{
  "best_metric": 15.015790814663829,
  "best_model_checkpoint": "./whisper-small-mix-fr/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": 10.08858585357666,
      "learning_rate": 4.6000000000000004e-07,
      "loss": 0.9513,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.1840314865112305,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.7499,
      "step": 50
    },
    {
      "epoch": 0.015,
      "grad_norm": 4.318074703216553,
      "learning_rate": 1.46e-06,
      "loss": 0.5574,
      "step": 75
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.8745129108428955,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.3836,
      "step": 100
    },
    {
      "epoch": 0.025,
      "grad_norm": 3.0574047565460205,
      "learning_rate": 2.46e-06,
      "loss": 0.3209,
      "step": 125
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.6434690952301025,
      "learning_rate": 2.96e-06,
      "loss": 0.3555,
      "step": 150
    },
    {
      "epoch": 0.035,
      "grad_norm": 3.391029119491577,
      "learning_rate": 3.46e-06,
      "loss": 0.3315,
      "step": 175
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.6835551261901855,
      "learning_rate": 3.96e-06,
      "loss": 0.325,
      "step": 200
    },
    {
      "epoch": 0.045,
      "grad_norm": 2.660156011581421,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.3011,
      "step": 225
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.9171078205108643,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.2927,
      "step": 250
    },
    {
      "epoch": 0.055,
      "grad_norm": 2.8251867294311523,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.2897,
      "step": 275
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.700000762939453,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.2666,
      "step": 300
    },
    {
      "epoch": 0.065,
      "grad_norm": 2.6077873706817627,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.2599,
      "step": 325
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.5077693462371826,
      "learning_rate": 6.96e-06,
      "loss": 0.2465,
      "step": 350
    },
    {
      "epoch": 0.075,
      "grad_norm": 2.5103046894073486,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.2513,
      "step": 375
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.961869716644287,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.2471,
      "step": 400
    },
    {
      "epoch": 0.085,
      "grad_norm": 2.522207498550415,
      "learning_rate": 8.46e-06,
      "loss": 0.2353,
      "step": 425
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.2522737979888916,
      "learning_rate": 8.96e-06,
      "loss": 0.2116,
      "step": 450
    },
    {
      "epoch": 0.095,
      "grad_norm": 2.7229421138763428,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.2182,
      "step": 475
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.5425918102264404,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.239,
      "step": 500
    },
    {
      "epoch": 0.105,
      "grad_norm": 2.1395082473754883,
      "learning_rate": 9.94888888888889e-06,
      "loss": 0.2252,
      "step": 525
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.3434157371520996,
      "learning_rate": 9.893333333333334e-06,
      "loss": 0.2193,
      "step": 550
    },
    {
      "epoch": 0.115,
      "grad_norm": 2.3833093643188477,
      "learning_rate": 9.837777777777778e-06,
      "loss": 0.2198,
      "step": 575
    },
    {
      "epoch": 0.12,
      "grad_norm": 3.546076536178589,
      "learning_rate": 9.782222222222222e-06,
      "loss": 0.3714,
      "step": 600
    },
    {
      "epoch": 0.125,
      "grad_norm": 2.898831605911255,
      "learning_rate": 9.726666666666668e-06,
      "loss": 0.4111,
      "step": 625
    },
    {
      "epoch": 0.13,
      "grad_norm": 3.033583164215088,
      "learning_rate": 9.671111111111112e-06,
      "loss": 0.3474,
      "step": 650
    },
    {
      "epoch": 0.135,
      "grad_norm": 2.8390543460845947,
      "learning_rate": 9.615555555555558e-06,
      "loss": 0.2983,
      "step": 675
