whisper-small-es / trainer_state.json
zuazo's picture
End of training
ee69829
raw
history blame
25.8 kB
{
"best_metric": 8.266774443952604,
"best_model_checkpoint": "./checkpoint-3000",
"epoch": 10.018,
"eval_steps": 1000,
"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-07,
"loss": 1.1567,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.400000000000001e-07,
"loss": 0.9708,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.44e-06,
"loss": 0.6529,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.94e-06,
"loss": 0.3622,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.4400000000000004e-06,
"loss": 0.2945,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.9400000000000002e-06,
"loss": 0.2799,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 3.44e-06,
"loss": 0.2632,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 3.94e-06,
"loss": 0.2575,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.440000000000001e-06,
"loss": 0.2707,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 4.94e-06,
"loss": 0.3628,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 5.4400000000000004e-06,
"loss": 0.3489,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 5.94e-06,
"loss": 0.3334,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 6.440000000000001e-06,
"loss": 0.2956,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 6.9400000000000005e-06,
"loss": 0.3098,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 7.440000000000001e-06,
"loss": 0.3203,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 7.94e-06,
"loss": 0.2852,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 8.44e-06,
"loss": 0.2591,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 8.94e-06,
"loss": 0.191,
"step": 450
},
{
"epoch": 0.1,
"learning_rate": 9.440000000000001e-06,
"loss": 0.1919,
"step": 475
},
{
"epoch": 1.0,
"learning_rate": 9.940000000000001e-06,
"loss": 0.2119,
"step": 500
},
{
"epoch": 1.01,
"learning_rate": 9.951111111111111e-06,
"loss": 0.2157,
"step": 525
},
{
"epoch": 1.01,
"learning_rate": 9.895555555555557e-06,
"loss": 0.2569,
"step": 550
},
{
"epoch": 1.02,
"learning_rate": 9.84e-06,
"loss": 0.1907,
"step": 575
},
{
"epoch": 1.02,
"learning_rate": 9.784444444444445e-06,
"loss": 0.1924,
"step": 600
},
{
"epoch": 1.03,
"learning_rate": 9.72888888888889e-06,
"loss": 0.1902,
"step": 625
},
{
"epoch": 1.03,
"learning_rate": 9.673333333333334e-06,
"loss": 0.1669,
"step": 650
},
{
"epoch": 1.04,
"learning_rate": 9.617777777777778e-06,
"loss": 0.1743,
"step": 675
},
{
"epoch": 1.04,
"learning_rate": 9.562222222222223e-06,
"loss": 0.1697,
"step": 700
},
{
"epoch": 1.05,
"learning_rate": 9.506666666666667e-06,
"loss": 0.2325,
"step": 725
},
{
"epoch": 1.05,
