{
  "best_metric": 6.840122206312234,
  "best_model_checkpoint": "./whisper-medium-mix-it/checkpoint-5000",
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005,
      "grad_norm": 7.006679058074951,
      "learning_rate": 4.6000000000000004e-07,
      "loss": 0.7121,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.37919282913208,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.5421,
      "step": 50
    },
    {
      "epoch": 0.015,
      "grad_norm": 3.432912826538086,
      "learning_rate": 1.46e-06,
      "loss": 0.3692,
      "step": 75
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.942035436630249,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.2601,
      "step": 100
    },
    {
      "epoch": 0.025,
      "grad_norm": 2.9149224758148193,
      "learning_rate": 2.46e-06,
      "loss": 0.2557,
      "step": 125
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.536968231201172,
      "learning_rate": 2.96e-06,
      "loss": 0.216,
      "step": 150
    },
    {
      "epoch": 0.035,
      "grad_norm": 4.307021617889404,
      "learning_rate": 3.46e-06,
      "loss": 0.2004,
      "step": 175
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.456475019454956,
      "learning_rate": 3.96e-06,
      "loss": 0.2118,
      "step": 200
    },
    {
      "epoch": 0.045,
      "grad_norm": 3.5058531761169434,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.2071,
      "step": 225
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.056401014328003,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.2103,
      "step": 250
    },
    {
      "epoch": 0.055,
      "grad_norm": 3.7942306995391846,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.2138,
      "step": 275
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.421769857406616,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.2114,
      "step": 300
    },
    {
      "epoch": 0.065,
      "grad_norm": 3.5847620964050293,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.1987,
      "step": 325
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.585036039352417,
      "learning_rate": 6.96e-06,
      "loss": 0.1886,
      "step": 350
    },
    {
      "epoch": 0.075,
      "grad_norm": 3.3235092163085938,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.1883,
      "step": 375
    },
    {
      "epoch": 0.08,
      "grad_norm": 4.223800182342529,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.2115,
      "step": 400
    },
    {
      "epoch": 0.085,
      "grad_norm": 3.307344675064087,
      "learning_rate": 8.46e-06,
      "loss": 0.1942,
      "step": 425
    },
    {
      "epoch": 0.09,
      "grad_norm": 3.1430752277374268,
      "learning_rate": 8.96e-06,
      "loss": 0.1796,
      "step": 450
    },
    {
      "epoch": 0.095,
      "grad_norm": 3.7946882247924805,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.2053,
      "step": 475
    },
    {
      "epoch": 0.1,
      "grad_norm": 3.383258581161499,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.1918,
      "step": 500
    },
    {
      "epoch": 0.105,
      "grad_norm": 3.1705129146575928,
      "learning_rate": 9.94888888888889e-06,
      "loss": 0.2061,
      "step": 525
    },
    {
      "epoch": 0.11,
      "grad_norm": 3.299571990966797,
      "learning_rate": 9.893333333333334e-06,
      "loss": 0.1891,
      "step": 550
    },
    {
      "epoch": 0.115,
      "grad_norm": 3.413519859313965,
      "learning_rate": 9.837777777777778e-06,
      "loss": 0.2221,
      "step": 575
    },
    {
      "epoch": 0.12,
      "grad_norm": 3.8573830127716064,
      "learning_rate": 9.782222222222222e-06,
      "loss": 0.2092,
      "step": 600
    },
    {
      "epoch": 0.125,
      "grad_norm": 2.475337266921997,
      "learning_rate": 9.726666666666668e-06,
      "loss": 0.2199,
      "step": 625
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.221552610397339,
      "learning_rate": 9.671111111111112e-06,
      "loss": 0.1787,
      "step": 650
    },
    {
      "epoch": 0.135,
      "grad_norm": 3.197284460067749,
      "learning_rate": 9.615555555555558e-06,
