{
  "best_metric": 0.753694919914759,
  "best_model_checkpoint": "./whisper-small-dialect_iraqi/checkpoint-3000",
  "epoch": 3.977724741447892,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 90.6798324584961,
      "learning_rate": 5.000000000000001e-07,
      "loss": 4.6998,
      "step": 25
    },
    {
      "epoch": 0.04,
      "grad_norm": 40.06618118286133,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 4.1109,
      "step": 50
    },
    {
      "epoch": 0.06,
      "grad_norm": 38.31050109863281,
      "learning_rate": 1.5e-06,
      "loss": 3.2388,
      "step": 75
    },
    {
      "epoch": 0.08,
      "grad_norm": 27.650033950805664,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.7981,
      "step": 100
    },
    {
      "epoch": 0.1,
      "grad_norm": 34.07307052612305,
      "learning_rate": 2.5e-06,
      "loss": 2.3226,
      "step": 125
    },
    {
      "epoch": 0.12,
      "grad_norm": 30.355411529541016,
      "learning_rate": 3e-06,
      "loss": 2.3817,
      "step": 150
    },
    {
      "epoch": 0.14,
      "grad_norm": 22.727617263793945,
      "learning_rate": 3.5e-06,
      "loss": 2.1619,
      "step": 175
    },
    {
      "epoch": 0.16,
      "grad_norm": 24.718725204467773,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.0719,
      "step": 200
    },
    {
      "epoch": 0.18,
      "grad_norm": 28.388458251953125,
      "learning_rate": 4.5e-06,
      "loss": 2.0441,
      "step": 225
    },
    {
      "epoch": 0.2,
      "grad_norm": 21.303863525390625,
      "learning_rate": 5e-06,
      "loss": 1.941,
      "step": 250
    },
    {
      "epoch": 0.22,
      "grad_norm": 27.430503845214844,
      "learning_rate": 5.500000000000001e-06,
      "loss": 1.9,
      "step": 275
    },
    {
      "epoch": 0.24,
      "grad_norm": 27.235267639160156,
      "learning_rate": 6e-06,
      "loss": 1.7483,
      "step": 300
    },
    {
      "epoch": 0.26,
      "grad_norm": 29.63104820251465,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 1.6439,
      "step": 325
    },
    {
      "epoch": 0.28,
      "grad_norm": 21.515504837036133,
      "learning_rate": 7e-06,
      "loss": 1.5067,
      "step": 350
    },
    {
      "epoch": 0.3,
      "grad_norm": 26.370962142944336,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.3685,
      "step": 375
    },
    {
      "epoch": 0.32,
      "grad_norm": 25.11650276184082,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4172,
      "step": 400
    },
    {
      "epoch": 0.34,
      "grad_norm": 25.06173324584961,
      "learning_rate": 8.5e-06,
      "loss": 1.5026,
      "step": 425
    },
    {
      "epoch": 0.36,
      "grad_norm": 24.87285614013672,
      "learning_rate": 9e-06,
      "loss": 1.449,
      "step": 450
    },
    {
      "epoch": 0.38,
      "grad_norm": 23.05367660522461,
      "learning_rate": 9.5e-06,
      "loss": 1.3377,
      "step": 475
    },
    {
      "epoch": 0.4,
      "grad_norm": 22.767385482788086,
      "learning_rate": 1e-05,
      "loss": 1.3833,
      "step": 500
    },
    {
      "epoch": 0.42,
      "grad_norm": 25.26249122619629,
      "learning_rate": 9.944444444444445e-06,
      "loss": 1.3704,
      "step": 525
    },
    {
      "epoch": 0.44,
      "grad_norm": 16.503665924072266,
      "learning_rate": 9.88888888888889e-06,
      "loss": 1.3313,
      "step": 550
    },
    {
      "epoch": 0.46,
      "grad_norm": 25.474660873413086,
      "learning_rate": 9.833333333333333e-06,
      "loss": 1.3546,
      "step": 575
    },
    {
      "epoch": 0.48,
      "grad_norm": 23.469463348388672,
      "learning_rate": 9.777777777777779e-06,
      "loss": 1.3364,
      "step": 600
    },
    {
      "epoch": 0.5,
      "grad_norm": 27.196256637573242,
      "learning_rate": 9.722222222222223e-06,
      "loss": 1.3881,
      "step": 625
    },
    {
      "epoch": 0.52,
      "grad_norm": 28.02334213256836,
      "learning_rate": 9.666666666666667e-06,
      "loss": 1.476,
      "step": 650
    },
    {
      "epoch": 0.54,
