{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 30.01,
  "eval_steps": 3000,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008333333333333333,
      "grad_norm": 44.35525131225586,
      "learning_rate": 4.2000000000000006e-07,
      "loss": 2.441,
      "step": 25
    },
    {
      "epoch": 0.016666666666666666,
      "grad_norm": 22.19506072998047,
      "learning_rate": 9.200000000000001e-07,
      "loss": 1.7682,
      "step": 50
    },
    {
      "epoch": 0.025,
      "grad_norm": 9.717977523803711,
      "learning_rate": 1.42e-06,
      "loss": 0.9699,
      "step": 75
    },
    {
      "epoch": 1.0003333333333333,
      "grad_norm": 7.14320182800293,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.5701,
      "step": 100
    },
    {
      "epoch": 1.0086666666666666,
      "grad_norm": 3.4074296951293945,
      "learning_rate": 2.42e-06,
      "loss": 0.3157,
      "step": 125
    },
    {
      "epoch": 1.017,
      "grad_norm": 3.311967372894287,
      "learning_rate": 2.92e-06,
      "loss": 0.2732,
      "step": 150
    },
    {
      "epoch": 1.0253333333333334,
      "grad_norm": 3.092350959777832,
      "learning_rate": 3.4200000000000007e-06,
      "loss": 0.2653,
      "step": 175
    },
    {
      "epoch": 2.0006666666666666,
      "grad_norm": 2.89245343208313,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.2502,
      "step": 200
    },
    {
      "epoch": 2.009,
      "grad_norm": 3.0979723930358887,
      "learning_rate": 4.42e-06,
      "loss": 0.1824,
      "step": 225
    },
    {
      "epoch": 2.017333333333333,
      "grad_norm": 3.1348249912261963,
      "learning_rate": 4.92e-06,
      "loss": 0.1828,
      "step": 250
    },
    {
      "epoch": 2.0256666666666665,
      "grad_norm": 2.990718364715576,
      "learning_rate": 5.420000000000001e-06,
      "loss": 0.187,
      "step": 275
    },
    {
      "epoch": 3.001,
      "grad_norm": 2.026869058609009,
      "learning_rate": 5.92e-06,
      "loss": 0.1859,
      "step": 300
    },
    {
      "epoch": 3.009333333333333,
      "grad_norm": 1.9744936227798462,
      "learning_rate": 6.42e-06,
      "loss": 0.1179,
      "step": 325
    },
    {
      "epoch": 3.0176666666666665,
      "grad_norm": 2.247405767440796,
      "learning_rate": 6.92e-06,
      "loss": 0.1157,
      "step": 350
    },
    {
      "epoch": 3.026,
      "grad_norm": 2.4352807998657227,
      "learning_rate": 7.420000000000001e-06,
      "loss": 0.1191,
      "step": 375
    },
    {
      "epoch": 4.001333333333333,
      "grad_norm": 1.5341213941574097,
      "learning_rate": 7.92e-06,
      "loss": 0.1136,
      "step": 400
    },
    {
      "epoch": 4.009666666666667,
      "grad_norm": 1.683750033378601,
      "learning_rate": 8.42e-06,
      "loss": 0.0566,
      "step": 425
    },
    {
      "epoch": 4.018,
      "grad_norm": 1.9196466207504272,
      "learning_rate": 8.920000000000001e-06,
      "loss": 0.0575,
      "step": 450
    },
    {
      "epoch": 4.0263333333333335,
      "grad_norm": 1.5782861709594727,
      "learning_rate": 9.42e-06,
      "loss": 0.0651,
      "step": 475
    },
    {
      "epoch": 5.001666666666667,
      "grad_norm": 1.724906325340271,
      "learning_rate": 9.920000000000002e-06,
      "loss": 0.0546,
      "step": 500
    },
    {
      "epoch": 5.01,
      "grad_norm": 1.6357895135879517,
      "learning_rate": 9.916000000000001e-06,
