whisper-medium-el-n2 / trainer_state.json
emilios's picture
End of training
fa3c142
raw
history blame
26.2 kB
{
"best_metric": 10.642644873699851,
"best_model_checkpoint": "./checkpoint-5000",
"epoch": 293.6470588235294,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 2.78,
"learning_rate": 5.0453611334320685e-06,
"loss": 0.6804,
"step": 25
},
{
"epoch": 5.56,
"learning_rate": 6.229195710491767e-06,
"loss": 0.1847,
"step": 50
},
{
"epoch": 8.33,
"learning_rate": 6.903829450223392e-06,
"loss": 0.0821,
"step": 75
},
{
"epoch": 11.11,
"learning_rate": 7.377725845391017e-06,
"loss": 0.0485,
"step": 100
},
{
"epoch": 13.89,
"learning_rate": 7.743343231239583e-06,
"loss": 0.0432,
"step": 125
},
{
"epoch": 16.67,
"learning_rate": 8.041073861170494e-06,
"loss": 0.0328,
"step": 150
},
{
"epoch": 19.44,
"learning_rate": 8.292222957399574e-06,
"loss": 0.0291,
"step": 175
},
{
"epoch": 22.22,
"learning_rate": 8.509413541357755e-06,
"loss": 0.0298,
"step": 200
},
{
"epoch": 25.0,
"learning_rate": 8.700744577655557e-06,
"loss": 0.0269,
"step": 225
},
{
"epoch": 27.78,
"learning_rate": 8.871723942761204e-06,
"loss": 0.0272,
"step": 250
},
{
"epoch": 30.56,
"learning_rate": 9.026267958246849e-06,
"loss": 0.027,
"step": 275
},
{
"epoch": 33.33,
"learning_rate": 9.16726106663399e-06,
"loss": 0.0213,
"step": 300
},
{
"epoch": 36.11,
"learning_rate": 9.296889251455016e-06,
"loss": 0.0215,
"step": 325
},
{
"epoch": 38.89,
"learning_rate": 9.416848797368692e-06,
"loss": 0.0195,
"step": 350
},
{
"epoch": 41.67,
"learning_rate": 9.528482449516371e-06,
"loss": 0.0167,
"step": 375
},
{
"epoch": 44.44,
"learning_rate": 9.632871309784314e-06,
"loss": 0.0184,
"step": 400
},
{
"epoch": 47.22,
"learning_rate": 9.73089868785391e-06,
"loss": 0.0159,
"step": 425
},
{
"epoch": 50.0,
"learning_rate": 9.823295589572114e-06,
"loss": 0.0172,
"step": 450
},
{
"epoch": 52.78,
"learning_rate": 9.910673836465484e-06,
"loss": 0.0123,
"step": 475
},
{
"epoch": 55.56,
"learning_rate": 9.993550644973805e-06,
"loss": 0.0144,
"step": 500
},
{
"epoch": 58.33,
"learning_rate": 9.951111111111111e-06,
"loss": 0.0135,
"step": 525
},
{
"epoch": 61.11,
"learning_rate": 9.895555555555557e-06,
"loss": 0.0128,
"step": 550
},
{
"epoch": 63.89,
"learning_rate": 9.84e-06,
"loss": 0.0115,
"step": 575
},
{
"epoch": 66.67,
"learning_rate": 9.784444444444445e-06,
"loss": 0.0105,
"step": 600
},
{
"epoch": 69.44,
"learning_rate": 9.72888888888889e-06,
"loss": 0.0104,
"step": 625
},
{
"epoch": 72.22,
"learning_rate": 9.673333333333334e-06,
"loss": 0.0087,
"step": 650
},
{
"epoch": 75.0,
"learning_rate": 9.617777777777778e-06,
"loss": 0.0091,
"step": 675
},
{
"epoch": 77.78,
"learning_rate": 9.562222222222223e-06,
"loss": 0.0085,
"step": 700
},
{
"epoch": 80.56,
"learning_rate": 9.506666666666667e-06,
