cv9-special-batch8-2lr-small / trainer_state.json
TheRains's picture
End of training
8d96059
raw
history blame
25.9 kB
{
"best_metric": 12.454566367609846,
"best_model_checkpoint": "./output/small/cv9-special-batch8-2lr-small/checkpoint-5000",
"epoch": 4.84027105517909,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 8.8e-07,
"loss": 1.2804,
"step": 25
},
{
"epoch": 0.05,
"learning_rate": 1.8800000000000002e-06,
"loss": 1.1403,
"step": 50
},
{
"epoch": 0.07,
"learning_rate": 2.88e-06,
"loss": 0.7479,
"step": 75
},
{
"epoch": 0.1,
"learning_rate": 3.88e-06,
"loss": 0.5417,
"step": 100
},
{
"epoch": 0.12,
"learning_rate": 4.880000000000001e-06,
"loss": 0.467,
"step": 125
},
{
"epoch": 0.15,
"learning_rate": 5.8800000000000005e-06,
"loss": 0.4304,
"step": 150
},
{
"epoch": 0.17,
"learning_rate": 6.88e-06,
"loss": 0.4375,
"step": 175
},
{
"epoch": 0.19,
"learning_rate": 7.88e-06,
"loss": 0.436,
"step": 200
},
{
"epoch": 0.22,
"learning_rate": 8.880000000000001e-06,
"loss": 0.3625,
"step": 225
},
{
"epoch": 0.24,
"learning_rate": 9.88e-06,
"loss": 0.3659,
"step": 250
},
{
"epoch": 0.27,
"learning_rate": 1.0880000000000001e-05,
"loss": 0.4069,
"step": 275
},
{
"epoch": 0.29,
"learning_rate": 1.188e-05,
"loss": 0.4047,
"step": 300
},
{
"epoch": 0.31,
"learning_rate": 1.2880000000000002e-05,
"loss": 0.3636,
"step": 325
},
{
"epoch": 0.34,
"learning_rate": 1.3880000000000001e-05,
"loss": 0.3875,
"step": 350
},
{
"epoch": 0.36,
"learning_rate": 1.4880000000000002e-05,
"loss": 0.4207,
"step": 375
},
{
"epoch": 0.39,
"learning_rate": 1.588e-05,
"loss": 0.4002,
"step": 400
},
{
"epoch": 0.41,
"learning_rate": 1.688e-05,
"loss": 0.379,
"step": 425
},
{
"epoch": 0.44,
"learning_rate": 1.788e-05,
"loss": 0.4032,
"step": 450
},
{
"epoch": 0.46,
"learning_rate": 1.8880000000000002e-05,
"loss": 0.3423,
"step": 475
},
{
"epoch": 0.48,
"learning_rate": 1.9880000000000003e-05,
"loss": 0.3676,
"step": 500
},
{
"epoch": 0.51,
"learning_rate": 1.9902222222222222e-05,
"loss": 0.4074,
"step": 525
},
{
"epoch": 0.53,
"learning_rate": 1.9791111111111113e-05,
"loss": 0.3594,
"step": 550
},
{
"epoch": 0.56,
"learning_rate": 1.968e-05,
"loss": 0.3835,
"step": 575
},
{
"epoch": 0.58,
"learning_rate": 1.956888888888889e-05,
"loss": 0.4108,
"step": 600
},
{
"epoch": 0.61,
"learning_rate": 1.945777777777778e-05,
"loss": 0.3742,
"step": 625
},
{
"epoch": 0.63,
"learning_rate": 1.934666666666667e-05,
"loss": 0.345,
"step": 650
},
{
"epoch": 0.65,
"learning_rate": 1.9235555555555557e-05,
"loss": 0.3801,
"step": 675
},
{
"epoch": 0.68,
"learning_rate": 1.9124444444444445e-05,
"loss": 0.3891,
"step": 700
},
{
"epoch": 0.7,
"learning_rate": 1.9013333333333333e-05,
"loss": 0.424,
"step": 725
},
{
"epoch": 0.73,