      "loss": 0.2044,
      "step": 675
    },
    {
      "epoch": 0.14,
      "grad_norm": 3.177967071533203,
      "learning_rate": 9.56e-06,
      "loss": 0.196,
      "step": 700
    },
    {
      "epoch": 0.145,
      "grad_norm": 3.4896106719970703,
      "learning_rate": 9.504444444444446e-06,
      "loss": 0.1887,
      "step": 725
    },
    {
      "epoch": 0.15,
      "grad_norm": 3.654257297515869,
      "learning_rate": 9.44888888888889e-06,
      "loss": 0.1979,
      "step": 750
    },
    {
      "epoch": 0.155,
      "grad_norm": 3.537205934524536,
      "learning_rate": 9.393333333333334e-06,
      "loss": 0.1855,
      "step": 775
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.230680227279663,
      "learning_rate": 9.33777777777778e-06,
      "loss": 0.1668,
      "step": 800
    },
    {
      "epoch": 0.165,
      "grad_norm": 2.3759679794311523,
      "learning_rate": 9.282222222222222e-06,
      "loss": 0.1643,
      "step": 825
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.8923466205596924,
      "learning_rate": 9.226666666666668e-06,
      "loss": 0.1781,
      "step": 850
    },
    {
      "epoch": 0.175,
      "grad_norm": 2.956648349761963,
      "learning_rate": 9.171111111111112e-06,
      "loss": 0.1708,
      "step": 875
    },
    {
      "epoch": 0.18,
      "grad_norm": 2.7545344829559326,
      "learning_rate": 9.115555555555556e-06,
      "loss": 0.161,
      "step": 900
    },
    {
      "epoch": 0.185,
      "grad_norm": 2.0833823680877686,
      "learning_rate": 9.060000000000001e-06,
      "loss": 0.1576,
      "step": 925
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.689265727996826,
      "learning_rate": 9.004444444444445e-06,
      "loss": 0.1702,
      "step": 950
    },
    {
      "epoch": 0.195,
      "grad_norm": 2.9991567134857178,
      "learning_rate": 8.94888888888889e-06,
      "loss": 0.1571,
      "step": 975
    },
    {
      "epoch": 0.2,
      "grad_norm": 2.6324071884155273,
      "learning_rate": 8.893333333333333e-06,
      "loss": 0.1502,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.17080245912075043,
      "eval_runtime": 1706.7457,
      "eval_samples_per_second": 8.879,
      "eval_steps_per_second": 1.11,
      "eval_wer": 9.092228813503402,
      "step": 1000
    },
    {
      "epoch": 0.205,
      "grad_norm": 3.2417123317718506,
      "learning_rate": 8.83777777777778e-06,
      "loss": 0.1723,
      "step": 1025
    },
    {
      "epoch": 0.21,
      "grad_norm": 4.017546653747559,
      "learning_rate": 8.782222222222223e-06,
      "loss": 0.1652,
      "step": 1050
    },
    {
      "epoch": 0.215,
      "grad_norm": 2.1816272735595703,
      "learning_rate": 8.726666666666667e-06,
      "loss": 0.1738,
      "step": 1075
    },
    {
      "epoch": 0.22,
      "grad_norm": 2.869568347930908,
      "learning_rate": 8.671111111111113e-06,
      "loss": 0.1975,
      "step": 1100
    },
    {
      "epoch": 0.225,
      "grad_norm": 3.047701835632324,
      "learning_rate": 8.615555555555555e-06,
      "loss": 0.2191,
      "step": 1125
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.380408763885498,
      "learning_rate": 8.560000000000001e-06,
      "loss": 0.2155,
      "step": 1150
    },
    {
      "epoch": 0.235,
      "grad_norm": 2.9988808631896973,
      "learning_rate": 8.504444444444445e-06,
      "loss": 0.2288,
      "step": 1175
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.327162027359009,
      "learning_rate": 8.448888888888889e-06,
      "loss": 0.2262,
      "step": 1200
    },
    {
      "epoch": 0.245,
      "grad_norm": 2.988764762878418,
      "learning_rate": 8.393333333333335e-06,
      "loss": 0.191,
      "step": 1225
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.3341264724731445,
      "learning_rate": 8.337777777777777e-06,
      "loss": 0.1974,
      "step": 1250
    },
    {
      "epoch": 0.255,
      "grad_norm": 3.1267290115356445,
      "learning_rate": 8.282222222222223e-06,
      "loss": 0.1951,
      "step": 1275
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.5510425567626953,
      "learning_rate": 8.226666666666667e-06,
