whisper-large-nob-ncc-s-lr5e-6 / trainer_state.json
versae's picture
End of training
d1529b9
{
"best_metric": 12.058465286236297,
"best_model_checkpoint": "./checkpoint-5000",
"epoch": 1.0,
"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": 2.2e-07,
"loss": 1.1015,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 4.6000000000000004e-07,
"loss": 1.0967,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 7.1e-07,
"loss": 1.0955,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 9.600000000000001e-07,
"loss": 0.9982,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 1.21e-06,
"loss": 0.8406,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 1.46e-06,
"loss": 0.7874,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 1.7100000000000004e-06,
"loss": 0.8137,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.7833,
"step": 200
},
{
"epoch": 0.04,
"learning_rate": 2.21e-06,
"loss": 0.7546,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 2.46e-06,
"loss": 0.7955,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 2.7100000000000003e-06,
"loss": 0.7452,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 2.96e-06,
"loss": 0.772,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 3.21e-06,
"loss": 0.7355,
"step": 325
},
{
"epoch": 0.07,
"learning_rate": 3.46e-06,
"loss": 0.7656,
"step": 350
},
{
"epoch": 0.07,
"learning_rate": 3.7100000000000005e-06,
"loss": 0.7343,
"step": 375
},
{
"epoch": 0.08,
"learning_rate": 3.96e-06,
"loss": 0.7333,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 4.2000000000000004e-06,
"loss": 0.735,
"step": 425
},
{
"epoch": 0.09,
"learning_rate": 4.450000000000001e-06,
"loss": 0.7541,
"step": 450
},
{
"epoch": 0.1,
"learning_rate": 4.7e-06,
"loss": 0.7517,
"step": 475
},
{
"epoch": 0.1,
"learning_rate": 4.95e-06,
"loss": 0.7158,
"step": 500
},
{
"epoch": 0.1,
"learning_rate": 4.978888888888889e-06,
"loss": 0.7319,
"step": 525
},
{
"epoch": 0.11,
"learning_rate": 4.951111111111111e-06,
"loss": 0.7119,
"step": 550
},
{
"epoch": 0.12,
"learning_rate": 4.923333333333334e-06,
"loss": 0.7451,
"step": 575
},
{
"epoch": 0.12,
"learning_rate": 4.895555555555556e-06,
"loss": 0.7517,
"step": 600
},
{
"epoch": 0.12,
"learning_rate": 4.867777777777778e-06,
"loss": 0.7012,
"step": 625
},
{
"epoch": 0.13,
"learning_rate": 4.84e-06,
"loss": 0.7418,
"step": 650
},
{
"epoch": 0.14,
"learning_rate": 4.812222222222222e-06,
"loss": 0.7383,
"step": 675
},
{
"epoch": 0.14,
"learning_rate": 4.784444444444445e-06,
"loss": 0.6722,
"step": 700
},
{
"epoch": 0.14,
"learning_rate": 4.756666666666667e-06,
"loss": 0.7277,
"step": 725
},
{
"epoch": 0.15,
"learning_rate": 4.728888888888889e-06,
"loss": 0.6973,
"step": 750
},
{
"epoch": 0.15,
