05,
"learning_rate": 2.1244444444444443e-06,
"loss": 0.0017,
"step": 4050
},
{
"epoch": 7.05,
"learning_rate": 2.0688888888888892e-06,
"loss": 0.0017,
"step": 4075
},
{
"epoch": 7.06,
"learning_rate": 2.0133333333333337e-06,
"loss": 0.0013,
"step": 4100
},
{
"epoch": 7.06,
"learning_rate": 1.9577777777777777e-06,
"loss": 0.0015,
"step": 4125
},
{
"epoch": 7.07,
"learning_rate": 1.9022222222222222e-06,
"loss": 0.0011,
"step": 4150
},
{
"epoch": 7.07,
"learning_rate": 1.8466666666666668e-06,
"loss": 0.0021,
"step": 4175
},
{
"epoch": 7.08,
"learning_rate": 1.7911111111111113e-06,
"loss": 0.0014,
"step": 4200
},
{
"epoch": 7.08,
"learning_rate": 1.7355555555555555e-06,
"loss": 0.001,
"step": 4225
},
{
"epoch": 7.09,
"learning_rate": 1.6800000000000002e-06,
"loss": 0.0019,
"step": 4250
},
{
"epoch": 7.09,
"learning_rate": 1.6244444444444447e-06,
"loss": 0.0014,
"step": 4275
},
{
"epoch": 7.1,
"learning_rate": 1.568888888888889e-06,
"loss": 0.0014,
"step": 4300
},
{
"epoch": 7.1,
"learning_rate": 1.5133333333333334e-06,
"loss": 0.0013,
"step": 4325
},
{
"epoch": 8.0,
"learning_rate": 1.457777777777778e-06,
"loss": 0.0016,
"step": 4350
},
{
"epoch": 8.01,
"learning_rate": 1.4022222222222223e-06,
"loss": 0.0011,
"step": 4375
},
{
"epoch": 8.01,
"learning_rate": 1.3466666666666668e-06,
"loss": 0.0011,
"step": 4400
},
{
"epoch": 8.02,
"learning_rate": 1.2911111111111112e-06,
"loss": 0.0015,
"step": 4425
},
{
"epoch": 8.02,
"learning_rate": 1.2355555555555557e-06,
"loss": 0.0014,
"step": 4450
},
{
"epoch": 8.03,
"learning_rate": 1.1800000000000001e-06,
"loss": 0.0013,
"step": 4475
},
{
"epoch": 8.03,
"learning_rate": 1.1244444444444446e-06,
"loss": 0.0011,
"step": 4500
},
{
"epoch": 8.04,
"learning_rate": 1.068888888888889e-06,
"loss": 0.0012,
"step": 4525
},
{
"epoch": 8.04,
"learning_rate": 1.0133333333333333e-06,
"loss": 0.001,
"step": 4550
},
{
"epoch": 8.05,
"learning_rate": 9.57777777777778e-07,
"loss": 0.0013,
"step": 4575
},
{
"epoch": 8.05,
"learning_rate": 9.022222222222222e-07,
"loss": 0.0009,
"step": 4600
},
{
"epoch": 8.06,
"learning_rate": 8.466666666666668e-07,
"loss": 0.0011,
"step": 4625
},
{
"epoch": 8.06,
"learning_rate": 7.911111111111111e-07,
"loss": 0.001,
"step": 4650
},
{
"epoch": 8.07,
"learning_rate": 7.355555555555556e-07,
"loss": 0.002,
"step": 4675
},
{
"epoch": 8.07,
"learning_rate": 6.800000000000001e-07,
"loss": 0.0008,
"step": 4700
},
{
"epoch": 8.08,
"learning_rate": 6.244444444444445e-07,
"loss": 0.001,
"step": 4725
},
{
"epoch": 8.08,
"learning_rate": 5.68888888888889e-07,
"loss": 0.0009,
"step": 4750
},
{
"epoch": 8.09,
"learning_rate": 5.133333333333334e-07,
"loss": 0.0009,
"step": 4775
},
{
"epoch": 8.09,
"learning_rate": 4.5777777777777784e-07,
"loss": 0.001,
"step": 4800
},
{
"epoch": 8.1,
"learning_rate": 4.0222222222222224e-07,
"loss": 0.0009,
"step": 4825
},
{
"epoch": 8.1,
"learning_rate": 3.466666666666667e-07,
"loss": 0.0008,
"step": 4850
},
{
"epoch": 8.11,
"learning_rate": 2.9111111111111116e-07,
"loss": 0.0009,
"step": 4875
},
{
"epoch": 9.0,
"learning_rate": 2.3555555555555556e-07,
"loss": 0.0009,
"step": 4900
},
{
"epoch": 9.01,
"learning_rate": 1.8e-07,
"loss": 0.0009,
"step": 4925
},
{
"epoch": 9.01,
"learning_rate": 1.2444444444444446e-07,
"loss": 0.0009,
"step": 4950
},
{
"epoch": 9.02,
"learning_rate": 6.888888888888889e-08,
"loss": 0.0009,
"step": 4975
},
{
"epoch": 9.02,
"learning_rate": 1.3333333333333334e-08,
"loss": 0.0009,
"step": 5000
},
{
"epoch": 9.02,
"eval_loss": 0.2821445167064667,
"eval_runtime": 8017.2735,
"eval_samples_per_second": 1.084,
"eval_steps_per_second": 0.136,
"eval_wer": 5.590020342630419,
"step": 5000
},
{
"epoch": 9.02,
"step": 5000,
"total_flos": 3.39707363328e+20,
"train_loss": 0.050828523495793344,
"train_runtime": 87374.8588,
"train_samples_per_second": 1.831,
"train_steps_per_second": 0.057
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 3.39707363328e+20,
"trial_name": null,
"trial_params": null
}