      "loss": 0.1787,
      "step": 1300
    },
    {
      "epoch": 0.265,
      "grad_norm": 2.5202274322509766,
      "learning_rate": 8.171111111111113e-06,
      "loss": 0.1549,
      "step": 1325
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.8440310955047607,
      "learning_rate": 8.115555555555557e-06,
      "loss": 0.1858,
      "step": 1350
    },
    {
      "epoch": 0.275,
      "grad_norm": 2.4390523433685303,
      "learning_rate": 8.06e-06,
      "loss": 0.1696,
      "step": 1375
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.6690866947174072,
      "learning_rate": 8.004444444444445e-06,
      "loss": 0.1502,
      "step": 1400
    },
    {
      "epoch": 0.285,
      "grad_norm": 2.090963125228882,
      "learning_rate": 7.948888888888889e-06,
      "loss": 0.1478,
      "step": 1425
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.3862431049346924,
      "learning_rate": 7.893333333333335e-06,
      "loss": 0.1563,
      "step": 1450
    },
    {
      "epoch": 0.295,
      "grad_norm": 2.7396063804626465,
      "learning_rate": 7.837777777777779e-06,
      "loss": 0.1589,
      "step": 1475
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.534780979156494,
      "learning_rate": 7.782222222222223e-06,
      "loss": 0.1488,
      "step": 1500
    },
    {
      "epoch": 0.305,
      "grad_norm": 2.8536627292633057,
      "learning_rate": 7.726666666666667e-06,
      "loss": 0.1731,
      "step": 1525
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.810988664627075,
      "learning_rate": 7.67111111111111e-06,
      "loss": 0.1465,
      "step": 1550
    },
    {
      "epoch": 0.315,
      "grad_norm": 3.3764379024505615,
      "learning_rate": 7.6155555555555564e-06,
      "loss": 0.2259,
      "step": 1575
    },
    {
      "epoch": 0.32,
      "grad_norm": 3.14015793800354,
      "learning_rate": 7.5600000000000005e-06,
      "loss": 0.238,
      "step": 1600
    },
    {
      "epoch": 0.325,
      "grad_norm": 2.997870683670044,
      "learning_rate": 7.504444444444445e-06,
      "loss": 0.2381,
      "step": 1625
    },
    {
      "epoch": 0.33,
      "grad_norm": 2.7839579582214355,
      "learning_rate": 7.44888888888889e-06,
      "loss": 0.2352,
      "step": 1650
    },
    {
      "epoch": 0.335,
      "grad_norm": 2.8877196311950684,
      "learning_rate": 7.393333333333333e-06,
      "loss": 0.2431,
      "step": 1675
    },
    {
      "epoch": 0.34,
      "grad_norm": 3.2856152057647705,
      "learning_rate": 7.337777777777778e-06,
      "loss": 0.2279,
      "step": 1700
    },
    {
      "epoch": 0.345,
      "grad_norm": 3.3398728370666504,
      "learning_rate": 7.282222222222222e-06,
      "loss": 0.2156,
      "step": 1725
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.628492593765259,
      "learning_rate": 7.226666666666667e-06,
      "loss": 0.1938,
      "step": 1750
    },
    {
      "epoch": 0.355,
      "grad_norm": 3.0386624336242676,
      "learning_rate": 7.171111111111112e-06,
      "loss": 0.1899,
      "step": 1775
    },
    {
      "epoch": 0.36,
      "grad_norm": 2.6567234992980957,
      "learning_rate": 7.115555555555557e-06,
      "loss": 0.1997,
      "step": 1800
    },
    {
      "epoch": 0.365,
      "grad_norm": 2.5155370235443115,
      "learning_rate": 7.06e-06,
      "loss": 0.1518,
      "step": 1825
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.6725313663482666,
      "learning_rate": 7.004444444444445e-06,
      "loss": 0.1554,
      "step": 1850
    },
    {
      "epoch": 0.375,
      "grad_norm": 2.657243013381958,
      "learning_rate": 6.948888888888889e-06,
      "loss": 0.1529,
      "step": 1875
    },
    {
      "epoch": 0.38,
      "grad_norm": 2.400428295135498,
      "learning_rate": 6.893333333333334e-06,
      "loss": 0.1556,
      "step": 1900
    },
    {
      "epoch": 0.385,
      "grad_norm": 2.190220594406128,
      "learning_rate": 6.837777777777779e-06,
      "loss": 0.1417,
      "step": 1925
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.150498867034912,
      "learning_rate": 6.782222222222222e-06,
