whisper-medium-it / trainer_state.json
EdoAbati's picture
End of training
550f569
raw
history blame
25.8 kB
{
"best_metric": 6.322427174144163,
"best_model_checkpoint": "./checkpoint-5000",
"epoch": 5.062,
"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": 0.9427,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.600000000000001e-07,
"loss": 0.7467,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.46e-06,
"loss": 0.4001,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.264,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 2.46e-06,
"loss": 0.2276,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 2.96e-06,
"loss": 0.1945,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 3.46e-06,
"loss": 0.1831,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 3.96e-06,
"loss": 0.1762,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 4.4600000000000005e-06,
"loss": 0.1657,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 4.960000000000001e-06,
"loss": 0.1656,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 5.460000000000001e-06,
"loss": 0.1668,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 5.9600000000000005e-06,
"loss": 0.1599,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 6.460000000000001e-06,
"loss": 0.1627,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 6.96e-06,
"loss": 0.167,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 7.4600000000000006e-06,
"loss": 0.1831,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 7.960000000000002e-06,
"loss": 0.1936,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 8.46e-06,
"loss": 0.1851,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 8.96e-06,
"loss": 0.1655,
"step": 450
},
{
"epoch": 0.1,
"learning_rate": 9.460000000000001e-06,
"loss": 0.1689,
"step": 475
},
{
"epoch": 0.1,
"learning_rate": 9.960000000000001e-06,
"loss": 0.1695,
"step": 500
},
{
"epoch": 0.1,
"learning_rate": 9.94888888888889e-06,
"loss": 0.1769,
"step": 525
},
{
"epoch": 0.11,
"learning_rate": 9.893333333333334e-06,
"loss": 0.1844,
"step": 550
},
{
"epoch": 0.12,
"learning_rate": 9.837777777777778e-06,
"loss": 0.1935,
"step": 575
},
{
"epoch": 0.12,
"learning_rate": 9.782222222222222e-06,
"loss": 0.198,
"step": 600
},
{
"epoch": 0.12,
"learning_rate": 9.726666666666668e-06,
"loss": 0.2206,
"step": 625
},
{
"epoch": 0.13,
"learning_rate": 9.671111111111112e-06,
"loss": 0.2206,
"step": 650
},
{
"epoch": 0.14,
"learning_rate": 9.615555555555558e-06,
"loss": 0.2211,
"step": 675
},
{
"epoch": 0.14,
"learning_rate": 9.56e-06,
"loss": 0.2038,
"step": 700
},
{
"epoch": 0.14,
"learning_rate": 9.504444444444446e-06,
"loss": 0.2103,
"step": 725
},
{
"epoch": 0.15,
"learning_rate": 9.44888888888889e-06,
"loss": 0.1813,
"step": 750
},
{
"epoch": 0.15,
"learning_rate": 9.393333333333334e-06,
"loss": 0.2052,
"step": 775
},
{
"epoch": 0.16,
"learning_rate": 9.33777777777778e-06,
"loss": 0.1975,
"step": 800
},
{
"epoch": 0.17,
"learning_rate": 9.282222222222222e-06,
"loss": 0.211,
"step": 825
},
{
"epoch": 0.17,
"learning_rate": 9.226666666666668e-06,
"loss": 0.1922,
"step": 850
},
{
"epoch": 0.17,
"learning_rate": 9.171111111111112e-06,
"loss": 0.1667,
