{
  "best_metric": 19.407016702629264,
  "best_model_checkpoint": "./checkpoint-2000",
  "epoch": 12.044,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.6000000000000004e-07,
      "loss": 1.8434,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.600000000000001e-07,
      "loss": 1.2165,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.46e-06,
      "loss": 0.7158,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.4012,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.46e-06,
      "loss": 0.312,
      "step": 125
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.96e-06,
      "loss": 0.2985,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.46e-06,
      "loss": 0.1981,
      "step": 175
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.96e-06,
      "loss": 0.1542,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.158,
      "step": 225
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.1559,
      "step": 250
    },
    {
      "epoch": 1.01,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.1534,
      "step": 275
    },
    {
      "epoch": 1.02,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.1715,
      "step": 300
    },
    {
      "epoch": 1.03,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.1562,
      "step": 325
    },
    {
      "epoch": 1.04,
      "learning_rate": 6.96e-06,
      "loss": 0.1489,
      "step": 350
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.1353,
      "step": 375
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.1028,
      "step": 400
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.46e-06,
      "loss": 0.0675,
      "step": 425
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.96e-06,
      "loss": 0.0597,
      "step": 450
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.0586,
      "step": 475
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.067,
      "step": 500
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.908e-06,
      "loss": 0.0656,
      "step": 525
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.808000000000002e-06,
      "loss": 0.0716,
      "step": 550
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.708000000000001e-06,
      "loss": 0.0632,
      "step": 575
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.608e-06,
      "loss": 0.0583,
      "step": 600
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.508000000000001e-06,
      "loss": 0.0499,
      "step": 625
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.408e-06,
      "loss": 0.0289,
      "step": 650
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.308e-06,
      "loss": 0.0236,
      "step": 675
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.208e-06,
      "loss": 0.0253,
      "step": 700
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.108000000000002e-06,
      "loss": 0.0224,
      "step": 725
    },
    {
      "epoch": 3.01,
      "learning_rate": 9.008e-06,
      "loss": 0.023,
      "step": 750
    },
    {
      "epoch": 3.02,
      "learning_rate": 8.908000000000002e-06,
      "loss": 0.0289,
      "step": 775
    },
    {
      "epoch": 3.03,
      "learning_rate": 8.808000000000001e-06,
      "loss": 0.0276,
      "step": 800
    },
    {
      "epoch": 3.04,
      "learning_rate": 8.708e-06,
      "loss": 0.0253,
      "step": 825
    },
    {
      "epoch": 3.04,
      "learning_rate": 8.608000000000001e-06,
      "loss": 0.0206,
      "step": 850
    },
    {
      "epoch": 3.05,
      "learning_rate": 8.508e-06,
      "loss": 0.0174,
      "step": 875
    },
    {
      "epoch": 3.06,