      "loss": 0.1512,
      "step": 1950
    },
    {
      "epoch": 0.395,
      "grad_norm": 2.2915945053100586,
      "learning_rate": 6.726666666666667e-06,
      "loss": 0.1559,
      "step": 1975
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.43928599357605,
      "learning_rate": 6.671111111111112e-06,
      "loss": 0.1584,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.1553979516029358,
      "eval_runtime": 1688.9101,
      "eval_samples_per_second": 8.973,
      "eval_steps_per_second": 1.122,
      "eval_wer": 8.175681471754636,
      "step": 2000
    },
    {
      "epoch": 0.405,
      "grad_norm": 2.3533153533935547,
      "learning_rate": 6.615555555555556e-06,
      "loss": 0.1576,
      "step": 2025
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.9315130710601807,
      "learning_rate": 6.560000000000001e-06,
      "loss": 0.1551,
      "step": 2050
    },
    {
      "epoch": 0.415,
      "grad_norm": 2.920414924621582,
      "learning_rate": 6.504444444444446e-06,
      "loss": 0.153,
      "step": 2075
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.833433985710144,
      "learning_rate": 6.448888888888889e-06,
      "loss": 0.1299,
      "step": 2100
    },
    {
      "epoch": 0.425,
      "grad_norm": 2.8404743671417236,
      "learning_rate": 6.393333333333334e-06,
      "loss": 0.1449,
      "step": 2125
    },
    {
      "epoch": 0.43,
      "grad_norm": 2.2315993309020996,
      "learning_rate": 6.3377777777777786e-06,
      "loss": 0.1469,
      "step": 2150
    },
    {
      "epoch": 0.435,
      "grad_norm": 3.2478203773498535,
      "learning_rate": 6.282222222222223e-06,
      "loss": 0.1611,
      "step": 2175
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.173109292984009,
      "learning_rate": 6.2266666666666675e-06,
      "loss": 0.1383,
      "step": 2200
    },
    {
      "epoch": 0.445,
      "grad_norm": 2.417632818222046,
      "learning_rate": 6.171111111111112e-06,
      "loss": 0.1473,
      "step": 2225
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.384589672088623,
      "learning_rate": 6.1155555555555555e-06,
      "loss": 0.128,
      "step": 2250
    },
    {
      "epoch": 0.455,
      "grad_norm": 2.1940202713012695,
      "learning_rate": 6.0600000000000004e-06,
      "loss": 0.1494,
      "step": 2275
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.8651658296585083,
      "learning_rate": 6.004444444444445e-06,
      "loss": 0.1411,
      "step": 2300
    },
    {
      "epoch": 0.465,
      "grad_norm": 2.0271828174591064,
      "learning_rate": 5.948888888888889e-06,
      "loss": 0.1373,
      "step": 2325
    },
    {
      "epoch": 0.47,
      "grad_norm": 2.5282301902770996,
      "learning_rate": 5.893333333333334e-06,
      "loss": 0.1382,
      "step": 2350
    },
    {
      "epoch": 0.475,
      "grad_norm": 2.3264193534851074,
      "learning_rate": 5.837777777777777e-06,
      "loss": 0.1297,
      "step": 2375
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.045041561126709,
      "learning_rate": 5.782222222222222e-06,
      "loss": 0.1337,
      "step": 2400
    },
    {
      "epoch": 0.485,
      "grad_norm": 1.7588191032409668,
      "learning_rate": 5.726666666666667e-06,
      "loss": 0.1274,
      "step": 2425
    },
    {
      "epoch": 0.49,
      "grad_norm": 2.436593770980835,
      "learning_rate": 5.671111111111112e-06,
      "loss": 0.1198,
      "step": 2450
    },
    {
      "epoch": 0.495,
      "grad_norm": 1.9072929620742798,
      "learning_rate": 5.615555555555556e-06,
      "loss": 0.1228,
      "step": 2475
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.490530490875244,
      "learning_rate": 5.560000000000001e-06,
      "loss": 0.1137,
      "step": 2500
    },
    {
      "epoch": 0.505,
      "grad_norm": 2.258913516998291,
      "learning_rate": 5.504444444444444e-06,
      "loss": 0.1389,
      "step": 2525
    },
    {
      "epoch": 0.51,
      "grad_norm": 2.099606990814209,
      "learning_rate": 5.448888888888889e-06,
      "loss": 0.1396,
      "step": 2550
    },
    {
      "epoch": 0.515,
      "grad_norm": 2.8082237243652344,
