cv9-special-batch12-lr4-small / trainer_state.json
TheRains's picture
End of training
dfa4f88
{
"best_metric": 17.593742811134117,
"best_model_checkpoint": "./output/small/cv9-special-batch12-lr4-small/checkpoint-5000",
"epoch": 7.256894049346879,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 4.4e-06,
"loss": 1.191,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 9.4e-06,
"loss": 0.6036,
"step": 50
},
{
"epoch": 0.11,
"learning_rate": 1.44e-05,
"loss": 0.4394,
"step": 75
},
{
"epoch": 0.15,
"learning_rate": 1.94e-05,
"loss": 0.42,
"step": 100
},
{
"epoch": 0.18,
"learning_rate": 2.44e-05,
"loss": 0.4484,
"step": 125
},
{
"epoch": 0.22,
"learning_rate": 2.94e-05,
"loss": 0.3933,
"step": 150
},
{
"epoch": 0.25,
"learning_rate": 3.4399999999999996e-05,
"loss": 0.4136,
"step": 175
},
{
"epoch": 0.29,
"learning_rate": 3.94e-05,
"loss": 0.4686,
"step": 200
},
{
"epoch": 0.33,
"learning_rate": 4.44e-05,
"loss": 0.4438,
"step": 225
},
{
"epoch": 0.36,
"learning_rate": 4.94e-05,
"loss": 0.5118,
"step": 250
},
{
"epoch": 0.4,
"learning_rate": 5.440000000000001e-05,
"loss": 0.4983,
"step": 275
},
{
"epoch": 0.44,
"learning_rate": 5.94e-05,
"loss": 0.4995,
"step": 300
},
{
"epoch": 0.47,
"learning_rate": 6.440000000000001e-05,
"loss": 0.4988,
"step": 325
},
{
"epoch": 0.51,
"learning_rate": 6.939999999999999e-05,
"loss": 0.5419,
"step": 350
},
{
"epoch": 0.54,
"learning_rate": 7.44e-05,
"loss": 0.5401,
"step": 375
},
{
"epoch": 0.58,
"learning_rate": 7.94e-05,
"loss": 0.5586,
"step": 400
},
{
"epoch": 0.62,
"learning_rate": 8.44e-05,
"loss": 0.5791,
"step": 425
},
{
"epoch": 0.65,
"learning_rate": 8.94e-05,
"loss": 0.6103,
"step": 450
},
{
"epoch": 0.69,
"learning_rate": 9.44e-05,
"loss": 0.6625,
"step": 475
},
{
"epoch": 0.73,
"learning_rate": 9.94e-05,
"loss": 0.6622,
"step": 500
},
{
"epoch": 0.76,
"learning_rate": 9.95111111111111e-05,
"loss": 0.6632,
"step": 525
},
{
"epoch": 0.8,
"learning_rate": 9.895555555555556e-05,
"loss": 0.657,
"step": 550
},
{
"epoch": 0.83,
"learning_rate": 9.84e-05,
"loss": 0.6526,
"step": 575
},
{
"epoch": 0.87,
"learning_rate": 9.784444444444444e-05,
"loss": 0.6615,
"step": 600
},
{
"epoch": 0.91,
"learning_rate": 9.72888888888889e-05,
"loss": 0.6762,
"step": 625
},
{
"epoch": 0.94,
"learning_rate": 9.673333333333334e-05,
"loss": 0.6791,
"step": 650
},
{
"epoch": 0.98,
"learning_rate": 9.617777777777778e-05,
"loss": 0.6629,
"step": 675
},
{
"epoch": 1.02,
"learning_rate": 9.562222222222222e-05,
"loss": 0.5772,
"step": 700
},
{
"epoch": 1.05,
"learning_rate": 9.506666666666667e-05,
"loss": 0.3396,
"step": 725
},
{
"epoch": 1.09,
"learning_rate": 9.451111111111112e-05,
"loss": 0.3575,
"step": 750
},
{
"epoch": 1.12,
"learning_rate": 9.395555555555557e-05,