"learning_rate": 1.8902222222222225e-05,
"loss": 0.3617,
"step": 750
},
{
"epoch": 0.75,
"learning_rate": 1.8791111111111113e-05,
"loss": 0.3896,
"step": 775
},
{
"epoch": 0.77,
"learning_rate": 1.8680000000000004e-05,
"loss": 0.3367,
"step": 800
},
{
"epoch": 0.8,
"learning_rate": 1.856888888888889e-05,
"loss": 0.3482,
"step": 825
},
{
"epoch": 0.82,
"learning_rate": 1.845777777777778e-05,
"loss": 0.3481,
"step": 850
},
{
"epoch": 0.85,
"learning_rate": 1.834666666666667e-05,
"loss": 0.3334,
"step": 875
},
{
"epoch": 0.87,
"learning_rate": 1.8235555555555556e-05,
"loss": 0.3513,
"step": 900
},
{
"epoch": 0.9,
"learning_rate": 1.8124444444444448e-05,
"loss": 0.3815,
"step": 925
},
{
"epoch": 0.92,
"learning_rate": 1.8013333333333333e-05,
"loss": 0.3367,
"step": 950
},
{
"epoch": 0.94,
"learning_rate": 1.7902222222222224e-05,
"loss": 0.3788,
"step": 975
},
{
"epoch": 0.97,
"learning_rate": 1.7791111111111112e-05,
"loss": 0.3346,
"step": 1000
},
{
"epoch": 0.97,
"eval_loss": 0.29453718662261963,
"eval_runtime": 2785.773,
"eval_samples_per_second": 1.3,
"eval_steps_per_second": 0.325,
"eval_wer": 17.115251897860592,
"step": 1000
},
{
"epoch": 0.99,
"learning_rate": 1.768e-05,
"loss": 0.4037,
"step": 1025
},
{
"epoch": 1.02,
"learning_rate": 1.756888888888889e-05,
"loss": 0.2413,
"step": 1050
},
{
"epoch": 1.04,
"learning_rate": 1.745777777777778e-05,
"loss": 0.1541,
"step": 1075
},
{
"epoch": 1.06,
"learning_rate": 1.7346666666666668e-05,
"loss": 0.1564,
"step": 1100
},
{
"epoch": 1.09,
"learning_rate": 1.7235555555555556e-05,
"loss": 0.1682,
"step": 1125
},
{
"epoch": 1.11,
"learning_rate": 1.7124444444444447e-05,
"loss": 0.158,
"step": 1150
},
{
"epoch": 1.14,
"learning_rate": 1.7013333333333335e-05,
"loss": 0.1546,
"step": 1175
},
{
"epoch": 1.16,
"learning_rate": 1.6902222222222223e-05,
"loss": 0.1525,
"step": 1200
},
{
"epoch": 1.19,
"learning_rate": 1.6791111111111115e-05,
"loss": 0.135,
"step": 1225
},
{
"epoch": 1.21,
"learning_rate": 1.668e-05,
"loss": 0.1567,
"step": 1250
},
{
"epoch": 1.23,
"learning_rate": 1.656888888888889e-05,
"loss": 0.1639,
"step": 1275
},
{
"epoch": 1.26,
"learning_rate": 1.645777777777778e-05,
"loss": 0.1502,
"step": 1300
},
{
"epoch": 1.28,
"learning_rate": 1.6346666666666667e-05,
"loss": 0.1556,
"step": 1325
},
{
"epoch": 1.31,
"learning_rate": 1.623555555555556e-05,
"loss": 0.1454,
"step": 1350
},
{
"epoch": 1.33,
"learning_rate": 1.6124444444444443e-05,
"loss": 0.1723,
"step": 1375
},
{
"epoch": 1.36,
"learning_rate": 1.6013333333333335e-05,
"loss": 0.1418,
"step": 1400
},
{
"epoch": 1.38,
"learning_rate": 1.5902222222222223e-05,
"loss": 0.1589,
"step": 1425
},
{
"epoch": 1.4,
"learning_rate": 1.5791111111111114e-05,
"loss": 0.1529,
"step": 1450
},
{
