whisper-large-v2-hi / trainer_state.json
tarunanand's picture
End of training
3ef6edf
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 12.224938875305623,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06,
"learning_rate": 4.2000000000000006e-07,
"loss": 0.6338,
"step": 25
},
{
"epoch": 0.12,
"learning_rate": 9.200000000000001e-07,
"loss": 0.3594,
"step": 50
},
{
"epoch": 0.18,
"learning_rate": 1.42e-06,
"loss": 0.2642,
"step": 75
},
{
"epoch": 0.24,
"learning_rate": 1.9000000000000002e-06,
"loss": 0.2519,
"step": 100
},
{
"epoch": 0.31,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.2017,
"step": 125
},
{
"epoch": 0.37,
"learning_rate": 2.9e-06,
"loss": 0.2021,
"step": 150
},
{
"epoch": 0.43,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.1983,
"step": 175
},
{
"epoch": 0.49,
"learning_rate": 3.900000000000001e-06,
"loss": 0.1861,
"step": 200
},
{
"epoch": 0.55,
"learning_rate": 4.4e-06,
"loss": 0.1672,
"step": 225
},
{
"epoch": 0.61,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.1659,
"step": 250
},
{
"epoch": 0.67,
"learning_rate": 5.400000000000001e-06,
"loss": 0.1758,
"step": 275
},
{
"epoch": 0.73,
"learning_rate": 5.9e-06,
"loss": 0.1516,
"step": 300
},
{
"epoch": 0.79,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.1639,
"step": 325
},
{
"epoch": 0.86,
"learning_rate": 6.9e-06,
"loss": 0.166,
"step": 350
},
{
"epoch": 0.92,
"learning_rate": 7.4e-06,
"loss": 0.1643,
"step": 375
},
{
"epoch": 0.98,
"learning_rate": 7.9e-06,
"loss": 0.1578,
"step": 400
},
{
"epoch": 1.04,
"learning_rate": 8.400000000000001e-06,
"loss": 0.1378,
"step": 425
},
{
"epoch": 1.1,
"learning_rate": 8.900000000000001e-06,
"loss": 0.1089,
"step": 450
},
{
"epoch": 1.16,
"learning_rate": 9.4e-06,
"loss": 0.1208,
"step": 475
},
{
"epoch": 1.22,
"learning_rate": 9.9e-06,
"loss": 0.1163,
"step": 500
},
{
"epoch": 1.28,
"learning_rate": 9.955555555555556e-06,
"loss": 0.1172,
"step": 525
},
{
"epoch": 1.34,
"learning_rate": 9.9e-06,
"loss": 0.1232,
"step": 550
},
{
"epoch": 1.41,
"learning_rate": 9.844444444444446e-06,
"loss": 0.1082,
"step": 575
},
{
"epoch": 1.47,
"learning_rate": 9.78888888888889e-06,
"loss": 0.1134,
"step": 600
},
{
"epoch": 1.53,
"learning_rate": 9.733333333333334e-06,
"loss": 0.1136,
"step": 625
},
{
"epoch": 1.59,
"learning_rate": 9.677777777777778e-06,
"loss": 0.113,
"step": 650
},
{
"epoch": 1.65,
"learning_rate": 9.622222222222222e-06,
"loss": 0.1131,
"step": 675
},
{
"epoch": 1.71,
"learning_rate": 9.566666666666668e-06,
"loss": 0.1071,
"step": 700
},
{
"epoch": 1.77,
"learning_rate": 9.511111111111112e-06,
"loss": 0.0999,
"step": 725
},
{
"epoch": 1.83,
"learning_rate": 9.455555555555557e-06,
"loss": 0.1175,
"step": 750
},
{
"epoch": 1.89,