      "grad_norm": 23.868146896362305,
      "learning_rate": 9.611111111111112e-06,
      "loss": 1.3418,
      "step": 675
    },
    {
      "epoch": 0.56,
      "grad_norm": 25.75493049621582,
      "learning_rate": 9.555555555555556e-06,
      "loss": 1.3658,
      "step": 700
    },
    {
      "epoch": 0.58,
      "grad_norm": 20.75641632080078,
      "learning_rate": 9.5e-06,
      "loss": 1.3521,
      "step": 725
    },
    {
      "epoch": 0.6,
      "grad_norm": 19.32164192199707,
      "learning_rate": 9.444444444444445e-06,
      "loss": 1.1973,
      "step": 750
    },
    {
      "epoch": 0.62,
      "grad_norm": 22.786579132080078,
      "learning_rate": 9.38888888888889e-06,
      "loss": 1.2699,
      "step": 775
    },
    {
      "epoch": 0.64,
      "grad_norm": 28.15203094482422,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.2875,
      "step": 800
    },
    {
      "epoch": 0.66,
      "grad_norm": 23.417707443237305,
      "learning_rate": 9.277777777777778e-06,
      "loss": 1.2642,
      "step": 825
    },
    {
      "epoch": 0.68,
      "grad_norm": 26.45885467529297,
      "learning_rate": 9.222222222222224e-06,
      "loss": 1.2766,
      "step": 850
    },
    {
      "epoch": 0.7,
      "grad_norm": 21.266761779785156,
      "learning_rate": 9.166666666666666e-06,
      "loss": 1.2071,
      "step": 875
    },
    {
      "epoch": 0.72,
      "grad_norm": 29.940473556518555,
      "learning_rate": 9.111111111111112e-06,
      "loss": 1.3464,
      "step": 900
    },
    {
      "epoch": 0.74,
      "grad_norm": 23.371288299560547,
      "learning_rate": 9.055555555555556e-06,
      "loss": 1.2047,
      "step": 925
    },
    {
      "epoch": 0.76,
      "grad_norm": 29.867013931274414,
      "learning_rate": 9e-06,
      "loss": 1.2475,
      "step": 950
    },
    {
      "epoch": 0.78,
      "grad_norm": 25.816246032714844,
      "learning_rate": 8.944444444444446e-06,
      "loss": 1.3114,
      "step": 975
    },
    {
      "epoch": 0.8,
      "grad_norm": 25.020904541015625,
      "learning_rate": 8.888888888888888e-06,
      "loss": 1.2314,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.2611169815063477,
      "eval_runtime": 759.5723,
      "eval_samples_per_second": 3.308,
      "eval_steps_per_second": 0.415,
      "eval_wer": 0.8315116518869871,
      "step": 1000
    },
    {
      "epoch": 0.82,
      "grad_norm": 26.749116897583008,
      "learning_rate": 8.833333333333334e-06,
      "loss": 1.3321,
      "step": 1025
    },
    {
      "epoch": 0.84,
      "grad_norm": 25.487159729003906,
      "learning_rate": 8.777777777777778e-06,
      "loss": 1.2774,
      "step": 1050
    },
    {
      "epoch": 0.86,
      "grad_norm": 24.627052307128906,
      "learning_rate": 8.722222222222224e-06,
      "loss": 1.2631,
      "step": 1075
    },
    {
      "epoch": 0.88,
      "grad_norm": 30.22520637512207,
      "learning_rate": 8.666666666666668e-06,
      "loss": 1.3112,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "grad_norm": 24.753944396972656,
      "learning_rate": 8.611111111111112e-06,
      "loss": 1.2371,
      "step": 1125
    },
    {
      "epoch": 0.91,
      "grad_norm": 28.411720275878906,
      "learning_rate": 8.555555555555556e-06,
      "loss": 1.2851,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "grad_norm": 15.954439163208008,
      "learning_rate": 8.5e-06,
      "loss": 1.1501,
      "step": 1175
    },
    {
      "epoch": 0.95,
      "grad_norm": 20.5186710357666,
      "learning_rate": 8.444444444444446e-06,
      "loss": 1.3305,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "grad_norm": 21.135791778564453,
      "learning_rate": 8.38888888888889e-06,
      "loss": 1.2152,
      "step": 1225
    },
    {
      "epoch": 0.99,
      "grad_norm": 23.05451202392578,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.2342,
      "step": 1250
    },
    {
      "epoch": 1.01,
      "grad_norm": 14.508014678955078,
      "learning_rate": 8.277777777777778e-06,
      "loss": 1.088,
      "step": 1275