"loss": 0.011,
"step": 725
},
{
"epoch": 83.33,
"learning_rate": 9.451111111111112e-06,
"loss": 0.0117,
"step": 750
},
{
"epoch": 86.11,
"learning_rate": 9.395555555555556e-06,
"loss": 0.0088,
"step": 775
},
{
"epoch": 88.89,
"learning_rate": 9.340000000000002e-06,
"loss": 0.0077,
"step": 800
},
{
"epoch": 91.67,
"learning_rate": 9.284444444444444e-06,
"loss": 0.0091,
"step": 825
},
{
"epoch": 94.44,
"learning_rate": 9.22888888888889e-06,
"loss": 0.0067,
"step": 850
},
{
"epoch": 97.22,
"learning_rate": 9.173333333333334e-06,
"loss": 0.0082,
"step": 875
},
{
"epoch": 100.0,
"learning_rate": 9.117777777777778e-06,
"loss": 0.0055,
"step": 900
},
{
"epoch": 102.78,
"learning_rate": 9.062222222222224e-06,
"loss": 0.0077,
"step": 925
},
{
"epoch": 105.56,
"learning_rate": 9.006666666666666e-06,
"loss": 0.0055,
"step": 950
},
{
"epoch": 108.33,
"learning_rate": 8.951111111111112e-06,
"loss": 0.005,
"step": 975
},
{
"epoch": 111.11,
"learning_rate": 8.895555555555556e-06,
"loss": 0.0066,
"step": 1000
},
{
"epoch": 111.11,
"eval_loss": 0.2357177734375,
"eval_runtime": 64.7785,
"eval_samples_per_second": 2.022,
"eval_steps_per_second": 0.139,
"eval_wer": 23.044096728307252,
"step": 1000
},
{
"epoch": 113.89,
"learning_rate": 8.844444444444445e-06,
"loss": 0.0057,
"step": 1025
},
{
"epoch": 116.67,
"learning_rate": 8.788888888888891e-06,
"loss": 0.0096,
"step": 1050
},
{
"epoch": 119.44,
"learning_rate": 8.733333333333333e-06,
"loss": 0.0063,
"step": 1075
},
{
"epoch": 122.22,
"learning_rate": 8.677777777777779e-06,
"loss": 0.0069,
"step": 1100
},
{
"epoch": 125.0,
"learning_rate": 8.622222222222223e-06,
"loss": 0.0069,
"step": 1125
},
{
"epoch": 127.78,
"learning_rate": 8.566666666666667e-06,
"loss": 0.0046,
"step": 1150
},
{
"epoch": 130.56,
"learning_rate": 8.511111111111113e-06,
"loss": 0.0051,
"step": 1175
},
{
"epoch": 133.33,
"learning_rate": 8.455555555555555e-06,
"loss": 0.0055,
"step": 1200
},
{
"epoch": 136.11,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0042,
"step": 1225
},
{
"epoch": 138.89,
"learning_rate": 8.344444444444445e-06,
"loss": 0.0042,
"step": 1250
},
{
"epoch": 141.67,
"learning_rate": 8.288888888888889e-06,
"loss": 0.005,
"step": 1275
},
{
"epoch": 144.44,
"learning_rate": 8.233333333333335e-06,
"loss": 0.0054,
"step": 1300
},
{
"epoch": 147.22,
"learning_rate": 8.177777777777779e-06,
"loss": 0.0052,
"step": 1325
},
{
"epoch": 150.0,
"learning_rate": 8.122222222222223e-06,
"loss": 0.0057,
"step": 1350
},
{
"epoch": 152.78,
"learning_rate": 8.066666666666667e-06,
"loss": 0.0039,
"step": 1375
},
{
"epoch": 155.56,
"learning_rate": 8.011111111111113e-06,
"loss": 0.0032,
"step": 1400
},
{
"epoch": 158.33,
"learning_rate": 7.955555555555557e-06,
"loss": 0.0034,
"step": 1425
},
{
"epoch": 161.11,
"learning_rate": 7.902222222222223e-06,
