{
  "best_metric": 10.033263447357072,
  "best_model_checkpoint": "./checkpoint-3000",
  "epoch": 3.187,
  "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": 0.9908,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.8221,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.46e-06,
      "loss": 0.4645,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.3184,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.46e-06,
      "loss": 0.2405,
      "step": 125
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.96e-06,
      "loss": 0.2118,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.46e-06,
      "loss": 0.1716,
      "step": 175
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.96e-06,
      "loss": 0.1416,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.1445,
      "step": 225
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.1076,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.0923,
      "step": 275
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.0711,
      "step": 300
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.077,
      "step": 325
    },
    {
      "epoch": 0.12,
      "learning_rate": 6.96e-06,
      "loss": 0.0675,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.0601,
      "step": 375
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.0573,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.46e-06,
      "loss": 0.0411,
      "step": 425
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.96e-06,
      "loss": 0.0359,
      "step": 450
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.0322,
      "step": 475
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.0391,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.908e-06,
      "loss": 0.0445,
      "step": 525
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.808000000000002e-06,
      "loss": 0.0427,
      "step": 550
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.708000000000001e-06,
      "loss": 0.0385,
      "step": 575
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.608e-06,
      "loss": 0.0449,
      "step": 600
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.508000000000001e-06,
      "loss": 0.0432,
      "step": 625
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.408e-06,
      "loss": 0.0412,
      "step": 650
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.308e-06,
      "loss": 0.0389,
      "step": 675
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.208e-06,
      "loss": 0.0339,
      "step": 700
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.108000000000002e-06,
      "loss": 0.0319,
      "step": 725
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.008e-06,
      "loss": 0.0274,
      "step": 750
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.908000000000002e-06,
      "loss": 0.034,
      "step": 775
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.808000000000001e-06,
      "loss": 0.0378,
      "step": 800
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.708e-06,
      "loss": 0.0401,
      "step": 825
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.608000000000001e-06,
      "loss": 0.0465,
      "step": 850
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.508e-06,
      "loss": 0.0424,
      "step": 875
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.408e-06,
      "loss": 0.0301,
      "step": 900
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.308e-06,
      "loss": 0.0299,
      "step": 925
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.208e-06,
      "loss": 0.0319,
      "step": 950
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.108000000000001e-06,
      "loss": 0.0257,
      "step": 975
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.008e-06,
      "loss": 0.0241,
      "step": 1000
    },
    {
      "epoch": 1.06,
      "eval_loss": 0.19960933923721313,
      "eval_runtime": 30342.5121,
      "eval_samples_per_second": 0.669,
      "eval_steps_per_second": 0.021,
      "eval_wer": 10.454287312492793,
      "step": 1000
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.908e-06,
      "loss": 0.0293,
      "step": 1025
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.808e-06,
      "loss": 0.0239,
      "step": 1050
    },
    {
      "epoch": 1.09,
      "learning_rate": 7.708000000000001e-06,
      "loss": 0.0258,
      "step": 1075
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.608000000000001e-06,
      "loss": 0.0197,
      "step": 1100
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.508000000000001e-06,
      "loss": 0.0197,
      "step": 1125