    },
    {
      "epoch": 1.03,
      "grad_norm": 22.216018676757812,
      "learning_rate": 8.222222222222222e-06,
      "loss": 0.9067,
      "step": 1300
    },
    {
      "epoch": 1.05,
      "grad_norm": 18.821016311645508,
      "learning_rate": 8.166666666666668e-06,
      "loss": 0.9273,
      "step": 1325
    },
    {
      "epoch": 1.07,
      "grad_norm": 14.82918930053711,
      "learning_rate": 8.111111111111112e-06,
      "loss": 0.8638,
      "step": 1350
    },
    {
      "epoch": 1.09,
      "grad_norm": 14.661232948303223,
      "learning_rate": 8.055555555555557e-06,
      "loss": 0.9233,
      "step": 1375
    },
    {
      "epoch": 1.11,
      "grad_norm": 12.39109992980957,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.9203,
      "step": 1400
    },
    {
      "epoch": 1.13,
      "grad_norm": 16.905513763427734,
      "learning_rate": 7.944444444444445e-06,
      "loss": 0.9121,
      "step": 1425
    },
    {
      "epoch": 1.15,
      "grad_norm": 19.73763656616211,
      "learning_rate": 7.88888888888889e-06,
      "loss": 0.9762,
      "step": 1450
    },
    {
      "epoch": 1.17,
      "grad_norm": 20.832565307617188,
      "learning_rate": 7.833333333333333e-06,
      "loss": 1.0234,
      "step": 1475
    },
    {
      "epoch": 1.19,
      "grad_norm": 20.59532356262207,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.9885,
      "step": 1500
    },
    {
      "epoch": 1.21,
      "grad_norm": 19.251754760742188,
      "learning_rate": 7.722222222222223e-06,
      "loss": 0.8777,
      "step": 1525
    },
    {
      "epoch": 1.23,
      "grad_norm": 22.859533309936523,
      "learning_rate": 7.666666666666667e-06,
      "loss": 1.0234,
      "step": 1550
    },
    {
      "epoch": 1.25,
      "grad_norm": 18.167465209960938,
      "learning_rate": 7.611111111111111e-06,
      "loss": 0.9787,
      "step": 1575
    },
    {
      "epoch": 1.27,
      "grad_norm": 19.155288696289062,
      "learning_rate": 7.555555555555556e-06,
      "loss": 0.9028,
      "step": 1600
    },
    {
      "epoch": 1.29,
      "grad_norm": 14.16364574432373,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.8999,
      "step": 1625
    },
    {
      "epoch": 1.31,
      "grad_norm": 19.69369125366211,
      "learning_rate": 7.444444444444445e-06,
      "loss": 0.9565,
      "step": 1650
    },
    {
      "epoch": 1.33,
      "grad_norm": 18.87957763671875,
      "learning_rate": 7.38888888888889e-06,
      "loss": 0.929,
      "step": 1675
    },
    {
      "epoch": 1.35,
      "grad_norm": 13.816971778869629,
      "learning_rate": 7.333333333333333e-06,
      "loss": 0.9804,
      "step": 1700
    },
    {
      "epoch": 1.37,
      "grad_norm": 21.371904373168945,
      "learning_rate": 7.277777777777778e-06,
      "loss": 0.9721,
      "step": 1725
    },
    {
      "epoch": 1.39,
      "grad_norm": 15.800612449645996,
      "learning_rate": 7.222222222222223e-06,
      "loss": 0.9186,
      "step": 1750
    },
    {
      "epoch": 1.41,
      "grad_norm": 24.303115844726562,
      "learning_rate": 7.166666666666667e-06,
      "loss": 0.9046,
      "step": 1775
    },
    {
      "epoch": 1.43,
      "grad_norm": 23.116640090942383,
      "learning_rate": 7.111111111111112e-06,
      "loss": 0.8983,
      "step": 1800
    },
    {
      "epoch": 1.45,
      "grad_norm": 20.9332332611084,
      "learning_rate": 7.055555555555557e-06,
      "loss": 0.9215,
      "step": 1825
    },
    {
      "epoch": 1.47,
      "grad_norm": 16.32611846923828,
      "learning_rate": 7e-06,
      "loss": 0.9387,
      "step": 1850
    },
    {
      "epoch": 1.49,
      "grad_norm": 12.740330696105957,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.8465,
      "step": 1875
    },
    {
      "epoch": 1.51,
      "grad_norm": 22.3834228515625,
      "learning_rate": 6.88888888888889e-06,
      "loss": 1.0026,
      "step": 1900
    },
    {
      "epoch": 1.53,
      "grad_norm": 15.032549858093262,
      "learning_rate": 6.833333333333334e-06,
      "loss": 0.9839,
