{
  "best_metric": 23.738796414852754,
  "best_model_checkpoint": "/app/data/hieunq/whisper_asr_train/whisper-large-chinese-pipyin/checkpoint-4410",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4410,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 6.144965986394558e-06,
      "loss": 2.1824,
      "step": 25
    },
    {
      "epoch": 0.06,
      "learning_rate": 6.109931972789116e-06,
      "loss": 0.8831,
      "step": 50
    },
    {
      "epoch": 0.09,
      "learning_rate": 6.074897959183673e-06,
      "loss": 0.6375,
      "step": 75
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.039863945578232e-06,
      "loss": 0.5378,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 6.004829931972789e-06,
      "loss": 0.4835,
      "step": 125
    },
    {
      "epoch": 0.17,
      "learning_rate": 5.969795918367347e-06,
      "loss": 0.5321,
      "step": 150
    },
    {
      "epoch": 0.2,
      "learning_rate": 5.934761904761905e-06,
      "loss": 0.5099,
      "step": 175
    },
    {
      "epoch": 0.23,
      "learning_rate": 5.899727891156462e-06,
      "loss": 0.5304,
      "step": 200
    },
    {
      "epoch": 0.26,
      "learning_rate": 5.864693877551021e-06,
      "loss": 0.4958,
      "step": 225
    },
    {
      "epoch": 0.28,
      "learning_rate": 5.8296598639455785e-06,
      "loss": 0.4695,
      "step": 250
    },
    {
      "epoch": 0.31,
      "learning_rate": 5.794625850340136e-06,
      "loss": 0.4351,
      "step": 275
    },
    {
      "epoch": 0.34,
      "learning_rate": 5.759591836734694e-06,
      "loss": 0.4394,
      "step": 300
    },
    {
      "epoch": 0.37,
      "learning_rate": 5.7245578231292515e-06,
      "loss": 0.414,
      "step": 325
    },
    {
      "epoch": 0.4,
      "learning_rate": 5.689523809523809e-06,
      "loss": 0.4521,
      "step": 350
    },
    {
      "epoch": 0.43,
      "learning_rate": 5.654489795918368e-06,
      "loss": 0.4334,
      "step": 375
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.619455782312925e-06,
      "loss": 0.4217,
      "step": 400
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.584421768707483e-06,
      "loss": 0.4108,
      "step": 425
    },
    {
      "epoch": 0.51,
      "learning_rate": 5.549387755102041e-06,
      "loss": 0.4005,
      "step": 450
    },
    {
      "epoch": 0.54,
      "learning_rate": 5.514353741496598e-06,
      "loss": 0.4097,
      "step": 475
    },
    {
      "epoch": 0.57,
      "learning_rate": 5.479319727891157e-06,
      "loss": 0.3855,
      "step": 500
    },
    {
      "epoch": 0.6,
      "learning_rate": 5.4442857142857146e-06,
      "loss": 0.4167,
      "step": 525
    },
    {
      "epoch": 0.62,
      "learning_rate": 5.409251700680272e-06,
      "loss": 0.3667,
      "step": 550
    },
    {
      "epoch": 0.65,
      "learning_rate": 5.37421768707483e-06,
      "loss": 0.4307,
      "step": 575
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.3391836734693876e-06,
      "loss": 0.3693,
      "step": 600
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.304149659863946e-06,
      "loss": 0.3089,
      "step": 625
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.269115646258504e-06,
      "loss": 0.3185,
      "step": 650
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.234081632653061e-06,
      "loss": 0.3783,
      "step": 675
    },
    {
      "epoch": 0.79,
      "learning_rate": 5.199047619047619e-06,
      "loss": 0.3681,
      "step": 700
    },
    {
      "epoch": 0.82,
      "learning_rate": 5.164013605442177e-06,
      "loss": 0.4317,
      "step": 725
    },
    {
      "epoch": 0.85,
      "learning_rate": 5.128979591836735e-06,
      "loss": 0.3815,
      "step": 750
    },
    {
      "epoch": 0.88,
      "learning_rate": 5.093945578231293e-06,
      "loss": 0.3618,
      "step": 775
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.05891156462585e-06,
      "loss": 0.3376,
      "step": 800
    },
    {
      "epoch": 0.94,
      "learning_rate": 5.023877551020408e-06,
      "loss": 0.3485,
      "step": 825
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.988843537414966e-06,
      "loss": 0.3397,
      "step": 850
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.9538095238095245e-06,
      "loss": 0.3836,
      "step": 875
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.342790424823761,
      "eval_runtime": 1384.9795,
      "eval_samples_per_second": 1.274,