"learning_rate": 9.4e-06,
"loss": 0.1137,
"step": 775
},
{
"epoch": 1.96,
"learning_rate": 9.344444444444446e-06,
"loss": 0.1063,
"step": 800
},
{
"epoch": 2.02,
"learning_rate": 9.28888888888889e-06,
"loss": 0.0836,
"step": 825
},
{
"epoch": 2.08,
"learning_rate": 9.233333333333334e-06,
"loss": 0.0666,
"step": 850
},
{
"epoch": 2.14,
"learning_rate": 9.17777777777778e-06,
"loss": 0.0589,
"step": 875
},
{
"epoch": 2.2,
"learning_rate": 9.122222222222223e-06,
"loss": 0.061,
"step": 900
},
{
"epoch": 2.26,
"learning_rate": 9.066666666666667e-06,
"loss": 0.063,
"step": 925
},
{
"epoch": 2.32,
"learning_rate": 9.011111111111111e-06,
"loss": 0.0623,
"step": 950
},
{
"epoch": 2.38,
"learning_rate": 8.955555555555555e-06,
"loss": 0.0637,
"step": 975
},
{
"epoch": 2.44,
"learning_rate": 8.900000000000001e-06,
"loss": 0.0693,
"step": 1000
},
{
"epoch": 2.44,
"eval_loss": 0.20400269329547882,
"eval_runtime": 933.5971,
"eval_samples_per_second": 3.1,
"eval_steps_per_second": 0.194,
"eval_wer": 0.24743926183018708,
"step": 1000
},
{
"epoch": 2.51,
"learning_rate": 8.844444444444445e-06,
"loss": 0.0671,
"step": 1025
},
{
"epoch": 2.57,
"learning_rate": 8.788888888888891e-06,
"loss": 0.0615,
"step": 1050
},
{
"epoch": 2.63,
"learning_rate": 8.733333333333333e-06,
"loss": 0.0675,
"step": 1075
},
{
"epoch": 2.69,
"learning_rate": 8.677777777777779e-06,
"loss": 0.0628,
"step": 1100
},
{
"epoch": 2.75,
"learning_rate": 8.622222222222223e-06,
"loss": 0.0638,
"step": 1125
},
{
"epoch": 2.81,
"learning_rate": 8.566666666666667e-06,
"loss": 0.0597,
"step": 1150
},
{
"epoch": 2.87,
"learning_rate": 8.511111111111113e-06,
"loss": 0.0641,
"step": 1175
},
{
"epoch": 2.93,
"learning_rate": 8.455555555555555e-06,
"loss": 0.0688,
"step": 1200
},
{
"epoch": 3.0,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0591,
"step": 1225
},
{
"epoch": 3.06,
"learning_rate": 8.344444444444445e-06,
"loss": 0.0338,
"step": 1250
},
{
"epoch": 3.12,
"learning_rate": 8.288888888888889e-06,
"loss": 0.0334,
"step": 1275
},
{
"epoch": 3.18,
"learning_rate": 8.233333333333335e-06,
"loss": 0.0332,
"step": 1300
},
{
"epoch": 3.24,
"learning_rate": 8.177777777777779e-06,
"loss": 0.0373,
"step": 1325
},
{
"epoch": 3.3,
"learning_rate": 8.122222222222223e-06,
"loss": 0.0398,
"step": 1350
},
{
"epoch": 3.36,
"learning_rate": 8.066666666666667e-06,
"loss": 0.0371,
"step": 1375
},
{
"epoch": 3.42,
"learning_rate": 8.011111111111113e-06,
"loss": 0.0353,
"step": 1400
},
{
"epoch": 3.48,
"learning_rate": 7.955555555555557e-06,
"loss": 0.0357,
"step": 1425
},
{
"epoch": 3.55,
"learning_rate": 7.9e-06,
"loss": 0.0316,
"step": 1450
},
{
"epoch": 3.61,
"learning_rate": 7.844444444444446e-06,
"loss": 0.0374,
"step": 1475
},
{
"epoch": 3.67,