"step": 875
},
{
"epoch": 0.18,
"learning_rate": 9.115555555555556e-06,
"loss": 0.1457,
"step": 900
},
{
"epoch": 0.18,
"learning_rate": 9.060000000000001e-06,
"loss": 0.1472,
"step": 925
},
{
"epoch": 1.0,
"learning_rate": 9.004444444444445e-06,
"loss": 0.1557,
"step": 950
},
{
"epoch": 1.01,
"learning_rate": 8.94888888888889e-06,
"loss": 0.157,
"step": 975
},
{
"epoch": 1.01,
"learning_rate": 8.893333333333333e-06,
"loss": 0.1609,
"step": 1000
},
{
"epoch": 1.01,
"eval_loss": 0.19324588775634766,
"eval_runtime": 9564.6276,
"eval_samples_per_second": 1.569,
"eval_steps_per_second": 0.098,
"eval_wer": 7.9271209866042955,
"step": 1000
},
{
"epoch": 1.02,
"learning_rate": 8.83777777777778e-06,
"loss": 0.181,
"step": 1025
},
{
"epoch": 1.02,
"learning_rate": 8.782222222222223e-06,
"loss": 0.1501,
"step": 1050
},
{
"epoch": 1.03,
"learning_rate": 8.726666666666667e-06,
"loss": 0.1636,
"step": 1075
},
{
"epoch": 1.03,
"learning_rate": 8.671111111111113e-06,
"loss": 0.1371,
"step": 1100
},
{
"epoch": 1.04,
"learning_rate": 8.615555555555555e-06,
"loss": 0.1385,
"step": 1125
},
{
"epoch": 1.04,
"learning_rate": 8.560000000000001e-06,
"loss": 0.1267,
"step": 1150
},
{
"epoch": 1.05,
"learning_rate": 8.504444444444445e-06,
"loss": 0.1096,
"step": 1175
},
{
"epoch": 1.05,
"learning_rate": 8.448888888888889e-06,
"loss": 0.1372,
"step": 1200
},
{
"epoch": 1.06,
"learning_rate": 8.393333333333335e-06,
"loss": 0.1185,
"step": 1225
},
{
"epoch": 1.06,
"learning_rate": 8.337777777777777e-06,
"loss": 0.13,
"step": 1250
},
{
"epoch": 1.07,
"learning_rate": 8.282222222222223e-06,
"loss": 0.1431,
"step": 1275
},
{
"epoch": 1.07,
"learning_rate": 8.226666666666667e-06,
"loss": 0.133,
"step": 1300
},
{
"epoch": 1.08,
"learning_rate": 8.171111111111113e-06,
"loss": 0.0968,
"step": 1325
},
{
"epoch": 1.08,
"learning_rate": 8.115555555555557e-06,
"loss": 0.0884,
"step": 1350
},
{
"epoch": 1.09,
"learning_rate": 8.06e-06,
"loss": 0.1118,
"step": 1375
},
{
"epoch": 1.09,
"learning_rate": 8.004444444444445e-06,
"loss": 0.109,
"step": 1400
},
{
"epoch": 1.1,
"learning_rate": 7.948888888888889e-06,
"loss": 0.1045,
"step": 1425
},
{
"epoch": 1.1,
"learning_rate": 7.893333333333335e-06,
"loss": 0.1106,
"step": 1450
},
{
"epoch": 1.11,
"learning_rate": 7.837777777777779e-06,
"loss": 0.1102,
"step": 1475
},
{
"epoch": 1.11,
"learning_rate": 7.782222222222223e-06,
"loss": 0.1074,
"step": 1500
},
{
"epoch": 1.12,
"learning_rate": 7.726666666666667e-06,
"loss": 0.1225,
"step": 1525
},
{
"epoch": 1.12,
"learning_rate": 7.67111111111111e-06,
"loss": 0.1189,
"step": 1550
},
{
"epoch": 1.13,
"learning_rate": 7.6155555555555564e-06,
"loss": 0.1091,
"step": 1575
},
{
"epoch": 1.13,
"learning_rate": 7.5600000000000005e-06,
"loss": 0.1218,
"step": 1600
},
{
"epoch": 1.14,
"learning_rate": 7.504444444444445e-06,
"loss": 0.1196,
"step": 1625
},
{
"epoch": 1.14,
"learning_rate": 7.44888888888889e-06,
"loss": 0.1288,
"step": 1650
},
{
"epoch": 1.15,
"learning_rate": 7.393333333333333e-06,
"loss": 0.1107,
"step": 1675
},
{
"epoch": 1.15,
"learning_rate": 7.337777777777778e-06,
"loss": 0.1147,
"step": 1700
},
{
"epoch": 1.16,
"learning_rate": 7.282222222222222e-06,