      "eval_steps_per_second": 0.16,
      "eval_wer": 27.76781903542467,
      "step": 882
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.918775510204081e-06,
      "loss": 0.2476,
      "step": 900
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.883741496598639e-06,
      "loss": 0.1817,
      "step": 925
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.8487074829931975e-06,
      "loss": 0.2218,
      "step": 950
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.813673469387755e-06,
      "loss": 0.1971,
      "step": 975
    },
    {
      "epoch": 1.13,
      "learning_rate": 4.778639455782314e-06,
      "loss": 0.2224,
      "step": 1000
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.7436054421768705e-06,
      "loss": 0.1781,
      "step": 1025
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.708571428571428e-06,
      "loss": 0.2359,
      "step": 1050
    },
    {
      "epoch": 1.22,
      "learning_rate": 4.673537414965987e-06,
      "loss": 0.1955,
      "step": 1075
    },
    {
      "epoch": 1.25,
      "learning_rate": 4.638503401360544e-06,
      "loss": 0.2005,
      "step": 1100
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.603469387755103e-06,
      "loss": 0.2131,
      "step": 1125
    },
    {
      "epoch": 1.3,
      "learning_rate": 4.56843537414966e-06,
      "loss": 0.179,
      "step": 1150
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.533401360544217e-06,
      "loss": 0.2156,
      "step": 1175
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.498367346938776e-06,
      "loss": 0.186,
      "step": 1200
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.4633333333333335e-06,
      "loss": 0.2004,
      "step": 1225
    },
    {
      "epoch": 1.42,
      "learning_rate": 4.428299319727891e-06,
      "loss": 0.2206,
      "step": 1250
    },
    {
      "epoch": 1.45,
      "learning_rate": 4.393265306122449e-06,
      "loss": 0.2029,
      "step": 1275
    },
    {
      "epoch": 1.47,
      "learning_rate": 4.3582312925170065e-06,
      "loss": 0.1952,
      "step": 1300
    },
    {
      "epoch": 1.5,
      "learning_rate": 4.323197278911565e-06,
      "loss": 0.2412,
      "step": 1325
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.288163265306123e-06,
      "loss": 0.2011,
      "step": 1350
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.25312925170068e-06,
      "loss": 0.1771,
      "step": 1375
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.218095238095238e-06,
      "loss": 0.2092,
      "step": 1400
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.183061224489796e-06,
      "loss": 0.1916,
      "step": 1425
    },
    {
      "epoch": 1.64,
      "learning_rate": 4.148027210884354e-06,
      "loss": 0.191,
      "step": 1450
    },
    {
      "epoch": 1.67,
      "learning_rate": 4.112993197278912e-06,
      "loss": 0.1907,
      "step": 1475
    },
    {
      "epoch": 1.7,
      "learning_rate": 4.0779591836734696e-06,
      "loss": 0.1984,
      "step": 1500
    },
    {
      "epoch": 1.73,
      "learning_rate": 4.042925170068027e-06,
      "loss": 0.1986,
      "step": 1525
    },
    {
      "epoch": 1.76,
      "learning_rate": 4.007891156462585e-06,
      "loss": 0.196,
      "step": 1550
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.972857142857143e-06,
      "loss": 0.1685,
      "step": 1575
    },
    {
      "epoch": 1.81,
      "learning_rate": 3.937823129251701e-06,
      "loss": 0.2104,
      "step": 1600
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.902789115646259e-06,
      "loss": 0.2079,
      "step": 1625
    },
    {
      "epoch": 1.87,
      "learning_rate": 3.8677551020408164e-06,
      "loss": 0.2119,
      "step": 1650
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.832721088435374e-06,
      "loss": 0.2,
      "step": 1675
    },
    {
      "epoch": 1.93,
      "learning_rate": 3.797687074829932e-06,
      "loss": 0.1765,
      "step": 1700
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.76265306122449e-06,
      "loss": 0.2026,
      "step": 1725
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.7276190476190475e-06,
      "loss": 0.1914,
      "step": 1750
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.32358798384666443,
      "eval_runtime": 1373.0201,
      "eval_samples_per_second": 1.285,
      "eval_steps_per_second": 0.161,
      "eval_wer": 25.761843790012804,
      "step": 1764
    },
    {
      "epoch": 2.01,