"learning_rate": 4.701111111111111e-06,
"loss": 0.7423,
"step": 775
},
{
"epoch": 0.16,
"learning_rate": 4.673333333333333e-06,
"loss": 0.6813,
"step": 800
},
{
"epoch": 0.17,
"learning_rate": 4.645555555555556e-06,
"loss": 0.7022,
"step": 825
},
{
"epoch": 0.17,
"learning_rate": 4.617777777777778e-06,
"loss": 0.7134,
"step": 850
},
{
"epoch": 0.17,
"learning_rate": 4.590000000000001e-06,
"loss": 0.7388,
"step": 875
},
{
"epoch": 0.18,
"learning_rate": 4.562222222222222e-06,
"loss": 0.7253,
"step": 900
},
{
"epoch": 0.18,
"learning_rate": 4.534444444444445e-06,
"loss": 0.7216,
"step": 925
},
{
"epoch": 0.19,
"learning_rate": 4.506666666666667e-06,
"loss": 0.709,
"step": 950
},
{
"epoch": 0.2,
"learning_rate": 4.478888888888889e-06,
"loss": 0.7006,
"step": 975
},
{
"epoch": 0.2,
"learning_rate": 4.451111111111112e-06,
"loss": 0.6755,
"step": 1000
},
{
"epoch": 0.2,
"eval_loss": 0.3108437657356262,
"eval_runtime": 383.2246,
"eval_samples_per_second": 0.42,
"eval_steps_per_second": 0.07,
"eval_wer": 14.311814859926919,
"step": 1000
},
{
"epoch": 0.2,
"learning_rate": 4.423333333333334e-06,
"loss": 0.7139,
"step": 1025
},
{
"epoch": 0.21,
"learning_rate": 4.395555555555556e-06,
"loss": 0.6889,
"step": 1050
},
{
"epoch": 0.21,
"learning_rate": 4.367777777777778e-06,
"loss": 0.6946,
"step": 1075
},
{
"epoch": 0.22,
"learning_rate": 4.34e-06,
"loss": 0.6745,
"step": 1100
},
{
"epoch": 0.23,
"learning_rate": 4.312222222222223e-06,
"loss": 0.6997,
"step": 1125
},
{
"epoch": 0.23,
"learning_rate": 4.284444444444445e-06,
"loss": 0.6902,
"step": 1150
},
{
"epoch": 0.23,
"learning_rate": 4.256666666666668e-06,
"loss": 0.6871,
"step": 1175
},
{
"epoch": 0.24,
"learning_rate": 4.228888888888889e-06,
"loss": 0.7103,
"step": 1200
},
{
"epoch": 0.24,
"learning_rate": 4.201111111111112e-06,
"loss": 0.7145,
"step": 1225
},
{
"epoch": 0.25,
"learning_rate": 4.173333333333334e-06,
"loss": 0.7069,
"step": 1250
},
{
"epoch": 0.26,
"learning_rate": 4.145555555555556e-06,
"loss": 0.6902,
"step": 1275
},
{
"epoch": 0.26,
"learning_rate": 4.117777777777779e-06,
"loss": 0.6999,
"step": 1300
},
{
"epoch": 0.27,
"learning_rate": 4.09e-06,
"loss": 0.682,
"step": 1325
},
{
"epoch": 0.27,
"learning_rate": 4.062222222222223e-06,
"loss": 0.6975,
"step": 1350
},
{
"epoch": 0.28,
"learning_rate": 4.034444444444445e-06,
"loss": 0.6974,
"step": 1375
},
{
"epoch": 0.28,
"learning_rate": 4.006666666666667e-06,
"loss": 0.7005,
"step": 1400
},
{
"epoch": 0.28,
"learning_rate": 3.9788888888888896e-06,
"loss": 0.641,
"step": 1425
},
{
"epoch": 0.29,
"learning_rate": 3.951111111111112e-06,
"loss": 0.6462,
"step": 1450
},
{
"epoch": 0.29,
"learning_rate": 3.923333333333334e-06,
"loss": 0.6554,
"step": 1475
},