"loss": 0.3685,
"step": 775
},
{
"epoch": 1.16,
"learning_rate": 9.340000000000001e-05,
"loss": 0.3711,
"step": 800
},
{
"epoch": 1.2,
"learning_rate": 9.284444444444444e-05,
"loss": 0.3574,
"step": 825
},
{
"epoch": 1.23,
"learning_rate": 9.22888888888889e-05,
"loss": 0.3793,
"step": 850
},
{
"epoch": 1.27,
"learning_rate": 9.173333333333333e-05,
"loss": 0.375,
"step": 875
},
{
"epoch": 1.31,
"learning_rate": 9.117777777777778e-05,
"loss": 0.4094,
"step": 900
},
{
"epoch": 1.34,
"learning_rate": 9.062222222222223e-05,
"loss": 0.3846,
"step": 925
},
{
"epoch": 1.38,
"learning_rate": 9.006666666666667e-05,
"loss": 0.3782,
"step": 950
},
{
"epoch": 1.42,
"learning_rate": 8.95111111111111e-05,
"loss": 0.3584,
"step": 975
},
{
"epoch": 1.45,
"learning_rate": 8.895555555555556e-05,
"loss": 0.3743,
"step": 1000
},
{
"epoch": 1.45,
"eval_loss": 0.5497691631317139,
"eval_runtime": 2341.8245,
"eval_samples_per_second": 1.547,
"eval_steps_per_second": 0.258,
"eval_wer": 28.27237175063262,
"step": 1000
},
{
"epoch": 1.49,
"learning_rate": 8.840000000000001e-05,
"loss": 0.3457,
"step": 1025
},
{
"epoch": 1.52,
"learning_rate": 8.784444444444446e-05,
"loss": 0.3747,
"step": 1050
},
{
"epoch": 1.56,
"learning_rate": 8.72888888888889e-05,
"loss": 0.3429,
"step": 1075
},
{
"epoch": 1.6,
"learning_rate": 8.673333333333333e-05,
"loss": 0.3562,
"step": 1100
},
{
"epoch": 1.63,
"learning_rate": 8.617777777777778e-05,
"loss": 0.3805,
"step": 1125
},
{
"epoch": 1.67,
"learning_rate": 8.562222222222222e-05,
"loss": 0.3905,
"step": 1150
},
{
"epoch": 1.71,
"learning_rate": 8.506666666666667e-05,
"loss": 0.3792,
"step": 1175
},
{
"epoch": 1.74,
"learning_rate": 8.451111111111112e-05,
"loss": 0.3614,
"step": 1200
},
{
"epoch": 1.78,
"learning_rate": 8.395555555555557e-05,
"loss": 0.3787,
"step": 1225
},
{
"epoch": 1.81,
"learning_rate": 8.34e-05,
"loss": 0.3381,
"step": 1250
},
{
"epoch": 1.85,
"learning_rate": 8.284444444444444e-05,
"loss": 0.3715,
"step": 1275
},
{
"epoch": 1.89,
"learning_rate": 8.22888888888889e-05,
"loss": 0.3818,
"step": 1300
},
{
"epoch": 1.92,
"learning_rate": 8.173333333333335e-05,
"loss": 0.3498,
"step": 1325
},
{
"epoch": 1.96,
"learning_rate": 8.117777777777778e-05,
"loss": 0.3778,
"step": 1350
},
{
"epoch": 2.0,
"learning_rate": 8.062222222222222e-05,
"loss": 0.3512,
"step": 1375
},
{
"epoch": 2.03,
"learning_rate": 8.006666666666667e-05,
"loss": 0.1781,
"step": 1400
},
{
"epoch": 2.07,
"learning_rate": 7.951111111111111e-05,
"loss": 0.1771,
"step": 1425
},
{
"epoch": 2.1,
"learning_rate": 7.895555555555556e-05,
"loss": 0.1765,
"step": 1450
},
{
"epoch": 2.14,
"learning_rate": 7.840000000000001e-05,
"loss": 0.1813,
"step": 1475
},
{
"epoch": 2.18,