      "loss": 0.0253,
      "step": 525
    },
    {
      "epoch": 5.0183333333333335,
      "grad_norm": 1.2953633069992065,
      "learning_rate": 9.816e-06,
      "loss": 0.0254,
      "step": 550
    },
    {
      "epoch": 5.026666666666666,
      "grad_norm": 1.3112106323242188,
      "learning_rate": 9.716000000000002e-06,
      "loss": 0.0281,
      "step": 575
    },
    {
      "epoch": 6.002,
      "grad_norm": 1.5078754425048828,
      "learning_rate": 9.616e-06,
      "loss": 0.0243,
      "step": 600
    },
    {
      "epoch": 6.0103333333333335,
      "grad_norm": 0.5965685248374939,
      "learning_rate": 9.516e-06,
      "loss": 0.0149,
      "step": 625
    },
    {
      "epoch": 6.018666666666666,
      "grad_norm": 0.7676089406013489,
      "learning_rate": 9.416000000000001e-06,
      "loss": 0.0133,
      "step": 650
    },
    {
      "epoch": 6.027,
      "grad_norm": 0.9765186905860901,
      "learning_rate": 9.316e-06,
      "loss": 0.0137,
      "step": 675
    },
    {
      "epoch": 7.0023333333333335,
      "grad_norm": 1.1814440488815308,
      "learning_rate": 9.216000000000001e-06,
      "loss": 0.0145,
      "step": 700
    },
    {
      "epoch": 7.010666666666666,
      "grad_norm": 0.573243260383606,
      "learning_rate": 9.116e-06,
      "loss": 0.0095,
      "step": 725
    },
    {
      "epoch": 7.019,
      "grad_norm": 0.9459896683692932,
      "learning_rate": 9.016e-06,
      "loss": 0.0066,
      "step": 750
    },
    {
      "epoch": 7.027333333333333,
      "grad_norm": 1.4142969846725464,
      "learning_rate": 8.916e-06,
      "loss": 0.0076,
      "step": 775
    },
    {
      "epoch": 8.002666666666666,
      "grad_norm": 0.46104899048805237,
      "learning_rate": 8.816000000000002e-06,
      "loss": 0.0078,
      "step": 800
    },
    {
      "epoch": 8.011,
      "grad_norm": 0.6141907572746277,
      "learning_rate": 8.716000000000001e-06,
      "loss": 0.0064,
      "step": 825
    },
    {
      "epoch": 8.019333333333334,
      "grad_norm": 0.41213399171829224,
      "learning_rate": 8.616000000000002e-06,
      "loss": 0.005,
      "step": 850
    },
    {
      "epoch": 8.027666666666667,
      "grad_norm": 3.1000335216522217,
      "learning_rate": 8.516000000000001e-06,
      "loss": 0.0049,
      "step": 875
    },
    {
      "epoch": 9.003,
      "grad_norm": 0.7775946259498596,
      "learning_rate": 8.416e-06,
      "loss": 0.0047,
      "step": 900
    },
    {
      "epoch": 9.011333333333333,
      "grad_norm": 0.1238495409488678,
      "learning_rate": 8.316000000000001e-06,
      "loss": 0.0039,
      "step": 925
    },
    {
      "epoch": 9.019666666666666,
      "grad_norm": 0.17236466705799103,
      "learning_rate": 8.216e-06,
      "loss": 0.0034,
      "step": 950
    },
    {
      "epoch": 9.028,
      "grad_norm": 0.5582916140556335,
      "learning_rate": 8.116e-06,
      "loss": 0.0039,
      "step": 975
    },
    {
      "epoch": 10.003333333333334,
      "grad_norm": 0.07784730941057205,
      "learning_rate": 8.016e-06,
      "loss": 0.0027,
      "step": 1000
    },
    {
      "epoch": 10.011666666666667,
      "grad_norm": 0.07262728363275528,
      "learning_rate": 7.916e-06,
      "loss": 0.002,
      "step": 1025
    },
    {
      "epoch": 10.02,
      "grad_norm": 0.08925045281648636,
      "learning_rate": 7.816000000000001e-06,