{
"epoch": 0.3,
"learning_rate": 3.895555555555556e-06,
"loss": 0.6634,
"step": 1500
},
{
"epoch": 0.3,
"learning_rate": 3.8677777777777785e-06,
"loss": 0.7454,
"step": 1525
},
{
"epoch": 0.31,
"learning_rate": 3.8400000000000005e-06,
"loss": 0.692,
"step": 1550
},
{
"epoch": 0.32,
"learning_rate": 3.8122222222222225e-06,
"loss": 0.6538,
"step": 1575
},
{
"epoch": 0.32,
"learning_rate": 3.784444444444445e-06,
"loss": 0.6935,
"step": 1600
},
{
"epoch": 0.33,
"learning_rate": 3.756666666666667e-06,
"loss": 0.6657,
"step": 1625
},
{
"epoch": 0.33,
"learning_rate": 3.728888888888889e-06,
"loss": 0.7138,
"step": 1650
},
{
"epoch": 0.34,
"learning_rate": 3.7011111111111114e-06,
"loss": 0.6927,
"step": 1675
},
{
"epoch": 0.34,
"learning_rate": 3.673333333333334e-06,
"loss": 0.6395,
"step": 1700
},
{
"epoch": 0.34,
"learning_rate": 3.645555555555556e-06,
"loss": 0.6686,
"step": 1725
},
{
"epoch": 0.35,
"learning_rate": 3.617777777777778e-06,
"loss": 0.676,
"step": 1750
},
{
"epoch": 0.35,
"learning_rate": 3.5900000000000004e-06,
"loss": 0.6424,
"step": 1775
},
{
"epoch": 0.36,
"learning_rate": 3.5622222222222224e-06,
"loss": 0.6794,
"step": 1800
},
{
"epoch": 0.36,
"learning_rate": 3.534444444444445e-06,
"loss": 0.683,
"step": 1825
},
{
"epoch": 0.37,
"learning_rate": 3.5066666666666673e-06,
"loss": 0.6575,
"step": 1850
},
{
"epoch": 0.38,
"learning_rate": 3.4788888888888893e-06,
"loss": 0.6529,
"step": 1875
},
{
"epoch": 0.38,
"learning_rate": 3.4511111111111113e-06,
"loss": 0.6695,
"step": 1900
},
{
"epoch": 0.39,
"learning_rate": 3.4233333333333333e-06,
"loss": 0.7013,
"step": 1925
},
{
"epoch": 0.39,
"learning_rate": 3.3955555555555558e-06,
"loss": 0.6638,
"step": 1950
},
{
"epoch": 0.4,
"learning_rate": 3.367777777777778e-06,
"loss": 0.5981,
"step": 1975
},
{
"epoch": 0.4,
"learning_rate": 3.3400000000000006e-06,
"loss": 0.673,
"step": 2000
},
{
"epoch": 0.4,
"eval_loss": 0.30037370324134827,
"eval_runtime": 381.5276,
"eval_samples_per_second": 0.422,
"eval_steps_per_second": 0.071,
"eval_wer": 13.45919610231425,
"step": 2000
},
{
"epoch": 0.41,
"learning_rate": 3.3122222222222222e-06,
"loss": 0.6845,
"step": 2025
},
{
"epoch": 0.41,
"learning_rate": 3.2844444444444447e-06,
"loss": 0.6491,
"step": 2050
},
{
"epoch": 0.41,
"learning_rate": 3.2566666666666667e-06,
"loss": 0.6749,
"step": 2075
},
{
"epoch": 0.42,
"learning_rate": 3.228888888888889e-06,
"loss": 0.6791,
"step": 2100
},
{
"epoch": 0.42,
"learning_rate": 3.2011111111111116e-06,
"loss": 0.6412,
"step": 2125
},
{
"epoch": 0.43,
"learning_rate": 3.173333333333334e-06,
"loss": 0.6369,
"step": 2150
},
{
"epoch": 0.43,
"learning_rate": 3.1455555555555556e-06,
"loss": 0.6315,
"step": 2175
},
{
"epoch": 0.44,
