{
  "best_metric": 11.806634378141277,
  "best_model_checkpoint": "./checkpoint-2000",
  "epoch": 8.09,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 2.1889,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.000000000000001e-07,
      "loss": 1.6717,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.4000000000000001e-06,
      "loss": 1.0615,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9000000000000002e-06,
      "loss": 0.7958,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.7462,
      "step": 125
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.9e-06,
      "loss": 0.6103,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.4000000000000005e-06,
      "loss": 0.5642,
      "step": 175
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.900000000000001e-06,
      "loss": 0.5419,
      "step": 200
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.4e-06,
      "loss": 0.5566,
      "step": 225
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 0.7147,
      "step": 250
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.400000000000001e-06,
      "loss": 0.7586,
      "step": 275
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.9e-06,
      "loss": 0.4333,
      "step": 300
    },
    {
      "epoch": 0.08,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.3043,
      "step": 325
    },
    {
      "epoch": 0.09,
      "learning_rate": 6.9e-06,
      "loss": 0.2495,
      "step": 350
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.4e-06,
      "loss": 0.3004,
      "step": 375
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.9e-06,
      "loss": 0.2729,
      "step": 400
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.3112,
      "step": 425
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.900000000000001e-06,
      "loss": 0.2401,
      "step": 450
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.4e-06,
      "loss": 0.4472,
      "step": 475
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.9e-06,
      "loss": 0.3378,
      "step": 500
    },
    {
      "epoch": 1.01,
      "eval_loss": 0.3505442440509796,
      "eval_runtime": 793.6653,
      "eval_samples_per_second": 3.224,
      "eval_steps_per_second": 0.101,
      "eval_wer": 12.44761817730839,
      "step": 500
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.942857142857145e-06,
      "loss": 0.2251,
      "step": 525
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.871428571428572e-06,
      "loss": 0.1785,
      "step": 550
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.800000000000001e-06,
      "loss": 0.1642,
      "step": 575
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.72857142857143e-06,
      "loss": 0.1424,
      "step": 600
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.657142857142859e-06,
      "loss": 0.2129,
      "step": 625
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.585714285714286e-06,
      "loss": 0.2079,
      "step": 650
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.514285714285715e-06,
      "loss": 0.1761,
      "step": 675
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.442857142857144e-06,
      "loss": 0.1913,
      "step": 700
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.371428571428572e-06,
      "loss": 0.3963,
      "step": 725
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.3e-06,
      "loss": 0.3194,
      "step": 750
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.22857142857143e-06,
      "loss": 0.1634,
      "step": 775
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.157142857142857e-06,
      "loss": 0.1781,
      "step": 800
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.085714285714286e-06,
      "loss": 0.1633,
      "step": 825
    },
    {
      "epoch": 1.1,
      "learning_rate": 9.014285714285715e-06,
      "loss": 0.1977,
      "step": 850
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.942857142857142e-06,
      "loss": 0.3465,
      "step": 875
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.871428571428571e-06,
      "loss": 0.2696,
      "step": 900
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.8e-06,