      "learning_rate": 5.393333333333334e-06,
      "loss": 0.1709,
      "step": 2575
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.9639875888824463,
      "learning_rate": 5.337777777777779e-06,
      "loss": 0.1857,
      "step": 2600
    },
    {
      "epoch": 0.525,
      "grad_norm": 2.2722744941711426,
      "learning_rate": 5.282222222222223e-06,
      "loss": 0.1761,
      "step": 2625
    },
    {
      "epoch": 0.53,
      "grad_norm": 3.0935170650482178,
      "learning_rate": 5.226666666666667e-06,
      "loss": 0.1566,
      "step": 2650
    },
    {
      "epoch": 0.535,
      "grad_norm": 1.6952731609344482,
      "learning_rate": 5.171111111111111e-06,
      "loss": 0.1426,
      "step": 2675
    },
    {
      "epoch": 0.54,
      "grad_norm": 2.025660991668701,
      "learning_rate": 5.115555555555556e-06,
      "loss": 0.1525,
      "step": 2700
    },
    {
      "epoch": 0.545,
      "grad_norm": 2.3472840785980225,
      "learning_rate": 5.060000000000001e-06,
      "loss": 0.1468,
      "step": 2725
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.9560490846633911,
      "learning_rate": 5.004444444444445e-06,
      "loss": 0.1391,
      "step": 2750
    },
    {
      "epoch": 0.555,
      "grad_norm": 2.2275073528289795,
      "learning_rate": 4.94888888888889e-06,
      "loss": 0.1444,
      "step": 2775
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.6753711700439453,
      "learning_rate": 4.893333333333334e-06,
      "loss": 0.1474,
      "step": 2800
    },
    {
      "epoch": 0.565,
      "grad_norm": 2.616377353668213,
      "learning_rate": 4.837777777777778e-06,
      "loss": 0.1266,
      "step": 2825
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.25164532661438,
      "learning_rate": 4.7822222222222226e-06,
      "loss": 0.1167,
      "step": 2850
    },
    {
      "epoch": 0.575,
      "grad_norm": 2.1922624111175537,
      "learning_rate": 4.7266666666666674e-06,
      "loss": 0.107,
      "step": 2875
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.9851558208465576,
      "learning_rate": 4.6711111111111115e-06,
      "loss": 0.1185,
      "step": 2900
    },
    {
      "epoch": 0.585,
      "grad_norm": 2.494908571243286,
      "learning_rate": 4.6155555555555555e-06,
      "loss": 0.121,
      "step": 2925
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.9508136510849,
      "learning_rate": 4.56e-06,
      "loss": 0.1134,
      "step": 2950
    },
    {
      "epoch": 0.595,
      "grad_norm": 2.1634881496429443,
      "learning_rate": 4.504444444444444e-06,
      "loss": 0.1155,
      "step": 2975
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.3343162536621094,
      "learning_rate": 4.448888888888889e-06,
      "loss": 0.1309,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.1425519436597824,
      "eval_runtime": 1689.609,
      "eval_samples_per_second": 8.97,
      "eval_steps_per_second": 1.122,
      "eval_wer": 7.414201534837376,
      "step": 3000
    },
    {
      "epoch": 0.605,
      "grad_norm": 1.8268561363220215,
      "learning_rate": 4.393333333333334e-06,
      "loss": 0.0938,
      "step": 3025
    },
    {
      "epoch": 0.61,
      "grad_norm": 2.4155433177948,
      "learning_rate": 4.337777777777778e-06,
      "loss": 0.1574,
      "step": 3050
    },
    {
      "epoch": 0.615,
      "grad_norm": 2.5877668857574463,
      "learning_rate": 4.282222222222222e-06,
      "loss": 0.165,
      "step": 3075
    },
    {
      "epoch": 0.62,
      "grad_norm": 2.576686382293701,
      "learning_rate": 4.226666666666667e-06,
      "loss": 0.1851,
      "step": 3100
    },
    {
      "epoch": 0.625,
      "grad_norm": 3.1633784770965576,
      "learning_rate": 4.171111111111111e-06,
      "loss": 0.1663,
      "step": 3125
    },
    {
      "epoch": 0.63,
      "grad_norm": 2.219902753829956,
      "learning_rate": 4.115555555555556e-06,
      "loss": 0.1451,
      "step": 3150
    },
    {
      "epoch": 0.635,
      "grad_norm": 1.9273529052734375,
      "learning_rate": 4.060000000000001e-06,
      "loss": 0.1177,
      "step": 3175
    },
    {