"epoch": 1.43,
"learning_rate": 1.5680000000000002e-05,
"loss": 0.151,
"step": 1475
},
{
"epoch": 1.45,
"learning_rate": 1.556888888888889e-05,
"loss": 0.1903,
"step": 1500
},
{
"epoch": 1.48,
"learning_rate": 1.545777777777778e-05,
"loss": 0.1357,
"step": 1525
},
{
"epoch": 1.5,
"learning_rate": 1.5346666666666667e-05,
"loss": 0.1739,
"step": 1550
},
{
"epoch": 1.52,
"learning_rate": 1.5235555555555556e-05,
"loss": 0.1593,
"step": 1575
},
{
"epoch": 1.55,
"learning_rate": 1.5124444444444446e-05,
"loss": 0.1578,
"step": 1600
},
{
"epoch": 1.57,
"learning_rate": 1.5013333333333336e-05,
"loss": 0.1504,
"step": 1625
},
{
"epoch": 1.6,
"learning_rate": 1.4902222222222222e-05,
"loss": 0.1408,
"step": 1650
},
{
"epoch": 1.62,
"learning_rate": 1.4791111111111112e-05,
"loss": 0.172,
"step": 1675
},
{
"epoch": 1.65,
"learning_rate": 1.4680000000000002e-05,
"loss": 0.173,
"step": 1700
},
{
"epoch": 1.67,
"learning_rate": 1.456888888888889e-05,
"loss": 0.1618,
"step": 1725
},
{
"epoch": 1.69,
"learning_rate": 1.445777777777778e-05,
"loss": 0.1406,
"step": 1750
},
{
"epoch": 1.72,
"learning_rate": 1.434666666666667e-05,
"loss": 0.1637,
"step": 1775
},
{
"epoch": 1.74,
"learning_rate": 1.4235555555555556e-05,
"loss": 0.1661,
"step": 1800
},
{
"epoch": 1.77,
"learning_rate": 1.4124444444444445e-05,
"loss": 0.153,
"step": 1825
},
{
"epoch": 1.79,
"learning_rate": 1.4013333333333334e-05,
"loss": 0.1421,
"step": 1850
},
{
"epoch": 1.82,
"learning_rate": 1.3902222222222223e-05,
"loss": 0.1561,
"step": 1875
},
{
"epoch": 1.84,
"learning_rate": 1.3791111111111113e-05,
"loss": 0.1402,
"step": 1900
},
{
"epoch": 1.86,
"learning_rate": 1.3680000000000003e-05,
"loss": 0.1485,
"step": 1925
},
{
"epoch": 1.89,
"learning_rate": 1.356888888888889e-05,
"loss": 0.1627,
"step": 1950
},
{
"epoch": 1.91,
"learning_rate": 1.3457777777777779e-05,
"loss": 0.1467,
"step": 1975
},
{
"epoch": 1.94,
"learning_rate": 1.3346666666666667e-05,
"loss": 0.1474,
"step": 2000
},
{
"epoch": 1.94,
"eval_loss": 0.2713002860546112,
"eval_runtime": 2604.138,
"eval_samples_per_second": 1.391,
"eval_steps_per_second": 0.348,
"eval_wer": 14.639981596503334,
"step": 2000
},
{
"epoch": 1.96,
"learning_rate": 1.3235555555555557e-05,
"loss": 0.1473,
"step": 2025
},
{
"epoch": 1.98,
"learning_rate": 1.3124444444444447e-05,
"loss": 0.1795,
"step": 2050
},
{
"epoch": 2.01,
"learning_rate": 1.3013333333333333e-05,
"loss": 0.1153,
"step": 2075
},
{
"epoch": 2.03,
"learning_rate": 1.2902222222222223e-05,
"loss": 0.045,
"step": 2100
},
{
"epoch": 2.06,
"learning_rate": 1.2791111111111112e-05,
"loss": 0.0451,
"step": 2125
},
{
"epoch": 2.08,
"learning_rate": 1.268e-05,
"loss": 0.0524,
"step": 2150
},
{
"epoch": 2.11,
"learning_rate": 1.256888888888889e-05,