"learning_rate": 3.117777777777778e-06,
"loss": 0.6481,
"step": 2200
},
{
"epoch": 0.45,
"learning_rate": 3.09e-06,
"loss": 0.6319,
"step": 2225
},
{
"epoch": 0.45,
"learning_rate": 3.0622222222222225e-06,
"loss": 0.6412,
"step": 2250
},
{
"epoch": 0.46,
"learning_rate": 3.034444444444445e-06,
"loss": 0.6676,
"step": 2275
},
{
"epoch": 0.46,
"learning_rate": 3.0066666666666674e-06,
"loss": 0.6318,
"step": 2300
},
{
"epoch": 0.47,
"learning_rate": 2.978888888888889e-06,
"loss": 0.6247,
"step": 2325
},
{
"epoch": 0.47,
"learning_rate": 2.9511111111111114e-06,
"loss": 0.6429,
"step": 2350
},
{
"epoch": 0.47,
"learning_rate": 2.9233333333333334e-06,
"loss": 0.6451,
"step": 2375
},
{
"epoch": 0.48,
"learning_rate": 2.895555555555556e-06,
"loss": 0.6472,
"step": 2400
},
{
"epoch": 0.48,
"learning_rate": 2.8677777777777783e-06,
"loss": 0.6246,
"step": 2425
},
{
"epoch": 0.49,
"learning_rate": 2.84e-06,
"loss": 0.6555,
"step": 2450
},
{
"epoch": 0.49,
"learning_rate": 2.8122222222222224e-06,
"loss": 0.6454,
"step": 2475
},
{
"epoch": 0.5,
"learning_rate": 2.784444444444445e-06,
"loss": 0.6462,
"step": 2500
},
{
"epoch": 0.51,
"learning_rate": 2.756666666666667e-06,
"loss": 0.6122,
"step": 2525
},
{
"epoch": 0.51,
"learning_rate": 2.7288888888888893e-06,
"loss": 0.6286,
"step": 2550
},
{
"epoch": 0.52,
"learning_rate": 2.7011111111111117e-06,
"loss": 0.6323,
"step": 2575
},
{
"epoch": 0.52,
"learning_rate": 2.6733333333333333e-06,
"loss": 0.6184,
"step": 2600
},
{
"epoch": 0.53,
"learning_rate": 2.6455555555555557e-06,
"loss": 0.6416,
"step": 2625
},
{
"epoch": 0.53,
"learning_rate": 2.617777777777778e-06,
"loss": 0.6403,
"step": 2650
},
{
"epoch": 0.54,
"learning_rate": 2.59e-06,
"loss": 0.6051,
"step": 2675
},
{
"epoch": 0.54,
"learning_rate": 2.5622222222222226e-06,
"loss": 0.61,
"step": 2700
},
{
"epoch": 0.55,
"learning_rate": 2.534444444444445e-06,
"loss": 0.6042,
"step": 2725
},
{
"epoch": 0.55,
"learning_rate": 2.5066666666666667e-06,
"loss": 0.6596,
"step": 2750
},
{
"epoch": 0.56,
"learning_rate": 2.478888888888889e-06,
"loss": 0.6214,
"step": 2775
},
{
"epoch": 0.56,
"learning_rate": 2.451111111111111e-06,
"loss": 0.6557,
"step": 2800
},
{
"epoch": 0.56,
"learning_rate": 2.4233333333333336e-06,
"loss": 0.6284,
"step": 2825
},
{
"epoch": 0.57,
"learning_rate": 2.3955555555555556e-06,
"loss": 0.587,
"step": 2850
},
{
"epoch": 0.57,
"learning_rate": 2.367777777777778e-06,
"loss": 0.6204,
"step": 2875
},
{
"epoch": 0.58,
"learning_rate": 2.3411111111111113e-06,
"loss": 0.6298,
"step": 2900
},
{
"epoch": 0.58,
"learning_rate": 2.3133333333333333e-06,
"loss": 0.6379,
"step": 2925
},
{
"epoch": 0.59,
"learning_rate": 2.285555555555556e-06,
"loss": 0.6392,
"step": 2950
},
{