"learning_rate": 7.788888888888889e-06,
"loss": 0.034,
"step": 1500
},
{
"epoch": 3.73,
"learning_rate": 7.733333333333334e-06,
"loss": 0.036,
"step": 1525
},
{
"epoch": 3.79,
"learning_rate": 7.677777777777778e-06,
"loss": 0.0416,
"step": 1550
},
{
"epoch": 3.85,
"learning_rate": 7.622222222222223e-06,
"loss": 0.038,
"step": 1575
},
{
"epoch": 3.91,
"learning_rate": 7.566666666666667e-06,
"loss": 0.0378,
"step": 1600
},
{
"epoch": 3.97,
"learning_rate": 7.511111111111111e-06,
"loss": 0.0351,
"step": 1625
},
{
"epoch": 4.03,
"learning_rate": 7.455555555555556e-06,
"loss": 0.0321,
"step": 1650
},
{
"epoch": 4.1,
"learning_rate": 7.4e-06,
"loss": 0.0207,
"step": 1675
},
{
"epoch": 4.16,
"learning_rate": 7.344444444444445e-06,
"loss": 0.0202,
"step": 1700
},
{
"epoch": 4.22,
"learning_rate": 7.28888888888889e-06,
"loss": 0.0189,
"step": 1725
},
{
"epoch": 4.28,
"learning_rate": 7.233333333333334e-06,
"loss": 0.018,
"step": 1750
},
{
"epoch": 4.34,
"learning_rate": 7.177777777777778e-06,
"loss": 0.0212,
"step": 1775
},
{
"epoch": 4.4,
"learning_rate": 7.122222222222222e-06,
"loss": 0.0204,
"step": 1800
},
{
"epoch": 4.46,
"learning_rate": 7.066666666666667e-06,
"loss": 0.0166,
"step": 1825
},
{
"epoch": 4.52,
"learning_rate": 7.011111111111112e-06,
"loss": 0.019,
"step": 1850
},
{
"epoch": 4.58,
"learning_rate": 6.955555555555557e-06,
"loss": 0.0206,
"step": 1875
},
{
"epoch": 4.65,
"learning_rate": 6.9e-06,
"loss": 0.0199,
"step": 1900
},
{
"epoch": 4.71,
"learning_rate": 6.844444444444445e-06,
"loss": 0.0211,
"step": 1925
},
{
"epoch": 4.77,
"learning_rate": 6.788888888888889e-06,
"loss": 0.0206,
"step": 1950
},
{
"epoch": 4.83,
"learning_rate": 6.733333333333334e-06,
"loss": 0.0222,
"step": 1975
},
{
"epoch": 4.89,
"learning_rate": 6.677777777777779e-06,
"loss": 0.0218,
"step": 2000
},
{
"epoch": 4.89,
"eval_loss": 0.22987425327301025,
"eval_runtime": 932.6088,
"eval_samples_per_second": 3.103,
"eval_steps_per_second": 0.194,
"eval_wer": 0.24134428172352493,
"step": 2000
},
{
"epoch": 4.95,
"learning_rate": 6.6222222222222236e-06,
"loss": 0.0192,
"step": 2025
},
{
"epoch": 5.01,
"learning_rate": 6.566666666666667e-06,
"loss": 0.0165,
"step": 2050
},
{
"epoch": 5.07,
"learning_rate": 6.511111111111112e-06,
"loss": 0.015,
"step": 2075
},
{
"epoch": 5.13,
"learning_rate": 6.455555555555556e-06,
"loss": 0.013,
"step": 2100
},
{
"epoch": 5.2,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.0119,
"step": 2125
},
{
"epoch": 5.26,
"learning_rate": 6.3444444444444454e-06,
"loss": 0.0123,
"step": 2150
},
{
"epoch": 5.32,
"learning_rate": 6.28888888888889e-06,
"loss": 0.0107,
"step": 2175
},
{
"epoch": 5.38,
"learning_rate": 6.2333333333333335e-06,
"loss": 0.011,
"step": 2200
},