"loss": 0.0955,
"step": 1725
},
{
"epoch": 1.16,
"learning_rate": 7.226666666666667e-06,
"loss": 0.0905,
"step": 1750
},
{
"epoch": 1.17,
"learning_rate": 7.171111111111112e-06,
"loss": 0.0854,
"step": 1775
},
{
"epoch": 1.17,
"learning_rate": 7.115555555555557e-06,
"loss": 0.1005,
"step": 1800
},
{
"epoch": 1.18,
"learning_rate": 7.06e-06,
"loss": 0.0995,
"step": 1825
},
{
"epoch": 1.18,
"learning_rate": 7.004444444444445e-06,
"loss": 0.1121,
"step": 1850
},
{
"epoch": 1.19,
"learning_rate": 6.948888888888889e-06,
"loss": 0.1036,
"step": 1875
},
{
"epoch": 2.0,
"learning_rate": 6.893333333333334e-06,
"loss": 0.1281,
"step": 1900
},
{
"epoch": 2.01,
"learning_rate": 6.837777777777779e-06,
"loss": 0.1173,
"step": 1925
},
{
"epoch": 2.01,
"learning_rate": 6.782222222222222e-06,
"loss": 0.1131,
"step": 1950
},
{
"epoch": 2.02,
"learning_rate": 6.726666666666667e-06,
"loss": 0.12,
"step": 1975
},
{
"epoch": 2.02,
"learning_rate": 6.671111111111112e-06,
"loss": 0.1529,
"step": 2000
},
{
"epoch": 2.02,
"eval_loss": 0.17985273897647858,
"eval_runtime": 9576.7899,
"eval_samples_per_second": 1.567,
"eval_steps_per_second": 0.098,
"eval_wer": 7.278598766744632,
"step": 2000
},
{
"epoch": 2.03,
"learning_rate": 6.615555555555556e-06,
"loss": 0.1303,
"step": 2025
},
{
"epoch": 2.03,
"learning_rate": 6.560000000000001e-06,
"loss": 0.1081,
"step": 2050
},
{
"epoch": 2.04,
"learning_rate": 6.504444444444446e-06,
"loss": 0.1227,
"step": 2075
},
{
"epoch": 2.04,
"learning_rate": 6.448888888888889e-06,
"loss": 0.1292,
"step": 2100
},
{
"epoch": 2.05,
"learning_rate": 6.393333333333334e-06,
"loss": 0.1052,
"step": 2125
},
{
"epoch": 2.05,
"learning_rate": 6.3377777777777786e-06,
"loss": 0.0946,
"step": 2150
},
{
"epoch": 2.06,
"learning_rate": 6.282222222222223e-06,
"loss": 0.1124,
"step": 2175
},
{
"epoch": 2.06,
"learning_rate": 6.2266666666666675e-06,
"loss": 0.1076,
"step": 2200
},
{
"epoch": 2.07,
"learning_rate": 6.171111111111112e-06,
"loss": 0.105,
"step": 2225
},
{
"epoch": 2.07,
"learning_rate": 6.1155555555555555e-06,
"loss": 0.087,
"step": 2250
},
{
"epoch": 2.08,
"learning_rate": 6.0600000000000004e-06,
"loss": 0.0951,
"step": 2275
},
{
"epoch": 2.08,
"learning_rate": 6.004444444444445e-06,
"loss": 0.0929,
"step": 2300
},
{
"epoch": 2.09,
"learning_rate": 5.948888888888889e-06,
"loss": 0.1039,
"step": 2325
},
{
"epoch": 2.09,
"learning_rate": 5.893333333333334e-06,
"loss": 0.0943,
"step": 2350
},
{
"epoch": 2.1,
"learning_rate": 5.837777777777777e-06,
"loss": 0.0906,
"step": 2375
},
{
"epoch": 2.1,
"learning_rate": 5.782222222222222e-06,
"loss": 0.1143,
"step": 2400
},
{
"epoch": 2.11,
"learning_rate": 5.726666666666667e-06,
"loss": 0.1012,
"step": 2425
},
{
"epoch": 2.11,
"learning_rate": 5.671111111111112e-06,
"loss": 0.1137,
"step": 2450
},
{
"epoch": 2.12,
"learning_rate": 5.615555555555556e-06,
"loss": 0.113,
"step": 2475
},
{
"epoch": 2.12,
"learning_rate": 5.560000000000001e-06,
"loss": 0.0924,
"step": 2500
},
{
"epoch": 2.13,
"learning_rate": 5.504444444444444e-06,
"loss": 0.0888,
"step": 2525
},
{
"epoch": 2.13,
"learning_rate": 5.448888888888889e-06,
"loss": 0.0926,
"step": 2550
},
{
"epoch": 2.14,