      "loss": 0.152,
      "step": 925
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.72857142857143e-06,
      "loss": 0.0684,
      "step": 950
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.657142857142858e-06,
      "loss": 0.078,
      "step": 975
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.585714285714286e-06,
      "loss": 0.1075,
      "step": 1000
    },
    {
      "epoch": 2.02,
      "eval_loss": 0.3609441816806793,
      "eval_runtime": 847.5808,
      "eval_samples_per_second": 3.019,
      "eval_steps_per_second": 0.094,
      "eval_wer": 11.81446717405778,
      "step": 1000
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.514285714285715e-06,
      "loss": 0.0839,
      "step": 1025
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.442857142857144e-06,
      "loss": 0.0686,
      "step": 1050
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.371428571428573e-06,
      "loss": 0.074,
      "step": 1075
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.3e-06,
      "loss": 0.0943,
      "step": 1100
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.22857142857143e-06,
      "loss": 0.1926,
      "step": 1125
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.157142857142858e-06,
      "loss": 0.1439,
      "step": 1150
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.085714285714287e-06,
      "loss": 0.2175,
      "step": 1175
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.014285714285715e-06,
      "loss": 0.2157,
      "step": 1200
    },
    {
      "epoch": 2.08,
      "learning_rate": 7.942857142857144e-06,
      "loss": 0.1455,
      "step": 1225
    },
    {
      "epoch": 2.08,
      "learning_rate": 7.871428571428573e-06,
      "loss": 0.2559,
      "step": 1250
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.800000000000002e-06,
      "loss": 0.2255,
      "step": 1275
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.72857142857143e-06,
      "loss": 0.1284,
      "step": 1300
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.657142857142858e-06,
      "loss": 0.0993,
      "step": 1325
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.585714285714286e-06,
      "loss": 0.091,
      "step": 1350
    },
    {
      "epoch": 3.0,
      "learning_rate": 7.514285714285715e-06,
      "loss": 0.0831,
      "step": 1375
    },
    {
      "epoch": 3.01,
      "learning_rate": 7.442857142857144e-06,
      "loss": 0.1121,
      "step": 1400
    },
    {
      "epoch": 3.02,
      "learning_rate": 7.371428571428571e-06,
      "loss": 0.1114,
      "step": 1425
    },
    {
      "epoch": 3.02,
      "learning_rate": 7.3e-06,
      "loss": 0.0556,
      "step": 1450
    },
    {
      "epoch": 3.03,
      "learning_rate": 7.2285714285714294e-06,
      "loss": 0.0415,
      "step": 1475
    },
    {
      "epoch": 3.03,
      "learning_rate": 7.1571428571428584e-06,
      "loss": 0.041,
      "step": 1500
    },
    {
      "epoch": 3.03,
      "eval_loss": 0.40608036518096924,
      "eval_runtime": 766.2413,
      "eval_samples_per_second": 3.34,
      "eval_steps_per_second": 0.104,
      "eval_wer": 11.657811255727731,
      "step": 1500
    },
    {
      "epoch": 3.04,
      "learning_rate": 7.085714285714286e-06,
      "loss": 0.0547,
      "step": 1525
    },
    {
      "epoch": 3.05,
      "learning_rate": 7.014285714285715e-06,
      "loss": 0.0312,
      "step": 1550
    },
    {
      "epoch": 3.05,
      "learning_rate": 6.942857142857144e-06,
      "loss": 0.0285,
      "step": 1575
    },
    {
      "epoch": 3.06,
      "learning_rate": 6.871428571428572e-06,
      "loss": 0.0401,
      "step": 1600
    },
    {
      "epoch": 3.06,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.0512,
      "step": 1625
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.728571428571429e-06,
      "loss": 0.0398,
      "step": 1650
    },
    {
      "epoch": 3.08,
      "learning_rate": 6.657142857142857e-06,
      "loss": 0.0443,
      "step": 1675
    },
    {
      "epoch": 3.08,
      "learning_rate": 6.585714285714286e-06,
      "loss": 0.108,