"epoch": 0.59,
"learning_rate": 2.257777777777778e-06,
"loss": 0.5836,
"step": 2975
},
{
"epoch": 0.6,
"learning_rate": 2.2300000000000002e-06,
"loss": 0.6378,
"step": 3000
},
{
"epoch": 0.6,
"eval_loss": 0.28651803731918335,
"eval_runtime": 381.9189,
"eval_samples_per_second": 0.422,
"eval_steps_per_second": 0.071,
"eval_wer": 13.002436053593177,
"step": 3000
},
{
"epoch": 0.6,
"learning_rate": 2.2022222222222227e-06,
"loss": 0.6564,
"step": 3025
},
{
"epoch": 0.61,
"learning_rate": 2.1744444444444447e-06,
"loss": 0.5728,
"step": 3050
},
{
"epoch": 0.61,
"learning_rate": 2.1466666666666667e-06,
"loss": 0.6299,
"step": 3075
},
{
"epoch": 0.62,
"learning_rate": 2.1188888888888887e-06,
"loss": 0.6285,
"step": 3100
},
{
"epoch": 0.62,
"learning_rate": 2.091111111111111e-06,
"loss": 0.5931,
"step": 3125
},
{
"epoch": 0.63,
"learning_rate": 2.0633333333333336e-06,
"loss": 0.6304,
"step": 3150
},
{
"epoch": 0.64,
"learning_rate": 2.0355555555555556e-06,
"loss": 0.5942,
"step": 3175
},
{
"epoch": 0.64,
"learning_rate": 2.007777777777778e-06,
"loss": 0.6283,
"step": 3200
},
{
"epoch": 0.65,
"learning_rate": 1.98e-06,
"loss": 0.6728,
"step": 3225
},
{
"epoch": 0.65,
"learning_rate": 1.952222222222222e-06,
"loss": 0.5897,
"step": 3250
},
{
"epoch": 0.66,
"learning_rate": 1.9244444444444446e-06,
"loss": 0.67,
"step": 3275
},
{
"epoch": 0.66,
"learning_rate": 1.896666666666667e-06,
"loss": 0.6138,
"step": 3300
},
{
"epoch": 0.67,
"learning_rate": 1.868888888888889e-06,
"loss": 0.5949,
"step": 3325
},
{
"epoch": 0.67,
"learning_rate": 1.8411111111111112e-06,
"loss": 0.665,
"step": 3350
},
{
"epoch": 0.68,
"learning_rate": 1.8133333333333337e-06,
"loss": 0.5992,
"step": 3375
},
{
"epoch": 0.68,
"learning_rate": 1.7855555555555557e-06,
"loss": 0.5849,
"step": 3400
},
{
"epoch": 0.69,
"learning_rate": 1.7588888888888892e-06,
"loss": 0.657,
"step": 3425
},
{
"epoch": 0.69,
"learning_rate": 1.7311111111111112e-06,
"loss": 0.5962,
"step": 3450
},
{
"epoch": 0.69,
"learning_rate": 1.7033333333333335e-06,
"loss": 0.6158,
"step": 3475
},
{
"epoch": 0.7,
"learning_rate": 1.675555555555556e-06,
"loss": 0.637,
"step": 3500
},
{
"epoch": 0.7,
"learning_rate": 1.647777777777778e-06,
"loss": 0.6314,
"step": 3525
},
{
"epoch": 0.71,
"learning_rate": 1.6200000000000002e-06,
"loss": 0.653,
"step": 3550
},
{
"epoch": 0.71,
"learning_rate": 1.5922222222222222e-06,
"loss": 0.5764,
"step": 3575
},
{
"epoch": 0.72,
"learning_rate": 1.5644444444444446e-06,
"loss": 0.5966,
"step": 3600
},
{
"epoch": 0.72,
"learning_rate": 1.5366666666666668e-06,
"loss": 0.6227,
"step": 3625
},
{
"epoch": 0.73,
"learning_rate": 1.5088888888888889e-06,
"loss": 0.5869,
"step": 3650
},
{
"epoch": 0.73,
