{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 11.389521640091116,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 467.0219,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 446.5955,
      "step": 200
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 393.423,
      "step": 300
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 360.4873,
      "step": 400
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 369.4461,
      "step": 500
    },
    {
      "epoch": 0.68,
      "learning_rate": 5e-05,
      "loss": 364.8889,
      "step": 600
    },
    {
      "epoch": 0.8,
      "learning_rate": 5e-05,
      "loss": 362.8324,
      "step": 700
    },
    {
      "epoch": 0.91,
      "learning_rate": 5e-05,
      "loss": 345.4677,
      "step": 800
    },
    {
      "epoch": 1.03,
      "learning_rate": 5e-05,
      "loss": 377.1422,
      "step": 900
    },
    {
      "epoch": 1.14,
      "learning_rate": 5e-05,
      "loss": 319.0323,
      "step": 1000
    },
    {
      "epoch": 1.25,
      "learning_rate": 5e-05,
      "loss": 298.1518,
      "step": 1100
    },
    {
      "epoch": 1.37,
      "learning_rate": 5e-05,
      "loss": 335.3308,
      "step": 1200
    },
    {
      "epoch": 1.48,
      "learning_rate": 5e-05,
      "loss": 290.9471,
      "step": 1300
    },
    {
      "epoch": 1.59,
      "learning_rate": 5e-05,
      "loss": 310.6187,
      "step": 1400
    },
    {
      "epoch": 1.71,
      "learning_rate": 5e-05,
      "loss": 333.6563,
      "step": 1500
    },
    {
      "epoch": 1.82,
      "learning_rate": 5e-05,
      "loss": 287.3142,
      "step": 1600
    },
    {
      "epoch": 1.94,
      "learning_rate": 5e-05,
      "loss": 304.1946,
      "step": 1700
    },
    {
      "epoch": 2.05,
      "learning_rate": 5e-05,
      "loss": 295.4456,
      "step": 1800
    },
    {
      "epoch": 2.16,
      "learning_rate": 5e-05,
      "loss": 288.6632,
      "step": 1900
    },
    {
      "epoch": 2.28,
      "learning_rate": 5e-05,
      "loss": 268.8452,
      "step": 2000
    },
    {
      "epoch": 2.28,
      "eval_cer": 0.1040043410507569,
      "eval_loss": 262.4808654785156,
      "eval_runtime": 146.2892,
      "eval_samples_per_second": 12.783,
      "eval_steps_per_second": 1.6,
      "eval_wer": 0.22893966022261278,
      "step": 2000
    },
    {
      "epoch": 2.39,
      "learning_rate": 5e-05,
      "loss": 279.8674,
      "step": 2100
    },
    {
      "epoch": 2.51,
      "learning_rate": 5e-05,
      "loss": 288.3673,
      "step": 2200
    },
    {
      "epoch": 2.62,
      "learning_rate": 5e-05,
      "loss": 266.2094,
      "step": 2300
    },
    {
      "epoch": 2.73,
      "learning_rate": 5e-05,
      "loss": 266.3993,
      "step": 2400
    },
    {
      "epoch": 2.85,
      "learning_rate": 5e-05,
      "loss": 265.0983,
      "step": 2500
    },
    {
      "epoch": 2.96,
      "learning_rate": 5e-05,
      "loss": 280.5026,
      "step": 2600
    },
    {
      "epoch": 3.08,
      "learning_rate": 5e-05,
      "loss": 253.5466,
      "step": 2700
    },
    {
      "epoch": 3.19,
      "learning_rate": 5e-05,
      "loss": 259.7672,
      "step": 2800
    },
    {
      "epoch": 3.3,
      "learning_rate": 5e-05,
      "loss": 241.1983,
      "step": 2900
    },
    {
      "epoch": 3.42,
      "learning_rate": 5e-05,
      "loss": 234.3184,
      "step": 3000
    },
    {
      "epoch": 3.53,
      "learning_rate": 5e-05,
      "loss": 241.2133,
      "step": 3100
    },
    {
      "epoch": 3.64,
      "learning_rate": 5e-05,
      "loss": 264.1351,
      "step": 3200
    },
    {
      "epoch": 3.76,
      "learning_rate": 5e-05,
      "loss": 247.2135,
      "step": 3300
    },
    {
      "epoch": 3.87,
      "learning_rate": 5e-05,
      "loss": 227.8616,
      "step": 3400
    },
    {
      "epoch": 3.99,
      "learning_rate": 5e-05,
      "loss": 236.7091,
      "step": 3500
    },
    {
      "epoch": 4.1,
      "learning_rate": 5e-05,
      "loss": 234.9673,
      "step": 3600
    },