      "step": 1700
    },
    {
      "epoch": 3.09,
      "learning_rate": 6.514285714285715e-06,
      "loss": 0.1373,
      "step": 1725
    },
    {
      "epoch": 3.1,
      "learning_rate": 6.442857142857143e-06,
      "loss": 0.0668,
      "step": 1750
    },
    {
      "epoch": 3.1,
      "learning_rate": 6.371428571428572e-06,
      "loss": 0.044,
      "step": 1775
    },
    {
      "epoch": 3.11,
      "learning_rate": 6.300000000000001e-06,
      "loss": 0.0875,
      "step": 1800
    },
    {
      "epoch": 4.0,
      "learning_rate": 6.22857142857143e-06,
      "loss": 0.0986,
      "step": 1825
    },
    {
      "epoch": 4.01,
      "learning_rate": 6.157142857142858e-06,
      "loss": 0.0126,
      "step": 1850
    },
    {
      "epoch": 4.01,
      "learning_rate": 6.085714285714286e-06,
      "loss": 0.0102,
      "step": 1875
    },
    {
      "epoch": 4.02,
      "learning_rate": 6.014285714285715e-06,
      "loss": 0.0163,
      "step": 1900
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.942857142857143e-06,
      "loss": 0.0161,
      "step": 1925
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.871428571428572e-06,
      "loss": 0.0398,
      "step": 1950
    },
    {
      "epoch": 4.04,
      "learning_rate": 5.8e-06,
      "loss": 0.0491,
      "step": 1975
    },
    {
      "epoch": 4.04,
      "learning_rate": 5.7285714285714285e-06,
      "loss": 0.0256,
      "step": 2000
    },
    {
      "epoch": 4.04,
      "eval_loss": 0.4597214162349701,
      "eval_runtime": 665.6344,
      "eval_samples_per_second": 3.844,
      "eval_steps_per_second": 0.12,
      "eval_wer": 11.806634378141277,
      "step": 2000
    },
    {
      "epoch": 4.05,
      "learning_rate": 5.6571428571428576e-06,
      "loss": 0.0561,
      "step": 2025
    },
    {
      "epoch": 4.06,
      "learning_rate": 5.5857142857142866e-06,
      "loss": 0.0666,
      "step": 2050
    },
    {
      "epoch": 4.06,
      "learning_rate": 5.514285714285714e-06,
      "loss": 0.0325,
      "step": 2075
    },
    {
      "epoch": 4.07,
      "learning_rate": 5.442857142857143e-06,
      "loss": 0.0516,
      "step": 2100
    },
    {
      "epoch": 4.08,
      "learning_rate": 5.371428571428572e-06,
      "loss": 0.0315,
      "step": 2125
    },
    {
      "epoch": 4.08,
      "learning_rate": 5.300000000000001e-06,
      "loss": 0.02,
      "step": 2150
    },
    {
      "epoch": 4.09,
      "learning_rate": 5.22857142857143e-06,
      "loss": 0.0175,
      "step": 2175
    },
    {
      "epoch": 4.09,
      "learning_rate": 5.157142857142857e-06,
      "loss": 0.0227,
      "step": 2200
    },
    {
      "epoch": 4.1,
      "learning_rate": 5.085714285714286e-06,
      "loss": 0.0203,
      "step": 2225
    },
    {
      "epoch": 4.11,
      "learning_rate": 5.014285714285715e-06,
      "loss": 0.019,
      "step": 2250
    },
    {
      "epoch": 4.11,
      "learning_rate": 4.9428571428571435e-06,
      "loss": 0.021,
      "step": 2275
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.871428571428572e-06,
      "loss": 0.006,
      "step": 2300
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.0077,
      "step": 2325
    },
    {
      "epoch": 5.02,
      "learning_rate": 4.728571428571429e-06,
      "loss": 0.0094,
      "step": 2350
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.657142857142857e-06,
      "loss": 0.0173,
      "step": 2375
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.585714285714286e-06,
      "loss": 0.0169,
      "step": 2400
    },
    {
      "epoch": 5.04,
      "learning_rate": 4.514285714285714e-06,
      "loss": 0.0069,
      "step": 2425
    },
    {
      "epoch": 5.04,
      "learning_rate": 4.442857142857143e-06,
      "loss": 0.0202,
      "step": 2450
    },
    {
      "epoch": 5.05,
      "learning_rate": 4.371428571428572e-06,
      "loss": 0.0224,
      "step": 2475
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.3e-06,
      "loss": 0.0099,
      "step": 2500
    },
    {
      "epoch": 5.06,
      "eval_loss": 0.5019838213920593,
      "eval_runtime": 701.182,
      "eval_samples_per_second": 3.65,