"loss": 0.0509,
"step": 2175
},
{
"epoch": 2.13,
"learning_rate": 1.245777777777778e-05,
"loss": 0.0506,
"step": 2200
},
{
"epoch": 2.15,
"learning_rate": 1.2346666666666666e-05,
"loss": 0.0497,
"step": 2225
},
{
"epoch": 2.18,
"learning_rate": 1.2235555555555556e-05,
"loss": 0.0536,
"step": 2250
},
{
"epoch": 2.2,
"learning_rate": 1.2124444444444446e-05,
"loss": 0.0535,
"step": 2275
},
{
"epoch": 2.23,
"learning_rate": 1.2013333333333334e-05,
"loss": 0.0538,
"step": 2300
},
{
"epoch": 2.25,
"learning_rate": 1.1902222222222224e-05,
"loss": 0.0507,
"step": 2325
},
{
"epoch": 2.27,
"learning_rate": 1.1791111111111114e-05,
"loss": 0.0485,
"step": 2350
},
{
"epoch": 2.3,
"learning_rate": 1.168e-05,
"loss": 0.0487,
"step": 2375
},
{
"epoch": 2.32,
"learning_rate": 1.156888888888889e-05,
"loss": 0.0635,
"step": 2400
},
{
"epoch": 2.35,
"learning_rate": 1.145777777777778e-05,
"loss": 0.0555,
"step": 2425
},
{
"epoch": 2.37,
"learning_rate": 1.1346666666666668e-05,
"loss": 0.0561,
"step": 2450
},
{
"epoch": 2.4,
"learning_rate": 1.1235555555555557e-05,
"loss": 0.0728,
"step": 2475
},
{
"epoch": 2.42,
"learning_rate": 1.1124444444444444e-05,
"loss": 0.0446,
"step": 2500
},
{
"epoch": 2.44,
"learning_rate": 1.1013333333333333e-05,
"loss": 0.0494,
"step": 2525
},
{
"epoch": 2.47,
"learning_rate": 1.0902222222222223e-05,
"loss": 0.0503,
"step": 2550
},
{
"epoch": 2.49,
"learning_rate": 1.0791111111111113e-05,
"loss": 0.0498,
"step": 2575
},
{
"epoch": 2.52,
"learning_rate": 1.0680000000000001e-05,
"loss": 0.0606,
"step": 2600
},
{
"epoch": 2.54,
"learning_rate": 1.0568888888888891e-05,
"loss": 0.0486,
"step": 2625
},
{
"epoch": 2.57,
"learning_rate": 1.0457777777777777e-05,
"loss": 0.0656,
"step": 2650
},
{
"epoch": 2.59,
"learning_rate": 1.0346666666666667e-05,
"loss": 0.0494,
"step": 2675
},
{
"epoch": 2.61,
"learning_rate": 1.0235555555555557e-05,
"loss": 0.0569,
"step": 2700
},
{
"epoch": 2.64,
"learning_rate": 1.0124444444444445e-05,
"loss": 0.0464,
"step": 2725
},
{
"epoch": 2.66,
"learning_rate": 1.0013333333333335e-05,
"loss": 0.0635,
"step": 2750
},
{
"epoch": 2.69,
"learning_rate": 9.902222222222223e-06,
"loss": 0.0499,
"step": 2775
},
{
"epoch": 2.71,
"learning_rate": 9.791111111111112e-06,
"loss": 0.0484,
"step": 2800
},
{
"epoch": 2.73,
"learning_rate": 9.68e-06,
"loss": 0.0484,
"step": 2825
},
{
"epoch": 2.76,
"learning_rate": 9.56888888888889e-06,
"loss": 0.0515,
"step": 2850
},
{
"epoch": 2.78,
"learning_rate": 9.457777777777778e-06,
"loss": 0.0473,
"step": 2875
},
{
"epoch": 2.81,
"learning_rate": 9.346666666666666e-06,
"loss": 0.0552,
"step": 2900
},
{
"epoch": 2.83,
"learning_rate": 9.235555555555556e-06,
"loss": 0.0477,
"step": 2925
},
{
"epoch": 2.86,
"learning_rate": 9.124444444444444e-06,