"learning_rate": 7.784444444444446e-05,
"loss": 0.1741,
"step": 1500
},
{
"epoch": 2.21,
"learning_rate": 7.728888888888888e-05,
"loss": 0.1679,
"step": 1525
},
{
"epoch": 2.25,
"learning_rate": 7.673333333333333e-05,
"loss": 0.1941,
"step": 1550
},
{
"epoch": 2.29,
"learning_rate": 7.617777777777778e-05,
"loss": 0.1667,
"step": 1575
},
{
"epoch": 2.32,
"learning_rate": 7.562222222222223e-05,
"loss": 0.1783,
"step": 1600
},
{
"epoch": 2.36,
"learning_rate": 7.506666666666667e-05,
"loss": 0.1625,
"step": 1625
},
{
"epoch": 2.39,
"learning_rate": 7.451111111111111e-05,
"loss": 0.1849,
"step": 1650
},
{
"epoch": 2.43,
"learning_rate": 7.395555555555556e-05,
"loss": 0.1872,
"step": 1675
},
{
"epoch": 2.47,
"learning_rate": 7.340000000000001e-05,
"loss": 0.1949,
"step": 1700
},
{
"epoch": 2.5,
"learning_rate": 7.284444444444445e-05,
"loss": 0.1759,
"step": 1725
},
{
"epoch": 2.54,
"learning_rate": 7.22888888888889e-05,
"loss": 0.182,
"step": 1750
},
{
"epoch": 2.58,
"learning_rate": 7.173333333333335e-05,
"loss": 0.1907,
"step": 1775
},
{
"epoch": 2.61,
"learning_rate": 7.117777777777777e-05,
"loss": 0.1982,
"step": 1800
},
{
"epoch": 2.65,
"learning_rate": 7.062222222222222e-05,
"loss": 0.1707,
"step": 1825
},
{
"epoch": 2.69,
"learning_rate": 7.006666666666667e-05,
"loss": 0.1818,
"step": 1850
},
{
"epoch": 2.72,
"learning_rate": 6.951111111111112e-05,
"loss": 0.1791,
"step": 1875
},
{
"epoch": 2.76,
"learning_rate": 6.895555555555556e-05,
"loss": 0.1866,
"step": 1900
},
{
"epoch": 2.79,
"learning_rate": 6.840000000000001e-05,
"loss": 0.1761,
"step": 1925
},
{
"epoch": 2.83,
"learning_rate": 6.784444444444445e-05,
"loss": 0.1766,
"step": 1950
},
{
"epoch": 2.87,
"learning_rate": 6.72888888888889e-05,
"loss": 0.1599,
"step": 1975
},
{
"epoch": 2.9,
"learning_rate": 6.673333333333333e-05,
"loss": 0.1633,
"step": 2000
},
{
"epoch": 2.9,
"eval_loss": 0.5009704828262329,
"eval_runtime": 2334.1804,
"eval_samples_per_second": 1.552,
"eval_steps_per_second": 0.259,
"eval_wer": 25.152979066022546,
"step": 2000
},
{
"epoch": 2.94,
"learning_rate": 6.617777777777778e-05,
"loss": 0.1863,
"step": 2025
},
{
"epoch": 2.98,
"learning_rate": 6.562222222222223e-05,
"loss": 0.182,
"step": 2050
},
{
"epoch": 3.01,
"learning_rate": 6.506666666666666e-05,
"loss": 0.149,
"step": 2075
},
{
"epoch": 3.05,
"learning_rate": 6.451111111111111e-05,
"loss": 0.0853,
"step": 2100
},
{
"epoch": 3.08,
"learning_rate": 6.395555555555556e-05,
"loss": 0.0913,
"step": 2125
},
{
"epoch": 3.12,
"learning_rate": 6.340000000000001e-05,
"loss": 0.0969,
"step": 2150
},
{
"epoch": 3.16,
"learning_rate": 6.284444444444445e-05,
"loss": 0.0838,
"step": 2175
},
{
"epoch": 3.19,
"learning_rate": 6.22888888888889e-05,
"loss": 0.0936,