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.408000000000001e-06,
      "loss": 0.0242,
      "step": 1150
    },
    {
      "epoch": 1.12,
      "learning_rate": 7.308000000000001e-06,
      "loss": 0.0197,
      "step": 1175
    },
    {
      "epoch": 1.13,
      "learning_rate": 7.208e-06,
      "loss": 0.019,
      "step": 1200
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.108e-06,
      "loss": 0.0149,
      "step": 1225
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.0080000000000005e-06,
      "loss": 0.013,
      "step": 1250
    },
    {
      "epoch": 1.15,
      "learning_rate": 6.908000000000001e-06,
      "loss": 0.0106,
      "step": 1275
    },
    {
      "epoch": 1.16,
      "learning_rate": 6.808e-06,
      "loss": 0.0101,
      "step": 1300
    },
    {
      "epoch": 1.17,
      "learning_rate": 6.708e-06,
      "loss": 0.0156,
      "step": 1325
    },
    {
      "epoch": 1.18,
      "learning_rate": 6.608000000000001e-06,
      "loss": 0.0151,
      "step": 1350
    },
    {
      "epoch": 1.19,
      "learning_rate": 6.508000000000001e-06,
      "loss": 0.017,
      "step": 1375
    },
    {
      "epoch": 1.2,
      "learning_rate": 6.408000000000001e-06,
      "loss": 0.015,
      "step": 1400
    },
    {
      "epoch": 1.2,
      "learning_rate": 6.308e-06,
      "loss": 0.0151,
      "step": 1425
    },
    {
      "epoch": 1.21,
      "learning_rate": 6.2080000000000005e-06,
      "loss": 0.0162,
      "step": 1450
    },
    {
      "epoch": 1.22,
      "learning_rate": 6.108000000000001e-06,
      "loss": 0.0131,
      "step": 1475
    },
    {
      "epoch": 1.23,
      "learning_rate": 6.008000000000001e-06,
      "loss": 0.0127,
      "step": 1500
    },
    {
      "epoch": 1.24,
      "learning_rate": 5.908e-06,
      "loss": 0.011,
      "step": 1525
    },
    {
      "epoch": 1.25,
      "learning_rate": 5.808e-06,
      "loss": 0.0103,
      "step": 1550
    },
    {
      "epoch": 1.25,
      "learning_rate": 5.708e-06,
      "loss": 0.0103,
      "step": 1575
    },
    {
      "epoch": 1.26,
      "learning_rate": 5.608e-06,
      "loss": 0.015,
      "step": 1600
    },
    {
      "epoch": 1.27,
      "learning_rate": 5.508e-06,
      "loss": 0.0167,
      "step": 1625
    },
    {
      "epoch": 2.01,
      "learning_rate": 5.408e-06,
      "loss": 0.015,
      "step": 1650
    },
    {
      "epoch": 2.02,
      "learning_rate": 5.308000000000001e-06,
      "loss": 0.017,
      "step": 1675
    },
    {
      "epoch": 2.02,
      "learning_rate": 5.208000000000001e-06,
      "loss": 0.0152,
      "step": 1700
    },
    {
      "epoch": 2.03,
      "learning_rate": 5.108000000000001e-06,
      "loss": 0.0144,
      "step": 1725
    },
    {
      "epoch": 2.04,
      "learning_rate": 5.008000000000001e-06,
      "loss": 0.0146,
      "step": 1750
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.908e-06,
      "loss": 0.0128,
      "step": 1775
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.808e-06,
      "loss": 0.0134,
      "step": 1800
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.7080000000000005e-06,
      "loss": 0.011,
      "step": 1825
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.608000000000001e-06,
      "loss": 0.0115,
      "step": 1850
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.508e-06,
      "loss": 0.0106,
      "step": 1875
    },
    {
      "epoch": 2.09,
      "learning_rate": 4.408000000000001e-06,
      "loss": 0.01,
      "step": 1900
    },
    {
      "epoch": 2.1,
      "learning_rate": 4.308000000000001e-06,
      "loss": 0.0085,
      "step": 1925
    },
    {
      "epoch": 2.11,
      "learning_rate": 4.208e-06,
      "loss": 0.0097,
      "step": 1950
    },
    {
      "epoch": 2.12,
      "learning_rate": 4.108e-06,
      "loss": 0.0108,
      "step": 1975
    },
    {
      "epoch": 2.12,
      "learning_rate": 4.008e-06,
      "loss": 0.009,
      "step": 2000
    },
    {
      "epoch": 2.12,
      "eval_loss": 0.2155761867761612,
      "eval_runtime": 30652.225,
      "eval_samples_per_second": 0.662,
      "eval_steps_per_second": 0.021,
      "eval_wer": 10.1152478105478,
      "step": 2000
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.9080000000000005e-06,
      "loss": 0.0075,
      "step": 2025
    },
    {
      "epoch": 2.14,
      "learning_rate": 3.8080000000000006e-06,
      "loss": 0.0062,
      "step": 2050
    },
    {
      "epoch": 2.15,
      "learning_rate": 3.7080000000000003e-06,
      "loss": 0.0056,
      "step": 2075
    },
    {
      "epoch": 2.16,
      "learning_rate": 3.6080000000000004e-06,
      "loss": 0.0047,