"learning_rate": 5.393333333333334e-06,
"loss": 0.093,
"step": 2575
},
{
"epoch": 2.14,
"learning_rate": 5.337777777777779e-06,
"loss": 0.0839,
"step": 2600
},
{
"epoch": 2.15,
"learning_rate": 5.282222222222223e-06,
"loss": 0.079,
"step": 2625
},
{
"epoch": 2.15,
"learning_rate": 5.226666666666667e-06,
"loss": 0.0813,
"step": 2650
},
{
"epoch": 2.16,
"learning_rate": 5.171111111111111e-06,
"loss": 0.0812,
"step": 2675
},
{
"epoch": 2.16,
"learning_rate": 5.115555555555556e-06,
"loss": 0.0728,
"step": 2700
},
{
"epoch": 2.17,
"learning_rate": 5.060000000000001e-06,
"loss": 0.0756,
"step": 2725
},
{
"epoch": 2.17,
"learning_rate": 5.004444444444445e-06,
"loss": 0.0749,
"step": 2750
},
{
"epoch": 2.18,
"learning_rate": 4.94888888888889e-06,
"loss": 0.1002,
"step": 2775
},
{
"epoch": 2.18,
"learning_rate": 4.893333333333334e-06,
"loss": 0.0857,
"step": 2800
},
{
"epoch": 3.0,
"learning_rate": 4.837777777777778e-06,
"loss": 0.0959,
"step": 2825
},
{
"epoch": 3.01,
"learning_rate": 4.7822222222222226e-06,
"loss": 0.0954,
"step": 2850
},
{
"epoch": 3.01,
"learning_rate": 4.7266666666666674e-06,
"loss": 0.0743,
"step": 2875
},
{
"epoch": 3.02,
"learning_rate": 4.6711111111111115e-06,
"loss": 0.072,
"step": 2900
},
{
"epoch": 3.02,
"learning_rate": 4.6155555555555555e-06,
"loss": 0.0816,
"step": 2925
},
{
"epoch": 3.03,
"learning_rate": 4.56e-06,
"loss": 0.0648,
"step": 2950
},
{
"epoch": 3.03,
"learning_rate": 4.504444444444444e-06,
"loss": 0.0604,
"step": 2975
},
{
"epoch": 3.04,
"learning_rate": 4.448888888888889e-06,
"loss": 0.0501,
"step": 3000
},
{
"epoch": 3.04,
"eval_loss": 0.17620889842510223,
"eval_runtime": 9690.7589,
"eval_samples_per_second": 1.548,
"eval_steps_per_second": 0.097,
"eval_wer": 6.781575590048905,
"step": 3000
},
{
"epoch": 3.04,
"learning_rate": 4.393333333333334e-06,
"loss": 0.0562,
"step": 3025
},
{
"epoch": 3.05,
"learning_rate": 4.337777777777778e-06,
"loss": 0.0454,
"step": 3050
},
{
"epoch": 3.05,
"learning_rate": 4.282222222222222e-06,
"loss": 0.0482,
"step": 3075
},
{
"epoch": 3.06,
"learning_rate": 4.226666666666667e-06,
"loss": 0.0478,
"step": 3100
},
{
"epoch": 3.06,
"learning_rate": 4.171111111111111e-06,
"loss": 0.0471,
"step": 3125
},
{
"epoch": 3.07,
"learning_rate": 4.115555555555556e-06,
"loss": 0.043,
"step": 3150
},
{
"epoch": 3.07,
"learning_rate": 4.060000000000001e-06,
"loss": 0.0456,
"step": 3175
},
{
"epoch": 3.08,
"learning_rate": 4.004444444444445e-06,
"loss": 0.0469,
"step": 3200
},
{
"epoch": 3.08,
"learning_rate": 3.948888888888889e-06,
"loss": 0.0486,
"step": 3225
},
{
"epoch": 3.09,
"learning_rate": 3.893333333333333e-06,
"loss": 0.0433,
"step": 3250
},
{
"epoch": 3.09,
"learning_rate": 3.837777777777778e-06,
"loss": 0.0391,
"step": 3275
},
{
"epoch": 3.1,
"learning_rate": 3.782222222222223e-06,
"loss": 0.0369,
"step": 3300
},
{
"epoch": 3.1,
"learning_rate": 3.726666666666667e-06,
"loss": 0.0402,
"step": 3325
},
{
"epoch": 3.11,
"learning_rate": 3.6711111111111113e-06,
"loss": 0.0437,
"step": 3350
},
{
"epoch": 3.11,
"learning_rate": 3.615555555555556e-06,
"loss": 0.0404,
"step": 3375
},
{
"epoch": 3.12,
"learning_rate": 3.5600000000000002e-06,
"loss": 0.0443,
"step": 3400
},