      "step": 1925
    },
    {
      "epoch": 1.55,
      "grad_norm": 16.8016357421875,
      "learning_rate": 6.777777777777779e-06,
      "loss": 0.9477,
      "step": 1950
    },
    {
      "epoch": 1.57,
      "grad_norm": 17.123817443847656,
      "learning_rate": 6.7222222222222235e-06,
      "loss": 0.9238,
      "step": 1975
    },
    {
      "epoch": 1.59,
      "grad_norm": 23.146198272705078,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.8997,
      "step": 2000
    },
    {
      "epoch": 1.59,
      "eval_loss": 1.1737741231918335,
      "eval_runtime": 720.7256,
      "eval_samples_per_second": 3.487,
      "eval_steps_per_second": 0.437,
      "eval_wer": 0.8439540798790128,
      "step": 2000
    },
    {
      "epoch": 1.61,
      "grad_norm": 14.144861221313477,
      "learning_rate": 6.6111111111111115e-06,
      "loss": 0.9259,
      "step": 2025
    },
    {
      "epoch": 1.63,
      "grad_norm": 23.315628051757812,
      "learning_rate": 6.555555555555556e-06,
      "loss": 0.918,
      "step": 2050
    },
    {
      "epoch": 1.65,
      "grad_norm": 19.603803634643555,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.8248,
      "step": 2075
    },
    {
      "epoch": 1.67,
      "grad_norm": 23.32997703552246,
      "learning_rate": 6.444444444444445e-06,
      "loss": 0.994,
      "step": 2100
    },
    {
      "epoch": 1.69,
      "grad_norm": 14.22852897644043,
      "learning_rate": 6.3888888888888885e-06,
      "loss": 0.8911,
      "step": 2125
    },
    {
      "epoch": 1.71,
      "grad_norm": 25.88995933532715,
      "learning_rate": 6.333333333333333e-06,
      "loss": 0.9571,
      "step": 2150
    },
    {
      "epoch": 1.73,
      "grad_norm": 14.300514221191406,
      "learning_rate": 6.277777777777778e-06,
      "loss": 0.9271,
      "step": 2175
    },
    {
      "epoch": 1.75,
      "grad_norm": 20.620363235473633,
      "learning_rate": 6.222222222222223e-06,
      "loss": 0.9448,
      "step": 2200
    },
    {
      "epoch": 1.77,
      "grad_norm": 22.74941062927246,
      "learning_rate": 6.166666666666667e-06,
      "loss": 0.8855,
      "step": 2225
    },
    {
      "epoch": 1.79,
      "grad_norm": 23.452943801879883,
      "learning_rate": 6.111111111111112e-06,
      "loss": 0.9769,
      "step": 2250
    },
    {
      "epoch": 1.81,
      "grad_norm": 18.53643035888672,
      "learning_rate": 6.055555555555555e-06,
      "loss": 0.9351,
      "step": 2275
    },
    {
      "epoch": 1.83,
      "grad_norm": 12.938403129577637,
      "learning_rate": 6e-06,
      "loss": 0.886,
      "step": 2300
    },
    {
      "epoch": 1.85,
      "grad_norm": 21.200199127197266,
      "learning_rate": 5.944444444444445e-06,
      "loss": 0.893,
      "step": 2325
    },
    {
      "epoch": 1.87,
      "grad_norm": 21.017866134643555,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.9721,
      "step": 2350
    },
    {
      "epoch": 1.89,
      "grad_norm": 14.922087669372559,
      "learning_rate": 5.833333333333334e-06,
      "loss": 0.9119,
      "step": 2375
    },
    {
      "epoch": 1.91,
      "grad_norm": 18.216527938842773,
      "learning_rate": 5.777777777777778e-06,
      "loss": 0.8708,
      "step": 2400
    },
    {
      "epoch": 1.93,
      "grad_norm": 19.921646118164062,
      "learning_rate": 5.722222222222222e-06,
      "loss": 0.913,
      "step": 2425
    },
    {
      "epoch": 1.95,
      "grad_norm": 19.92595863342285,
      "learning_rate": 5.666666666666667e-06,
      "loss": 0.9116,
      "step": 2450
    },
    {
      "epoch": 1.97,
      "grad_norm": 21.927392959594727,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.9042,
      "step": 2475
    },
    {
      "epoch": 1.99,
      "grad_norm": 16.251340866088867,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.8799,
      "step": 2500
    },
    {
      "epoch": 2.01,
      "grad_norm": 12.66164493560791,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.6893,
      "step": 2525
    },
    {
      "epoch": 2.03,