      "epoch": 0.64,
      "grad_norm": 2.305210590362549,
      "learning_rate": 4.004444444444445e-06,
      "loss": 0.1125,
      "step": 3200
    },
    {
      "epoch": 0.645,
      "grad_norm": 2.5342929363250732,
      "learning_rate": 3.948888888888889e-06,
      "loss": 0.1072,
      "step": 3225
    },
    {
      "epoch": 0.65,
      "grad_norm": 2.423017978668213,
      "learning_rate": 3.893333333333333e-06,
      "loss": 0.1262,
      "step": 3250
    },
    {
      "epoch": 0.655,
      "grad_norm": 2.373019218444824,
      "learning_rate": 3.837777777777778e-06,
      "loss": 0.118,
      "step": 3275
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.5576426982879639,
      "learning_rate": 3.782222222222223e-06,
      "loss": 0.097,
      "step": 3300
    },
    {
      "epoch": 0.665,
      "grad_norm": 1.6943529844284058,
      "learning_rate": 3.726666666666667e-06,
      "loss": 0.1067,
      "step": 3325
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.728334903717041,
      "learning_rate": 3.6711111111111113e-06,
      "loss": 0.1024,
      "step": 3350
    },
    {
      "epoch": 0.675,
      "grad_norm": 2.135171890258789,
      "learning_rate": 3.615555555555556e-06,
      "loss": 0.1093,
      "step": 3375
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.618424415588379,
      "learning_rate": 3.5600000000000002e-06,
      "loss": 0.1017,
      "step": 3400
    },
    {
      "epoch": 0.685,
      "grad_norm": 1.7630062103271484,
      "learning_rate": 3.5044444444444447e-06,
      "loss": 0.1177,
      "step": 3425
    },
    {
      "epoch": 0.69,
      "grad_norm": 3.093877077102661,
      "learning_rate": 3.4488888888888896e-06,
      "loss": 0.1212,
      "step": 3450
    },
    {
      "epoch": 0.695,
      "grad_norm": 2.9288673400878906,
      "learning_rate": 3.3933333333333336e-06,
      "loss": 0.1296,
      "step": 3475
    },
    {
      "epoch": 0.7,
      "grad_norm": 2.8730647563934326,
      "learning_rate": 3.337777777777778e-06,
      "loss": 0.1247,
      "step": 3500
    },
    {
      "epoch": 0.705,
      "grad_norm": 2.404902696609497,
      "learning_rate": 3.282222222222223e-06,
      "loss": 0.118,
      "step": 3525
    },
    {
      "epoch": 0.71,
      "grad_norm": 2.426954984664917,
      "learning_rate": 3.226666666666667e-06,
      "loss": 0.1381,
      "step": 3550
    },
    {
      "epoch": 0.715,
      "grad_norm": 2.6250641345977783,
      "learning_rate": 3.1711111111111114e-06,
      "loss": 0.1375,
      "step": 3575
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.184222459793091,
      "learning_rate": 3.1155555555555555e-06,
      "loss": 0.1227,
      "step": 3600
    },
    {
      "epoch": 0.725,
      "grad_norm": 2.2613589763641357,
      "learning_rate": 3.0600000000000003e-06,
      "loss": 0.1225,
      "step": 3625
    },
    {
      "epoch": 0.73,
      "grad_norm": 2.7714171409606934,
      "learning_rate": 3.004444444444445e-06,
      "loss": 0.1263,
      "step": 3650
    },
    {
      "epoch": 0.735,
      "grad_norm": 1.5858285427093506,
      "learning_rate": 2.948888888888889e-06,
      "loss": 0.1101,
      "step": 3675
    },
    {
      "epoch": 0.74,
      "grad_norm": 2.5831713676452637,
      "learning_rate": 2.8933333333333337e-06,
      "loss": 0.0983,
      "step": 3700
    },
    {
      "epoch": 0.745,
      "grad_norm": 2.260019302368164,
      "learning_rate": 2.837777777777778e-06,
      "loss": 0.1016,
      "step": 3725
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.7528841495513916,
      "learning_rate": 2.7822222222222222e-06,
      "loss": 0.0913,
      "step": 3750
    },
    {
      "epoch": 0.755,
      "grad_norm": 2.207277536392212,
      "learning_rate": 2.726666666666667e-06,
      "loss": 0.106,
      "step": 3775
    },
    {
      "epoch": 0.76,
      "grad_norm": 2.2939164638519287,
      "learning_rate": 2.6711111111111116e-06,
      "loss": 0.1077,
      "step": 3800
    },
    {
      "epoch": 0.765,
      "grad_norm": 3.1979310512542725,
      "learning_rate": 2.6155555555555556e-06,
      "loss": 0.1117,
      "step": 3825
    },