"loss": 0.0068,
"step": 1450
},
{
"epoch": 163.89,
"learning_rate": 7.846666666666667e-06,
"loss": 0.0034,
"step": 1475
},
{
"epoch": 166.67,
"learning_rate": 7.791111111111111e-06,
"loss": 0.0026,
"step": 1500
},
{
"epoch": 169.44,
"learning_rate": 7.735555555555557e-06,
"loss": 0.0036,
"step": 1525
},
{
"epoch": 172.22,
"learning_rate": 7.680000000000001e-06,
"loss": 0.0033,
"step": 1550
},
{
"epoch": 175.0,
"learning_rate": 7.624444444444445e-06,
"loss": 0.0021,
"step": 1575
},
{
"epoch": 177.78,
"learning_rate": 7.56888888888889e-06,
"loss": 0.0033,
"step": 1600
},
{
"epoch": 180.56,
"learning_rate": 7.513333333333334e-06,
"loss": 0.0037,
"step": 1625
},
{
"epoch": 183.33,
"learning_rate": 7.457777777777778e-06,
"loss": 0.0032,
"step": 1650
},
{
"epoch": 186.11,
"learning_rate": 7.402222222222223e-06,
"loss": 0.0037,
"step": 1675
},
{
"epoch": 188.89,
"learning_rate": 7.346666666666668e-06,
"loss": 0.0022,
"step": 1700
},
{
"epoch": 191.67,
"learning_rate": 7.291111111111112e-06,
"loss": 0.0024,
"step": 1725
},
{
"epoch": 194.44,
"learning_rate": 7.235555555555556e-06,
"loss": 0.0026,
"step": 1750
},
{
"epoch": 197.22,
"learning_rate": 7.180000000000001e-06,
"loss": 0.0022,
"step": 1775
},
{
"epoch": 200.0,
"learning_rate": 7.124444444444445e-06,
"loss": 0.0026,
"step": 1800
},
{
"epoch": 202.78,
"learning_rate": 7.06888888888889e-06,
"loss": 0.0032,
"step": 1825
},
{
"epoch": 205.56,
"learning_rate": 7.0133333333333345e-06,
"loss": 0.0033,
"step": 1850
},
{
"epoch": 208.33,
"learning_rate": 6.9577777777777785e-06,
"loss": 0.0027,
"step": 1875
},
{
"epoch": 211.11,
"learning_rate": 6.902222222222223e-06,
"loss": 0.0043,
"step": 1900
},
{
"epoch": 213.89,
"learning_rate": 6.846666666666667e-06,
"loss": 0.0028,
"step": 1925
},
{
"epoch": 216.67,
"learning_rate": 6.7911111111111115e-06,
"loss": 0.0012,
"step": 1950
},
{
"epoch": 219.44,
"learning_rate": 6.735555555555556e-06,
"loss": 0.0015,
"step": 1975
},
{
"epoch": 222.22,
"learning_rate": 6.680000000000001e-06,
"loss": 0.0024,
"step": 2000
},
{
"epoch": 222.22,
"eval_loss": 0.2607421875,
"eval_runtime": 57.0802,
"eval_samples_per_second": 2.295,
"eval_steps_per_second": 0.158,
"eval_wer": 19.665718349928877,
"step": 2000
},
{
"epoch": 225.0,
"learning_rate": 6.6244444444444445e-06,
"loss": 0.0029,
"step": 2025
},
{
"epoch": 227.78,
"learning_rate": 6.568888888888889e-06,
"loss": 0.0021,
"step": 2050
},
{
"epoch": 230.56,
"learning_rate": 6.513333333333333e-06,
"loss": 0.0022,
"step": 2075
},
{
"epoch": 233.33,
"learning_rate": 6.457777777777778e-06,
"loss": 0.0022,
"step": 2100
},
{
"epoch": 236.11,
"learning_rate": 6.402222222222223e-06,
"loss": 0.0011,
"step": 2125
},
{
"epoch": 238.89,
"learning_rate": 6.346666666666668e-06,
"loss": 0.0026,
"step": 2150
},
{
"epoch": 241.67,