"learning_rate": 9.451111111111112e-06,
"loss": 0.2245,
"step": 750
},
{
"epoch": 1.06,
"learning_rate": 9.395555555555556e-06,
"loss": 0.2235,
"step": 775
},
{
"epoch": 1.06,
"learning_rate": 9.340000000000002e-06,
"loss": 0.1839,
"step": 800
},
{
"epoch": 1.07,
"learning_rate": 9.284444444444444e-06,
"loss": 0.1892,
"step": 825
},
{
"epoch": 1.07,
"learning_rate": 9.22888888888889e-06,
"loss": 0.1874,
"step": 850
},
{
"epoch": 1.08,
"learning_rate": 9.173333333333334e-06,
"loss": 0.1812,
"step": 875
},
{
"epoch": 1.08,
"learning_rate": 9.117777777777778e-06,
"loss": 0.1625,
"step": 900
},
{
"epoch": 1.09,
"learning_rate": 9.062222222222224e-06,
"loss": 0.111,
"step": 925
},
{
"epoch": 1.09,
"learning_rate": 9.006666666666666e-06,
"loss": 0.103,
"step": 950
},
{
"epoch": 1.1,
"learning_rate": 8.951111111111112e-06,
"loss": 0.1066,
"step": 975
},
{
"epoch": 2.0,
"learning_rate": 8.895555555555556e-06,
"loss": 0.132,
"step": 1000
},
{
"epoch": 2.0,
"eval_loss": 0.24614261090755463,
"eval_runtime": 2190.8426,
"eval_samples_per_second": 7.17,
"eval_steps_per_second": 0.224,
"eval_wer": 9.526742239073116,
"step": 1000
},
{
"epoch": 2.01,
"learning_rate": 8.84e-06,
"loss": 0.1606,
"step": 1025
},
{
"epoch": 2.01,
"learning_rate": 8.784444444444446e-06,
"loss": 0.2656,
"step": 1050
},
{
"epoch": 2.02,
"learning_rate": 8.72888888888889e-06,
"loss": 0.214,
"step": 1075
},
{
"epoch": 2.02,
"learning_rate": 8.673333333333334e-06,
"loss": 0.2065,
"step": 1100
},
{
"epoch": 2.03,
"learning_rate": 8.617777777777778e-06,
"loss": 0.1623,
"step": 1125
},
{
"epoch": 2.03,
"learning_rate": 8.562222222222224e-06,
"loss": 0.1839,
"step": 1150
},
{
"epoch": 2.04,
"learning_rate": 8.506666666666668e-06,
"loss": 0.1938,
"step": 1175
},
{
"epoch": 2.04,
"learning_rate": 8.451111111111112e-06,
"loss": 0.1909,
"step": 1200
},
{
"epoch": 2.05,
"learning_rate": 8.395555555555557e-06,
"loss": 0.1257,
"step": 1225
},
{
"epoch": 2.05,
"learning_rate": 8.34e-06,
"loss": 0.1262,
"step": 1250
},
{
"epoch": 2.06,
"learning_rate": 8.284444444444446e-06,
"loss": 0.1334,
"step": 1275
},
{
"epoch": 2.06,
"learning_rate": 8.22888888888889e-06,
"loss": 0.1661,
"step": 1300
},
{
"epoch": 2.07,
"learning_rate": 8.173333333333334e-06,
"loss": 0.2135,
"step": 1325
},
{
"epoch": 2.07,
"learning_rate": 8.11777777777778e-06,
"loss": 0.2401,
"step": 1350
},
{
"epoch": 2.08,
"learning_rate": 8.062222222222222e-06,
"loss": 0.1723,
"step": 1375
},
{
"epoch": 2.08,
"learning_rate": 8.006666666666667e-06,
"loss": 0.1362,
"step": 1400
},
{
"epoch": 2.09,
"learning_rate": 7.951111111111111e-06,
"loss": 0.1439,
"step": 1425
},
{
"epoch": 2.09,
"learning_rate": 7.895555555555557e-06,
"loss": 0.1279,
"step": 1450
},
{
"epoch": 3.0,
"learning_rate": 7.840000000000001e-06,