    {
      "epoch": 4.21,
      "learning_rate": 5e-05,
      "loss": 211.579,
      "step": 3700
    },
    {
      "epoch": 4.33,
      "learning_rate": 5e-05,
      "loss": 226.7139,
      "step": 3800
    },
    {
      "epoch": 4.44,
      "learning_rate": 5e-05,
      "loss": 239.6923,
      "step": 3900
    },
    {
      "epoch": 4.56,
      "learning_rate": 5e-05,
      "loss": 212.4134,
      "step": 4000
    },
    {
      "epoch": 4.56,
      "eval_cer": 0.09499526936776491,
      "eval_loss": 273.2842712402344,
      "eval_runtime": 145.12,
      "eval_samples_per_second": 12.886,
      "eval_steps_per_second": 1.612,
      "eval_wer": 0.20917789494239405,
      "step": 4000
    },
    {
      "epoch": 4.67,
      "learning_rate": 5e-05,
      "loss": 222.3978,
      "step": 4100
    },
    {
      "epoch": 4.78,
      "learning_rate": 5e-05,
      "loss": 220.115,
      "step": 4200
    },
    {
      "epoch": 4.9,
      "learning_rate": 5e-05,
      "loss": 214.0393,
      "step": 4300
    },
    {
      "epoch": 5.01,
      "learning_rate": 5e-05,
      "loss": 217.6124,
      "step": 4400
    },
    {
      "epoch": 5.13,
      "learning_rate": 5e-05,
      "loss": 207.3773,
      "step": 4500
    },
    {
      "epoch": 5.24,
      "learning_rate": 5e-05,
      "loss": 210.2402,
      "step": 4600
    },
    {
      "epoch": 5.35,
      "learning_rate": 5e-05,
      "loss": 183.2794,
      "step": 4700
    },
    {
      "epoch": 5.47,
      "learning_rate": 5e-05,
      "loss": 212.013,
      "step": 4800
    },
    {
      "epoch": 5.58,
      "learning_rate": 5e-05,
      "loss": 201.9256,
      "step": 4900
    },
    {
      "epoch": 5.69,
      "learning_rate": 5e-05,
      "loss": 204.919,
      "step": 5000
    },
    {
      "epoch": 5.81,
      "learning_rate": 5e-05,
      "loss": 200.8947,
      "step": 5100
    },
    {
      "epoch": 5.92,
      "learning_rate": 5e-05,
      "loss": 205.1183,
      "step": 5200
    },
    {
      "epoch": 6.04,
      "learning_rate": 5e-05,
      "loss": 183.53,
      "step": 5300
    },
    {
      "epoch": 6.15,
      "learning_rate": 5e-05,
      "loss": 176.2721,
      "step": 5400
    },
    {
      "epoch": 6.26,
      "learning_rate": 5e-05,
      "loss": 182.444,
      "step": 5500
    },
    {
      "epoch": 6.38,
      "learning_rate": 5e-05,
      "loss": 179.9177,
      "step": 5600
    },
    {
      "epoch": 6.49,
      "learning_rate": 5e-05,
      "loss": 184.7364,
      "step": 5700
    },
    {
      "epoch": 6.61,
      "learning_rate": 5e-05,
      "loss": 170.2941,
      "step": 5800
    },
    {
      "epoch": 6.72,
      "learning_rate": 5e-05,
      "loss": 184.6765,
      "step": 5900
    },
    {
      "epoch": 6.83,
      "learning_rate": 5e-05,
      "loss": 200.3965,
      "step": 6000
    },
    {
      "epoch": 6.83,
      "eval_cer": 0.09548224621549421,
      "eval_loss": 279.6430358886719,
      "eval_runtime": 147.4008,
      "eval_samples_per_second": 12.687,
      "eval_steps_per_second": 1.588,
      "eval_wer": 0.2134348760007811,
      "step": 6000
    },
    {
      "epoch": 6.95,
      "learning_rate": 5e-05,
      "loss": 203.8473,
      "step": 6100
    },
    {
      "epoch": 7.06,
      "learning_rate": 5e-05,
      "loss": 180.4022,
      "step": 6200
    },
    {
      "epoch": 7.18,
      "learning_rate": 5e-05,
      "loss": 167.2582,
      "step": 6300
    },
    {
      "epoch": 7.29,
      "learning_rate": 5e-05,
      "loss": 171.7429,
      "step": 6400
    },
    {
      "epoch": 7.4,
      "learning_rate": 5e-05,
      "loss": 161.9806,
      "step": 6500
    },
    {
      "epoch": 7.52,
      "learning_rate": 5e-05,
      "loss": 181.7546,
      "step": 6600
    },
    {
      "epoch": 7.63,
      "learning_rate": 5e-05,
      "loss": 166.6461,
      "step": 6700
    },
    {
      "epoch": 7.74,
      "learning_rate": 5e-05,
      "loss": 156.8281,
      "step": 6800
    },
    {
      "epoch": 7.86,
      "learning_rate": 5e-05,
      "loss": 162.8639,