"learning_rate": 6.291111111111111e-06,
"loss": 0.0021,
"step": 2175
},
{
"epoch": 244.44,
"learning_rate": 6.235555555555556e-06,
"loss": 0.0016,
"step": 2200
},
{
"epoch": 247.22,
"learning_rate": 6.18e-06,
"loss": 0.0024,
"step": 2225
},
{
"epoch": 250.0,
"learning_rate": 6.124444444444445e-06,
"loss": 0.0046,
"step": 2250
},
{
"epoch": 252.78,
"learning_rate": 6.06888888888889e-06,
"loss": 0.0018,
"step": 2275
},
{
"epoch": 255.56,
"learning_rate": 6.013333333333335e-06,
"loss": 0.0012,
"step": 2300
},
{
"epoch": 258.33,
"learning_rate": 5.957777777777778e-06,
"loss": 0.0014,
"step": 2325
},
{
"epoch": 261.11,
"learning_rate": 5.902222222222223e-06,
"loss": 0.0007,
"step": 2350
},
{
"epoch": 263.89,
"learning_rate": 5.846666666666667e-06,
"loss": 0.0014,
"step": 2375
},
{
"epoch": 266.67,
"learning_rate": 5.791111111111112e-06,
"loss": 0.0009,
"step": 2400
},
{
"epoch": 269.44,
"learning_rate": 5.735555555555557e-06,
"loss": 0.0008,
"step": 2425
},
{
"epoch": 272.22,
"learning_rate": 5.68e-06,
"loss": 0.0028,
"step": 2450
},
{
"epoch": 275.0,
"learning_rate": 5.624444444444445e-06,
"loss": 0.002,
"step": 2475
},
{
"epoch": 277.78,
"learning_rate": 5.56888888888889e-06,
"loss": 0.0011,
"step": 2500
},
{
"epoch": 280.56,
"learning_rate": 5.513333333333334e-06,
"loss": 0.001,
"step": 2525
},
{
"epoch": 283.33,
"learning_rate": 5.4577777777777785e-06,
"loss": 0.0007,
"step": 2550
},
{
"epoch": 286.11,
"learning_rate": 5.402222222222223e-06,
"loss": 0.0007,
"step": 2575
},
{
"epoch": 288.89,
"learning_rate": 5.346666666666667e-06,
"loss": 0.0008,
"step": 2600
},
{
"epoch": 291.67,
"learning_rate": 5.2911111111111115e-06,
"loss": 0.0012,
"step": 2625
},
{
"epoch": 294.44,
"learning_rate": 5.235555555555556e-06,
"loss": 0.0016,
"step": 2650
},
{
"epoch": 297.22,
"learning_rate": 5.18e-06,
"loss": 0.0012,
"step": 2675
},
{
"epoch": 300.0,
"learning_rate": 5.124444444444445e-06,
"loss": 0.001,
"step": 2700
},
{
"epoch": 302.78,
"learning_rate": 5.06888888888889e-06,
"loss": 0.0012,
"step": 2725
},
{
"epoch": 305.56,
"learning_rate": 5.013333333333333e-06,
"loss": 0.001,
"step": 2750
},
{
"epoch": 308.33,
"learning_rate": 4.957777777777778e-06,
"loss": 0.0013,
"step": 2775
},
{
"epoch": 311.11,
"learning_rate": 4.902222222222222e-06,
"loss": 0.0015,
"step": 2800
},
{
"epoch": 313.89,
"learning_rate": 4.846666666666667e-06,
"loss": 0.0014,
"step": 2825
},
{
"epoch": 316.67,
"learning_rate": 4.791111111111111e-06,
"loss": 0.0007,
"step": 2850
},
{
"epoch": 319.44,
"learning_rate": 4.735555555555556e-06,
"loss": 0.0009,
"step": 2875
},
{
"epoch": 322.22,
"learning_rate": 4.680000000000001e-06,
"loss": 0.0021,
"step": 2900
},
{
"epoch": 325.0,
"learning_rate": 4.624444444444445e-06,
"loss": 0.0015,
"step": 2925
},
{
"epoch": 327.78,