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.839262008666992,
      "learning_rate": 9.56e-06,
      "loss": 0.3226,
      "step": 700
    },
    {
      "epoch": 0.145,
      "grad_norm": 2.5362398624420166,
      "learning_rate": 9.504444444444446e-06,
      "loss": 0.2798,
      "step": 725
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.3447251319885254,
      "learning_rate": 9.44888888888889e-06,
      "loss": 0.2298,
      "step": 750
    },
    {
      "epoch": 0.155,
      "grad_norm": 2.498056411743164,
      "learning_rate": 9.393333333333334e-06,
      "loss": 0.2775,
      "step": 775
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.353714942932129,
      "learning_rate": 9.33777777777778e-06,
      "loss": 0.2355,
      "step": 800
    },
    {
      "epoch": 0.165,
      "grad_norm": 2.372309923171997,
      "learning_rate": 9.282222222222222e-06,
      "loss": 0.2285,
      "step": 825
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.3350610733032227,
      "learning_rate": 9.226666666666668e-06,
      "loss": 0.2159,
      "step": 850
    },
    {
      "epoch": 0.175,
      "grad_norm": 2.5473477840423584,
      "learning_rate": 9.171111111111112e-06,
      "loss": 0.1899,
      "step": 875
    },
    {
      "epoch": 0.18,
      "grad_norm": 2.363154649734497,
      "learning_rate": 9.115555555555556e-06,
      "loss": 0.2044,
      "step": 900
    },
    {
      "epoch": 0.185,
      "grad_norm": 1.9945369958877563,
      "learning_rate": 9.060000000000001e-06,
      "loss": 0.1781,
      "step": 925
    },
    {
      "epoch": 0.19,
      "grad_norm": 2.254298448562622,
      "learning_rate": 9.004444444444445e-06,
      "loss": 0.1707,
      "step": 950
    },
    {
      "epoch": 0.195,
      "grad_norm": 2.3024325370788574,
      "learning_rate": 8.94888888888889e-06,
      "loss": 0.1829,
      "step": 975
    },
    {
      "epoch": 0.2,
      "grad_norm": 2.001993417739868,
      "learning_rate": 8.893333333333333e-06,
      "loss": 0.187,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.3652883768081665,
      "eval_runtime": 728.8609,
      "eval_samples_per_second": 22.17,
      "eval_steps_per_second": 2.771,
      "eval_wer": 17.349750250342183,
      "step": 1000
    },
    {
      "epoch": 0.205,
      "grad_norm": 2.180095672607422,
      "learning_rate": 8.83777777777778e-06,
      "loss": 0.1777,
      "step": 1025
    },
    {
      "epoch": 0.21,
      "grad_norm": 2.2858028411865234,
      "learning_rate": 8.782222222222223e-06,
      "loss": 0.1726,
      "step": 1050
    },
    {
      "epoch": 0.215,
      "grad_norm": 1.9933851957321167,
      "learning_rate": 8.726666666666667e-06,
      "loss": 0.1802,
      "step": 1075
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.7360317707061768,
      "learning_rate": 8.671111111111113e-06,
      "loss": 0.1673,
      "step": 1100
    },
    {
      "epoch": 0.225,
      "grad_norm": 2.6692724227905273,
      "learning_rate": 8.615555555555555e-06,
      "loss": 0.1633,
      "step": 1125
    },
    {
      "epoch": 0.23,
      "grad_norm": 2.456207275390625,
      "learning_rate": 8.560000000000001e-06,
      "loss": 0.2043,
      "step": 1150
    },
    {
      "epoch": 0.235,
      "grad_norm": 2.5965209007263184,
      "learning_rate": 8.504444444444445e-06,
      "loss": 0.1786,
      "step": 1175
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.5705783367156982,
      "learning_rate": 8.448888888888889e-06,
      "loss": 0.2001,
      "step": 1200
    },
    {
      "epoch": 0.245,
      "grad_norm": 2.3910598754882812,
      "learning_rate": 8.393333333333335e-06,
      "loss": 0.2014,
      "step": 1225
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.1060574054718018,
      "learning_rate": 8.337777777777777e-06,
      "loss": 0.1898,
      "step": 1250
    },
    {
      "epoch": 0.255,
      "grad_norm": 1.8726638555526733,
      "learning_rate": 8.282222222222223e-06,
      "loss": 0.1763,
      "step": 1275
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.0807666778564453,
      "learning_rate": 8.226666666666667e-06,