"loss": 0.0504,
"step": 2950
},
{
"epoch": 2.88,
"learning_rate": 9.013333333333334e-06,
"loss": 0.0527,
"step": 2975
},
{
"epoch": 2.9,
"learning_rate": 8.902222222222224e-06,
"loss": 0.052,
"step": 3000
},
{
"epoch": 2.9,
"eval_loss": 0.27132648229599,
"eval_runtime": 2600.2688,
"eval_samples_per_second": 1.393,
"eval_steps_per_second": 0.348,
"eval_wer": 14.124683689901083,
"step": 3000
},
{
"epoch": 2.93,
"learning_rate": 8.791111111111112e-06,
"loss": 0.0431,
"step": 3025
},
{
"epoch": 2.95,
"learning_rate": 8.68e-06,
"loss": 0.0576,
"step": 3050
},
{
"epoch": 2.98,
"learning_rate": 8.56888888888889e-06,
"loss": 0.045,
"step": 3075
},
{
"epoch": 3.0,
"learning_rate": 8.457777777777778e-06,
"loss": 0.0468,
"step": 3100
},
{
"epoch": 3.03,
"learning_rate": 8.346666666666668e-06,
"loss": 0.0196,
"step": 3125
},
{
"epoch": 3.05,
"learning_rate": 8.235555555555557e-06,
"loss": 0.0201,
"step": 3150
},
{
"epoch": 3.07,
"learning_rate": 8.124444444444445e-06,
"loss": 0.0224,
"step": 3175
},
{
"epoch": 3.1,
"learning_rate": 8.013333333333333e-06,
"loss": 0.0251,
"step": 3200
},
{
"epoch": 3.12,
"learning_rate": 7.902222222222223e-06,
"loss": 0.0287,
"step": 3225
},
{
"epoch": 3.15,
"learning_rate": 7.791111111111111e-06,
"loss": 0.0164,
"step": 3250
},
{
"epoch": 3.17,
"learning_rate": 7.680000000000001e-06,
"loss": 0.0215,
"step": 3275
},
{
"epoch": 3.19,
"learning_rate": 7.56888888888889e-06,
"loss": 0.0178,
"step": 3300
},
{
"epoch": 3.22,
"learning_rate": 7.457777777777778e-06,
"loss": 0.0162,
"step": 3325
},
{
"epoch": 3.24,
"learning_rate": 7.346666666666668e-06,
"loss": 0.0125,
"step": 3350
},
{
"epoch": 3.27,
"learning_rate": 7.235555555555556e-06,
"loss": 0.0252,
"step": 3375
},
{
"epoch": 3.29,
"learning_rate": 7.124444444444445e-06,
"loss": 0.0193,
"step": 3400
},
{
"epoch": 3.32,
"learning_rate": 7.0133333333333345e-06,
"loss": 0.0144,
"step": 3425
},
{
"epoch": 3.34,
"learning_rate": 6.902222222222223e-06,
"loss": 0.0133,
"step": 3450
},
{
"epoch": 3.36,
"learning_rate": 6.7911111111111115e-06,
"loss": 0.0187,
"step": 3475
},
{
"epoch": 3.39,
"learning_rate": 6.680000000000001e-06,
"loss": 0.0208,
"step": 3500
},
{
"epoch": 3.41,
"learning_rate": 6.568888888888889e-06,
"loss": 0.0141,
"step": 3525
},
{
"epoch": 3.44,
"learning_rate": 6.457777777777778e-06,
"loss": 0.0165,
"step": 3550
},
{
"epoch": 3.46,
"learning_rate": 6.346666666666668e-06,
"loss": 0.0157,
"step": 3575
},
{
"epoch": 3.48,
"learning_rate": 6.235555555555556e-06,
"loss": 0.0188,
"step": 3600
},
{
"epoch": 3.51,
"learning_rate": 6.124444444444445e-06,
"loss": 0.0172,
"step": 3625
},
{
"epoch": 3.53,
"learning_rate": 6.013333333333335e-06,
"loss": 0.0189,
"step": 3650
},
{
"epoch": 3.56,