"learning_rate": 4.568888888888889e-06,
"loss": 0.0012,
"step": 2950
},
{
"epoch": 330.56,
"learning_rate": 4.513333333333333e-06,
"loss": 0.0009,
"step": 2975
},
{
"epoch": 333.33,
"learning_rate": 4.457777777777778e-06,
"loss": 0.0011,
"step": 3000
},
{
"epoch": 333.33,
"eval_loss": 0.277099609375,
"eval_runtime": 58.1634,
"eval_samples_per_second": 2.252,
"eval_steps_per_second": 0.155,
"eval_wer": 20.874822190611663,
"step": 3000
},
{
"epoch": 177.47,
"learning_rate": 1.760888888888889e-06,
"loss": 0.5801,
"step": 3025
},
{
"epoch": 178.94,
"learning_rate": 1.7386666666666666e-06,
"loss": 0.1501,
"step": 3050
},
{
"epoch": 180.41,
"learning_rate": 1.7164444444444444e-06,
"loss": 0.0789,
"step": 3075
},
{
"epoch": 181.88,
"learning_rate": 1.6942222222222222e-06,
"loss": 0.0531,
"step": 3100
},
{
"epoch": 183.35,
"learning_rate": 1.6719999999999998e-06,
"loss": 0.0409,
"step": 3125
},
{
"epoch": 184.82,
"learning_rate": 1.6497777777777777e-06,
"loss": 0.032,
"step": 3150
},
{
"epoch": 186.29,
"learning_rate": 1.6275555555555555e-06,
"loss": 0.0251,
"step": 3175
},
{
"epoch": 187.76,
"learning_rate": 1.6053333333333333e-06,
"loss": 0.0203,
"step": 3200
},
{
"epoch": 189.24,
"learning_rate": 1.5831111111111111e-06,
"loss": 0.0167,
"step": 3225
},
{
"epoch": 190.71,
"learning_rate": 1.560888888888889e-06,
"loss": 0.0159,
"step": 3250
},
{
"epoch": 192.18,
"learning_rate": 1.5386666666666666e-06,
"loss": 0.0137,
"step": 3275
},
{
"epoch": 193.65,
"learning_rate": 1.5164444444444444e-06,
"loss": 0.0122,
"step": 3300
},
{
"epoch": 195.12,
"learning_rate": 1.494222222222222e-06,
"loss": 0.0106,
"step": 3325
},
{
"epoch": 196.59,
"learning_rate": 1.4719999999999998e-06,
"loss": 0.0094,
"step": 3350
},
{
"epoch": 198.06,
"learning_rate": 1.4497777777777777e-06,
"loss": 0.009,
"step": 3375
},
{
"epoch": 199.53,
"learning_rate": 1.4275555555555555e-06,
"loss": 0.0104,
"step": 3400
},
{
"epoch": 201.0,
"learning_rate": 1.4053333333333333e-06,
"loss": 0.0069,
"step": 3425
},
{
"epoch": 202.47,
"learning_rate": 1.3848888888888889e-06,
"loss": 0.0073,
"step": 3450
},
{
"epoch": 203.94,
"learning_rate": 1.3626666666666667e-06,
"loss": 0.0073,
"step": 3475
},
{
"epoch": 205.41,
"learning_rate": 1.3404444444444445e-06,
"loss": 0.0063,
"step": 3500
},
{
"epoch": 206.88,
"learning_rate": 1.3182222222222221e-06,
"loss": 0.007,
"step": 3525
},
{
"epoch": 208.35,
"learning_rate": 1.296e-06,
"loss": 0.0061,
"step": 3550
},
{
"epoch": 209.82,
"learning_rate": 1.2737777777777776e-06,
"loss": 0.0053,
"step": 3575
},
{
"epoch": 211.29,
"learning_rate": 1.2515555555555554e-06,
"loss": 0.0056,
"step": 3600
},
{
"epoch": 212.76,
"learning_rate": 1.2293333333333334e-06,
"loss": 0.005,
"step": 3625
},
{
"epoch": 214.24,
"learning_rate": 1.207111111111111e-06,
"loss": 0.0047,