      "grad_norm": 16.812175750732422,
      "learning_rate": 5.444444444444445e-06,
      "loss": 0.5815,
      "step": 2550
    },
    {
      "epoch": 2.05,
      "grad_norm": 24.777708053588867,
      "learning_rate": 5.388888888888889e-06,
      "loss": 0.6612,
      "step": 2575
    },
    {
      "epoch": 2.07,
      "grad_norm": 18.211021423339844,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.6015,
      "step": 2600
    },
    {
      "epoch": 2.09,
      "grad_norm": 16.52762222290039,
      "learning_rate": 5.2777777777777785e-06,
      "loss": 0.6056,
      "step": 2625
    },
    {
      "epoch": 2.11,
      "grad_norm": 14.783591270446777,
      "learning_rate": 5.2222222222222226e-06,
      "loss": 0.6516,
      "step": 2650
    },
    {
      "epoch": 2.13,
      "grad_norm": 14.879830360412598,
      "learning_rate": 5.1666666666666675e-06,
      "loss": 0.6334,
      "step": 2675
    },
    {
      "epoch": 2.15,
      "grad_norm": 12.205127716064453,
      "learning_rate": 5.1111111111111115e-06,
      "loss": 0.5561,
      "step": 2700
    },
    {
      "epoch": 2.17,
      "grad_norm": 21.069787979125977,
      "learning_rate": 5.0555555555555555e-06,
      "loss": 0.6318,
      "step": 2725
    },
    {
      "epoch": 2.19,
      "grad_norm": 19.374557495117188,
      "learning_rate": 5e-06,
      "loss": 0.6581,
      "step": 2750
    },
    {
      "epoch": 2.21,
      "grad_norm": 12.187582015991211,
      "learning_rate": 4.944444444444445e-06,
      "loss": 0.5867,
      "step": 2775
    },
    {
      "epoch": 2.23,
      "grad_norm": 10.92504596710205,
      "learning_rate": 4.888888888888889e-06,
      "loss": 0.6629,
      "step": 2800
    },
    {
      "epoch": 2.25,
      "grad_norm": 16.010311126708984,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.6236,
      "step": 2825
    },
    {
      "epoch": 2.27,
      "grad_norm": 17.286380767822266,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.6444,
      "step": 2850
    },
    {
      "epoch": 2.29,
      "grad_norm": 20.500370025634766,
      "learning_rate": 4.722222222222222e-06,
      "loss": 0.6486,
      "step": 2875
    },
    {
      "epoch": 2.31,
      "grad_norm": 8.12575912475586,
      "learning_rate": 4.666666666666667e-06,
      "loss": 0.647,
      "step": 2900
    },
    {
      "epoch": 2.33,
      "grad_norm": 14.205204010009766,
      "learning_rate": 4.611111111111112e-06,
      "loss": 0.6264,
      "step": 2925
    },
    {
      "epoch": 2.35,
      "grad_norm": 19.79119300842285,
      "learning_rate": 4.555555555555556e-06,
      "loss": 0.6556,
      "step": 2950
    },
    {
      "epoch": 2.37,
      "grad_norm": 16.10354995727539,
      "learning_rate": 4.5e-06,
      "loss": 0.724,
      "step": 2975
    },
    {
      "epoch": 2.39,
      "grad_norm": 15.188028335571289,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.6654,
      "step": 3000
    },
    {
      "epoch": 2.39,
      "eval_loss": 1.176345705986023,
      "eval_runtime": 739.8672,
      "eval_samples_per_second": 3.397,
      "eval_steps_per_second": 0.426,
      "eval_wer": 0.753694919914759,
      "step": 3000
    },
    {
      "epoch": 2.41,
      "grad_norm": 12.310816764831543,
      "learning_rate": 4.388888888888889e-06,
      "loss": 0.6626,
      "step": 3025
    },
    {
      "epoch": 2.43,
      "grad_norm": 16.34236717224121,
      "learning_rate": 4.333333333333334e-06,
      "loss": 0.5884,
      "step": 3050
    },
    {
      "epoch": 2.45,
      "grad_norm": 21.42963981628418,
      "learning_rate": 4.277777777777778e-06,
      "loss": 0.6561,
      "step": 3075
    },
    {
      "epoch": 2.47,
      "grad_norm": 13.815032958984375,
      "learning_rate": 4.222222222222223e-06,
      "loss": 0.5719,
      "step": 3100
    },
    {
      "epoch": 2.49,
      "grad_norm": 17.99472999572754,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6684,
      "step": 3125
    },
    {
      "epoch": 2.51,
      "grad_norm": 16.833436965942383,
      "learning_rate": 4.111111111111111e-06,