{
"epoch": 5.44,
"learning_rate": 6.177777777777778e-06,
"loss": 0.0112,
"step": 2225
},
{
"epoch": 5.5,
"learning_rate": 6.1222222222222224e-06,
"loss": 0.0112,
"step": 2250
},
{
"epoch": 5.56,
"learning_rate": 6.066666666666667e-06,
"loss": 0.0114,
"step": 2275
},
{
"epoch": 5.62,
"learning_rate": 6.011111111111112e-06,
"loss": 0.0096,
"step": 2300
},
{
"epoch": 5.68,
"learning_rate": 5.955555555555555e-06,
"loss": 0.0082,
"step": 2325
},
{
"epoch": 5.75,
"learning_rate": 5.9e-06,
"loss": 0.0124,
"step": 2350
},
{
"epoch": 5.81,
"learning_rate": 5.844444444444445e-06,
"loss": 0.0107,
"step": 2375
},
{
"epoch": 5.87,
"learning_rate": 5.788888888888889e-06,
"loss": 0.0121,
"step": 2400
},
{
"epoch": 5.93,
"learning_rate": 5.733333333333334e-06,
"loss": 0.0105,
"step": 2425
},
{
"epoch": 5.99,
"learning_rate": 5.677777777777779e-06,
"loss": 0.01,
"step": 2450
},
{
"epoch": 6.05,
"learning_rate": 5.622222222222222e-06,
"loss": 0.0073,
"step": 2475
},
{
"epoch": 6.11,
"learning_rate": 5.566666666666667e-06,
"loss": 0.0046,
"step": 2500
},
{
"epoch": 6.17,
"learning_rate": 5.511111111111112e-06,
"loss": 0.0034,
"step": 2525
},
{
"epoch": 6.23,
"learning_rate": 5.455555555555556e-06,
"loss": 0.0052,
"step": 2550
},
{
"epoch": 6.3,
"learning_rate": 5.400000000000001e-06,
"loss": 0.0043,
"step": 2575
},
{
"epoch": 6.36,
"learning_rate": 5.344444444444446e-06,
"loss": 0.0046,
"step": 2600
},
{
"epoch": 6.42,
"learning_rate": 5.288888888888889e-06,
"loss": 0.0052,
"step": 2625
},
{
"epoch": 6.48,
"learning_rate": 5.233333333333334e-06,
"loss": 0.0053,
"step": 2650
},
{
"epoch": 6.54,
"learning_rate": 5.177777777777779e-06,
"loss": 0.0068,
"step": 2675
},
{
"epoch": 6.6,
"learning_rate": 5.122222222222223e-06,
"loss": 0.0063,
"step": 2700
},
{
"epoch": 6.66,
"learning_rate": 5.0666666666666676e-06,
"loss": 0.0048,
"step": 2725
},
{
"epoch": 6.72,
"learning_rate": 5.011111111111111e-06,
"loss": 0.0077,
"step": 2750
},
{
"epoch": 6.78,
"learning_rate": 4.9555555555555565e-06,
"loss": 0.0039,
"step": 2775
},
{
"epoch": 6.85,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.0058,
"step": 2800
},
{
"epoch": 6.91,
"learning_rate": 4.8444444444444446e-06,
"loss": 0.0065,
"step": 2825
},
{
"epoch": 6.97,
"learning_rate": 4.7888888888888894e-06,
"loss": 0.0053,
"step": 2850
},
{
"epoch": 7.03,
"learning_rate": 4.7333333333333335e-06,
"loss": 0.0045,
"step": 2875
},
{
"epoch": 7.09,
"learning_rate": 4.677777777777778e-06,
"loss": 0.0025,
"step": 2900
},
{
"epoch": 7.15,
"learning_rate": 4.622222222222222e-06,
"loss": 0.0031,
"step": 2925
},
{
"epoch": 7.21,
"learning_rate": 4.566666666666667e-06,
"loss": 0.0025,
"step": 2950
},
{
"epoch": 7.27,
"learning_rate": 4.511111111111111e-06,