      "loss": 0.1841,
      "step": 1300
    },
    {
      "epoch": 0.265,
      "grad_norm": 2.084286689758301,
      "learning_rate": 8.171111111111113e-06,
      "loss": 0.1675,
      "step": 1325
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.35015606880188,
      "learning_rate": 8.115555555555557e-06,
      "loss": 0.1623,
      "step": 1350
    },
    {
      "epoch": 0.275,
      "grad_norm": 2.1124699115753174,
      "learning_rate": 8.06e-06,
      "loss": 0.1607,
      "step": 1375
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.906491994857788,
      "learning_rate": 8.004444444444445e-06,
      "loss": 0.1542,
      "step": 1400
    },
    {
      "epoch": 0.285,
      "grad_norm": 2.005571126937866,
      "learning_rate": 7.948888888888889e-06,
      "loss": 0.1749,
      "step": 1425
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.8412165641784668,
      "learning_rate": 7.893333333333335e-06,
      "loss": 0.176,
      "step": 1450
    },
    {
      "epoch": 0.295,
      "grad_norm": 2.491558313369751,
      "learning_rate": 7.837777777777779e-06,
      "loss": 0.1645,
      "step": 1475
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.0074329376220703,
      "learning_rate": 7.782222222222223e-06,
      "loss": 0.1654,
      "step": 1500
    },
    {
      "epoch": 0.305,
      "grad_norm": 2.0252339839935303,
      "learning_rate": 7.726666666666667e-06,
      "loss": 0.1769,
      "step": 1525
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.3819239139556885,
      "learning_rate": 7.67111111111111e-06,
      "loss": 0.1534,
      "step": 1550
    },
    {
      "epoch": 0.315,
      "grad_norm": 2.0175583362579346,
      "learning_rate": 7.6155555555555564e-06,
      "loss": 0.1553,
      "step": 1575
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.1074907779693604,
      "learning_rate": 7.5600000000000005e-06,
      "loss": 0.1659,
      "step": 1600
    },
    {
      "epoch": 0.325,
      "grad_norm": 1.6283425092697144,
      "learning_rate": 7.504444444444445e-06,
      "loss": 0.1646,
      "step": 1625
    },
    {
      "epoch": 0.33,
      "grad_norm": 2.038062810897827,
      "learning_rate": 7.44888888888889e-06,
      "loss": 0.1689,
      "step": 1650
    },
    {
      "epoch": 0.335,
      "grad_norm": 2.047210931777954,
      "learning_rate": 7.393333333333333e-06,
      "loss": 0.1802,
      "step": 1675
    },
    {
      "epoch": 0.34,
      "grad_norm": 2.040024757385254,
      "learning_rate": 7.337777777777778e-06,
      "loss": 0.1665,
      "step": 1700
    },
    {
      "epoch": 0.345,
      "grad_norm": 1.7998043298721313,
      "learning_rate": 7.282222222222222e-06,
      "loss": 0.1586,
      "step": 1725
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.252765417098999,
      "learning_rate": 7.226666666666667e-06,
      "loss": 0.1361,
      "step": 1750
    },
    {
      "epoch": 0.355,
      "grad_norm": 2.491612672805786,
      "learning_rate": 7.171111111111112e-06,
      "loss": 0.183,
      "step": 1775
    },
    {
      "epoch": 0.36,
      "grad_norm": 2.3502893447875977,
      "learning_rate": 7.115555555555557e-06,
      "loss": 0.2771,
      "step": 1800
    },
    {
      "epoch": 0.365,
      "grad_norm": 1.9661980867385864,
      "learning_rate": 7.06e-06,
      "loss": 0.219,
      "step": 1825
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.2753665447235107,
      "learning_rate": 7.004444444444445e-06,
      "loss": 0.1976,
      "step": 1850
    },
    {
      "epoch": 0.375,
      "grad_norm": 2.1407172679901123,
      "learning_rate": 6.948888888888889e-06,
      "loss": 0.2095,
      "step": 1875
    },
    {
      "epoch": 0.38,
      "grad_norm": 2.26143479347229,
      "learning_rate": 6.893333333333334e-06,
      "loss": 0.1887,
      "step": 1900
    },
    {
      "epoch": 0.385,
      "grad_norm": 1.8551641702651978,
      "learning_rate": 6.837777777777779e-06,
      "loss": 0.1815,
      "step": 1925
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.3487014770507812,
      "learning_rate": 6.782222222222222e-06,
      "loss": 0.1885,