      "learning_rate": 3.6925850340136056e-06,
      "loss": 0.1853,
      "step": 1775
    },
    {
      "epoch": 2.04,
      "learning_rate": 3.6575510204081633e-06,
      "loss": 0.104,
      "step": 1800
    },
    {
      "epoch": 2.07,
      "learning_rate": 3.622517006802721e-06,
      "loss": 0.1111,
      "step": 1825
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.587482993197279e-06,
      "loss": 0.1044,
      "step": 1850
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.5524489795918367e-06,
      "loss": 0.0938,
      "step": 1875
    },
    {
      "epoch": 2.15,
      "learning_rate": 3.517414965986395e-06,
      "loss": 0.1344,
      "step": 1900
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.4823809523809525e-06,
      "loss": 0.1153,
      "step": 1925
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.44734693877551e-06,
      "loss": 0.0989,
      "step": 1950
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.4123129251700682e-06,
      "loss": 0.1086,
      "step": 1975
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.377278911564626e-06,
      "loss": 0.1255,
      "step": 2000
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.342244897959184e-06,
      "loss": 0.1093,
      "step": 2025
    },
    {
      "epoch": 2.32,
      "learning_rate": 3.3072108843537417e-06,
      "loss": 0.0984,
      "step": 2050
    },
    {
      "epoch": 2.35,
      "learning_rate": 3.2721768707482993e-06,
      "loss": 0.0984,
      "step": 2075
    },
    {
      "epoch": 2.38,
      "learning_rate": 3.2371428571428574e-06,
      "loss": 0.1139,
      "step": 2100
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.202108843537415e-06,
      "loss": 0.1043,
      "step": 2125
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.167074829931973e-06,
      "loss": 0.0983,
      "step": 2150
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.132040816326531e-06,
      "loss": 0.1194,
      "step": 2175
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.097006802721088e-06,
      "loss": 0.0958,
      "step": 2200
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.061972789115646e-06,
      "loss": 0.0857,
      "step": 2225
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.026938775510204e-06,
      "loss": 0.1022,
      "step": 2250
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.991904761904762e-06,
      "loss": 0.1118,
      "step": 2275
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.95687074829932e-06,
      "loss": 0.092,
      "step": 2300
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.9218367346938777e-06,
      "loss": 0.1063,
      "step": 2325
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.8868027210884354e-06,
      "loss": 0.1028,
      "step": 2350
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.851768707482993e-06,
      "loss": 0.1077,
      "step": 2375
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.816734693877551e-06,
      "loss": 0.0974,
      "step": 2400
    },
    {
      "epoch": 2.75,
      "learning_rate": 2.781700680272109e-06,
      "loss": 0.108,
      "step": 2425
    },
    {
      "epoch": 2.78,
      "learning_rate": 2.7466666666666665e-06,
      "loss": 0.0852,
      "step": 2450
    },
    {
      "epoch": 2.81,
      "learning_rate": 2.7116326530612246e-06,
      "loss": 0.1009,
      "step": 2475
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.6765986394557822e-06,
      "loss": 0.1127,
      "step": 2500
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.6415646258503403e-06,
      "loss": 0.1072,
      "step": 2525
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.606530612244898e-06,
      "loss": 0.0893,
      "step": 2550
    },
    {
      "epoch": 2.92,
      "learning_rate": 2.5714965986394557e-06,
      "loss": 0.0945,
      "step": 2575
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.5364625850340138e-06,
      "loss": 0.1149,
      "step": 2600
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.5014285714285714e-06,
      "loss": 0.0844,
      "step": 2625
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.33547088503837585,
      "eval_runtime": 1375.3795,
      "eval_samples_per_second": 1.283,
      "eval_steps_per_second": 0.161,
      "eval_wer": 24.609475032010245,
      "step": 2646
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.4663945578231295e-06,
      "loss": 0.0889,
      "step": 2650