      "step": 6900
    },
    {
      "epoch": 7.97,
      "learning_rate": 5e-05,
      "loss": 177.6029,
      "step": 7000
    },
    {
      "epoch": 8.09,
      "learning_rate": 5e-05,
      "loss": 162.6876,
      "step": 7100
    },
    {
      "epoch": 8.2,
      "learning_rate": 5e-05,
      "loss": 154.6953,
      "step": 7200
    },
    {
      "epoch": 8.31,
      "learning_rate": 5e-05,
      "loss": 157.7532,
      "step": 7300
    },
    {
      "epoch": 8.43,
      "learning_rate": 5e-05,
      "loss": 155.4144,
      "step": 7400
    },
    {
      "epoch": 8.54,
      "learning_rate": 5e-05,
      "loss": 160.9859,
      "step": 7500
    },
    {
      "epoch": 8.66,
      "learning_rate": 5e-05,
      "loss": 157.5087,
      "step": 7600
    },
    {
      "epoch": 8.77,
      "learning_rate": 5e-05,
      "loss": 158.3729,
      "step": 7700
    },
    {
      "epoch": 8.88,
      "learning_rate": 5e-05,
      "loss": 161.7107,
      "step": 7800
    },
    {
      "epoch": 9.0,
      "learning_rate": 5e-05,
      "loss": 158.3218,
      "step": 7900
    },
    {
      "epoch": 9.11,
      "learning_rate": 5e-05,
      "loss": 130.7524,
      "step": 8000
    },
    {
      "epoch": 9.11,
      "eval_cer": 0.09586487088156723,
      "eval_loss": 320.47552490234375,
      "eval_runtime": 144.7814,
      "eval_samples_per_second": 12.916,
      "eval_steps_per_second": 1.616,
      "eval_wer": 0.20667838312829526,
      "step": 8000
    },
    {
      "epoch": 9.23,
      "learning_rate": 5e-05,
      "loss": 142.4579,
      "step": 8100
    },
    {
      "epoch": 9.34,
      "learning_rate": 5e-05,
      "loss": 147.8841,
      "step": 8200
    },
    {
      "epoch": 9.45,
      "learning_rate": 5e-05,
      "loss": 158.0805,
      "step": 8300
    },
    {
      "epoch": 9.57,
      "learning_rate": 5e-05,
      "loss": 145.2056,
      "step": 8400
    },
    {
      "epoch": 9.68,
      "learning_rate": 5e-05,
      "loss": 144.9573,
      "step": 8500
    },
    {
      "epoch": 9.79,
      "learning_rate": 5e-05,
      "loss": 148.1379,
      "step": 8600
    },
    {
      "epoch": 9.91,
      "learning_rate": 5e-05,
      "loss": 148.8498,
      "step": 8700
    },
    {
      "epoch": 10.02,
      "learning_rate": 5e-05,
      "loss": 160.1041,
      "step": 8800
    },
    {
      "epoch": 10.14,
      "learning_rate": 5e-05,
      "loss": 141.2527,
      "step": 8900
    },
    {
      "epoch": 10.25,
      "learning_rate": 5e-05,
      "loss": 127.7468,
      "step": 9000
    },
    {
      "epoch": 10.36,
      "learning_rate": 5e-05,
      "loss": 136.7523,
      "step": 9100
    },
    {
      "epoch": 10.48,
      "learning_rate": 5e-05,
      "loss": 134.6625,
      "step": 9200
    },
    {
      "epoch": 10.59,
      "learning_rate": 5e-05,
      "loss": 138.4917,
      "step": 9300
    },
    {
      "epoch": 10.71,
      "learning_rate": 5e-05,
      "loss": 135.3308,
      "step": 9400
    },
    {
      "epoch": 10.82,
      "learning_rate": 5e-05,
      "loss": 143.3156,
      "step": 9500
    },
    {
      "epoch": 10.93,
      "learning_rate": 5e-05,
      "loss": 130.6054,
      "step": 9600
    },
    {
      "epoch": 11.05,
      "learning_rate": 5e-05,
      "loss": 138.9151,
      "step": 9700
    },
    {
      "epoch": 11.16,
      "learning_rate": 5e-05,
      "loss": 125.203,
      "step": 9800
    },
    {
      "epoch": 11.28,
      "learning_rate": 5e-05,
      "loss": 131.7745,
      "step": 9900
    },
    {
      "epoch": 11.39,
      "learning_rate": 5e-05,
      "loss": 119.4062,
      "step": 10000
    },
    {
      "epoch": 11.39,
      "eval_cer": 0.09683186776491541,
      "eval_loss": 327.7448425292969,
      "eval_runtime": 145.6399,
      "eval_samples_per_second": 12.84,
      "eval_steps_per_second": 1.607,
      "eval_wer": 0.20937316930287053,
      "step": 10000
    }
  ],
  "max_steps": 10000,
  "num_train_epochs": 12,
  "total_flos": 1.2933525176475093e+19,
  "trial_name": null,
  "trial_params": null
}