"step": 3650
},
{
"epoch": 215.71,
"learning_rate": 1.1848888888888889e-06,
"loss": 0.0052,
"step": 3675
},
{
"epoch": 217.18,
"learning_rate": 1.1626666666666667e-06,
"loss": 0.0044,
"step": 3700
},
{
"epoch": 218.65,
"learning_rate": 1.1404444444444443e-06,
"loss": 0.0046,
"step": 3725
},
{
"epoch": 220.12,
"learning_rate": 1.1182222222222221e-06,
"loss": 0.0045,
"step": 3750
},
{
"epoch": 221.59,
"learning_rate": 1.096e-06,
"loss": 0.0041,
"step": 3775
},
{
"epoch": 223.06,
"learning_rate": 1.0737777777777776e-06,
"loss": 0.0054,
"step": 3800
},
{
"epoch": 224.53,
"learning_rate": 1.0515555555555556e-06,
"loss": 0.0038,
"step": 3825
},
{
"epoch": 226.0,
"learning_rate": 1.0293333333333334e-06,
"loss": 0.0038,
"step": 3850
},
{
"epoch": 227.47,
"learning_rate": 1.007111111111111e-06,
"loss": 0.004,
"step": 3875
},
{
"epoch": 228.94,
"learning_rate": 9.848888888888889e-07,
"loss": 0.0036,
"step": 3900
},
{
"epoch": 230.41,
"learning_rate": 9.626666666666667e-07,
"loss": 0.0041,
"step": 3925
},
{
"epoch": 231.88,
"learning_rate": 9.404444444444443e-07,
"loss": 0.0032,
"step": 3950
},
{
"epoch": 233.35,
"learning_rate": 9.182222222222223e-07,
"loss": 0.0038,
"step": 3975
},
{
"epoch": 234.82,
"learning_rate": 8.96e-07,
"loss": 0.0043,
"step": 4000
},
{
"epoch": 234.82,
"eval_loss": 0.45361328125,
"eval_runtime": 157.593,
"eval_samples_per_second": 1.726,
"eval_steps_per_second": 0.108,
"eval_wer": 10.707652303120357,
"step": 4000
},
{
"epoch": 236.29,
"learning_rate": 8.737777777777777e-07,
"loss": 0.004,
"step": 4025
},
{
"epoch": 237.76,
"learning_rate": 8.515555555555555e-07,
"loss": 0.0029,
"step": 4050
},
{
"epoch": 239.24,
"learning_rate": 8.293333333333333e-07,
"loss": 0.0034,
"step": 4075
},
{
"epoch": 240.71,
"learning_rate": 8.071111111111111e-07,
"loss": 0.0032,
"step": 4100
},
{
"epoch": 242.18,
"learning_rate": 7.848888888888888e-07,
"loss": 0.003,
"step": 4125
},
{
"epoch": 243.65,
"learning_rate": 7.626666666666667e-07,
"loss": 0.0034,
"step": 4150
},
{
"epoch": 245.12,
"learning_rate": 7.404444444444444e-07,
"loss": 0.0032,
"step": 4175
},
{
"epoch": 246.59,
"learning_rate": 7.182222222222222e-07,
"loss": 0.0032,
"step": 4200
},
{
"epoch": 248.06,
"learning_rate": 6.959999999999999e-07,
"loss": 0.0028,
"step": 4225
},
{
"epoch": 249.53,
"learning_rate": 6.737777777777778e-07,
"loss": 0.0028,
"step": 4250
},
{
"epoch": 251.0,
"learning_rate": 6.515555555555555e-07,
"loss": 0.0025,
"step": 4275
},
{
"epoch": 252.47,
"learning_rate": 6.293333333333333e-07,
"loss": 0.0026,
"step": 4300
},
{
"epoch": 253.94,
"learning_rate": 6.071111111111111e-07,
"loss": 0.003,
"step": 4325
},
{
"epoch": 255.41,
"learning_rate": 5.848888888888889e-07,
"loss": 0.0026,
"step": 4350
},
{
"epoch": 256.88,
"learning_rate": 5.626666666666666e-07,
"loss": 0.0027,