      "eval_steps_per_second": 0.114,
      "eval_wer": 12.096447827051866,
      "step": 2500
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.228571428571429e-06,
      "loss": 0.0196,
      "step": 2525
    },
    {
      "epoch": 5.07,
      "learning_rate": 4.1571428571428575e-06,
      "loss": 0.0145,
      "step": 2550
    },
    {
      "epoch": 5.08,
      "learning_rate": 4.0857142857142865e-06,
      "loss": 0.0081,
      "step": 2575
    },
    {
      "epoch": 5.08,
      "learning_rate": 4.014285714285715e-06,
      "loss": 0.0052,
      "step": 2600
    },
    {
      "epoch": 5.09,
      "learning_rate": 3.942857142857143e-06,
      "loss": 0.0103,
      "step": 2625
    },
    {
      "epoch": 5.09,
      "learning_rate": 3.871428571428572e-06,
      "loss": 0.0056,
      "step": 2650
    },
    {
      "epoch": 5.1,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.0091,
      "step": 2675
    },
    {
      "epoch": 5.11,
      "learning_rate": 3.7285714285714286e-06,
      "loss": 0.0052,
      "step": 2700
    },
    {
      "epoch": 5.11,
      "learning_rate": 3.6571428571428576e-06,
      "loss": 0.007,
      "step": 2725
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.5857142857142862e-06,
      "loss": 0.0099,
      "step": 2750
    },
    {
      "epoch": 6.01,
      "learning_rate": 3.5142857142857144e-06,
      "loss": 0.004,
      "step": 2775
    },
    {
      "epoch": 6.02,
      "learning_rate": 3.4428571428571434e-06,
      "loss": 0.0019,
      "step": 2800
    },
    {
      "epoch": 6.02,
      "learning_rate": 3.3714285714285716e-06,
      "loss": 0.0031,
      "step": 2825
    },
    {
      "epoch": 6.03,
      "learning_rate": 3.3000000000000006e-06,
      "loss": 0.0029,
      "step": 2850
    },
    {
      "epoch": 6.04,
      "learning_rate": 3.2285714285714288e-06,
      "loss": 0.0039,
      "step": 2875
    },
    {
      "epoch": 6.04,
      "learning_rate": 3.1571428571428573e-06,
      "loss": 0.0026,
      "step": 2900
    },
    {
      "epoch": 6.05,
      "learning_rate": 3.085714285714286e-06,
      "loss": 0.0022,
      "step": 2925
    },
    {
      "epoch": 6.05,
      "learning_rate": 3.0142857142857145e-06,
      "loss": 0.0032,
      "step": 2950
    },
    {
      "epoch": 6.06,
      "learning_rate": 2.9428571428571427e-06,
      "loss": 0.0018,
      "step": 2975
    },
    {
      "epoch": 6.07,
      "learning_rate": 2.8714285714285717e-06,
      "loss": 0.0028,
      "step": 3000
    },
    {
      "epoch": 6.07,
      "eval_loss": 0.5476312637329102,
      "eval_runtime": 703.0169,
      "eval_samples_per_second": 3.64,
      "eval_steps_per_second": 0.114,
      "eval_wer": 11.981566820276496,
      "step": 3000
    },
    {
      "epoch": 6.07,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.0052,
      "step": 3025
    },
    {
      "epoch": 6.08,
      "learning_rate": 2.728571428571429e-06,
      "loss": 0.0093,
      "step": 3050
    },
    {
      "epoch": 6.09,
      "learning_rate": 2.6571428571428575e-06,
      "loss": 0.0086,
      "step": 3075
    },
    {
      "epoch": 6.09,
      "learning_rate": 2.5857142857142856e-06,
      "loss": 0.0064,
      "step": 3100
    },
    {
      "epoch": 6.1,
      "learning_rate": 2.5142857142857147e-06,
      "loss": 0.0029,
      "step": 3125
    },
    {
      "epoch": 6.11,
      "learning_rate": 2.442857142857143e-06,
      "loss": 0.0025,
      "step": 3150
    },
    {
      "epoch": 6.11,
      "learning_rate": 2.371428571428572e-06,
      "loss": 0.0025,
      "step": 3175
    },
    {
      "epoch": 7.0,
      "learning_rate": 2.3000000000000004e-06,
      "loss": 0.0021,
      "step": 3200
    },
    {
      "epoch": 7.01,
      "learning_rate": 2.228571428571429e-06,
      "loss": 0.0009,
      "step": 3225
    },
    {
      "epoch": 7.02,
      "learning_rate": 2.157142857142857e-06,
      "loss": 0.0023,
      "step": 3250
    },
    {
      "epoch": 7.02,
      "learning_rate": 2.0857142857142858e-06,
      "loss": 0.0035,
      "step": 3275
    },
    {
      "epoch": 7.03,
      "learning_rate": 2.0142857142857144e-06,