      "step": 1950
    },
    {
      "epoch": 0.395,
      "grad_norm": 1.6881073713302612,
      "learning_rate": 6.726666666666667e-06,
      "loss": 0.141,
      "step": 1975
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.776225209236145,
      "learning_rate": 6.671111111111112e-06,
      "loss": 0.1445,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.337929904460907,
      "eval_runtime": 715.6286,
      "eval_samples_per_second": 22.58,
      "eval_steps_per_second": 2.823,
      "eval_wer": 16.047970895128845,
      "step": 2000
    },
    {
      "epoch": 0.405,
      "grad_norm": 1.7904995679855347,
      "learning_rate": 6.615555555555556e-06,
      "loss": 0.155,
      "step": 2025
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.136061191558838,
      "learning_rate": 6.560000000000001e-06,
      "loss": 0.2267,
      "step": 2050
    },
    {
      "epoch": 0.415,
      "grad_norm": 1.9061017036437988,
      "learning_rate": 6.504444444444446e-06,
      "loss": 0.2217,
      "step": 2075
    },
    {
      "epoch": 0.42,
      "grad_norm": 2.478090524673462,
      "learning_rate": 6.448888888888889e-06,
      "loss": 0.2131,
      "step": 2100
    },
    {
      "epoch": 0.425,
      "grad_norm": 1.8599025011062622,
      "learning_rate": 6.393333333333334e-06,
      "loss": 0.2291,
      "step": 2125
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.7993838787078857,
      "learning_rate": 6.3377777777777786e-06,
      "loss": 0.2152,
      "step": 2150
    },
    {
      "epoch": 0.435,
      "grad_norm": 1.959346890449524,
      "learning_rate": 6.282222222222223e-06,
      "loss": 0.2315,
      "step": 2175
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.289241313934326,
      "learning_rate": 6.2266666666666675e-06,
      "loss": 0.2424,
      "step": 2200
    },
    {
      "epoch": 0.445,
      "grad_norm": 2.3349103927612305,
      "learning_rate": 6.171111111111112e-06,
      "loss": 0.2149,
      "step": 2225
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.915467381477356,
      "learning_rate": 6.1155555555555555e-06,
      "loss": 0.1946,
      "step": 2250
    },
    {
      "epoch": 0.455,
      "grad_norm": 1.8727192878723145,
      "learning_rate": 6.0600000000000004e-06,
      "loss": 0.195,
      "step": 2275
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.938673734664917,
      "learning_rate": 6.004444444444445e-06,
      "loss": 0.1997,
      "step": 2300
    },
    {
      "epoch": 0.465,
      "grad_norm": 2.3887195587158203,
      "learning_rate": 5.948888888888889e-06,
      "loss": 0.2204,
      "step": 2325
    },
    {
      "epoch": 0.47,
      "grad_norm": 2.328936815261841,
      "learning_rate": 5.893333333333334e-06,
      "loss": 0.2219,
      "step": 2350
    },
    {
      "epoch": 0.475,
      "grad_norm": 1.9731664657592773,
      "learning_rate": 5.837777777777777e-06,
      "loss": 0.1999,
      "step": 2375
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.532160520553589,
      "learning_rate": 5.782222222222222e-06,
      "loss": 0.1592,
      "step": 2400
    },
    {
      "epoch": 0.485,
      "grad_norm": 2.0180513858795166,
      "learning_rate": 5.726666666666667e-06,
      "loss": 0.1483,
      "step": 2425
    },
    {
      "epoch": 0.49,
      "grad_norm": 2.120114326477051,
      "learning_rate": 5.671111111111112e-06,
      "loss": 0.1432,
      "step": 2450
    },
    {
      "epoch": 0.495,
      "grad_norm": 2.1796326637268066,
      "learning_rate": 5.615555555555556e-06,
      "loss": 0.1409,
      "step": 2475
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.18973708152771,
      "learning_rate": 5.560000000000001e-06,
      "loss": 0.1423,
      "step": 2500
    },
    {
      "epoch": 0.505,
      "grad_norm": 1.9231557846069336,
      "learning_rate": 5.504444444444444e-06,
      "loss": 0.1392,
      "step": 2525
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.988946795463562,
      "learning_rate": 5.448888888888889e-06,
      "loss": 0.171,
      "step": 2550
    },
    {
      "epoch": 0.515,
      "grad_norm": 2.1168086528778076,