"loss": 0.1213,
"step": 1475
},
{
"epoch": 3.01,
"learning_rate": 7.784444444444445e-06,
"loss": 0.1892,
"step": 1500
},
{
"epoch": 3.01,
"learning_rate": 7.72888888888889e-06,
"loss": 0.1561,
"step": 1525
},
{
"epoch": 3.02,
"learning_rate": 7.673333333333333e-06,
"loss": 0.1589,
"step": 1550
},
{
"epoch": 3.02,
"learning_rate": 7.617777777777778e-06,
"loss": 0.1468,
"step": 1575
},
{
"epoch": 3.03,
"learning_rate": 7.562222222222223e-06,
"loss": 0.137,
"step": 1600
},
{
"epoch": 3.03,
"learning_rate": 7.506666666666668e-06,
"loss": 0.128,
"step": 1625
},
{
"epoch": 3.04,
"learning_rate": 7.451111111111111e-06,
"loss": 0.1372,
"step": 1650
},
{
"epoch": 3.04,
"learning_rate": 7.395555555555556e-06,
"loss": 0.1443,
"step": 1675
},
{
"epoch": 3.05,
"learning_rate": 7.340000000000001e-06,
"loss": 0.1312,
"step": 1700
},
{
"epoch": 3.05,
"learning_rate": 7.284444444444445e-06,
"loss": 0.1352,
"step": 1725
},
{
"epoch": 3.06,
"learning_rate": 7.22888888888889e-06,
"loss": 0.1437,
"step": 1750
},
{
"epoch": 3.06,
"learning_rate": 7.173333333333335e-06,
"loss": 0.1327,
"step": 1775
},
{
"epoch": 3.07,
"learning_rate": 7.117777777777778e-06,
"loss": 0.1319,
"step": 1800
},
{
"epoch": 3.07,
"learning_rate": 7.062222222222223e-06,
"loss": 0.1389,
"step": 1825
},
{
"epoch": 3.08,
"learning_rate": 7.006666666666667e-06,
"loss": 0.1423,
"step": 1850
},
{
"epoch": 3.08,
"learning_rate": 6.951111111111112e-06,
"loss": 0.1418,
"step": 1875
},
{
"epoch": 3.09,
"learning_rate": 6.8955555555555565e-06,
"loss": 0.1389,
"step": 1900
},
{
"epoch": 3.09,
"learning_rate": 6.8400000000000014e-06,
"loss": 0.1292,
"step": 1925
},
{
"epoch": 3.1,
"learning_rate": 6.784444444444445e-06,
"loss": 0.127,
"step": 1950
},
{
"epoch": 4.0,
"learning_rate": 6.7288888888888895e-06,
"loss": 0.1264,
"step": 1975
},
{
"epoch": 4.01,
"learning_rate": 6.6733333333333335e-06,
"loss": 0.1288,
"step": 2000
},
{
"epoch": 4.01,
"eval_loss": 0.22509153187274933,
"eval_runtime": 2168.1365,
"eval_samples_per_second": 7.245,
"eval_steps_per_second": 0.226,
"eval_wer": 8.52146243512513,
"step": 2000
},
{
"epoch": 4.01,
"learning_rate": 6.617777777777778e-06,
"loss": 0.1398,
"step": 2025
},
{
"epoch": 4.02,
"learning_rate": 6.562222222222223e-06,
"loss": 0.1284,
"step": 2050
},
{
"epoch": 4.02,
"learning_rate": 6.5066666666666665e-06,
"loss": 0.1248,
"step": 2075
},
{
"epoch": 4.03,
"learning_rate": 6.451111111111111e-06,
"loss": 0.1335,
"step": 2100
},
{
"epoch": 4.03,
"learning_rate": 6.395555555555556e-06,
"loss": 0.1103,
"step": 2125
},
{
"epoch": 4.04,
"learning_rate": 6.34e-06,
"loss": 0.1176,
"step": 2150
},
{
"epoch": 4.04,
"learning_rate": 6.284444444444445e-06,
"loss": 0.1317,
"step": 2175
},
{
"epoch": 4.05,