"learning_rate": 1.4811111111111113e-06,
"loss": 0.6246,
"step": 3675
},
{
"epoch": 0.74,
"learning_rate": 1.4533333333333335e-06,
"loss": 0.6616,
"step": 3700
},
{
"epoch": 0.74,
"learning_rate": 1.4255555555555556e-06,
"loss": 0.5766,
"step": 3725
},
{
"epoch": 0.75,
"learning_rate": 1.397777777777778e-06,
"loss": 0.6168,
"step": 3750
},
{
"epoch": 0.76,
"learning_rate": 1.3700000000000002e-06,
"loss": 0.5761,
"step": 3775
},
{
"epoch": 0.76,
"learning_rate": 1.3422222222222222e-06,
"loss": 0.5984,
"step": 3800
},
{
"epoch": 0.77,
"learning_rate": 1.3144444444444447e-06,
"loss": 0.5923,
"step": 3825
},
{
"epoch": 0.77,
"learning_rate": 1.286666666666667e-06,
"loss": 0.5842,
"step": 3850
},
{
"epoch": 0.78,
"learning_rate": 1.258888888888889e-06,
"loss": 0.62,
"step": 3875
},
{
"epoch": 0.78,
"learning_rate": 1.2311111111111112e-06,
"loss": 0.6256,
"step": 3900
},
{
"epoch": 0.79,
"learning_rate": 1.2033333333333334e-06,
"loss": 0.6228,
"step": 3925
},
{
"epoch": 0.79,
"learning_rate": 1.1755555555555556e-06,
"loss": 0.6086,
"step": 3950
},
{
"epoch": 0.8,
"learning_rate": 1.1477777777777778e-06,
"loss": 0.5963,
"step": 3975
},
{
"epoch": 0.8,
"learning_rate": 1.12e-06,
"loss": 0.5776,
"step": 4000
},
{
"epoch": 0.8,
"eval_loss": 0.28093358874320984,
"eval_runtime": 381.875,
"eval_samples_per_second": 0.422,
"eval_steps_per_second": 0.071,
"eval_wer": 12.667478684531059,
"step": 4000
},
{
"epoch": 0.81,
"learning_rate": 1.0922222222222223e-06,
"loss": 0.6044,
"step": 4025
},
{
"epoch": 0.81,
"learning_rate": 1.0644444444444445e-06,
"loss": 0.5753,
"step": 4050
},
{
"epoch": 0.81,
"learning_rate": 1.0366666666666668e-06,
"loss": 0.5783,
"step": 4075
},
{
"epoch": 0.82,
"learning_rate": 1.008888888888889e-06,
"loss": 0.6202,
"step": 4100
},
{
"epoch": 0.82,
"learning_rate": 9.811111111111112e-07,
"loss": 0.6246,
"step": 4125
},
{
"epoch": 0.83,
"learning_rate": 9.533333333333335e-07,
"loss": 0.613,
"step": 4150
},
{
"epoch": 0.83,
"learning_rate": 9.255555555555557e-07,
"loss": 0.6398,
"step": 4175
},
{
"epoch": 0.84,
"learning_rate": 8.977777777777778e-07,
"loss": 0.6215,
"step": 4200
},
{
"epoch": 0.84,
"learning_rate": 8.7e-07,
"loss": 0.583,
"step": 4225
},
{
"epoch": 0.85,
"learning_rate": 8.422222222222224e-07,
"loss": 0.6226,
"step": 4250
},
{
"epoch": 0.85,
"learning_rate": 8.144444444444445e-07,
"loss": 0.6161,
"step": 4275
},
{
"epoch": 0.86,
"learning_rate": 7.866666666666667e-07,
"loss": 0.6145,
"step": 4300
},
{
"epoch": 0.86,
"learning_rate": 7.588888888888891e-07,
"loss": 0.5933,
"step": 4325
},
{
"epoch": 0.87,
"learning_rate": 7.311111111111112e-07,
"loss": 0.6017,
"step": 4350
},
{
"epoch": 0.88,
"learning_rate": 7.033333333333334e-07,
"loss": 0.5815,
"step": 4375
},