      "loss": 0.0028,
      "step": 1050
    },
    {
      "epoch": 10.028333333333334,
      "grad_norm": 0.39910703897476196,
      "learning_rate": 7.716e-06,
      "loss": 0.002,
      "step": 1075
    },
    {
      "epoch": 11.003666666666666,
      "grad_norm": 0.0856546089053154,
      "learning_rate": 7.616000000000001e-06,
      "loss": 0.0015,
      "step": 1100
    },
    {
      "epoch": 11.012,
      "grad_norm": 0.8903342485427856,
      "learning_rate": 7.516000000000001e-06,
      "loss": 0.004,
      "step": 1125
    },
    {
      "epoch": 11.020333333333333,
      "grad_norm": 0.06683026254177094,
      "learning_rate": 7.416000000000001e-06,
      "loss": 0.0017,
      "step": 1150
    },
    {
      "epoch": 11.028666666666666,
      "grad_norm": 0.3519723415374756,
      "learning_rate": 7.316000000000001e-06,
      "loss": 0.0013,
      "step": 1175
    },
    {
      "epoch": 12.004,
      "grad_norm": 0.05283417925238609,
      "learning_rate": 7.216000000000001e-06,
      "loss": 0.0012,
      "step": 1200
    },
    {
      "epoch": 12.012333333333334,
      "grad_norm": 0.797774076461792,
      "learning_rate": 7.116000000000001e-06,
      "loss": 0.0018,
      "step": 1225
    },
    {
      "epoch": 12.020666666666667,
      "grad_norm": 0.26817482709884644,
      "learning_rate": 7.016e-06,
      "loss": 0.0022,
      "step": 1250
    },
    {
      "epoch": 12.029,
      "grad_norm": 0.04436012730002403,
      "learning_rate": 6.916e-06,
      "loss": 0.0013,
      "step": 1275
    },
    {
      "epoch": 13.004333333333333,
      "grad_norm": 0.039572861045598984,
      "learning_rate": 6.8160000000000005e-06,
      "loss": 0.0012,
      "step": 1300
    },
    {
      "epoch": 13.012666666666666,
      "grad_norm": 0.06484813243150711,
      "learning_rate": 6.716000000000001e-06,
      "loss": 0.0009,
      "step": 1325
    },
    {
      "epoch": 13.021,
      "grad_norm": 0.02630372904241085,
      "learning_rate": 6.616e-06,
      "loss": 0.0011,
      "step": 1350
    },
    {
      "epoch": 13.029333333333334,
      "grad_norm": 0.04369799420237541,
      "learning_rate": 6.516e-06,
      "loss": 0.0013,
      "step": 1375
    },
    {
      "epoch": 14.004666666666667,
      "grad_norm": 0.036155689507722855,
      "learning_rate": 6.416e-06,
      "loss": 0.0009,
      "step": 1400
    },
    {
      "epoch": 14.013,
      "grad_norm": 0.024141667410731316,
      "learning_rate": 6.316000000000001e-06,
      "loss": 0.0006,
      "step": 1425
    },
    {
      "epoch": 14.021333333333333,
      "grad_norm": 0.1784670650959015,
      "learning_rate": 6.216000000000001e-06,
      "loss": 0.0007,
      "step": 1450
    },
    {
      "epoch": 14.029666666666667,
      "grad_norm": 0.023341720923781395,
      "learning_rate": 6.116000000000001e-06,
      "loss": 0.0012,
      "step": 1475
    },
    {
      "epoch": 15.005,
      "grad_norm": 0.018143419176340103,
      "learning_rate": 6.0160000000000005e-06,
      "loss": 0.0008,
      "step": 1500
    },
    {
      "epoch": 15.013333333333334,
      "grad_norm": 0.02102494426071644,
      "learning_rate": 5.916000000000001e-06,
      "loss": 0.0007,
      "step": 1525
    },
    {
      "epoch": 15.021666666666667,
      "grad_norm": 0.023690875619649887,
      "learning_rate": 5.816000000000001e-06,
      "loss": 0.0006,
      "step": 1550
    },