"loss": 0.0023,
"step": 2975
},
{
"epoch": 7.33,
"learning_rate": 4.455555555555555e-06,
"loss": 0.0023,
"step": 3000
},
{
"epoch": 7.33,
"eval_loss": 0.2687150537967682,
"eval_runtime": 936.0119,
"eval_samples_per_second": 3.092,
"eval_steps_per_second": 0.193,
"eval_wer": 0.22115466012020654,
"step": 3000
},
{
"epoch": 7.4,
"learning_rate": 4.4e-06,
"loss": 0.0024,
"step": 3025
},
{
"epoch": 7.46,
"learning_rate": 4.344444444444445e-06,
"loss": 0.0023,
"step": 3050
},
{
"epoch": 7.52,
"learning_rate": 4.288888888888889e-06,
"loss": 0.0013,
"step": 3075
},
{
"epoch": 7.58,
"learning_rate": 4.233333333333334e-06,
"loss": 0.0015,
"step": 3100
},
{
"epoch": 7.64,
"learning_rate": 4.177777777777778e-06,
"loss": 0.0032,
"step": 3125
},
{
"epoch": 7.7,
"learning_rate": 4.122222222222222e-06,
"loss": 0.0035,
"step": 3150
},
{
"epoch": 7.76,
"learning_rate": 4.066666666666667e-06,
"loss": 0.0026,
"step": 3175
},
{
"epoch": 7.82,
"learning_rate": 4.011111111111111e-06,
"loss": 0.0023,
"step": 3200
},
{
"epoch": 7.89,
"learning_rate": 3.955555555555556e-06,
"loss": 0.0021,
"step": 3225
},
{
"epoch": 7.95,
"learning_rate": 3.900000000000001e-06,
"loss": 0.0022,
"step": 3250
},
{
"epoch": 8.01,
"learning_rate": 3.844444444444445e-06,
"loss": 0.0023,
"step": 3275
},
{
"epoch": 8.07,
"learning_rate": 3.7888888888888893e-06,
"loss": 0.001,
"step": 3300
},
{
"epoch": 8.13,
"learning_rate": 3.7333333333333337e-06,
"loss": 0.001,
"step": 3325
},
{
"epoch": 8.19,
"learning_rate": 3.6777777777777778e-06,
"loss": 0.0015,
"step": 3350
},
{
"epoch": 8.25,
"learning_rate": 3.6222222222222226e-06,
"loss": 0.0015,
"step": 3375
},
{
"epoch": 8.31,
"learning_rate": 3.566666666666667e-06,
"loss": 0.0016,
"step": 3400
},
{
"epoch": 8.37,
"learning_rate": 3.511111111111111e-06,
"loss": 0.001,
"step": 3425
},
{
"epoch": 8.44,
"learning_rate": 3.455555555555556e-06,
"loss": 0.0009,
"step": 3450
},
{
"epoch": 8.5,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.0016,
"step": 3475
},
{
"epoch": 8.56,
"learning_rate": 3.3444444444444445e-06,
"loss": 0.0005,
"step": 3500
},
{
"epoch": 8.62,
"learning_rate": 3.2888888888888894e-06,
"loss": 0.0009,
"step": 3525
},
{
"epoch": 8.68,
"learning_rate": 3.2333333333333334e-06,
"loss": 0.0018,
"step": 3550
},
{
"epoch": 8.74,
"learning_rate": 3.177777777777778e-06,
"loss": 0.0011,
"step": 3575
},
{
"epoch": 8.8,
"learning_rate": 3.1222222222222228e-06,
"loss": 0.0008,
"step": 3600
},
{
"epoch": 8.86,
"learning_rate": 3.066666666666667e-06,
"loss": 0.0007,
"step": 3625
},
{
"epoch": 8.92,
"learning_rate": 3.0111111111111113e-06,
"loss": 0.0007,
"step": 3650
},
{
"epoch": 8.99,
"learning_rate": 2.955555555555556e-06,
"loss": 0.0006,
"step": 3675
},
{
"epoch": 9.05,