      "loss": 0.6566,
      "step": 3150
    },
    {
      "epoch": 2.53,
      "grad_norm": 18.723281860351562,
      "learning_rate": 4.055555555555556e-06,
      "loss": 0.5795,
      "step": 3175
    },
    {
      "epoch": 2.55,
      "grad_norm": 14.792190551757812,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.6239,
      "step": 3200
    },
    {
      "epoch": 2.57,
      "grad_norm": 12.845802307128906,
      "learning_rate": 3.944444444444445e-06,
      "loss": 0.6182,
      "step": 3225
    },
    {
      "epoch": 2.59,
      "grad_norm": 15.569611549377441,
      "learning_rate": 3.88888888888889e-06,
      "loss": 0.63,
      "step": 3250
    },
    {
      "epoch": 2.61,
      "grad_norm": 11.007694244384766,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.6106,
      "step": 3275
    },
    {
      "epoch": 2.63,
      "grad_norm": 18.574846267700195,
      "learning_rate": 3.777777777777778e-06,
      "loss": 0.5999,
      "step": 3300
    },
    {
      "epoch": 2.65,
      "grad_norm": 15.045463562011719,
      "learning_rate": 3.7222222222222225e-06,
      "loss": 0.6264,
      "step": 3325
    },
    {
      "epoch": 2.67,
      "grad_norm": 14.46103572845459,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.6492,
      "step": 3350
    },
    {
      "epoch": 2.68,
      "grad_norm": 21.062015533447266,
      "learning_rate": 3.6111111111111115e-06,
      "loss": 0.6262,
      "step": 3375
    },
    {
      "epoch": 2.7,
      "grad_norm": 18.670812606811523,
      "learning_rate": 3.555555555555556e-06,
      "loss": 0.6268,
      "step": 3400
    },
    {
      "epoch": 2.72,
      "grad_norm": 15.329480171203613,
      "learning_rate": 3.5e-06,
      "loss": 0.5885,
      "step": 3425
    },
    {
      "epoch": 2.74,
      "grad_norm": 13.511859893798828,
      "learning_rate": 3.444444444444445e-06,
      "loss": 0.625,
      "step": 3450
    },
    {
      "epoch": 2.76,
      "grad_norm": 18.378982543945312,
      "learning_rate": 3.3888888888888893e-06,
      "loss": 0.588,
      "step": 3475
    },
    {
      "epoch": 2.78,
      "grad_norm": 10.714431762695312,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.5832,
      "step": 3500
    },
    {
      "epoch": 2.8,
      "grad_norm": 13.219440460205078,
      "learning_rate": 3.277777777777778e-06,
      "loss": 0.6165,
      "step": 3525
    },
    {
      "epoch": 2.82,
      "grad_norm": 20.18632698059082,
      "learning_rate": 3.2222222222222227e-06,
      "loss": 0.5918,
      "step": 3550
    },
    {
      "epoch": 2.84,
      "grad_norm": 17.847423553466797,
      "learning_rate": 3.1666666666666667e-06,
      "loss": 0.5514,
      "step": 3575
    },
    {
      "epoch": 2.86,
      "grad_norm": 10.973837852478027,
      "learning_rate": 3.1111111111111116e-06,
      "loss": 0.6074,
      "step": 3600
    },
    {
      "epoch": 2.88,
      "grad_norm": 10.063940048217773,
      "learning_rate": 3.055555555555556e-06,
      "loss": 0.5841,
      "step": 3625
    },
    {
      "epoch": 2.9,
      "grad_norm": 19.3330020904541,
      "learning_rate": 3e-06,
      "loss": 0.6031,
      "step": 3650
    },
    {
      "epoch": 2.92,
      "grad_norm": 12.685822486877441,
      "learning_rate": 2.944444444444445e-06,
      "loss": 0.5543,
      "step": 3675
    },
    {
      "epoch": 2.94,
      "grad_norm": 13.237887382507324,
      "learning_rate": 2.888888888888889e-06,
      "loss": 0.5575,
      "step": 3700
    },
    {
      "epoch": 2.96,
      "grad_norm": 14.186014175415039,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.5701,
      "step": 3725
    },
    {
      "epoch": 2.98,
      "grad_norm": 11.545550346374512,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.5855,
      "step": 3750
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.496350288391113,
      "learning_rate": 2.7222222222222224e-06,
      "loss": 0.6069,
      "step": 3775
    },
    {
      "epoch": 3.02,
      "grad_norm": 10.237105369567871,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.3926,