{
"epoch": 0.88,
"learning_rate": 6.755555555555555e-07,
"loss": 0.6134,
"step": 4400
},
{
"epoch": 0.89,
"learning_rate": 6.477777777777779e-07,
"loss": 0.624,
"step": 4425
},
{
"epoch": 0.89,
"learning_rate": 6.200000000000001e-07,
"loss": 0.63,
"step": 4450
},
{
"epoch": 0.9,
"learning_rate": 5.922222222222223e-07,
"loss": 0.6042,
"step": 4475
},
{
"epoch": 0.9,
"learning_rate": 5.644444444444445e-07,
"loss": 0.6324,
"step": 4500
},
{
"epoch": 0.91,
"learning_rate": 5.366666666666667e-07,
"loss": 0.635,
"step": 4525
},
{
"epoch": 0.91,
"learning_rate": 5.088888888888889e-07,
"loss": 0.5908,
"step": 4550
},
{
"epoch": 0.92,
"learning_rate": 4.811111111111111e-07,
"loss": 0.6066,
"step": 4575
},
{
"epoch": 0.92,
"learning_rate": 4.533333333333334e-07,
"loss": 0.6337,
"step": 4600
},
{
"epoch": 0.93,
"learning_rate": 4.266666666666667e-07,
"loss": 0.6122,
"step": 4625
},
{
"epoch": 0.93,
"learning_rate": 3.9888888888888896e-07,
"loss": 0.6011,
"step": 4650
},
{
"epoch": 0.94,
"learning_rate": 3.7111111111111113e-07,
"loss": 0.5917,
"step": 4675
},
{
"epoch": 0.94,
"learning_rate": 3.4333333333333336e-07,
"loss": 0.5827,
"step": 4700
},
{
"epoch": 0.94,
"learning_rate": 3.155555555555556e-07,
"loss": 0.589,
"step": 4725
},
{
"epoch": 0.95,
"learning_rate": 2.8777777777777777e-07,
"loss": 0.6073,
"step": 4750
},
{
"epoch": 0.95,
"learning_rate": 2.6e-07,
"loss": 0.6126,
"step": 4775
},
{
"epoch": 0.96,
"learning_rate": 2.3222222222222223e-07,
"loss": 0.6161,
"step": 4800
},
{
"epoch": 0.96,
"learning_rate": 2.0444444444444448e-07,
"loss": 0.6203,
"step": 4825
},
{
"epoch": 0.97,
"learning_rate": 1.766666666666667e-07,
"loss": 0.5948,
"step": 4850
},
{
"epoch": 0.97,
"learning_rate": 1.488888888888889e-07,
"loss": 0.5922,
"step": 4875
},
{
"epoch": 0.98,
"learning_rate": 1.2111111111111112e-07,
"loss": 0.6041,
"step": 4900
},
{
"epoch": 0.98,
"learning_rate": 9.333333333333335e-08,
"loss": 0.6396,
"step": 4925
},
{
"epoch": 0.99,
"learning_rate": 6.555555555555556e-08,
"loss": 0.5767,
"step": 4950
},
{
"epoch": 0.99,
"learning_rate": 3.777777777777778e-08,
"loss": 0.6239,
"step": 4975
},
{
"epoch": 1.0,
"learning_rate": 1e-08,
"loss": 0.5962,
"step": 5000
},
{
"epoch": 1.0,
"eval_loss": 0.27835896611213684,
"eval_runtime": 384.629,
"eval_samples_per_second": 0.419,
"eval_steps_per_second": 0.07,
"eval_wer": 12.058465286236297,
"step": 5000
},
{
"epoch": 1.0,
"step": 5000,
"total_flos": 1.27390261248e+20,
"train_loss": 0.6619791774749756,
"train_runtime": 31611.5865,
"train_samples_per_second": 1.898,
"train_steps_per_second": 0.158
}
],
"max_steps": 5000,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.27390261248e+20,
"trial_name": null,
"trial_params": null
}