"step": 2200
},
{
"epoch": 3.23,
"learning_rate": 6.173333333333333e-05,
"loss": 0.0939,
"step": 2225
},
{
"epoch": 3.27,
"learning_rate": 6.117777777777778e-05,
"loss": 0.0937,
"step": 2250
},
{
"epoch": 3.3,
"learning_rate": 6.062222222222222e-05,
"loss": 0.0939,
"step": 2275
},
{
"epoch": 3.34,
"learning_rate": 6.006666666666667e-05,
"loss": 0.0963,
"step": 2300
},
{
"epoch": 3.37,
"learning_rate": 5.9511111111111116e-05,
"loss": 0.0912,
"step": 2325
},
{
"epoch": 3.41,
"learning_rate": 5.8955555555555566e-05,
"loss": 0.1011,
"step": 2350
},
{
"epoch": 3.45,
"learning_rate": 5.8399999999999997e-05,
"loss": 0.1081,
"step": 2375
},
{
"epoch": 3.48,
"learning_rate": 5.784444444444445e-05,
"loss": 0.0984,
"step": 2400
},
{
"epoch": 3.52,
"learning_rate": 5.728888888888889e-05,
"loss": 0.0938,
"step": 2425
},
{
"epoch": 3.56,
"learning_rate": 5.673333333333334e-05,
"loss": 0.0957,
"step": 2450
},
{
"epoch": 3.59,
"learning_rate": 5.6177777777777785e-05,
"loss": 0.0971,
"step": 2475
},
{
"epoch": 3.63,
"learning_rate": 5.562222222222222e-05,
"loss": 0.0863,
"step": 2500
},
{
"epoch": 3.66,
"learning_rate": 5.5066666666666666e-05,
"loss": 0.0897,
"step": 2525
},
{
"epoch": 3.7,
"learning_rate": 5.451111111111111e-05,
"loss": 0.0954,
"step": 2550
},
{
"epoch": 3.74,
"learning_rate": 5.395555555555556e-05,
"loss": 0.0792,
"step": 2575
},
{
"epoch": 3.77,
"learning_rate": 5.3400000000000004e-05,
"loss": 0.0961,
"step": 2600
},
{
"epoch": 3.81,
"learning_rate": 5.2844444444444454e-05,
"loss": 0.0887,
"step": 2625
},
{
"epoch": 3.85,
"learning_rate": 5.2288888888888884e-05,
"loss": 0.0887,
"step": 2650
},
{
"epoch": 3.88,
"learning_rate": 5.1733333333333335e-05,
"loss": 0.0843,
"step": 2675
},
{
"epoch": 3.92,
"learning_rate": 5.117777777777778e-05,
"loss": 0.0798,
"step": 2700
},
{
"epoch": 3.96,
"learning_rate": 5.062222222222223e-05,
"loss": 0.0869,
"step": 2725
},
{
"epoch": 3.99,
"learning_rate": 5.006666666666667e-05,
"loss": 0.085,
"step": 2750
},
{
"epoch": 4.03,
"learning_rate": 4.951111111111112e-05,
"loss": 0.0477,
"step": 2775
},
{
"epoch": 4.06,
"learning_rate": 4.895555555555556e-05,
"loss": 0.0518,
"step": 2800
},
{
"epoch": 4.1,
"learning_rate": 4.8400000000000004e-05,
"loss": 0.0501,
"step": 2825
},
{
"epoch": 4.14,
"learning_rate": 4.784444444444445e-05,
"loss": 0.0394,
"step": 2850
},
{
"epoch": 4.17,
"learning_rate": 4.728888888888889e-05,
"loss": 0.046,
"step": 2875
},
{
"epoch": 4.21,
"learning_rate": 4.6733333333333335e-05,
"loss": 0.0405,
"step": 2900
},
{
"epoch": 4.25,
"learning_rate": 4.617777777777778e-05,
"loss": 0.0525,
"step": 2925
},
{
"epoch": 4.28,
"learning_rate": 4.562222222222222e-05,
"loss": 0.0443,
"step": 2950
},
{
"epoch": 4.32,