    {
      "epoch": 15.03,
      "grad_norm": 0.04580727219581604,
      "learning_rate": 5.716000000000001e-06,
      "loss": 0.001,
      "step": 1575
    },
    {
      "epoch": 16.005333333333333,
      "grad_norm": 0.018508635461330414,
      "learning_rate": 5.616e-06,
      "loss": 0.0005,
      "step": 1600
    },
    {
      "epoch": 16.013666666666666,
      "grad_norm": 0.019894391298294067,
      "learning_rate": 5.516e-06,
      "loss": 0.0005,
      "step": 1625
    },
    {
      "epoch": 16.022,
      "grad_norm": 0.022277018055319786,
      "learning_rate": 5.416e-06,
      "loss": 0.0005,
      "step": 1650
    },
    {
      "epoch": 16.030333333333335,
      "grad_norm": 0.02107921987771988,
      "learning_rate": 5.3160000000000004e-06,
      "loss": 0.0006,
      "step": 1675
    },
    {
      "epoch": 17.005666666666666,
      "grad_norm": 0.0199015811085701,
      "learning_rate": 5.216e-06,
      "loss": 0.0004,
      "step": 1700
    },
    {
      "epoch": 17.014,
      "grad_norm": 0.014493563212454319,
      "learning_rate": 5.116000000000001e-06,
      "loss": 0.0004,
      "step": 1725
    },
    {
      "epoch": 17.022333333333332,
      "grad_norm": 0.017925165593624115,
      "learning_rate": 5.016000000000001e-06,
      "loss": 0.0008,
      "step": 1750
    },
    {
      "epoch": 17.030666666666665,
      "grad_norm": 0.014746254310011864,
      "learning_rate": 4.916e-06,
      "loss": 0.0004,
      "step": 1775
    },
    {
      "epoch": 18.006,
      "grad_norm": 0.01628812402486801,
      "learning_rate": 4.816e-06,
      "loss": 0.0004,
      "step": 1800
    },
    {
      "epoch": 18.014333333333333,
      "grad_norm": 0.015700463205575943,
      "learning_rate": 4.716e-06,
      "loss": 0.0005,
      "step": 1825
    },
    {
      "epoch": 18.022666666666666,
      "grad_norm": 0.012755331583321095,
      "learning_rate": 4.616e-06,
      "loss": 0.0005,
      "step": 1850
    },
    {
      "epoch": 18.031,
      "grad_norm": 0.018821561709046364,
      "learning_rate": 4.5160000000000005e-06,
      "loss": 0.0004,
      "step": 1875
    },
    {
      "epoch": 19.006333333333334,
      "grad_norm": 0.016657203435897827,
      "learning_rate": 4.416000000000001e-06,
      "loss": 0.0004,
      "step": 1900
    },
    {
      "epoch": 19.014666666666667,
      "grad_norm": 0.016449443995952606,
      "learning_rate": 4.316e-06,
      "loss": 0.0005,
      "step": 1925
    },
    {
      "epoch": 19.023,
      "grad_norm": 0.015951134264469147,
      "learning_rate": 4.216e-06,
      "loss": 0.0008,
      "step": 1950
    },
    {
      "epoch": 19.031333333333333,
      "grad_norm": 0.013866595923900604,
      "learning_rate": 4.116000000000001e-06,
      "loss": 0.0004,
      "step": 1975
    },
    {
      "epoch": 20.006666666666668,
      "grad_norm": 0.01308687124401331,
      "learning_rate": 4.016e-06,
      "loss": 0.0006,
      "step": 2000
    },
    {
      "epoch": 20.015,
      "grad_norm": 0.012808864936232567,
      "learning_rate": 3.916e-06,
      "loss": 0.0004,
      "step": 2025
    },
    {
      "epoch": 20.023333333333333,
      "grad_norm": 0.013972673565149307,
      "learning_rate": 3.816e-06,
      "loss": 0.0004,
      "step": 2050
    },
    {
      "epoch": 20.031666666666666,
      "grad_norm": 0.01527604553848505,
      "learning_rate": 3.716e-06,
      "loss": 0.0003,