"learning_rate": 2.9e-06,
"loss": 0.0006,
"step": 3700
},
{
"epoch": 9.11,
"learning_rate": 2.8444444444444446e-06,
"loss": 0.0004,
"step": 3725
},
{
"epoch": 9.17,
"learning_rate": 2.788888888888889e-06,
"loss": 0.0004,
"step": 3750
},
{
"epoch": 9.23,
"learning_rate": 2.7333333333333336e-06,
"loss": 0.0006,
"step": 3775
},
{
"epoch": 9.29,
"learning_rate": 2.677777777777778e-06,
"loss": 0.0003,
"step": 3800
},
{
"epoch": 9.35,
"learning_rate": 2.6222222222222225e-06,
"loss": 0.0003,
"step": 3825
},
{
"epoch": 9.41,
"learning_rate": 2.566666666666667e-06,
"loss": 0.0005,
"step": 3850
},
{
"epoch": 9.47,
"learning_rate": 2.5111111111111114e-06,
"loss": 0.0005,
"step": 3875
},
{
"epoch": 9.54,
"learning_rate": 2.455555555555556e-06,
"loss": 0.0006,
"step": 3900
},
{
"epoch": 9.6,
"learning_rate": 2.4000000000000003e-06,
"loss": 0.0006,
"step": 3925
},
{
"epoch": 9.66,
"learning_rate": 2.3444444444444448e-06,
"loss": 0.0005,
"step": 3950
},
{
"epoch": 9.72,
"learning_rate": 2.2888888888888892e-06,
"loss": 0.0007,
"step": 3975
},
{
"epoch": 9.78,
"learning_rate": 2.2333333333333333e-06,
"loss": 0.0003,
"step": 4000
},
{
"epoch": 9.78,
"eval_loss": 0.2887499928474426,
"eval_runtime": 946.1244,
"eval_samples_per_second": 3.059,
"eval_steps_per_second": 0.191,
"eval_wer": 0.21603318378058073,
"step": 4000
},
{
"epoch": 9.84,
"learning_rate": 2.1777777777777777e-06,
"loss": 0.0003,
"step": 4025
},
{
"epoch": 9.9,
"learning_rate": 2.1222222222222226e-06,
"loss": 0.0002,
"step": 4050
},
{
"epoch": 9.96,
"learning_rate": 2.0666666666666666e-06,
"loss": 0.0002,
"step": 4075
},
{
"epoch": 10.02,
"learning_rate": 2.011111111111111e-06,
"loss": 0.0002,
"step": 4100
},
{
"epoch": 10.09,
"learning_rate": 1.955555555555556e-06,
"loss": 0.0002,
"step": 4125
},
{
"epoch": 10.15,
"learning_rate": 1.9000000000000002e-06,
"loss": 0.0002,
"step": 4150
},
{
"epoch": 10.21,
"learning_rate": 1.8444444444444445e-06,
"loss": 0.0001,
"step": 4175
},
{
"epoch": 10.27,
"learning_rate": 1.788888888888889e-06,
"loss": 0.0001,
"step": 4200
},
{
"epoch": 10.33,
"learning_rate": 1.7333333333333336e-06,
"loss": 0.0001,
"step": 4225
},
{
"epoch": 10.39,
"learning_rate": 1.6777777777777779e-06,
"loss": 0.0001,
"step": 4250
},
{
"epoch": 10.45,
"learning_rate": 1.6222222222222223e-06,
"loss": 0.0001,
"step": 4275
},
{
"epoch": 10.51,
"learning_rate": 1.566666666666667e-06,
"loss": 0.0001,
"step": 4300
},
{
"epoch": 10.57,
"learning_rate": 1.5111111111111112e-06,
"loss": 0.0002,
"step": 4325
},
{
"epoch": 10.64,
"learning_rate": 1.4555555555555557e-06,
"loss": 0.0001,
"step": 4350
},
{
"epoch": 10.7,
"learning_rate": 1.4000000000000001e-06,
"loss": 0.0006,
"step": 4375
},
{
"epoch": 10.76,
"learning_rate": 1.3444444444444446e-06,