"learning_rate": 5.902222222222223e-06,
"loss": 0.0176,
"step": 3675
},
{
"epoch": 3.58,
"learning_rate": 5.791111111111112e-06,
"loss": 0.0202,
"step": 3700
},
{
"epoch": 3.61,
"learning_rate": 5.68e-06,
"loss": 0.0155,
"step": 3725
},
{
"epoch": 3.63,
"learning_rate": 5.56888888888889e-06,
"loss": 0.0176,
"step": 3750
},
{
"epoch": 3.65,
"learning_rate": 5.4577777777777785e-06,
"loss": 0.0167,
"step": 3775
},
{
"epoch": 3.68,
"learning_rate": 5.346666666666667e-06,
"loss": 0.0143,
"step": 3800
},
{
"epoch": 3.7,
"learning_rate": 5.235555555555556e-06,
"loss": 0.0196,
"step": 3825
},
{
"epoch": 3.73,
"learning_rate": 5.124444444444445e-06,
"loss": 0.0228,
"step": 3850
},
{
"epoch": 3.75,
"learning_rate": 5.013333333333333e-06,
"loss": 0.016,
"step": 3875
},
{
"epoch": 3.78,
"learning_rate": 4.902222222222222e-06,
"loss": 0.0124,
"step": 3900
},
{
"epoch": 3.8,
"learning_rate": 4.791111111111111e-06,
"loss": 0.0139,
"step": 3925
},
{
"epoch": 3.82,
"learning_rate": 4.680000000000001e-06,
"loss": 0.0126,
"step": 3950
},
{
"epoch": 3.85,
"learning_rate": 4.568888888888889e-06,
"loss": 0.0154,
"step": 3975
},
{
"epoch": 3.87,
"learning_rate": 4.457777777777778e-06,
"loss": 0.0142,
"step": 4000
},
{
"epoch": 3.87,
"eval_loss": 0.2881431579589844,
"eval_runtime": 2586.854,
"eval_samples_per_second": 1.4,
"eval_steps_per_second": 0.35,
"eval_wer": 12.82723717506326,
"step": 4000
},
{
"epoch": 3.9,
"learning_rate": 4.346666666666667e-06,
"loss": 0.0147,
"step": 4025
},
{
"epoch": 3.92,
"learning_rate": 4.235555555555556e-06,
"loss": 0.0153,
"step": 4050
},
{
"epoch": 3.94,
"learning_rate": 4.124444444444445e-06,
"loss": 0.0133,
"step": 4075
},
{
"epoch": 3.97,
"learning_rate": 4.013333333333334e-06,
"loss": 0.0122,
"step": 4100
},
{
"epoch": 3.99,
"learning_rate": 3.9022222222222225e-06,
"loss": 0.0141,
"step": 4125
},
{
"epoch": 4.02,
"learning_rate": 3.7911111111111114e-06,
"loss": 0.0072,
"step": 4150
},
{
"epoch": 4.04,
"learning_rate": 3.6800000000000003e-06,
"loss": 0.0051,
"step": 4175
},
{
"epoch": 4.07,
"learning_rate": 3.568888888888889e-06,
"loss": 0.0051,
"step": 4200
},
{
"epoch": 4.09,
"learning_rate": 3.457777777777778e-06,
"loss": 0.0062,
"step": 4225
},
{
"epoch": 4.11,
"learning_rate": 3.346666666666667e-06,
"loss": 0.0053,
"step": 4250
},
{
"epoch": 4.14,
"learning_rate": 3.2355555555555556e-06,
"loss": 0.0059,
"step": 4275
},
{
"epoch": 4.16,
"learning_rate": 3.124444444444445e-06,
"loss": 0.0046,
"step": 4300
},
{
"epoch": 4.19,
"learning_rate": 3.013333333333334e-06,
"loss": 0.0034,
"step": 4325
},
{
"epoch": 4.21,
"learning_rate": 2.9022222222222223e-06,
"loss": 0.0058,
"step": 4350
},
{
"epoch": 4.24,
"learning_rate": 2.7911111111111113e-06,
"loss": 0.0041,
"step": 4375
},