      "step": 2075
    },
    {
      "epoch": 21.007,
      "grad_norm": 0.03139733895659447,
      "learning_rate": 3.616e-06,
      "loss": 0.0003,
      "step": 2100
    },
    {
      "epoch": 21.015333333333334,
      "grad_norm": 0.011607775464653969,
      "learning_rate": 3.5160000000000007e-06,
      "loss": 0.0003,
      "step": 2125
    },
    {
      "epoch": 21.023666666666667,
      "grad_norm": 0.012830990366637707,
      "learning_rate": 3.4160000000000004e-06,
      "loss": 0.0004,
      "step": 2150
    },
    {
      "epoch": 21.032,
      "grad_norm": 0.013420588336884975,
      "learning_rate": 3.3160000000000005e-06,
      "loss": 0.0004,
      "step": 2175
    },
    {
      "epoch": 22.00733333333333,
      "grad_norm": 0.013782207854092121,
      "learning_rate": 3.216e-06,
      "loss": 0.0003,
      "step": 2200
    },
    {
      "epoch": 22.015666666666668,
      "grad_norm": 0.009835809469223022,
      "learning_rate": 3.1160000000000003e-06,
      "loss": 0.0003,
      "step": 2225
    },
    {
      "epoch": 22.024,
      "grad_norm": 0.012228522449731827,
      "learning_rate": 3.016e-06,
      "loss": 0.0003,
      "step": 2250
    },
    {
      "epoch": 22.032333333333334,
      "grad_norm": 0.010861766524612904,
      "learning_rate": 2.9160000000000005e-06,
      "loss": 0.0004,
      "step": 2275
    },
    {
      "epoch": 23.007666666666665,
      "grad_norm": 0.0120700066909194,
      "learning_rate": 2.8160000000000002e-06,
      "loss": 0.0003,
      "step": 2300
    },
    {
      "epoch": 23.016,
      "grad_norm": 0.013041774742305279,
      "learning_rate": 2.7160000000000003e-06,
      "loss": 0.0003,
      "step": 2325
    },
    {
      "epoch": 23.024333333333335,
      "grad_norm": 0.01479949988424778,
      "learning_rate": 2.616e-06,
      "loss": 0.0003,
      "step": 2350
    },
    {
      "epoch": 23.032666666666668,
      "grad_norm": 0.010125643573701382,
      "learning_rate": 2.516e-06,
      "loss": 0.0003,
      "step": 2375
    },
    {
      "epoch": 24.008,
      "grad_norm": 0.011404238641262054,
      "learning_rate": 2.4160000000000002e-06,
      "loss": 0.0003,
      "step": 2400
    },
    {
      "epoch": 24.016333333333332,
      "grad_norm": 0.01367497444152832,
      "learning_rate": 2.3160000000000004e-06,
      "loss": 0.0003,
      "step": 2425
    },
    {
      "epoch": 24.02466666666667,
      "grad_norm": 0.0132884681224823,
      "learning_rate": 2.216e-06,
      "loss": 0.0003,
      "step": 2450
    },
    {
      "epoch": 24.033,
      "grad_norm": 0.01917206309735775,
      "learning_rate": 2.116e-06,
      "loss": 0.0003,
      "step": 2475
    },
    {
      "epoch": 25.008333333333333,
      "grad_norm": 0.012251622974872589,
      "learning_rate": 2.0160000000000003e-06,
      "loss": 0.0003,
      "step": 2500
    },
    {
      "epoch": 25.016666666666666,
      "grad_norm": 0.20725885033607483,
      "learning_rate": 1.916e-06,
      "loss": 0.0003,
      "step": 2525
    },
    {
      "epoch": 25.025,
      "grad_norm": 0.01084393635392189,
      "learning_rate": 1.8160000000000003e-06,
      "loss": 0.0003,
      "step": 2550
    },
    {
      "epoch": 26.000333333333334,
      "grad_norm": 0.01044237520545721,
      "learning_rate": 1.7160000000000002e-06,
      "loss": 0.0003,
      "step": 2575
    },
    {
      "epoch": 26.008666666666667,