"learning_rate": 6.22888888888889e-06,
"loss": 0.1396,
"step": 2200
},
{
"epoch": 4.05,
"learning_rate": 6.173333333333333e-06,
"loss": 0.1315,
"step": 2225
},
{
"epoch": 4.06,
"learning_rate": 6.117777777777778e-06,
"loss": 0.1224,
"step": 2250
},
{
"epoch": 4.06,
"learning_rate": 6.062222222222223e-06,
"loss": 0.1291,
"step": 2275
},
{
"epoch": 4.07,
"learning_rate": 6.006666666666667e-06,
"loss": 0.1344,
"step": 2300
},
{
"epoch": 4.07,
"learning_rate": 5.951111111111112e-06,
"loss": 0.1238,
"step": 2325
},
{
"epoch": 4.08,
"learning_rate": 5.895555555555557e-06,
"loss": 0.1359,
"step": 2350
},
{
"epoch": 4.08,
"learning_rate": 5.84e-06,
"loss": 0.1252,
"step": 2375
},
{
"epoch": 4.09,
"learning_rate": 5.784444444444445e-06,
"loss": 0.1228,
"step": 2400
},
{
"epoch": 4.09,
"learning_rate": 5.72888888888889e-06,
"loss": 0.1095,
"step": 2425
},
{
"epoch": 4.1,
"learning_rate": 5.673333333333334e-06,
"loss": 0.1103,
"step": 2450
},
{
"epoch": 5.0,
"learning_rate": 5.617777777777779e-06,
"loss": 0.1154,
"step": 2475
},
{
"epoch": 5.01,
"learning_rate": 5.562222222222222e-06,
"loss": 0.1126,
"step": 2500
},
{
"epoch": 5.01,
"learning_rate": 5.506666666666667e-06,
"loss": 0.1652,
"step": 2525
},
{
"epoch": 5.02,
"learning_rate": 5.451111111111112e-06,
"loss": 0.1523,
"step": 2550
},
{
"epoch": 5.02,
"learning_rate": 5.3955555555555565e-06,
"loss": 0.1207,
"step": 2575
},
{
"epoch": 5.03,
"learning_rate": 5.3400000000000005e-06,
"loss": 0.1096,
"step": 2600
},
{
"epoch": 5.03,
"learning_rate": 5.2844444444444454e-06,
"loss": 0.0985,
"step": 2625
},
{
"epoch": 5.04,
"learning_rate": 5.228888888888889e-06,
"loss": 0.1494,
"step": 2650
},
{
"epoch": 5.04,
"learning_rate": 5.1733333333333335e-06,
"loss": 0.1289,
"step": 2675
},
{
"epoch": 5.05,
"learning_rate": 5.117777777777778e-06,
"loss": 0.1039,
"step": 2700
},
{
"epoch": 5.05,
"learning_rate": 5.062222222222222e-06,
"loss": 0.1195,
"step": 2725
},
{
"epoch": 5.06,
"learning_rate": 5.006666666666667e-06,
"loss": 0.1218,
"step": 2750
},
{
"epoch": 5.06,
"learning_rate": 4.951111111111111e-06,
"loss": 0.0945,
"step": 2775
},
{
"epoch": 5.07,
"learning_rate": 4.895555555555556e-06,
"loss": 0.132,
"step": 2800
},
{
"epoch": 5.07,
"learning_rate": 4.84e-06,
"loss": 0.292,
"step": 2825
},
{
"epoch": 5.08,
"learning_rate": 4.784444444444445e-06,
"loss": 0.1441,
"step": 2850
},
{
"epoch": 5.08,
"learning_rate": 4.728888888888889e-06,
"loss": 0.2019,
"step": 2875
},
{
"epoch": 5.09,
"learning_rate": 4.673333333333333e-06,
"loss": 0.1407,
"step": 2900
},
{
"epoch": 5.09,
"learning_rate": 4.617777777777778e-06,
"loss": 0.1426,
"step": 2925
},
{
"epoch": 6.0,
"learning_rate": 4.562222222222222e-06,
"loss": 0.1862,
"step": 2950
},
{
"epoch": 6.01,