"learning_rate": 4.5066666666666667e-05,
"loss": 0.0444,
"step": 2975
},
{
"epoch": 4.35,
"learning_rate": 4.451111111111112e-05,
"loss": 0.0505,
"step": 3000
},
{
"epoch": 4.35,
"eval_loss": 0.5048635005950928,
"eval_runtime": 2311.8612,
"eval_samples_per_second": 1.567,
"eval_steps_per_second": 0.261,
"eval_wer": 22.167011732229124,
"step": 3000
},
{
"epoch": 4.39,
"learning_rate": 4.3955555555555554e-05,
"loss": 0.0371,
"step": 3025
},
{
"epoch": 4.43,
"learning_rate": 4.3400000000000005e-05,
"loss": 0.0496,
"step": 3050
},
{
"epoch": 4.46,
"learning_rate": 4.284444444444445e-05,
"loss": 0.043,
"step": 3075
},
{
"epoch": 4.5,
"learning_rate": 4.228888888888889e-05,
"loss": 0.0517,
"step": 3100
},
{
"epoch": 4.54,
"learning_rate": 4.1733333333333336e-05,
"loss": 0.0509,
"step": 3125
},
{
"epoch": 4.57,
"learning_rate": 4.117777777777778e-05,
"loss": 0.0409,
"step": 3150
},
{
"epoch": 4.61,
"learning_rate": 4.062222222222222e-05,
"loss": 0.0389,
"step": 3175
},
{
"epoch": 4.64,
"learning_rate": 4.006666666666667e-05,
"loss": 0.0426,
"step": 3200
},
{
"epoch": 4.68,
"learning_rate": 3.951111111111112e-05,
"loss": 0.0414,
"step": 3225
},
{
"epoch": 4.72,
"learning_rate": 3.8955555555555555e-05,
"loss": 0.0457,
"step": 3250
},
{
"epoch": 4.75,
"learning_rate": 3.8400000000000005e-05,
"loss": 0.0378,
"step": 3275
},
{
"epoch": 4.79,
"learning_rate": 3.784444444444445e-05,
"loss": 0.0413,
"step": 3300
},
{
"epoch": 4.83,
"learning_rate": 3.728888888888889e-05,
"loss": 0.0435,
"step": 3325
},
{
"epoch": 4.86,
"learning_rate": 3.6733333333333336e-05,
"loss": 0.0454,
"step": 3350
},
{
"epoch": 4.9,
"learning_rate": 3.617777777777778e-05,
"loss": 0.0369,
"step": 3375
},
{
"epoch": 4.93,
"learning_rate": 3.5622222222222224e-05,
"loss": 0.0391,
"step": 3400
},
{
"epoch": 4.97,
"learning_rate": 3.506666666666667e-05,
"loss": 0.0354,
"step": 3425
},
{
"epoch": 5.01,
"learning_rate": 3.451111111111111e-05,
"loss": 0.0336,
"step": 3450
},
{
"epoch": 5.04,
"learning_rate": 3.3955555555555555e-05,
"loss": 0.0251,
"step": 3475
},
{
"epoch": 5.08,
"learning_rate": 3.3400000000000005e-05,
"loss": 0.0131,
"step": 3500
},
{
"epoch": 5.12,
"learning_rate": 3.284444444444444e-05,
"loss": 0.0214,
"step": 3525
},
{
"epoch": 5.15,
"learning_rate": 3.228888888888889e-05,
"loss": 0.0191,
"step": 3550
},
{
"epoch": 5.19,
"learning_rate": 3.173333333333334e-05,
"loss": 0.0245,
"step": 3575
},
{
"epoch": 5.22,
"learning_rate": 3.117777777777778e-05,
"loss": 0.0132,
"step": 3600
},
{
"epoch": 5.26,
"learning_rate": 3.0622222222222224e-05,
"loss": 0.0213,
"step": 3625
},
{
"epoch": 5.3,
"learning_rate": 3.006666666666667e-05,
"loss": 0.0196,
"step": 3650
},
{
"epoch": 5.33,
"learning_rate": 2.951111111111111e-05,