      "grad_norm": 0.010727889835834503,
      "learning_rate": 1.616e-06,
      "loss": 0.0003,
      "step": 2600
    },
    {
      "epoch": 26.017,
      "grad_norm": 0.009735923260450363,
      "learning_rate": 1.5160000000000002e-06,
      "loss": 0.0003,
      "step": 2625
    },
    {
      "epoch": 26.025333333333332,
      "grad_norm": 0.010894649662077427,
      "learning_rate": 1.416e-06,
      "loss": 0.0003,
      "step": 2650
    },
    {
      "epoch": 27.000666666666667,
      "grad_norm": 0.009900378994643688,
      "learning_rate": 1.316e-06,
      "loss": 0.0002,
      "step": 2675
    },
    {
      "epoch": 27.009,
      "grad_norm": 0.010404952801764011,
      "learning_rate": 1.216e-06,
      "loss": 0.0002,
      "step": 2700
    },
    {
      "epoch": 27.017333333333333,
      "grad_norm": 0.009290441870689392,
      "learning_rate": 1.1160000000000002e-06,
      "loss": 0.0002,
      "step": 2725
    },
    {
      "epoch": 27.025666666666666,
      "grad_norm": 0.011125192977488041,
      "learning_rate": 1.016e-06,
      "loss": 0.0003,
      "step": 2750
    },
    {
      "epoch": 28.001,
      "grad_norm": 0.0103826392441988,
      "learning_rate": 9.160000000000001e-07,
      "loss": 0.0003,
      "step": 2775
    },
    {
      "epoch": 28.009333333333334,
      "grad_norm": 0.010299935936927795,
      "learning_rate": 8.160000000000001e-07,
      "loss": 0.0002,
      "step": 2800
    },
    {
      "epoch": 28.017666666666667,
      "grad_norm": 0.009835168719291687,
      "learning_rate": 7.16e-07,
      "loss": 0.0002,
      "step": 2825
    },
    {
      "epoch": 28.026,
      "grad_norm": 0.009076902642846107,
      "learning_rate": 6.160000000000001e-07,
      "loss": 0.0003,
      "step": 2850
    },
    {
      "epoch": 29.001333333333335,
      "grad_norm": 0.009419168345630169,
      "learning_rate": 5.16e-07,
      "loss": 0.0003,
      "step": 2875
    },
    {
      "epoch": 29.009666666666668,
      "grad_norm": 0.010406610555946827,
      "learning_rate": 4.16e-07,
      "loss": 0.0002,
      "step": 2900
    },
    {
      "epoch": 29.018,
      "grad_norm": 0.009127378463745117,
      "learning_rate": 3.160000000000001e-07,
      "loss": 0.0002,
      "step": 2925
    },
    {
      "epoch": 29.026333333333334,
      "grad_norm": 0.007270839996635914,
      "learning_rate": 2.1600000000000003e-07,
      "loss": 0.0002,
      "step": 2950
    },
    {
      "epoch": 30.001666666666665,
      "grad_norm": 0.009513996541500092,
      "learning_rate": 1.16e-07,
      "loss": 0.0003,
      "step": 2975
    },
    {
      "epoch": 30.01,
      "grad_norm": 0.008983739651739597,
      "learning_rate": 1.6e-08,
      "loss": 0.0002,
      "step": 3000
    },
    {
      "epoch": 30.01,
      "eval_loss": 1.3570799827575684,
      "eval_runtime": 1096.6232,
      "eval_samples_per_second": 2.963,
      "eval_steps_per_second": 0.371,
      "eval_wer": 32.550952630658976,
      "step": 3000
    },
    {
      "epoch": 30.01,
      "step": 3000,
      "total_flos": 2.76089652486144e+19,
      "train_loss": 0.07142243345221505,
      "train_runtime": 18533.1767,
      "train_samples_per_second": 5.18,
      "train_steps_per_second": 0.162
    }
  ],
  "logging_steps": 25,
  "max_steps": 3000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.76089652486144e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}