"learning_rate": 4.506666666666667e-06,
"loss": 0.0754,
"step": 2975
},
{
"epoch": 6.01,
"learning_rate": 4.451111111111112e-06,
"loss": 0.0814,
"step": 3000
},
{
"epoch": 6.01,
"eval_loss": 0.22123335301876068,
"eval_runtime": 2909.2254,
"eval_samples_per_second": 5.399,
"eval_steps_per_second": 0.169,
"eval_wer": 8.266774443952604,
"step": 3000
},
{
"epoch": 6.02,
"learning_rate": 4.395555555555556e-06,
"loss": 0.0762,
"step": 3025
},
{
"epoch": 6.02,
"learning_rate": 4.34e-06,
"loss": 0.0662,
"step": 3050
},
{
"epoch": 6.03,
"learning_rate": 4.284444444444445e-06,
"loss": 0.0663,
"step": 3075
},
{
"epoch": 6.03,
"learning_rate": 4.228888888888889e-06,
"loss": 0.0642,
"step": 3100
},
{
"epoch": 6.04,
"learning_rate": 4.173333333333334e-06,
"loss": 0.0711,
"step": 3125
},
{
"epoch": 6.04,
"learning_rate": 4.117777777777779e-06,
"loss": 0.063,
"step": 3150
},
{
"epoch": 6.05,
"learning_rate": 4.062222222222223e-06,
"loss": 0.0672,
"step": 3175
},
{
"epoch": 6.05,
"learning_rate": 4.006666666666667e-06,
"loss": 0.0708,
"step": 3200
},
{
"epoch": 6.06,
"learning_rate": 3.951111111111112e-06,
"loss": 0.0715,
"step": 3225
},
{
"epoch": 6.06,
"learning_rate": 3.895555555555556e-06,
"loss": 0.0643,
"step": 3250
},
{
"epoch": 6.07,
"learning_rate": 3.8400000000000005e-06,
"loss": 0.0644,
"step": 3275
},
{
"epoch": 6.07,
"learning_rate": 3.784444444444445e-06,
"loss": 0.0753,
"step": 3300
},
{
"epoch": 6.08,
"learning_rate": 3.728888888888889e-06,
"loss": 0.0675,
"step": 3325
},
{
"epoch": 6.08,
"learning_rate": 3.673333333333334e-06,
"loss": 0.0735,
"step": 3350
},
{
"epoch": 6.09,
"learning_rate": 3.617777777777778e-06,
"loss": 0.0678,
"step": 3375
},
{
"epoch": 6.09,
"learning_rate": 3.5622222222222224e-06,
"loss": 0.0623,
"step": 3400
},
{
"epoch": 6.1,
"learning_rate": 3.5066666666666673e-06,
"loss": 0.0667,
"step": 3425
},
{
"epoch": 7.0,
"learning_rate": 3.4511111111111113e-06,
"loss": 0.0677,
"step": 3450
},
{
"epoch": 7.01,
"learning_rate": 3.3955555555555558e-06,
"loss": 0.0564,
"step": 3475
},
{
"epoch": 7.01,
"learning_rate": 3.3400000000000006e-06,
"loss": 0.1228,
"step": 3500
},
{
"epoch": 7.02,
"learning_rate": 3.2844444444444447e-06,
"loss": 0.1077,
"step": 3525
},
{
"epoch": 7.02,
"learning_rate": 3.228888888888889e-06,
"loss": 0.0834,
"step": 3550
},
{
"epoch": 7.03,
"learning_rate": 3.173333333333334e-06,
"loss": 0.0781,
"step": 3575
},
{
"epoch": 7.03,
"learning_rate": 3.117777777777778e-06,
"loss": 0.0781,
"step": 3600
},
{
"epoch": 7.04,
"learning_rate": 3.0622222222222225e-06,
"loss": 0.0883,
"step": 3625
},
{
"epoch": 7.04,
"learning_rate": 3.0066666666666674e-06,
"loss": 0.0916,
"step": 3650
},
{
"epoch": 7.05,
"learning_rate": 2.9511111111111114e-06,
"loss": 0.0531,
"step": 3675