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.4313605442176868e-06,
      "loss": 0.0382,
      "step": 2675
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.396326530612245e-06,
      "loss": 0.0627,
      "step": 2700
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.361292517006803e-06,
      "loss": 0.0562,
      "step": 2725
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.3262585034013606e-06,
      "loss": 0.049,
      "step": 2750
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.2912244897959187e-06,
      "loss": 0.0432,
      "step": 2775
    },
    {
      "epoch": 3.17,
      "learning_rate": 2.256190476190476e-06,
      "loss": 0.0594,
      "step": 2800
    },
    {
      "epoch": 3.2,
      "learning_rate": 2.221156462585034e-06,
      "loss": 0.0415,
      "step": 2825
    },
    {
      "epoch": 3.23,
      "learning_rate": 2.1861224489795917e-06,
      "loss": 0.0451,
      "step": 2850
    },
    {
      "epoch": 3.26,
      "learning_rate": 2.15108843537415e-06,
      "loss": 0.045,
      "step": 2875
    },
    {
      "epoch": 3.29,
      "learning_rate": 2.1160544217687075e-06,
      "loss": 0.0503,
      "step": 2900
    },
    {
      "epoch": 3.32,
      "learning_rate": 2.081020408163265e-06,
      "loss": 0.0368,
      "step": 2925
    },
    {
      "epoch": 3.34,
      "learning_rate": 2.0459863945578232e-06,
      "loss": 0.0447,
      "step": 2950
    },
    {
      "epoch": 3.37,
      "learning_rate": 2.010952380952381e-06,
      "loss": 0.041,
      "step": 2975
    },
    {
      "epoch": 3.4,
      "learning_rate": 1.975918367346939e-06,
      "loss": 0.054,
      "step": 3000
    },
    {
      "epoch": 3.43,
      "learning_rate": 1.9408843537414967e-06,
      "loss": 0.0409,
      "step": 3025
    },
    {
      "epoch": 3.46,
      "learning_rate": 1.9058503401360543e-06,
      "loss": 0.0423,
      "step": 3050
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.8708163265306122e-06,
      "loss": 0.0485,
      "step": 3075
    },
    {
      "epoch": 3.51,
      "learning_rate": 1.8357823129251701e-06,
      "loss": 0.048,
      "step": 3100
    },
    {
      "epoch": 3.54,
      "learning_rate": 1.8007482993197282e-06,
      "loss": 0.0509,
      "step": 3125
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.7657142857142857e-06,
      "loss": 0.0525,
      "step": 3150
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.7306802721088435e-06,
      "loss": 0.0539,
      "step": 3175
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.6956462585034014e-06,
      "loss": 0.0417,
      "step": 3200
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.6606122448979593e-06,
      "loss": 0.0415,
      "step": 3225
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.625578231292517e-06,
      "loss": 0.0557,
      "step": 3250
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.5905442176870748e-06,
      "loss": 0.0458,
      "step": 3275
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.5555102040816327e-06,
      "loss": 0.0425,
      "step": 3300
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.5204761904761904e-06,
      "loss": 0.0505,
      "step": 3325
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.4854421768707483e-06,
      "loss": 0.0505,
      "step": 3350
    },
    {
      "epoch": 3.83,
      "learning_rate": 1.4504081632653062e-06,
      "loss": 0.0445,
      "step": 3375
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.415374149659864e-06,
      "loss": 0.0462,
      "step": 3400
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.380340136054422e-06,
      "loss": 0.0556,
      "step": 3425
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.3453061224489796e-06,
      "loss": 0.0459,
      "step": 3450
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.3102721088435375e-06,
      "loss": 0.0439,
      "step": 3475
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.2752380952380951e-06,
      "loss": 0.0586,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.240204081632653e-06,
      "loss": 0.0418,
      "step": 3525
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.37163934111595154,
      "eval_runtime": 1376.5549,
      "eval_samples_per_second": 1.281,
      "eval_steps_per_second": 0.161,
      "eval_wer": 23.781476739223216,
      "step": 3528
    },
    {
      "epoch": 4.02,
      "learning_rate": 1.205170068027211e-06,
      "loss": 0.0216,
      "step": 3550