      "learning_rate": 5.393333333333334e-06,
      "loss": 0.1573,
      "step": 2575
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.9249277114868164,
      "learning_rate": 5.337777777777779e-06,
      "loss": 0.1947,
      "step": 2600
    },
    {
      "epoch": 0.525,
      "grad_norm": 1.9734596014022827,
      "learning_rate": 5.282222222222223e-06,
      "loss": 0.2473,
      "step": 2625
    },
    {
      "epoch": 0.53,
      "grad_norm": 2.1487271785736084,
      "learning_rate": 5.226666666666667e-06,
      "loss": 0.2281,
      "step": 2650
    },
    {
      "epoch": 0.535,
      "grad_norm": 2.1245877742767334,
      "learning_rate": 5.171111111111111e-06,
      "loss": 0.2103,
      "step": 2675
    },
    {
      "epoch": 0.54,
      "grad_norm": 2.845296859741211,
      "learning_rate": 5.115555555555556e-06,
      "loss": 0.1695,
      "step": 2700
    },
    {
      "epoch": 0.545,
      "grad_norm": 1.894158124923706,
      "learning_rate": 5.060000000000001e-06,
      "loss": 0.1626,
      "step": 2725
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.936550259590149,
      "learning_rate": 5.004444444444445e-06,
      "loss": 0.1296,
      "step": 2750
    },
    {
      "epoch": 0.555,
      "grad_norm": 1.6843122243881226,
      "learning_rate": 4.94888888888889e-06,
      "loss": 0.141,
      "step": 2775
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.5642544031143188,
      "learning_rate": 4.893333333333334e-06,
      "loss": 0.1503,
      "step": 2800
    },
    {
      "epoch": 0.565,
      "grad_norm": 1.9369734525680542,
      "learning_rate": 4.837777777777778e-06,
      "loss": 0.1251,
      "step": 2825
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.033722400665283,
      "learning_rate": 4.7822222222222226e-06,
      "loss": 0.1464,
      "step": 2850
    },
    {
      "epoch": 0.575,
      "grad_norm": 1.9096460342407227,
      "learning_rate": 4.7266666666666674e-06,
      "loss": 0.1292,
      "step": 2875
    },
    {
      "epoch": 0.58,
      "grad_norm": 2.1668505668640137,
      "learning_rate": 4.6711111111111115e-06,
      "loss": 0.1495,
      "step": 2900
    },
    {
      "epoch": 0.585,
      "grad_norm": 1.7556490898132324,
      "learning_rate": 4.6155555555555555e-06,
      "loss": 0.1201,
      "step": 2925
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.806643009185791,
      "learning_rate": 4.56e-06,
      "loss": 0.1559,
      "step": 2950
    },
    {
      "epoch": 0.595,
      "grad_norm": 2.2240452766418457,
      "learning_rate": 4.504444444444444e-06,
      "loss": 0.1916,
      "step": 2975
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.006112813949585,
      "learning_rate": 4.448888888888889e-06,
      "loss": 0.1659,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.3254563808441162,
      "eval_runtime": 713.8266,
      "eval_samples_per_second": 22.637,
      "eval_steps_per_second": 2.83,
      "eval_wer": 15.377231600590155,
      "step": 3000
    },
    {
      "epoch": 0.605,
      "grad_norm": 2.583686590194702,
      "learning_rate": 4.393333333333334e-06,
      "loss": 0.18,
      "step": 3025
    },
    {
      "epoch": 0.61,
      "grad_norm": 2.474371910095215,
      "learning_rate": 4.337777777777778e-06,
      "loss": 0.1708,
      "step": 3050
    },
    {
      "epoch": 0.615,
      "grad_norm": 1.7476117610931396,
      "learning_rate": 4.282222222222222e-06,
      "loss": 0.1694,
      "step": 3075
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.8390570878982544,
      "learning_rate": 4.226666666666667e-06,
      "loss": 0.1527,
      "step": 3100
    },
    {
      "epoch": 0.625,
      "grad_norm": 2.105689287185669,
      "learning_rate": 4.171111111111111e-06,
      "loss": 0.1589,
      "step": 3125
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.7435396909713745,
      "learning_rate": 4.115555555555556e-06,
      "loss": 0.142,
      "step": 3150
    },
    {
      "epoch": 0.635,
      "grad_norm": 1.6353161334991455,
      "learning_rate": 4.060000000000001e-06,
      "loss": 0.1432,
      "step": 3175