},
{
"epoch": 7.05,
"learning_rate": 2.895555555555556e-06,
"loss": 0.0517,
"step": 3700
},
{
"epoch": 7.06,
"learning_rate": 2.84e-06,
"loss": 0.0513,
"step": 3725
},
{
"epoch": 7.06,
"learning_rate": 2.784444444444445e-06,
"loss": 0.0736,
"step": 3750
},
{
"epoch": 7.07,
"learning_rate": 2.7288888888888893e-06,
"loss": 0.0973,
"step": 3775
},
{
"epoch": 7.07,
"learning_rate": 2.6733333333333333e-06,
"loss": 0.1317,
"step": 3800
},
{
"epoch": 7.08,
"learning_rate": 2.617777777777778e-06,
"loss": 0.0844,
"step": 3825
},
{
"epoch": 7.08,
"learning_rate": 2.5622222222222226e-06,
"loss": 0.0614,
"step": 3850
},
{
"epoch": 7.09,
"learning_rate": 2.5066666666666667e-06,
"loss": 0.0664,
"step": 3875
},
{
"epoch": 7.09,
"learning_rate": 2.451111111111111e-06,
"loss": 0.0577,
"step": 3900
},
{
"epoch": 7.1,
"learning_rate": 2.3955555555555556e-06,
"loss": 0.053,
"step": 3925
},
{
"epoch": 8.0,
"learning_rate": 2.3400000000000005e-06,
"loss": 0.0306,
"step": 3950
},
{
"epoch": 8.01,
"learning_rate": 2.2844444444444445e-06,
"loss": 0.0447,
"step": 3975
},
{
"epoch": 8.01,
"learning_rate": 2.228888888888889e-06,
"loss": 0.0905,
"step": 4000
},
{
"epoch": 8.01,
"eval_loss": 0.23096486926078796,
"eval_runtime": 2944.0525,
"eval_samples_per_second": 5.336,
"eval_steps_per_second": 0.167,
"eval_wer": 8.499650365352169,
"step": 4000
},
{
"epoch": 8.02,
"learning_rate": 2.1733333333333334e-06,
"loss": 0.0618,
"step": 4025
},
{
"epoch": 8.02,
"learning_rate": 2.117777777777778e-06,
"loss": 0.056,
"step": 4050
},
{
"epoch": 8.03,
"learning_rate": 2.0622222222222223e-06,
"loss": 0.0479,
"step": 4075
},
{
"epoch": 8.03,
"learning_rate": 2.006666666666667e-06,
"loss": 0.0552,
"step": 4100
},
{
"epoch": 8.04,
"learning_rate": 1.9511111111111113e-06,
"loss": 0.0599,
"step": 4125
},
{
"epoch": 8.04,
"learning_rate": 1.8955555555555557e-06,
"loss": 0.0523,
"step": 4150
},
{
"epoch": 8.05,
"learning_rate": 1.8400000000000002e-06,
"loss": 0.0312,
"step": 4175
},
{
"epoch": 8.05,
"learning_rate": 1.7844444444444444e-06,
"loss": 0.032,
"step": 4200
},
{
"epoch": 8.06,
"learning_rate": 1.728888888888889e-06,
"loss": 0.0371,
"step": 4225
},
{
"epoch": 8.06,
"learning_rate": 1.6733333333333335e-06,
"loss": 0.0606,
"step": 4250
},
{
"epoch": 8.07,
"learning_rate": 1.6177777777777778e-06,
"loss": 0.0743,
"step": 4275
},
{
"epoch": 8.07,
"learning_rate": 1.5622222222222225e-06,
"loss": 0.0912,
"step": 4300
},
{
"epoch": 8.08,
"learning_rate": 1.506666666666667e-06,
"loss": 0.049,
"step": 4325
},
{
"epoch": 8.08,
"learning_rate": 1.4511111111111112e-06,
"loss": 0.0393,
"step": 4350
},
{
"epoch": 8.09,
"learning_rate": 1.3955555555555556e-06,
"loss": 0.0459,
"step": 4375
},
{
"epoch": 8.09,
"learning_rate": 1.34e-06,