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.1701360544217688e-06,
      "loss": 0.0167,
      "step": 3575
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.1351020408163267e-06,
      "loss": 0.0179,
      "step": 3600
    },
    {
      "epoch": 4.11,
      "learning_rate": 1.1000680272108843e-06,
      "loss": 0.016,
      "step": 3625
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.0650340136054422e-06,
      "loss": 0.0157,
      "step": 3650
    },
    {
      "epoch": 4.17,
      "learning_rate": 1.0299999999999999e-06,
      "loss": 0.019,
      "step": 3675
    },
    {
      "epoch": 4.2,
      "learning_rate": 9.949659863945578e-07,
      "loss": 0.017,
      "step": 3700
    },
    {
      "epoch": 4.22,
      "learning_rate": 9.599319727891156e-07,
      "loss": 0.0193,
      "step": 3725
    },
    {
      "epoch": 4.25,
      "learning_rate": 9.248979591836735e-07,
      "loss": 0.0144,
      "step": 3750
    },
    {
      "epoch": 4.28,
      "learning_rate": 8.898639455782314e-07,
      "loss": 0.0161,
      "step": 3775
    },
    {
      "epoch": 4.31,
      "learning_rate": 8.548299319727891e-07,
      "loss": 0.018,
      "step": 3800
    },
    {
      "epoch": 4.34,
      "learning_rate": 8.197959183673471e-07,
      "loss": 0.0173,
      "step": 3825
    },
    {
      "epoch": 4.37,
      "learning_rate": 7.847619047619047e-07,
      "loss": 0.013,
      "step": 3850
    },
    {
      "epoch": 4.39,
      "learning_rate": 7.497278911564626e-07,
      "loss": 0.018,
      "step": 3875
    },
    {
      "epoch": 4.42,
      "learning_rate": 7.146938775510205e-07,
      "loss": 0.012,
      "step": 3900
    },
    {
      "epoch": 4.45,
      "learning_rate": 6.796598639455783e-07,
      "loss": 0.0224,
      "step": 3925
    },
    {
      "epoch": 4.48,
      "learning_rate": 6.44625850340136e-07,
      "loss": 0.0142,
      "step": 3950
    },
    {
      "epoch": 4.51,
      "learning_rate": 6.095918367346939e-07,
      "loss": 0.0167,
      "step": 3975
    },
    {
      "epoch": 4.54,
      "learning_rate": 5.745578231292517e-07,
      "loss": 0.0136,
      "step": 4000
    },
    {
      "epoch": 4.56,
      "learning_rate": 5.395238095238095e-07,
      "loss": 0.0184,
      "step": 4025
    },
    {
      "epoch": 4.59,
      "learning_rate": 5.044897959183673e-07,
      "loss": 0.0168,
      "step": 4050
    },
    {
      "epoch": 4.62,
      "learning_rate": 4.6945578231292523e-07,
      "loss": 0.0137,
      "step": 4075
    },
    {
      "epoch": 4.65,
      "learning_rate": 4.34421768707483e-07,
      "loss": 0.0133,
      "step": 4100
    },
    {
      "epoch": 4.68,
      "learning_rate": 3.9938775510204083e-07,
      "loss": 0.0141,
      "step": 4125
    },
    {
      "epoch": 4.71,
      "learning_rate": 3.6435374149659866e-07,
      "loss": 0.0125,
      "step": 4150
    },
    {
      "epoch": 4.73,
      "learning_rate": 3.293197278911565e-07,
      "loss": 0.0261,
      "step": 4175
    },
    {
      "epoch": 4.76,
      "learning_rate": 2.9428571428571426e-07,
      "loss": 0.0153,
      "step": 4200
    },
    {
      "epoch": 4.79,
      "learning_rate": 2.5925170068027214e-07,
      "loss": 0.0115,
      "step": 4225
    },
    {
      "epoch": 4.82,
      "learning_rate": 2.2421768707482994e-07,
      "loss": 0.0155,
      "step": 4250
    },
    {
      "epoch": 4.85,
      "learning_rate": 1.8918367346938774e-07,
      "loss": 0.0126,
      "step": 4275
    },
    {
      "epoch": 4.88,
      "learning_rate": 1.541496598639456e-07,
      "loss": 0.0146,
      "step": 4300
    },
    {
      "epoch": 4.9,
      "learning_rate": 1.191156462585034e-07,
      "loss": 0.0199,
      "step": 4325
    },
    {
      "epoch": 4.93,
      "learning_rate": 8.408163265306122e-08,
      "loss": 0.0149,
      "step": 4350
    },
    {
      "epoch": 4.96,
      "learning_rate": 4.9047619047619045e-08,
      "loss": 0.0104,
      "step": 4375
    },
    {
      "epoch": 4.99,
      "learning_rate": 1.4013605442176871e-08,
      "loss": 0.0121,
      "step": 4400
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.4422626793384552,
      "eval_runtime": 1364.6517,
      "eval_samples_per_second": 1.293,
      "eval_steps_per_second": 0.162,
      "eval_wer": 23.738796414852754,
      "step": 4410
    }
  ],
  "logging_steps": 25,
  "max_steps": 4410,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 7.4863010193408e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}