"loss": 0.0001,
"step": 4400
},
{
"epoch": 10.82,
"learning_rate": 1.288888888888889e-06,
"loss": 0.0001,
"step": 4425
},
{
"epoch": 10.88,
"learning_rate": 1.2333333333333335e-06,
"loss": 0.0001,
"step": 4450
},
{
"epoch": 10.94,
"learning_rate": 1.1777777777777778e-06,
"loss": 0.0002,
"step": 4475
},
{
"epoch": 11.0,
"learning_rate": 1.1222222222222222e-06,
"loss": 0.0001,
"step": 4500
},
{
"epoch": 11.06,
"learning_rate": 1.066666666666667e-06,
"loss": 0.0001,
"step": 4525
},
{
"epoch": 11.12,
"learning_rate": 1.0111111111111111e-06,
"loss": 0.0001,
"step": 4550
},
{
"epoch": 11.19,
"learning_rate": 9.555555555555556e-07,
"loss": 0.0001,
"step": 4575
},
{
"epoch": 11.25,
"learning_rate": 9.000000000000001e-07,
"loss": 0.0001,
"step": 4600
},
{
"epoch": 11.31,
"learning_rate": 8.444444444444445e-07,
"loss": 0.0001,
"step": 4625
},
{
"epoch": 11.37,
"learning_rate": 7.888888888888889e-07,
"loss": 0.0001,
"step": 4650
},
{
"epoch": 11.43,
"learning_rate": 7.333333333333334e-07,
"loss": 0.0001,
"step": 4675
},
{
"epoch": 11.49,
"learning_rate": 6.777777777777779e-07,
"loss": 0.0001,
"step": 4700
},
{
"epoch": 11.55,
"learning_rate": 6.222222222222223e-07,
"loss": 0.0001,
"step": 4725
},
{
"epoch": 11.61,
"learning_rate": 5.666666666666667e-07,
"loss": 0.0001,
"step": 4750
},
{
"epoch": 11.67,
"learning_rate": 5.111111111111112e-07,
"loss": 0.0001,
"step": 4775
},
{
"epoch": 11.74,
"learning_rate": 4.5555555555555563e-07,
"loss": 0.0001,
"step": 4800
},
{
"epoch": 11.8,
"learning_rate": 4.0000000000000003e-07,
"loss": 0.0001,
"step": 4825
},
{
"epoch": 11.86,
"learning_rate": 3.444444444444445e-07,
"loss": 0.0001,
"step": 4850
},
{
"epoch": 11.92,
"learning_rate": 2.888888888888889e-07,
"loss": 0.0001,
"step": 4875
},
{
"epoch": 11.98,
"learning_rate": 2.3333333333333336e-07,
"loss": 0.0001,
"step": 4900
},
{
"epoch": 12.04,
"learning_rate": 1.777777777777778e-07,
"loss": 0.0001,
"step": 4925
},
{
"epoch": 12.1,
"learning_rate": 1.2222222222222225e-07,
"loss": 0.0001,
"step": 4950
},
{
"epoch": 12.16,
"learning_rate": 6.666666666666668e-08,
"loss": 0.0001,
"step": 4975
},
{
"epoch": 12.22,
"learning_rate": 1.1111111111111112e-08,
"loss": 0.0001,
"step": 5000
},
{
"epoch": 12.22,
"eval_loss": 0.31141459941864014,
"eval_runtime": 937.1973,
"eval_samples_per_second": 3.088,
"eval_steps_per_second": 0.193,
"eval_wer": 0.21476339625835944,
"step": 5000
},
{
"epoch": 12.22,
"step": 5000,
"total_flos": 1.6975177089986593e+20,
"train_loss": 0.039045126114599406,
"train_runtime": 22755.7778,
"train_samples_per_second": 3.516,
"train_steps_per_second": 0.22
}
],
"max_steps": 5000,
"num_train_epochs": 13,
"total_flos": 1.6975177089986593e+20,
"trial_name": null,
"trial_params": null
}