    {
      "epoch": 0.77,
      "grad_norm": 2.3320703506469727,
      "learning_rate": 2.56e-06,
      "loss": 0.1517,
      "step": 3850
    },
    {
      "epoch": 0.775,
      "grad_norm": 2.1557302474975586,
      "learning_rate": 2.504444444444445e-06,
      "loss": 0.1413,
      "step": 3875
    },
    {
      "epoch": 0.78,
      "grad_norm": 2.062872886657715,
      "learning_rate": 2.448888888888889e-06,
      "loss": 0.131,
      "step": 3900
    },
    {
      "epoch": 0.785,
      "grad_norm": 1.8930553197860718,
      "learning_rate": 2.3933333333333334e-06,
      "loss": 0.1329,
      "step": 3925
    },
    {
      "epoch": 0.79,
      "grad_norm": 2.496727228164673,
      "learning_rate": 2.337777777777778e-06,
      "loss": 0.1188,
      "step": 3950
    },
    {
      "epoch": 0.795,
      "grad_norm": 2.155111312866211,
      "learning_rate": 2.2822222222222223e-06,
      "loss": 0.0917,
      "step": 3975
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.717517137527466,
      "learning_rate": 2.226666666666667e-06,
      "loss": 0.0984,
      "step": 4000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.13701671361923218,
      "eval_runtime": 1725.4047,
      "eval_samples_per_second": 8.783,
      "eval_steps_per_second": 1.098,
      "eval_wer": 7.129801315763426,
      "step": 4000
    },
    {
      "epoch": 0.805,
      "grad_norm": 1.4921413660049438,
      "learning_rate": 2.1711111111111113e-06,
      "loss": 0.0898,
      "step": 4025
    },
    {
      "epoch": 0.81,
      "grad_norm": 2.083139657974243,
      "learning_rate": 2.1155555555555557e-06,
      "loss": 0.0922,
      "step": 4050
    },
    {
      "epoch": 0.815,
      "grad_norm": 2.0208919048309326,
      "learning_rate": 2.06e-06,
      "loss": 0.0958,
      "step": 4075
    },
    {
      "epoch": 0.82,
      "grad_norm": 2.9720261096954346,
      "learning_rate": 2.0044444444444446e-06,
      "loss": 0.1046,
      "step": 4100
    },
    {
      "epoch": 0.825,
      "grad_norm": 2.5535240173339844,
      "learning_rate": 1.948888888888889e-06,
      "loss": 0.1217,
      "step": 4125
    },
    {
      "epoch": 0.83,
      "grad_norm": 2.354639768600464,
      "learning_rate": 1.8933333333333333e-06,
      "loss": 0.1314,
      "step": 4150
    },
    {
      "epoch": 0.835,
      "grad_norm": 2.3064558506011963,
      "learning_rate": 1.837777777777778e-06,
      "loss": 0.1135,
      "step": 4175
    },
    {
      "epoch": 0.84,
      "grad_norm": 3.168377161026001,
      "learning_rate": 1.7822222222222225e-06,
      "loss": 0.1533,
      "step": 4200
    },
    {
      "epoch": 0.845,
      "grad_norm": 1.9521429538726807,
      "learning_rate": 1.728888888888889e-06,
      "loss": 0.2723,
      "step": 4225
    },
    {
      "epoch": 0.85,
      "grad_norm": 2.471975326538086,
      "learning_rate": 1.6733333333333335e-06,
      "loss": 0.2678,
      "step": 4250
    },
    {
      "epoch": 0.855,
      "grad_norm": 2.516317367553711,
      "learning_rate": 1.6177777777777778e-06,
      "loss": 0.2768,
      "step": 4275
    },
    {
      "epoch": 0.86,
      "grad_norm": 2.5380780696868896,
      "learning_rate": 1.5622222222222225e-06,
      "loss": 0.2323,
      "step": 4300
    },
    {
      "epoch": 0.865,
      "grad_norm": 1.7794257402420044,
      "learning_rate": 1.506666666666667e-06,
      "loss": 0.2052,
      "step": 4325
    },
    {
      "epoch": 0.87,
      "grad_norm": 2.1327309608459473,
      "learning_rate": 1.4511111111111112e-06,
      "loss": 0.1311,
      "step": 4350
    },
    {
      "epoch": 0.875,
      "grad_norm": 1.5396376848220825,
      "learning_rate": 1.3955555555555556e-06,
      "loss": 0.087,
      "step": 4375
    },
    {
      "epoch": 0.88,
      "grad_norm": 2.444408416748047,
      "learning_rate": 1.34e-06,
      "loss": 0.1117,
      "step": 4400
    },
    {
      "epoch": 0.885,
      "grad_norm": 1.7176605463027954,
      "learning_rate": 1.2844444444444445e-06,
      "loss": 0.0818,
      "step": 4425
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.6177054643630981,
      "learning_rate": 1.228888888888889e-06,