{
"epoch": 4.26,
"learning_rate": 2.68e-06,
"loss": 0.0041,
"step": 4400
},
{
"epoch": 4.28,
"learning_rate": 2.568888888888889e-06,
"loss": 0.003,
"step": 4425
},
{
"epoch": 4.31,
"learning_rate": 2.457777777777778e-06,
"loss": 0.0038,
"step": 4450
},
{
"epoch": 4.33,
"learning_rate": 2.346666666666667e-06,
"loss": 0.0054,
"step": 4475
},
{
"epoch": 4.36,
"learning_rate": 2.235555555555556e-06,
"loss": 0.0045,
"step": 4500
},
{
"epoch": 4.38,
"learning_rate": 2.1244444444444443e-06,
"loss": 0.0034,
"step": 4525
},
{
"epoch": 4.4,
"learning_rate": 2.0133333333333337e-06,
"loss": 0.0031,
"step": 4550
},
{
"epoch": 4.43,
"learning_rate": 1.9022222222222222e-06,
"loss": 0.0036,
"step": 4575
},
{
"epoch": 4.45,
"learning_rate": 1.7911111111111113e-06,
"loss": 0.007,
"step": 4600
},
{
"epoch": 4.48,
"learning_rate": 1.6800000000000002e-06,
"loss": 0.004,
"step": 4625
},
{
"epoch": 4.5,
"learning_rate": 1.568888888888889e-06,
"loss": 0.0034,
"step": 4650
},
{
"epoch": 4.53,
"learning_rate": 1.457777777777778e-06,
"loss": 0.0071,
"step": 4675
},
{
"epoch": 4.55,
"learning_rate": 1.3466666666666668e-06,
"loss": 0.0066,
"step": 4700
},
{
"epoch": 4.57,
"learning_rate": 1.2355555555555557e-06,
"loss": 0.0052,
"step": 4725
},
{
"epoch": 4.6,
"learning_rate": 1.1244444444444446e-06,
"loss": 0.0036,
"step": 4750
},
{
"epoch": 4.62,
"learning_rate": 1.0133333333333333e-06,
"loss": 0.0058,
"step": 4775
},
{
"epoch": 4.65,
"learning_rate": 9.022222222222222e-07,
"loss": 0.0054,
"step": 4800
},
{
"epoch": 4.67,
"learning_rate": 7.911111111111111e-07,
"loss": 0.0028,
"step": 4825
},
{
"epoch": 4.7,
"learning_rate": 6.800000000000001e-07,
"loss": 0.0034,
"step": 4850
},
{
"epoch": 4.72,
"learning_rate": 5.68888888888889e-07,
"loss": 0.0049,
"step": 4875
},
{
"epoch": 4.74,
"learning_rate": 4.5777777777777784e-07,
"loss": 0.0035,
"step": 4900
},
{
"epoch": 4.77,
"learning_rate": 3.466666666666667e-07,
"loss": 0.0038,
"step": 4925
},
{
"epoch": 4.79,
"learning_rate": 2.3555555555555556e-07,
"loss": 0.0039,
"step": 4950
},
{
"epoch": 4.82,
"learning_rate": 1.2444444444444446e-07,
"loss": 0.0038,
"step": 4975
},
{
"epoch": 4.84,
"learning_rate": 1.3333333333333334e-08,
"loss": 0.0034,
"step": 5000
},
{
"epoch": 4.84,
"eval_loss": 0.3003479242324829,
"eval_runtime": 2589.8653,
"eval_samples_per_second": 1.399,
"eval_steps_per_second": 0.35,
"eval_wer": 12.454566367609846,
"step": 5000
},
{
"epoch": 4.84,
"step": 5000,
"total_flos": 1.153995298799616e+19,
"train_loss": 0.13730322830528022,
"train_runtime": 39511.7663,
"train_samples_per_second": 1.012,
"train_steps_per_second": 0.127
}
],
"max_steps": 5000,
"num_train_epochs": 5,
"total_flos": 1.153995298799616e+19,
"trial_name": null,
"trial_params": null
}