      "learning_rate": 8.408e-06,
      "loss": 0.0114,
      "step": 900
    },
    {
      "epoch": 3.07,
      "learning_rate": 8.308e-06,
      "loss": 0.0116,
      "step": 925
    },
    {
      "epoch": 3.08,
      "learning_rate": 8.208e-06,
      "loss": 0.0113,
      "step": 950
    },
    {
      "epoch": 4.01,
      "learning_rate": 8.108000000000001e-06,
      "loss": 0.0117,
      "step": 975
    },
    {
      "epoch": 4.01,
      "learning_rate": 8.008e-06,
      "loss": 0.011,
      "step": 1000
    },
    {
      "epoch": 4.01,
      "eval_loss": 0.32340896129608154,
      "eval_runtime": 8572.581,
      "eval_samples_per_second": 1.08,
      "eval_steps_per_second": 0.034,
      "eval_wer": 20.597778310557022,
      "step": 1000
    },
    {
      "epoch": 4.02,
      "learning_rate": 7.908e-06,
      "loss": 0.0134,
      "step": 1025
    },
    {
      "epoch": 4.03,
      "learning_rate": 7.808e-06,
      "loss": 0.0106,
      "step": 1050
    },
    {
      "epoch": 4.04,
      "learning_rate": 7.708000000000001e-06,
      "loss": 0.0104,
      "step": 1075
    },
    {
      "epoch": 4.05,
      "learning_rate": 7.608000000000001e-06,
      "loss": 0.0105,
      "step": 1100
    },
    {
      "epoch": 4.06,
      "learning_rate": 7.508000000000001e-06,
      "loss": 0.007,
      "step": 1125
    },
    {
      "epoch": 4.06,
      "learning_rate": 7.408000000000001e-06,
      "loss": 0.0057,
      "step": 1150
    },
    {
      "epoch": 4.07,
      "learning_rate": 7.308000000000001e-06,
      "loss": 0.0053,
      "step": 1175
    },
    {
      "epoch": 5.0,
      "learning_rate": 7.208e-06,
      "loss": 0.0062,
      "step": 1200
    },
    {
      "epoch": 5.01,
      "learning_rate": 7.108e-06,
      "loss": 0.0055,
      "step": 1225
    },
    {
      "epoch": 5.02,
      "learning_rate": 7.0080000000000005e-06,
      "loss": 0.0064,
      "step": 1250
    },
    {
      "epoch": 5.03,
      "learning_rate": 6.908000000000001e-06,
      "loss": 0.0068,
      "step": 1275
    },
    {
      "epoch": 5.04,
      "learning_rate": 6.808e-06,
      "loss": 0.0053,
      "step": 1300
    },
    {
      "epoch": 5.04,
      "learning_rate": 6.708e-06,
      "loss": 0.005,
      "step": 1325
    },
    {
      "epoch": 5.05,
      "learning_rate": 6.608000000000001e-06,
      "loss": 0.0044,
      "step": 1350
    },
    {
      "epoch": 5.06,
      "learning_rate": 6.508000000000001e-06,
      "loss": 0.0033,
      "step": 1375
    },
    {
      "epoch": 5.07,
      "learning_rate": 6.408000000000001e-06,
      "loss": 0.0034,
      "step": 1400
    },
    {
      "epoch": 5.08,
      "learning_rate": 6.308e-06,
      "loss": 0.004,
      "step": 1425
    },
    {
      "epoch": 6.01,
      "learning_rate": 6.2080000000000005e-06,
      "loss": 0.0034,
      "step": 1450
    },
    {
      "epoch": 6.01,
      "learning_rate": 6.108000000000001e-06,
      "loss": 0.0027,
      "step": 1475
    },
    {
      "epoch": 6.02,
      "learning_rate": 6.008000000000001e-06,
      "loss": 0.0039,
      "step": 1500
    },
    {
      "epoch": 6.03,
      "learning_rate": 5.908e-06,
      "loss": 0.0029,
      "step": 1525
    },
    {
      "epoch": 6.04,
      "learning_rate": 5.808e-06,
      "loss": 0.0031,
      "step": 1550
    },
    {
      "epoch": 6.05,
      "learning_rate": 5.708e-06,
      "loss": 0.0027,
      "step": 1575
    },
    {
      "epoch": 6.06,
      "learning_rate": 5.608e-06,
      "loss": 0.0021,
      "step": 1600
    },
    {
      "epoch": 6.06,
      "learning_rate": 5.508e-06,