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.6273831129074097,
      "learning_rate": 4.004444444444445e-06,
      "loss": 0.1357,
      "step": 3200
    },
    {
      "epoch": 0.645,
      "grad_norm": 1.8896713256835938,
      "learning_rate": 3.948888888888889e-06,
      "loss": 0.1283,
      "step": 3225
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.4946659803390503,
      "learning_rate": 3.893333333333333e-06,
      "loss": 0.1138,
      "step": 3250
    },
    {
      "epoch": 0.655,
      "grad_norm": 1.7483470439910889,
      "learning_rate": 3.837777777777778e-06,
      "loss": 0.1196,
      "step": 3275
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.8768987655639648,
      "learning_rate": 3.782222222222223e-06,
      "loss": 0.1269,
      "step": 3300
    },
    {
      "epoch": 0.665,
      "grad_norm": 1.7175124883651733,
      "learning_rate": 3.726666666666667e-06,
      "loss": 0.146,
      "step": 3325
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.719070553779602,
      "learning_rate": 3.6711111111111113e-06,
      "loss": 0.1364,
      "step": 3350
    },
    {
      "epoch": 0.675,
      "grad_norm": 1.8970329761505127,
      "learning_rate": 3.615555555555556e-06,
      "loss": 0.1345,
      "step": 3375
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.9742193222045898,
      "learning_rate": 3.5600000000000002e-06,
      "loss": 0.1271,
      "step": 3400
    },
    {
      "epoch": 0.685,
      "grad_norm": 1.763426423072815,
      "learning_rate": 3.5044444444444447e-06,
      "loss": 0.1366,
      "step": 3425
    },
    {
      "epoch": 0.69,
      "grad_norm": 2.260113477706909,
      "learning_rate": 3.4488888888888896e-06,
      "loss": 0.151,
      "step": 3450
    },
    {
      "epoch": 0.695,
      "grad_norm": 1.8039120435714722,
      "learning_rate": 3.3933333333333336e-06,
      "loss": 0.1399,
      "step": 3475
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.8737797737121582,
      "learning_rate": 3.337777777777778e-06,
      "loss": 0.1357,
      "step": 3500
    },
    {
      "epoch": 0.705,
      "grad_norm": 1.9594708681106567,
      "learning_rate": 3.282222222222223e-06,
      "loss": 0.1381,
      "step": 3525
    },
    {
      "epoch": 0.71,
      "grad_norm": 2.234194278717041,
      "learning_rate": 3.226666666666667e-06,
      "loss": 0.1216,
      "step": 3550
    },
    {
      "epoch": 0.715,
      "grad_norm": 2.309079647064209,
      "learning_rate": 3.1711111111111114e-06,
      "loss": 0.209,
      "step": 3575
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.2596418857574463,
      "learning_rate": 3.1155555555555555e-06,
      "loss": 0.1934,
      "step": 3600
    },
    {
      "epoch": 0.725,
      "grad_norm": 2.279223918914795,
      "learning_rate": 3.0600000000000003e-06,
      "loss": 0.2069,
      "step": 3625
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.8627978563308716,
      "learning_rate": 3.004444444444445e-06,
      "loss": 0.2093,
      "step": 3650
    },
    {
      "epoch": 0.735,
      "grad_norm": 2.2204179763793945,
      "learning_rate": 2.948888888888889e-06,
      "loss": 0.2281,
      "step": 3675
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.9367740154266357,
      "learning_rate": 2.8933333333333337e-06,
      "loss": 0.2236,
      "step": 3700
    },
    {
      "epoch": 0.745,
      "grad_norm": 2.149648427963257,
      "learning_rate": 2.837777777777778e-06,
      "loss": 0.2187,
      "step": 3725
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.0620782375335693,
      "learning_rate": 2.7822222222222222e-06,
      "loss": 0.2031,
      "step": 3750
    },
    {
      "epoch": 0.755,
      "grad_norm": 1.87172532081604,
      "learning_rate": 2.726666666666667e-06,
      "loss": 0.185,
      "step": 3775
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.9918657541275024,
      "learning_rate": 2.6711111111111116e-06,
      "loss": 0.1866,
      "step": 3800
    },
    {
      "epoch": 0.765,
      "grad_norm": 1.9320180416107178,
      "learning_rate": 2.6155555555555556e-06,
      "loss": 0.1862,
      "step": 3825
    },