      "step": 3800
    },
    {
      "epoch": 3.04,
      "grad_norm": 6.904445648193359,
      "learning_rate": 2.6111111111111113e-06,
      "loss": 0.3931,
      "step": 3825
    },
    {
      "epoch": 3.06,
      "grad_norm": 9.90048599243164,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.3966,
      "step": 3850
    },
    {
      "epoch": 3.08,
      "grad_norm": 9.313446044921875,
      "learning_rate": 2.5e-06,
      "loss": 0.4119,
      "step": 3875
    },
    {
      "epoch": 3.1,
      "grad_norm": 9.53442668914795,
      "learning_rate": 2.4444444444444447e-06,
      "loss": 0.4221,
      "step": 3900
    },
    {
      "epoch": 3.12,
      "grad_norm": 15.683548927307129,
      "learning_rate": 2.388888888888889e-06,
      "loss": 0.423,
      "step": 3925
    },
    {
      "epoch": 3.14,
      "grad_norm": 13.14923095703125,
      "learning_rate": 2.3333333333333336e-06,
      "loss": 0.3881,
      "step": 3950
    },
    {
      "epoch": 3.16,
      "grad_norm": 7.727118015289307,
      "learning_rate": 2.277777777777778e-06,
      "loss": 0.3926,
      "step": 3975
    },
    {
      "epoch": 3.18,
      "grad_norm": 6.784303665161133,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.3151,
      "step": 4000
    },
    {
      "epoch": 3.18,
      "eval_loss": 1.20012366771698,
      "eval_runtime": 741.1631,
      "eval_samples_per_second": 3.391,
      "eval_steps_per_second": 0.425,
      "eval_wer": 0.7703306523681859,
      "step": 4000
    },
    {
      "epoch": 3.2,
      "grad_norm": 11.838973045349121,
      "learning_rate": 2.166666666666667e-06,
      "loss": 0.4212,
      "step": 4025
    },
    {
      "epoch": 3.22,
      "grad_norm": 9.963013648986816,
      "learning_rate": 2.1111111111111114e-06,
      "loss": 0.5183,
      "step": 4050
    },
    {
      "epoch": 3.24,
      "grad_norm": 11.655789375305176,
      "learning_rate": 2.0555555555555555e-06,
      "loss": 0.3753,
      "step": 4075
    },
    {
      "epoch": 3.26,
      "grad_norm": 12.085322380065918,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.3888,
      "step": 4100
    },
    {
      "epoch": 3.28,
      "grad_norm": 16.148874282836914,
      "learning_rate": 1.944444444444445e-06,
      "loss": 0.4098,
      "step": 4125
    },
    {
      "epoch": 3.3,
      "grad_norm": 14.727580070495605,
      "learning_rate": 1.888888888888889e-06,
      "loss": 0.4313,
      "step": 4150
    },
    {
      "epoch": 3.32,
      "grad_norm": 7.685079574584961,
      "learning_rate": 1.8333333333333333e-06,
      "loss": 0.4227,
      "step": 4175
    },
    {
      "epoch": 3.34,
      "grad_norm": 12.381622314453125,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.4043,
      "step": 4200
    },
    {
      "epoch": 3.36,
      "grad_norm": 12.150275230407715,
      "learning_rate": 1.7222222222222224e-06,
      "loss": 0.4428,
      "step": 4225
    },
    {
      "epoch": 3.38,
      "grad_norm": 14.8292875289917,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.4276,
      "step": 4250
    },
    {
      "epoch": 3.4,
      "grad_norm": 18.160255432128906,
      "learning_rate": 1.6111111111111113e-06,
      "loss": 0.4544,
      "step": 4275
    },
    {
      "epoch": 3.42,
      "grad_norm": 8.906795501708984,
      "learning_rate": 1.5555555555555558e-06,
      "loss": 0.3873,
      "step": 4300
    },
    {
      "epoch": 3.44,
      "grad_norm": 9.33066463470459,
      "learning_rate": 1.5e-06,
      "loss": 0.417,
      "step": 4325
    },
    {
      "epoch": 3.46,
      "grad_norm": 13.509571075439453,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.4384,
      "step": 4350
    },
    {
      "epoch": 3.48,
      "grad_norm": 14.600696563720703,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.471,
      "step": 4375
    },
    {
      "epoch": 3.5,
      "grad_norm": 15.88110065460205,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.3886,
      "step": 4400
    },
    {
      "epoch": 3.52,
      "grad_norm": 21.162866592407227,