"loss": 0.0222,
"step": 3675
},
{
"epoch": 5.37,
"learning_rate": 2.895555555555556e-05,
"loss": 0.0173,
"step": 3700
},
{
"epoch": 5.41,
"learning_rate": 2.84e-05,
"loss": 0.0194,
"step": 3725
},
{
"epoch": 5.44,
"learning_rate": 2.7844444444444446e-05,
"loss": 0.017,
"step": 3750
},
{
"epoch": 5.48,
"learning_rate": 2.728888888888889e-05,
"loss": 0.0177,
"step": 3775
},
{
"epoch": 5.52,
"learning_rate": 2.6733333333333334e-05,
"loss": 0.0176,
"step": 3800
},
{
"epoch": 5.55,
"learning_rate": 2.6177777777777777e-05,
"loss": 0.0199,
"step": 3825
},
{
"epoch": 5.59,
"learning_rate": 2.5622222222222225e-05,
"loss": 0.0168,
"step": 3850
},
{
"epoch": 5.62,
"learning_rate": 2.5066666666666665e-05,
"loss": 0.0268,
"step": 3875
},
{
"epoch": 5.66,
"learning_rate": 2.4511111111111112e-05,
"loss": 0.0175,
"step": 3900
},
{
"epoch": 5.7,
"learning_rate": 2.3955555555555556e-05,
"loss": 0.0142,
"step": 3925
},
{
"epoch": 5.73,
"learning_rate": 2.3400000000000003e-05,
"loss": 0.0153,
"step": 3950
},
{
"epoch": 5.77,
"learning_rate": 2.2844444444444447e-05,
"loss": 0.0138,
"step": 3975
},
{
"epoch": 5.81,
"learning_rate": 2.228888888888889e-05,
"loss": 0.0136,
"step": 4000
},
{
"epoch": 5.81,
"eval_loss": 0.4631120264530182,
"eval_runtime": 2311.1233,
"eval_samples_per_second": 1.567,
"eval_steps_per_second": 0.261,
"eval_wer": 18.633540372670808,
"step": 4000
},
{
"epoch": 5.84,
"learning_rate": 2.1733333333333334e-05,
"loss": 0.0166,
"step": 4025
},
{
"epoch": 5.88,
"learning_rate": 2.117777777777778e-05,
"loss": 0.0152,
"step": 4050
},
{
"epoch": 5.91,
"learning_rate": 2.062222222222222e-05,
"loss": 0.0099,
"step": 4075
},
{
"epoch": 5.95,
"learning_rate": 2.0066666666666665e-05,
"loss": 0.0158,
"step": 4100
},
{
"epoch": 5.99,
"learning_rate": 1.9511111111111113e-05,
"loss": 0.0131,
"step": 4125
},
{
"epoch": 6.02,
"learning_rate": 1.8955555555555556e-05,
"loss": 0.01,
"step": 4150
},
{
"epoch": 6.06,
"learning_rate": 1.84e-05,
"loss": 0.005,
"step": 4175
},
{
"epoch": 6.1,
"learning_rate": 1.7844444444444444e-05,
"loss": 0.0083,
"step": 4200
},
{
"epoch": 6.13,
"learning_rate": 1.728888888888889e-05,
"loss": 0.0047,
"step": 4225
},
{
"epoch": 6.17,
"learning_rate": 1.6733333333333335e-05,
"loss": 0.0067,
"step": 4250
},
{
"epoch": 6.2,
"learning_rate": 1.617777777777778e-05,
"loss": 0.0029,
"step": 4275
},
{
"epoch": 6.24,
"learning_rate": 1.5622222222222225e-05,
"loss": 0.002,
"step": 4300
},
{
"epoch": 6.28,
"learning_rate": 1.5066666666666668e-05,
"loss": 0.0031,
"step": 4325
},
{
"epoch": 6.31,
"learning_rate": 1.4511111111111111e-05,
"loss": 0.0036,
"step": 4350
},
{
"epoch": 6.35,
"learning_rate": 1.3955555555555555e-05,
"loss": 0.0041,
"step": 4375
},
{
"epoch": 6.39,