      "loss": 0.0023,
      "step": 1625
    },
    {
      "epoch": 6.07,
      "learning_rate": 5.408e-06,
      "loss": 0.0026,
      "step": 1650
    },
    {
      "epoch": 7.0,
      "learning_rate": 5.308000000000001e-06,
      "loss": 0.0024,
      "step": 1675
    },
    {
      "epoch": 7.01,
      "learning_rate": 5.208000000000001e-06,
      "loss": 0.0019,
      "step": 1700
    },
    {
      "epoch": 7.02,
      "learning_rate": 5.108000000000001e-06,
      "loss": 0.002,
      "step": 1725
    },
    {
      "epoch": 7.03,
      "learning_rate": 5.008000000000001e-06,
      "loss": 0.0028,
      "step": 1750
    },
    {
      "epoch": 7.03,
      "learning_rate": 4.908e-06,
      "loss": 0.0018,
      "step": 1775
    },
    {
      "epoch": 7.04,
      "learning_rate": 4.808e-06,
      "loss": 0.0017,
      "step": 1800
    },
    {
      "epoch": 7.05,
      "learning_rate": 4.7080000000000005e-06,
      "loss": 0.0015,
      "step": 1825
    },
    {
      "epoch": 7.06,
      "learning_rate": 4.608000000000001e-06,
      "loss": 0.0011,
      "step": 1850
    },
    {
      "epoch": 7.07,
      "learning_rate": 4.508e-06,
      "loss": 0.001,
      "step": 1875
    },
    {
      "epoch": 7.08,
      "learning_rate": 4.408000000000001e-06,
      "loss": 0.001,
      "step": 1900
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.308000000000001e-06,
      "loss": 0.0013,
      "step": 1925
    },
    {
      "epoch": 8.01,
      "learning_rate": 4.208e-06,
      "loss": 0.001,
      "step": 1950
    },
    {
      "epoch": 8.02,
      "learning_rate": 4.108e-06,
      "loss": 0.0012,
      "step": 1975
    },
    {
      "epoch": 8.03,
      "learning_rate": 4.008e-06,
      "loss": 0.0011,
      "step": 2000
    },
    {
      "epoch": 8.03,
      "eval_loss": 0.3650129437446594,
      "eval_runtime": 8583.7923,
      "eval_samples_per_second": 1.079,
      "eval_steps_per_second": 0.034,
      "eval_wer": 19.407016702629264,
      "step": 2000
    },
    {
      "epoch": 8.04,
      "learning_rate": 3.9080000000000005e-06,
      "loss": 0.001,
      "step": 2025
    },
    {
      "epoch": 8.05,
      "learning_rate": 3.8080000000000006e-06,
      "loss": 0.001,
      "step": 2050
    },
    {
      "epoch": 8.05,
      "learning_rate": 3.7080000000000003e-06,
      "loss": 0.0008,
      "step": 2075
    },
    {
      "epoch": 8.06,
      "learning_rate": 3.6080000000000004e-06,
      "loss": 0.0007,
      "step": 2100
    },
    {
      "epoch": 8.07,
      "learning_rate": 3.508e-06,
      "loss": 0.0007,
      "step": 2125
    },
    {
      "epoch": 8.08,
      "learning_rate": 3.4080000000000002e-06,
      "loss": 0.0007,
      "step": 2150
    },
    {
      "epoch": 9.01,
      "learning_rate": 3.308e-06,
      "loss": 0.0008,
      "step": 2175
    },
    {
      "epoch": 9.02,
      "learning_rate": 3.208e-06,
      "loss": 0.0008,
      "step": 2200
    },
    {
      "epoch": 9.02,
      "learning_rate": 3.1080000000000006e-06,
      "loss": 0.0008,
      "step": 2225
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.0080000000000003e-06,
      "loss": 0.0008,
      "step": 2250
    },
    {
      "epoch": 9.04,
      "learning_rate": 2.9080000000000004e-06,
      "loss": 0.0008,
      "step": 2275
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.808e-06,
      "loss": 0.0008,
      "step": 2300
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.708e-06,
      "loss": 0.0006,
      "step": 2325
    },
    {
      "epoch": 9.07,
      "learning_rate": 2.608e-06,
      "loss": 0.0006,