"loss": 0.0383,
"step": 4400
},
{
"epoch": 9.0,
"learning_rate": 1.2844444444444445e-06,
"loss": 0.0389,
"step": 4425
},
{
"epoch": 9.01,
"learning_rate": 1.228888888888889e-06,
"loss": 0.0409,
"step": 4450
},
{
"epoch": 9.01,
"learning_rate": 1.1733333333333335e-06,
"loss": 0.0458,
"step": 4475
},
{
"epoch": 9.02,
"learning_rate": 1.117777777777778e-06,
"loss": 0.0423,
"step": 4500
},
{
"epoch": 9.02,
"learning_rate": 1.0622222222222222e-06,
"loss": 0.0405,
"step": 4525
},
{
"epoch": 9.03,
"learning_rate": 1.0066666666666668e-06,
"loss": 0.0334,
"step": 4550
},
{
"epoch": 9.03,
"learning_rate": 9.511111111111111e-07,
"loss": 0.0355,
"step": 4575
},
{
"epoch": 9.04,
"learning_rate": 8.955555555555557e-07,
"loss": 0.0406,
"step": 4600
},
{
"epoch": 9.04,
"learning_rate": 8.400000000000001e-07,
"loss": 0.035,
"step": 4625
},
{
"epoch": 9.05,
"learning_rate": 7.844444444444445e-07,
"loss": 0.0388,
"step": 4650
},
{
"epoch": 9.05,
"learning_rate": 7.28888888888889e-07,
"loss": 0.0402,
"step": 4675
},
{
"epoch": 9.06,
"learning_rate": 6.733333333333334e-07,
"loss": 0.0401,
"step": 4700
},
{
"epoch": 9.06,
"learning_rate": 6.177777777777778e-07,
"loss": 0.0361,
"step": 4725
},
{
"epoch": 9.07,
"learning_rate": 5.622222222222223e-07,
"loss": 0.0358,
"step": 4750
},
{
"epoch": 9.07,
"learning_rate": 5.066666666666667e-07,
"loss": 0.0441,
"step": 4775
},
{
"epoch": 9.08,
"learning_rate": 4.511111111111111e-07,
"loss": 0.0413,
"step": 4800
},
{
"epoch": 9.08,
"learning_rate": 3.9555555555555557e-07,
"loss": 0.0408,
"step": 4825
},
{
"epoch": 9.09,
"learning_rate": 3.4000000000000003e-07,
"loss": 0.0408,
"step": 4850
},
{
"epoch": 9.09,
"learning_rate": 2.844444444444445e-07,
"loss": 0.037,
"step": 4875
},
{
"epoch": 9.1,
"learning_rate": 2.2888888888888892e-07,
"loss": 0.0414,
"step": 4900
},
{
"epoch": 10.0,
"learning_rate": 1.7333333333333335e-07,
"loss": 0.0425,
"step": 4925
},
{
"epoch": 10.01,
"learning_rate": 1.1777777777777778e-07,
"loss": 0.0307,
"step": 4950
},
{
"epoch": 10.01,
"learning_rate": 6.222222222222223e-08,
"loss": 0.0371,
"step": 4975
},
{
"epoch": 10.02,
"learning_rate": 6.666666666666667e-09,
"loss": 0.0319,
"step": 5000
},
{
"epoch": 10.02,
"eval_loss": 0.23580369353294373,
"eval_runtime": 2946.1038,
"eval_samples_per_second": 5.332,
"eval_steps_per_second": 0.167,
"eval_wer": 8.534293064403343,
"step": 5000
},
{
"epoch": 10.02,
"step": 5000,
"total_flos": 9.23242412703744e+19,
"train_loss": 0.136722675549984,
"train_runtime": 59535.1806,
"train_samples_per_second": 5.375,
"train_steps_per_second": 0.084
}
],
"logging_steps": 25,
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"save_steps": 1000,
"total_flos": 9.23242412703744e+19,
"trial_name": null,
"trial_params": null
}