"step": 4375
},
{
"epoch": 258.35,
"learning_rate": 5.404444444444443e-07,
"loss": 0.003,
"step": 4400
},
{
"epoch": 259.82,
"learning_rate": 5.182222222222223e-07,
"loss": 0.0027,
"step": 4425
},
{
"epoch": 261.29,
"learning_rate": 4.977777777777777e-07,
"loss": 0.0026,
"step": 4450
},
{
"epoch": 262.76,
"learning_rate": 4.7555555555555554e-07,
"loss": 0.0023,
"step": 4475
},
{
"epoch": 264.24,
"learning_rate": 4.5333333333333326e-07,
"loss": 0.0021,
"step": 4500
},
{
"epoch": 265.71,
"learning_rate": 4.311111111111111e-07,
"loss": 0.0022,
"step": 4525
},
{
"epoch": 267.18,
"learning_rate": 4.088888888888889e-07,
"loss": 0.0034,
"step": 4550
},
{
"epoch": 268.65,
"learning_rate": 3.8666666666666664e-07,
"loss": 0.0023,
"step": 4575
},
{
"epoch": 270.12,
"learning_rate": 3.6444444444444446e-07,
"loss": 0.0022,
"step": 4600
},
{
"epoch": 271.59,
"learning_rate": 3.422222222222222e-07,
"loss": 0.0022,
"step": 4625
},
{
"epoch": 273.06,
"learning_rate": 3.2e-07,
"loss": 0.0024,
"step": 4650
},
{
"epoch": 274.53,
"learning_rate": 2.9777777777777773e-07,
"loss": 0.0031,
"step": 4675
},
{
"epoch": 276.0,
"learning_rate": 2.7555555555555555e-07,
"loss": 0.0022,
"step": 4700
},
{
"epoch": 277.47,
"learning_rate": 2.533333333333333e-07,
"loss": 0.0022,
"step": 4725
},
{
"epoch": 278.94,
"learning_rate": 2.311111111111111e-07,
"loss": 0.0021,
"step": 4750
},
{
"epoch": 280.41,
"learning_rate": 2.088888888888889e-07,
"loss": 0.0023,
"step": 4775
},
{
"epoch": 281.88,
"learning_rate": 1.8666666666666667e-07,
"loss": 0.0025,
"step": 4800
},
{
"epoch": 283.35,
"learning_rate": 1.6444444444444444e-07,
"loss": 0.0022,
"step": 4825
},
{
"epoch": 284.82,
"learning_rate": 1.4222222222222222e-07,
"loss": 0.0022,
"step": 4850
},
{
"epoch": 286.29,
"learning_rate": 1.2e-07,
"loss": 0.0021,
"step": 4875
},
{
"epoch": 287.76,
"learning_rate": 9.777777777777778e-08,
"loss": 0.0023,
"step": 4900
},
{
"epoch": 289.24,
"learning_rate": 7.555555555555555e-08,
"loss": 0.002,
"step": 4925
},
{
"epoch": 290.71,
"learning_rate": 5.3333333333333334e-08,
"loss": 0.0025,
"step": 4950
},
{
"epoch": 292.18,
"learning_rate": 3.111111111111111e-08,
"loss": 0.002,
"step": 4975
},
{
"epoch": 293.65,
"learning_rate": 8.888888888888889e-09,
"loss": 0.0024,
"step": 5000
},
{
"epoch": 293.65,
"eval_loss": 0.465576171875,
"eval_runtime": 158.123,
"eval_samples_per_second": 1.72,
"eval_steps_per_second": 0.108,
"eval_wer": 10.642644873699851,
"step": 5000
},
{
"epoch": 293.65,
"step": 5000,
"total_flos": 1.555167153934817e+20,
"train_loss": 0.006518007612228393,
"train_runtime": 7844.7939,
"train_samples_per_second": 20.396,
"train_steps_per_second": 0.637
}
],
"max_steps": 5000,
"num_train_epochs": 295,
"total_flos": 1.555167153934817e+20,
"trial_name": null,
"trial_params": null
}