      "learning_rate": 1.2777777777777779e-06,
      "loss": 0.4064,
      "step": 4425
    },
    {
      "epoch": 3.54,
      "grad_norm": 16.15275764465332,
      "learning_rate": 1.2222222222222223e-06,
      "loss": 0.4125,
      "step": 4450
    },
    {
      "epoch": 3.56,
      "grad_norm": 10.692975997924805,
      "learning_rate": 1.1666666666666668e-06,
      "loss": 0.3805,
      "step": 4475
    },
    {
      "epoch": 3.58,
      "grad_norm": 13.422977447509766,
      "learning_rate": 1.111111111111111e-06,
      "loss": 0.4478,
      "step": 4500
    },
    {
      "epoch": 3.6,
      "grad_norm": 13.359613418579102,
      "learning_rate": 1.0555555555555557e-06,
      "loss": 0.3822,
      "step": 4525
    },
    {
      "epoch": 3.62,
      "grad_norm": 15.324664115905762,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.4295,
      "step": 4550
    },
    {
      "epoch": 3.64,
      "grad_norm": 12.166446685791016,
      "learning_rate": 9.444444444444445e-07,
      "loss": 0.4018,
      "step": 4575
    },
    {
      "epoch": 3.66,
      "grad_norm": 11.992637634277344,
      "learning_rate": 8.88888888888889e-07,
      "loss": 0.3979,
      "step": 4600
    },
    {
      "epoch": 3.68,
      "grad_norm": 19.6129150390625,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.4456,
      "step": 4625
    },
    {
      "epoch": 3.7,
      "grad_norm": 9.729543685913086,
      "learning_rate": 7.777777777777779e-07,
      "loss": 0.3642,
      "step": 4650
    },
    {
      "epoch": 3.72,
      "grad_norm": 10.414878845214844,
      "learning_rate": 7.222222222222222e-07,
      "loss": 0.3569,
      "step": 4675
    },
    {
      "epoch": 3.74,
      "grad_norm": 12.819910049438477,
      "learning_rate": 6.666666666666667e-07,
      "loss": 0.4157,
      "step": 4700
    },
    {
      "epoch": 3.76,
      "grad_norm": 17.679868698120117,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.4457,
      "step": 4725
    },
    {
      "epoch": 3.78,
      "grad_norm": 14.453750610351562,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.3788,
      "step": 4750
    },
    {
      "epoch": 3.8,
      "grad_norm": 11.410884857177734,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4198,
      "step": 4775
    },
    {
      "epoch": 3.82,
      "grad_norm": 16.757707595825195,
      "learning_rate": 4.444444444444445e-07,
      "loss": 0.3561,
      "step": 4800
    },
    {
      "epoch": 3.84,
      "grad_norm": 10.525169372558594,
      "learning_rate": 3.8888888888888895e-07,
      "loss": 0.3851,
      "step": 4825
    },
    {
      "epoch": 3.86,
      "grad_norm": 16.068998336791992,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.4166,
      "step": 4850
    },
    {
      "epoch": 3.88,
      "grad_norm": 15.054728507995605,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.386,
      "step": 4875
    },
    {
      "epoch": 3.9,
      "grad_norm": 15.278282165527344,
      "learning_rate": 2.2222222222222224e-07,
      "loss": 0.408,
      "step": 4900
    },
    {
      "epoch": 3.92,
      "grad_norm": 18.32455062866211,
      "learning_rate": 1.6666666666666668e-07,
      "loss": 0.4419,
      "step": 4925
    },
    {
      "epoch": 3.94,
      "grad_norm": 13.800189971923828,
      "learning_rate": 1.1111111111111112e-07,
      "loss": 0.4077,
      "step": 4950
    },
    {
      "epoch": 3.96,
      "grad_norm": 17.295827865600586,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.3982,
      "step": 4975
    },
    {
      "epoch": 3.98,
      "grad_norm": 20.146503448486328,
      "learning_rate": 0.0,
      "loss": 0.4063,
      "step": 5000
    },
    {
      "epoch": 3.98,
      "eval_loss": 1.197590947151184,
      "eval_runtime": 665.6332,
      "eval_samples_per_second": 3.775,
      "eval_steps_per_second": 0.473,
      "eval_wer": 0.7633876400632433,
      "step": 5000
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "total_flos": 1.15390872317952e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}