{
"epoch": 3.12,
"learning_rate": 3.5044444444444447e-06,
"loss": 0.0402,
"step": 3425
},
{
"epoch": 3.13,
"learning_rate": 3.4488888888888896e-06,
"loss": 0.0613,
"step": 3450
},
{
"epoch": 3.13,
"learning_rate": 3.3933333333333336e-06,
"loss": 0.0606,
"step": 3475
},
{
"epoch": 3.14,
"learning_rate": 3.337777777777778e-06,
"loss": 0.0505,
"step": 3500
},
{
"epoch": 3.14,
"learning_rate": 3.282222222222223e-06,
"loss": 0.051,
"step": 3525
},
{
"epoch": 3.15,
"learning_rate": 3.226666666666667e-06,
"loss": 0.048,
"step": 3550
},
{
"epoch": 3.15,
"learning_rate": 3.1711111111111114e-06,
"loss": 0.0437,
"step": 3575
},
{
"epoch": 3.16,
"learning_rate": 3.1155555555555555e-06,
"loss": 0.0529,
"step": 3600
},
{
"epoch": 3.16,
"learning_rate": 3.0600000000000003e-06,
"loss": 0.0527,
"step": 3625
},
{
"epoch": 3.17,
"learning_rate": 3.004444444444445e-06,
"loss": 0.0684,
"step": 3650
},
{
"epoch": 3.17,
"learning_rate": 2.948888888888889e-06,
"loss": 0.0492,
"step": 3675
},
{
"epoch": 3.18,
"learning_rate": 2.8933333333333337e-06,
"loss": 0.0361,
"step": 3700
},
{
"epoch": 3.18,
"learning_rate": 2.837777777777778e-06,
"loss": 0.0297,
"step": 3725
},
{
"epoch": 3.19,
"learning_rate": 2.7822222222222222e-06,
"loss": 0.0308,
"step": 3750
},
{
"epoch": 4.0,
"learning_rate": 2.726666666666667e-06,
"loss": 0.0414,
"step": 3775
},
{
"epoch": 4.01,
"learning_rate": 2.6711111111111116e-06,
"loss": 0.044,
"step": 3800
},
{
"epoch": 4.01,
"learning_rate": 2.6155555555555556e-06,
"loss": 0.0367,
"step": 3825
},
{
"epoch": 4.02,
"learning_rate": 2.56e-06,
"loss": 0.0367,
"step": 3850
},
{
"epoch": 4.02,
"learning_rate": 2.504444444444445e-06,
"loss": 0.057,
"step": 3875
},
{
"epoch": 4.03,
"learning_rate": 2.448888888888889e-06,
"loss": 0.0426,
"step": 3900
},
{
"epoch": 4.03,
"learning_rate": 2.3933333333333334e-06,
"loss": 0.0352,
"step": 3925
},
{
"epoch": 4.04,
"learning_rate": 2.337777777777778e-06,
"loss": 0.0347,
"step": 3950
},
{
"epoch": 4.04,
"learning_rate": 2.2822222222222223e-06,
"loss": 0.0395,
"step": 3975
},
{
"epoch": 4.05,
"learning_rate": 2.226666666666667e-06,
"loss": 0.0358,
"step": 4000
},
{
"epoch": 4.05,
"eval_loss": 0.17835251986980438,
"eval_runtime": 9824.4069,
"eval_samples_per_second": 1.527,
"eval_steps_per_second": 0.095,
"eval_wer": 6.518445672974697,
"step": 4000
},
{
"epoch": 4.05,
"learning_rate": 2.1711111111111113e-06,
"loss": 0.0304,
"step": 4025
},
{
"epoch": 4.06,
"learning_rate": 2.1155555555555557e-06,
"loss": 0.0349,
"step": 4050
},
{
"epoch": 4.06,
"learning_rate": 2.06e-06,
"loss": 0.0339,
"step": 4075
},
{
"epoch": 4.07,
"learning_rate": 2.0044444444444446e-06,
"loss": 0.0337,
"step": 4100
},
{
"epoch": 4.07,
"learning_rate": 1.948888888888889e-06,
"loss": 0.028,
"step": 4125
},
{
"epoch": 4.08,
"learning_rate": 1.8933333333333333e-06,
"loss": 0.0288,
"step": 4150
},
{
"epoch": 4.08,
"learning_rate": 1.837777777777778e-06,
"loss": 0.0285,
"step": 4175
},
{
"epoch": 4.09,
"learning_rate": 1.7822222222222225e-06,
"loss": 0.0325,
"step": 4200
},
{
"epoch": 4.09,
"learning_rate": 1.7266666666666667e-06,
"loss": 0.027,
"step": 4225
},
{
"epoch": 4.1,
"learning_rate": 1.6711111111111112e-06,