      "loss": 0.0777,
      "step": 4450
    },
    {
      "epoch": 0.895,
      "grad_norm": 2.648822069168091,
      "learning_rate": 1.1733333333333335e-06,
      "loss": 0.0901,
      "step": 4475
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.8845601081848145,
      "learning_rate": 1.117777777777778e-06,
      "loss": 0.0755,
      "step": 4500
    },
    {
      "epoch": 0.905,
      "grad_norm": 2.268307685852051,
      "learning_rate": 1.0622222222222222e-06,
      "loss": 0.0878,
      "step": 4525
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.6278784275054932,
      "learning_rate": 1.0066666666666668e-06,
      "loss": 0.1034,
      "step": 4550
    },
    {
      "epoch": 0.915,
      "grad_norm": 1.8028929233551025,
      "learning_rate": 9.511111111111111e-07,
      "loss": 0.0883,
      "step": 4575
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.7106696367263794,
      "learning_rate": 8.955555555555557e-07,
      "loss": 0.0709,
      "step": 4600
    },
    {
      "epoch": 0.925,
      "grad_norm": 1.5025192499160767,
      "learning_rate": 8.400000000000001e-07,
      "loss": 0.0815,
      "step": 4625
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.554764747619629,
      "learning_rate": 7.844444444444445e-07,
      "loss": 0.1045,
      "step": 4650
    },
    {
      "epoch": 0.935,
      "grad_norm": 2.26786732673645,
      "learning_rate": 7.28888888888889e-07,
      "loss": 0.1028,
      "step": 4675
    },
    {
      "epoch": 0.94,
      "grad_norm": 2.1246206760406494,
      "learning_rate": 6.733333333333334e-07,
      "loss": 0.11,
      "step": 4700
    },
    {
      "epoch": 0.945,
      "grad_norm": 2.038825750350952,
      "learning_rate": 6.177777777777778e-07,
      "loss": 0.1106,
      "step": 4725
    },
    {
      "epoch": 0.95,
      "grad_norm": 2.3029303550720215,
      "learning_rate": 5.622222222222223e-07,
      "loss": 0.1064,
      "step": 4750
    },
    {
      "epoch": 0.955,
      "grad_norm": 1.985775113105774,
      "learning_rate": 5.066666666666667e-07,
      "loss": 0.0967,
      "step": 4775
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.736402988433838,
      "learning_rate": 4.511111111111111e-07,
      "loss": 0.1028,
      "step": 4800
    },
    {
      "epoch": 0.965,
      "grad_norm": 2.5676920413970947,
      "learning_rate": 3.9555555555555557e-07,
      "loss": 0.1014,
      "step": 4825
    },
    {
      "epoch": 0.97,
      "grad_norm": 2.4653942584991455,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 0.1041,
      "step": 4850
    },
    {
      "epoch": 0.975,
      "grad_norm": 2.217700242996216,
      "learning_rate": 2.844444444444445e-07,
      "loss": 0.1098,
      "step": 4875
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.7745823860168457,
      "learning_rate": 2.2888888888888892e-07,
      "loss": 0.1013,
      "step": 4900
    },
    {
      "epoch": 0.985,
      "grad_norm": 1.8243005275726318,
      "learning_rate": 1.7333333333333335e-07,
      "loss": 0.1014,
      "step": 4925
    },
    {
      "epoch": 0.99,
      "grad_norm": 2.028550386428833,
      "learning_rate": 1.1777777777777778e-07,
      "loss": 0.0992,
      "step": 4950
    },
    {
      "epoch": 0.995,
      "grad_norm": 1.5399481058120728,
      "learning_rate": 6.222222222222223e-08,
      "loss": 0.0927,
      "step": 4975
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.7385917901992798,
      "learning_rate": 6.666666666666667e-09,
      "loss": 0.0933,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.13182643055915833,
      "eval_runtime": 1689.9204,
      "eval_samples_per_second": 8.968,
      "eval_steps_per_second": 1.121,
      "eval_wer": 6.840122206312234,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "step": 5000,
      "total_flos": 1.632967852032e+20,
      "train_loss": 0.15716482088565825,
      "train_runtime": 22600.0372,
      "train_samples_per_second": 7.08,
      "train_steps_per_second": 0.221
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.632967852032e+20,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}