      "step": 2350
    },
    {
      "epoch": 9.07,
      "learning_rate": 2.5080000000000004e-06,
      "loss": 0.0006,
      "step": 2375
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.408e-06,
      "loss": 0.0006,
      "step": 2400
    },
    {
      "epoch": 10.01,
      "learning_rate": 2.308e-06,
      "loss": 0.0007,
      "step": 2425
    },
    {
      "epoch": 10.02,
      "learning_rate": 2.2080000000000003e-06,
      "loss": 0.0007,
      "step": 2450
    },
    {
      "epoch": 10.03,
      "learning_rate": 2.108e-06,
      "loss": 0.0007,
      "step": 2475
    },
    {
      "epoch": 10.04,
      "learning_rate": 2.008e-06,
      "loss": 0.0007,
      "step": 2500
    },
    {
      "epoch": 10.04,
      "learning_rate": 1.908e-06,
      "loss": 0.0007,
      "step": 2525
    },
    {
      "epoch": 10.05,
      "learning_rate": 1.808e-06,
      "loss": 0.0006,
      "step": 2550
    },
    {
      "epoch": 10.06,
      "learning_rate": 1.7080000000000002e-06,
      "loss": 0.0006,
      "step": 2575
    },
    {
      "epoch": 10.07,
      "learning_rate": 1.608e-06,
      "loss": 0.0006,
      "step": 2600
    },
    {
      "epoch": 10.08,
      "learning_rate": 1.508e-06,
      "loss": 0.0006,
      "step": 2625
    },
    {
      "epoch": 11.01,
      "learning_rate": 1.4080000000000001e-06,
      "loss": 0.0006,
      "step": 2650
    },
    {
      "epoch": 11.02,
      "learning_rate": 1.308e-06,
      "loss": 0.0006,
      "step": 2675
    },
    {
      "epoch": 11.02,
      "learning_rate": 1.2080000000000001e-06,
      "loss": 0.0006,
      "step": 2700
    },
    {
      "epoch": 11.03,
      "learning_rate": 1.108e-06,
      "loss": 0.0006,
      "step": 2725
    },
    {
      "epoch": 11.04,
      "learning_rate": 1.0080000000000001e-06,
      "loss": 0.0007,
      "step": 2750
    },
    {
      "epoch": 11.05,
      "learning_rate": 9.080000000000001e-07,
      "loss": 0.0007,
      "step": 2775
    },
    {
      "epoch": 11.06,
      "learning_rate": 8.08e-07,
      "loss": 0.0005,
      "step": 2800
    },
    {
      "epoch": 11.07,
      "learning_rate": 7.08e-07,
      "loss": 0.0005,
      "step": 2825
    },
    {
      "epoch": 11.07,
      "learning_rate": 6.08e-07,
      "loss": 0.0005,
      "step": 2850
    },
    {
      "epoch": 12.0,
      "learning_rate": 5.08e-07,
      "loss": 0.0005,
      "step": 2875
    },
    {
      "epoch": 12.01,
      "learning_rate": 4.0800000000000005e-07,
      "loss": 0.0006,
      "step": 2900
    },
    {
      "epoch": 12.02,
      "learning_rate": 3.0800000000000006e-07,
      "loss": 0.0006,
      "step": 2925
    },
    {
      "epoch": 12.03,
      "learning_rate": 2.08e-07,
      "loss": 0.0006,
      "step": 2950
    },
    {
      "epoch": 12.04,
      "learning_rate": 1.0800000000000001e-07,
      "loss": 0.0006,
      "step": 2975
    },
    {
      "epoch": 12.04,
      "learning_rate": 8e-09,
      "loss": 0.0006,
      "step": 3000
    },
    {
      "epoch": 12.04,
      "eval_loss": 0.3830030858516693,
      "eval_runtime": 8302.5919,
      "eval_samples_per_second": 1.115,
      "eval_steps_per_second": 0.035,
      "eval_wer": 19.51730200591385,
      "step": 3000
    },
    {
      "epoch": 12.04,
      "step": 3000,
      "total_flos": 1.9557647721824256e+20,
      "train_loss": 0.061419941114572184,
      "train_runtime": 79200.2446,
      "train_samples_per_second": 2.424,
      "train_steps_per_second": 0.038
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.9557647721824256e+20,
  "trial_name": null,
  "trial_params": null
}