      "step": 2100
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.508e-06,
      "loss": 0.0066,
      "step": 2125
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.4080000000000002e-06,
      "loss": 0.0078,
      "step": 2150
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.308e-06,
      "loss": 0.0086,
      "step": 2175
    },
    {
      "epoch": 2.19,
      "learning_rate": 3.208e-06,
      "loss": 0.0085,
      "step": 2200
    },
    {
      "epoch": 2.2,
      "learning_rate": 3.1080000000000006e-06,
      "loss": 0.008,
      "step": 2225
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.0080000000000003e-06,
      "loss": 0.007,
      "step": 2250
    },
    {
      "epoch": 2.22,
      "learning_rate": 2.9080000000000004e-06,
      "loss": 0.0064,
      "step": 2275
    },
    {
      "epoch": 2.22,
      "learning_rate": 2.808e-06,
      "loss": 0.0057,
      "step": 2300
    },
    {
      "epoch": 2.23,
      "learning_rate": 2.708e-06,
      "loss": 0.0054,
      "step": 2325
    },
    {
      "epoch": 2.24,
      "learning_rate": 2.608e-06,
      "loss": 0.0054,
      "step": 2350
    },
    {
      "epoch": 2.25,
      "learning_rate": 2.5080000000000004e-06,
      "loss": 0.0044,
      "step": 2375
    },
    {
      "epoch": 2.26,
      "learning_rate": 2.408e-06,
      "loss": 0.006,
      "step": 2400
    },
    {
      "epoch": 2.27,
      "learning_rate": 2.308e-06,
      "loss": 0.0078,
      "step": 2425
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.2080000000000003e-06,
      "loss": 0.0076,
      "step": 2450
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.108e-06,
      "loss": 0.0075,
      "step": 2475
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.008e-06,
      "loss": 0.0094,
      "step": 2500
    },
    {
      "epoch": 3.03,
      "learning_rate": 1.908e-06,
      "loss": 0.0079,
      "step": 2525
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.808e-06,
      "loss": 0.0075,
      "step": 2550
    },
    {
      "epoch": 3.05,
      "learning_rate": 1.7080000000000002e-06,
      "loss": 0.0074,
      "step": 2575
    },
    {
      "epoch": 3.05,
      "learning_rate": 1.608e-06,
      "loss": 0.0079,
      "step": 2600
    },
    {
      "epoch": 3.06,
      "learning_rate": 1.508e-06,
      "loss": 0.007,
      "step": 2625
    },
    {
      "epoch": 3.07,
      "learning_rate": 1.4080000000000001e-06,
      "loss": 0.0047,
      "step": 2650
    },
    {
      "epoch": 3.08,
      "learning_rate": 1.308e-06,
      "loss": 0.0062,
      "step": 2675
    },
    {
      "epoch": 3.09,
      "learning_rate": 1.2080000000000001e-06,
      "loss": 0.0042,
      "step": 2700
    },
    {
      "epoch": 3.1,
      "learning_rate": 1.108e-06,
      "loss": 0.005,
      "step": 2725
    },
    {
      "epoch": 3.1,
      "learning_rate": 1.0080000000000001e-06,
      "loss": 0.0042,
      "step": 2750
    },
    {
      "epoch": 3.11,
      "learning_rate": 9.080000000000001e-07,
      "loss": 0.0048,
      "step": 2775
    },
    {
      "epoch": 3.12,
      "learning_rate": 8.08e-07,
      "loss": 0.0041,
      "step": 2800
    },
    {
      "epoch": 3.13,
      "learning_rate": 7.08e-07,
      "loss": 0.0038,
      "step": 2825
    },
    {
      "epoch": 3.14,
      "learning_rate": 6.08e-07,
      "loss": 0.0026,
      "step": 2850
    },
    {
      "epoch": 3.15,
      "learning_rate": 5.08e-07,
      "loss": 0.0027,
      "step": 2875
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.0800000000000005e-07,
      "loss": 0.0018,
      "step": 2900
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.0800000000000006e-07,
      "loss": 0.002,
      "step": 2925
    },
    {
      "epoch": 3.17,
      "learning_rate": 2.08e-07,
      "loss": 0.0037,
      "step": 2950
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.0800000000000001e-07,
      "loss": 0.004,
      "step": 2975
    },
    {
      "epoch": 3.19,
      "learning_rate": 8e-09,
      "loss": 0.0045,
      "step": 3000
    },
    {
      "epoch": 3.19,
      "eval_loss": 0.24060867726802826,
      "eval_runtime": 28649.195,
      "eval_samples_per_second": 0.708,
      "eval_steps_per_second": 0.022,
      "eval_wer": 10.033263447357072,
      "step": 3000
    },
    {
      "epoch": 3.19,
      "step": 3000,
      "total_flos": 1.9578161880465408e+20,
      "train_loss": 0.04787538333858053,
      "train_runtime": 139856.9298,
      "train_samples_per_second": 1.373,
      "train_steps_per_second": 0.021
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.9578161880465408e+20,
  "trial_name": null,
  "trial_params": null
}