    {
      "epoch": 0.77,
      "grad_norm": 2.31062650680542,
      "learning_rate": 2.56e-06,
      "loss": 0.2484,
      "step": 3850
    },
    {
      "epoch": 0.775,
      "grad_norm": 2.489675283432007,
      "learning_rate": 2.504444444444445e-06,
      "loss": 0.2913,
      "step": 3875
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.7794305086135864,
      "learning_rate": 2.448888888888889e-06,
      "loss": 0.2571,
      "step": 3900
    },
    {
      "epoch": 0.785,
      "grad_norm": 1.6303505897521973,
      "learning_rate": 2.3933333333333334e-06,
      "loss": 0.1488,
      "step": 3925
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.9400599002838135,
      "learning_rate": 2.337777777777778e-06,
      "loss": 0.1121,
      "step": 3950
    },
    {
      "epoch": 0.795,
      "grad_norm": 1.8359787464141846,
      "learning_rate": 2.2822222222222223e-06,
      "loss": 0.1516,
      "step": 3975
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.1083908081054688,
      "learning_rate": 2.226666666666667e-06,
      "loss": 0.1594,
      "step": 4000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.31362661719322205,
      "eval_runtime": 714.1383,
      "eval_samples_per_second": 22.627,
      "eval_steps_per_second": 2.829,
      "eval_wer": 15.195918681748424,
      "step": 4000
    },
    {
      "epoch": 0.805,
      "grad_norm": 2.0564794540405273,
      "learning_rate": 2.1711111111111113e-06,
      "loss": 0.1503,
      "step": 4025
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.9138249158859253,
      "learning_rate": 2.1155555555555557e-06,
      "loss": 0.1554,
      "step": 4050
    },
    {
      "epoch": 0.815,
      "grad_norm": 2.1700758934020996,
      "learning_rate": 2.06e-06,
      "loss": 0.1744,
      "step": 4075
    },
    {
      "epoch": 0.82,
      "grad_norm": 2.3517487049102783,
      "learning_rate": 2.0044444444444446e-06,
      "loss": 0.1572,
      "step": 4100
    },
    {
      "epoch": 0.825,
      "grad_norm": 2.2419686317443848,
      "learning_rate": 1.948888888888889e-06,
      "loss": 0.1772,
      "step": 4125
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.8083751201629639,
      "learning_rate": 1.8933333333333333e-06,
      "loss": 0.1927,
      "step": 4150
    },
    {
      "epoch": 0.835,
      "grad_norm": 1.8185911178588867,
      "learning_rate": 1.837777777777778e-06,
      "loss": 0.1648,
      "step": 4175
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.9195200204849243,
      "learning_rate": 1.7822222222222225e-06,
      "loss": 0.1487,
      "step": 4200
    },
    {
      "epoch": 0.845,
      "grad_norm": 1.9267535209655762,
      "learning_rate": 1.7266666666666667e-06,
      "loss": 0.1351,
      "step": 4225
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.8880534172058105,
      "learning_rate": 1.6711111111111112e-06,
      "loss": 0.1314,
      "step": 4250
    },
    {
      "epoch": 0.855,
      "grad_norm": 1.6127861738204956,
      "learning_rate": 1.6155555555555559e-06,
      "loss": 0.1276,
      "step": 4275
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.9241377115249634,
      "learning_rate": 1.56e-06,
      "loss": 0.1328,
      "step": 4300
    },
    {
      "epoch": 0.865,
      "grad_norm": 1.9935392141342163,
      "learning_rate": 1.5044444444444446e-06,
      "loss": 0.1336,
      "step": 4325
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.5158283710479736,
      "learning_rate": 1.4488888888888892e-06,
      "loss": 0.1275,
      "step": 4350
    },
    {
      "epoch": 0.875,
      "grad_norm": 1.600599765777588,
      "learning_rate": 1.3933333333333335e-06,
      "loss": 0.1245,
      "step": 4375
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.8442989587783813,
      "learning_rate": 1.337777777777778e-06,
      "loss": 0.1321,
      "step": 4400
    },
    {
      "epoch": 0.885,
      "grad_norm": 2.629706621170044,
      "learning_rate": 1.2822222222222222e-06,
      "loss": 0.2734,
      "step": 4425
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.6115596294403076,
      "learning_rate": 1.2266666666666666e-06,