"learning_rate": 1.3400000000000002e-05,
"loss": 0.0033,
"step": 4400
},
{
"epoch": 6.42,
"learning_rate": 1.2844444444444446e-05,
"loss": 0.0062,
"step": 4425
},
{
"epoch": 6.46,
"learning_rate": 1.228888888888889e-05,
"loss": 0.0035,
"step": 4450
},
{
"epoch": 6.49,
"learning_rate": 1.1733333333333333e-05,
"loss": 0.004,
"step": 4475
},
{
"epoch": 6.53,
"learning_rate": 1.1177777777777779e-05,
"loss": 0.0048,
"step": 4500
},
{
"epoch": 6.57,
"learning_rate": 1.0622222222222223e-05,
"loss": 0.0041,
"step": 4525
},
{
"epoch": 6.6,
"learning_rate": 1.0066666666666668e-05,
"loss": 0.0063,
"step": 4550
},
{
"epoch": 6.64,
"learning_rate": 9.511111111111112e-06,
"loss": 0.0047,
"step": 4575
},
{
"epoch": 6.68,
"learning_rate": 8.955555555555555e-06,
"loss": 0.0039,
"step": 4600
},
{
"epoch": 6.71,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0019,
"step": 4625
},
{
"epoch": 6.75,
"learning_rate": 7.844444444444445e-06,
"loss": 0.0033,
"step": 4650
},
{
"epoch": 6.79,
"learning_rate": 7.288888888888889e-06,
"loss": 0.0045,
"step": 4675
},
{
"epoch": 6.82,
"learning_rate": 6.733333333333333e-06,
"loss": 0.0051,
"step": 4700
},
{
"epoch": 6.86,
"learning_rate": 6.177777777777778e-06,
"loss": 0.0068,
"step": 4725
},
{
"epoch": 6.89,
"learning_rate": 5.622222222222222e-06,
"loss": 0.0022,
"step": 4750
},
{
"epoch": 6.93,
"learning_rate": 5.066666666666667e-06,
"loss": 0.0024,
"step": 4775
},
{
"epoch": 6.97,
"learning_rate": 4.511111111111111e-06,
"loss": 0.0027,
"step": 4800
},
{
"epoch": 7.0,
"learning_rate": 3.955555555555555e-06,
"loss": 0.007,
"step": 4825
},
{
"epoch": 7.04,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.0018,
"step": 4850
},
{
"epoch": 7.08,
"learning_rate": 2.8444444444444446e-06,
"loss": 0.0006,
"step": 4875
},
{
"epoch": 7.11,
"learning_rate": 2.2888888888888892e-06,
"loss": 0.0007,
"step": 4900
},
{
"epoch": 7.15,
"learning_rate": 1.7333333333333334e-06,
"loss": 0.0005,
"step": 4925
},
{
"epoch": 7.18,
"learning_rate": 1.1777777777777778e-06,
"loss": 0.0007,
"step": 4950
},
{
"epoch": 7.22,
"learning_rate": 6.222222222222223e-07,
"loss": 0.0008,
"step": 4975
},
{
"epoch": 7.26,
"learning_rate": 6.666666666666667e-08,
"loss": 0.0005,
"step": 5000
},
{
"epoch": 7.26,
"eval_loss": 0.456723153591156,
"eval_runtime": 2309.2911,
"eval_samples_per_second": 1.568,
"eval_steps_per_second": 0.262,
"eval_wer": 17.593742811134117,
"step": 5000
},
{
"epoch": 7.26,
"step": 5000,
"total_flos": 1.730098333458432e+19,
"train_loss": 0.17717905388567598,
"train_runtime": 47932.271,
"train_samples_per_second": 1.252,
"train_steps_per_second": 0.104
}
],
"max_steps": 5000,
"num_train_epochs": 8,
"total_flos": 1.730098333458432e+19,
"trial_name": null,
"trial_params": null
}