"loss": 0.0298,
"step": 4250
},
{
"epoch": 4.1,
"learning_rate": 1.6155555555555559e-06,
"loss": 0.0361,
"step": 4275
},
{
"epoch": 4.11,
"learning_rate": 1.56e-06,
"loss": 0.0323,
"step": 4300
},
{
"epoch": 4.11,
"learning_rate": 1.5044444444444446e-06,
"loss": 0.0325,
"step": 4325
},
{
"epoch": 4.12,
"learning_rate": 1.4488888888888892e-06,
"loss": 0.0372,
"step": 4350
},
{
"epoch": 4.12,
"learning_rate": 1.3933333333333335e-06,
"loss": 0.0308,
"step": 4375
},
{
"epoch": 4.13,
"learning_rate": 1.337777777777778e-06,
"loss": 0.0301,
"step": 4400
},
{
"epoch": 4.13,
"learning_rate": 1.2822222222222222e-06,
"loss": 0.0286,
"step": 4425
},
{
"epoch": 4.14,
"learning_rate": 1.2266666666666666e-06,
"loss": 0.0286,
"step": 4450
},
{
"epoch": 4.14,
"learning_rate": 1.171111111111111e-06,
"loss": 0.0257,
"step": 4475
},
{
"epoch": 4.15,
"learning_rate": 1.1155555555555558e-06,
"loss": 0.0239,
"step": 4500
},
{
"epoch": 4.15,
"learning_rate": 1.06e-06,
"loss": 0.028,
"step": 4525
},
{
"epoch": 4.16,
"learning_rate": 1.0044444444444445e-06,
"loss": 0.0283,
"step": 4550
},
{
"epoch": 4.16,
"learning_rate": 9.488888888888889e-07,
"loss": 0.0257,
"step": 4575
},
{
"epoch": 4.17,
"learning_rate": 8.933333333333334e-07,
"loss": 0.0277,
"step": 4600
},
{
"epoch": 4.17,
"learning_rate": 8.37777777777778e-07,
"loss": 0.023,
"step": 4625
},
{
"epoch": 4.18,
"learning_rate": 7.822222222222223e-07,
"loss": 0.0384,
"step": 4650
},
{
"epoch": 4.18,
"learning_rate": 7.266666666666668e-07,
"loss": 0.0252,
"step": 4675
},
{
"epoch": 5.0,
"learning_rate": 6.711111111111111e-07,
"loss": 0.0274,
"step": 4700
},
{
"epoch": 5.01,
"learning_rate": 6.155555555555556e-07,
"loss": 0.0316,
"step": 4725
},
{
"epoch": 5.01,
"learning_rate": 5.6e-07,
"loss": 0.034,
"step": 4750
},
{
"epoch": 5.02,
"learning_rate": 5.044444444444445e-07,
"loss": 0.0312,
"step": 4775
},
{
"epoch": 5.02,
"learning_rate": 4.488888888888889e-07,
"loss": 0.0339,
"step": 4800
},
{
"epoch": 5.03,
"learning_rate": 3.9333333333333336e-07,
"loss": 0.0326,
"step": 4825
},
{
"epoch": 5.03,
"learning_rate": 3.3777777777777777e-07,
"loss": 0.0332,
"step": 4850
},
{
"epoch": 5.04,
"learning_rate": 2.822222222222222e-07,
"loss": 0.0267,
"step": 4875
},
{
"epoch": 5.04,
"learning_rate": 2.266666666666667e-07,
"loss": 0.0271,
"step": 4900
},
{
"epoch": 5.05,
"learning_rate": 1.7111111111111114e-07,
"loss": 0.024,
"step": 4925
},
{
"epoch": 5.05,
"learning_rate": 1.1555555555555556e-07,
"loss": 0.0315,
"step": 4950
},
{
"epoch": 5.06,
"learning_rate": 6.000000000000001e-08,
"loss": 0.0286,
"step": 4975
},
{
"epoch": 5.06,
"learning_rate": 4.444444444444445e-09,
"loss": 0.0305,
"step": 5000
},
{
"epoch": 5.06,
"eval_loss": 0.18036045134067535,
"eval_runtime": 9975.9091,
"eval_samples_per_second": 1.504,
"eval_steps_per_second": 0.094,
"eval_wer": 6.322427174144163,
"step": 5000
},
{
"epoch": 5.06,
"step": 5000,
"total_flos": 1.631845186633728e+20,
"train_loss": 0.10211840506792068,
"train_runtime": 80044.0796,
"train_samples_per_second": 1.999,
"train_steps_per_second": 0.062
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.631845186633728e+20,
"trial_name": null,
"trial_params": null
}