      "loss": 0.0034,
      "step": 3300
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.942857142857143e-06,
      "loss": 0.0031,
      "step": 3325
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.8714285714285715e-06,
      "loss": 0.0016,
      "step": 3350
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.0015,
      "step": 3375
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.7285714285714287e-06,
      "loss": 0.0041,
      "step": 3400
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.657142857142857e-06,
      "loss": 0.0034,
      "step": 3425
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.5857142857142857e-06,
      "loss": 0.0018,
      "step": 3450
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.5142857142857145e-06,
      "loss": 0.0022,
      "step": 3475
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.442857142857143e-06,
      "loss": 0.0015,
      "step": 3500
    },
    {
      "epoch": 7.08,
      "eval_loss": 0.5869272947311401,
      "eval_runtime": 690.7949,
      "eval_samples_per_second": 3.704,
      "eval_steps_per_second": 0.116,
      "eval_wer": 11.71002989517108,
      "step": 3500
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.3714285714285717e-06,
      "loss": 0.0012,
      "step": 3525
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.3e-06,
      "loss": 0.0014,
      "step": 3550
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.2285714285714286e-06,
      "loss": 0.0011,
      "step": 3575
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.1571428571428572e-06,
      "loss": 0.0014,
      "step": 3600
    },
    {
      "epoch": 7.11,
      "learning_rate": 1.0857142857142858e-06,
      "loss": 0.0019,
      "step": 3625
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.0142857142857144e-06,
      "loss": 0.001,
      "step": 3650
    },
    {
      "epoch": 8.01,
      "learning_rate": 9.42857142857143e-07,
      "loss": 0.0016,
      "step": 3675
    },
    {
      "epoch": 8.02,
      "learning_rate": 8.714285714285716e-07,
      "loss": 0.0033,
      "step": 3700
    },
    {
      "epoch": 8.02,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.0011,
      "step": 3725
    },
    {
      "epoch": 8.03,
      "learning_rate": 7.285714285714287e-07,
      "loss": 0.001,
      "step": 3750
    },
    {
      "epoch": 8.03,
      "learning_rate": 6.571428571428571e-07,
      "loss": 0.0011,
      "step": 3775
    },
    {
      "epoch": 8.04,
      "learning_rate": 5.857142857142857e-07,
      "loss": 0.001,
      "step": 3800
    },
    {
      "epoch": 8.05,
      "learning_rate": 5.142857142857143e-07,
      "loss": 0.001,
      "step": 3825
    },
    {
      "epoch": 8.05,
      "learning_rate": 4.4285714285714286e-07,
      "loss": 0.003,
      "step": 3850
    },
    {
      "epoch": 8.06,
      "learning_rate": 3.7142857142857145e-07,
      "loss": 0.0022,
      "step": 3875
    },
    {
      "epoch": 8.06,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.0013,
      "step": 3900
    },
    {
      "epoch": 8.07,
      "learning_rate": 2.285714285714286e-07,
      "loss": 0.0013,
      "step": 3925
    },
    {
      "epoch": 8.08,
      "learning_rate": 1.5714285714285717e-07,
      "loss": 0.0011,
      "step": 3950
    },
    {
      "epoch": 8.08,
      "learning_rate": 8.571428571428573e-08,
      "loss": 0.0019,
      "step": 3975
    },
    {
      "epoch": 8.09,
      "learning_rate": 1.4285714285714288e-08,
      "loss": 0.0018,
      "step": 4000
    },
    {
      "epoch": 8.09,
      "eval_loss": 0.5812159180641174,
      "eval_runtime": 687.9522,
      "eval_samples_per_second": 3.72,
      "eval_steps_per_second": 0.116,
      "eval_wer": 11.942402840693985,
      "step": 4000
    },
    {
      "epoch": 8.09,
      "step": 4000,
      "total_flos": 1.3062109848403968e+20,
      "train_loss": 0.13265852789208293,
      "train_runtime": 45123.3628,
      "train_samples_per_second": 2.837,
      "train_steps_per_second": 0.089
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.3062109848403968e+20,
  "trial_name": null,
  "trial_params": null
}