      "loss": 0.322,
      "step": 4450
    },
    {
      "epoch": 0.895,
      "grad_norm": 2.2008018493652344,
      "learning_rate": 1.171111111111111e-06,
      "loss": 0.3303,
      "step": 4475
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.9677278995513916,
      "learning_rate": 1.1155555555555558e-06,
      "loss": 0.2398,
      "step": 4500
    },
    {
      "epoch": 0.905,
      "grad_norm": 1.6797411441802979,
      "learning_rate": 1.06e-06,
      "loss": 0.1363,
      "step": 4525
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.017742156982422,
      "learning_rate": 1.0044444444444445e-06,
      "loss": 0.1638,
      "step": 4550
    },
    {
      "epoch": 0.915,
      "grad_norm": 2.1174466609954834,
      "learning_rate": 9.488888888888889e-07,
      "loss": 0.1516,
      "step": 4575
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.573102593421936,
      "learning_rate": 8.933333333333334e-07,
      "loss": 0.1523,
      "step": 4600
    },
    {
      "epoch": 0.925,
      "grad_norm": 2.056255578994751,
      "learning_rate": 8.37777777777778e-07,
      "loss": 0.1539,
      "step": 4625
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.6231415271759033,
      "learning_rate": 7.822222222222223e-07,
      "loss": 0.1549,
      "step": 4650
    },
    {
      "epoch": 0.935,
      "grad_norm": 1.890496850013733,
      "learning_rate": 7.266666666666668e-07,
      "loss": 0.1548,
      "step": 4675
    },
    {
      "epoch": 0.94,
      "grad_norm": 1.7783973217010498,
      "learning_rate": 6.711111111111111e-07,
      "loss": 0.1373,
      "step": 4700
    },
    {
      "epoch": 0.945,
      "grad_norm": 2.0187580585479736,
      "learning_rate": 6.155555555555556e-07,
      "loss": 0.15,
      "step": 4725
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.9916925430297852,
      "learning_rate": 5.6e-07,
      "loss": 0.1395,
      "step": 4750
    },
    {
      "epoch": 0.955,
      "grad_norm": 1.9438104629516602,
      "learning_rate": 5.044444444444445e-07,
      "loss": 0.1187,
      "step": 4775
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.21679425239563,
      "learning_rate": 4.488888888888889e-07,
      "loss": 0.1349,
      "step": 4800
    },
    {
      "epoch": 0.965,
      "grad_norm": 1.6693533658981323,
      "learning_rate": 3.9333333333333336e-07,
      "loss": 0.1444,
      "step": 4825
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.4958335161209106,
      "learning_rate": 3.3777777777777777e-07,
      "loss": 0.1357,
      "step": 4850
    },
    {
      "epoch": 0.975,
      "grad_norm": 2.0575404167175293,
      "learning_rate": 2.822222222222222e-07,
      "loss": 0.1274,
      "step": 4875
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.9877169132232666,
      "learning_rate": 2.266666666666667e-07,
      "loss": 0.125,
      "step": 4900
    },
    {
      "epoch": 0.985,
      "grad_norm": 2.0602986812591553,
      "learning_rate": 1.7111111111111114e-07,
      "loss": 0.1416,
      "step": 4925
    },
    {
      "epoch": 0.99,
      "grad_norm": 2.5087714195251465,
      "learning_rate": 1.1555555555555556e-07,
      "loss": 0.1439,
      "step": 4950
    },
    {
      "epoch": 0.995,
      "grad_norm": 2.067476749420166,
      "learning_rate": 6.000000000000001e-08,
      "loss": 0.1382,
      "step": 4975
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.8874026536941528,
      "learning_rate": 4.444444444444445e-09,
      "loss": 0.1371,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.3092347979545593,
      "eval_runtime": 711.7473,
      "eval_samples_per_second": 22.703,
      "eval_steps_per_second": 2.838,
      "eval_wer": 15.015790814663829,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "step": 5000,
      "total_flos": 9.23473281024e+19,
      "train_loss": 0.1970298689365387,
      "train_runtime": 18494.1427,
      "train_samples_per_second": 17.303,
      "train_steps_per_second": 0.27
    }
  ],
  "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": 9.23473281024e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}