{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6622340425531914,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.5e-06,
      "loss": 1.7289,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-06,
      "loss": 1.5064,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.3393,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1e-05,
      "loss": 1.2335,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 9.948979591836737e-06,
      "loss": 1.1482,
      "step": 125
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.89795918367347e-06,
      "loss": 1.1069,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.846938775510205e-06,
      "loss": 1.0776,
      "step": 175
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.795918367346939e-06,
      "loss": 1.045,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.744897959183674e-06,
      "loss": 1.011,
      "step": 225
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.693877551020408e-06,
      "loss": 1.0032,
      "step": 250
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.0074882507324219,
      "eval_runtime": 73.4748,
      "eval_samples_per_second": 13.501,
      "eval_steps_per_second": 0.109,
      "eval_wer": 0.5062510418403067,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.642857142857144e-06,
      "loss": 0.9854,
      "step": 275
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.591836734693878e-06,
      "loss": 0.9543,
      "step": 300
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.540816326530612e-06,
      "loss": 0.9635,
      "step": 325
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.489795918367348e-06,
      "loss": 0.9388,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.438775510204082e-06,
      "loss": 0.9272,
      "step": 375
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.387755102040818e-06,
      "loss": 0.9151,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.336734693877552e-06,
      "loss": 0.9148,
      "step": 425
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.285714285714288e-06,
      "loss": 0.9065,
      "step": 450
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.234693877551022e-06,
      "loss": 0.8953,
      "step": 475
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.183673469387756e-06,
      "loss": 0.8983,
      "step": 500
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.8945361971855164,
      "eval_runtime": 71.6473,
      "eval_samples_per_second": 13.846,
      "eval_steps_per_second": 0.112,
      "eval_wer": 0.4649478616806505,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.13265306122449e-06,
      "loss": 0.8662,
      "step": 525
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.081632653061225e-06,
      "loss": 0.8682,
      "step": 550
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.03061224489796e-06,
      "loss": 0.8663,
      "step": 575
    },
    {
      "epoch": 0.2,
      "learning_rate": 8.979591836734695e-06,
      "loss": 0.8552,
      "step": 600
    },
    {
      "epoch": 0.21,
      "learning_rate": 8.92857142857143e-06,
      "loss": 0.8561,
      "step": 625
    },
    {
      "epoch": 0.22,
      "learning_rate": 8.877551020408163e-06,
      "loss": 0.8401,
      "step": 650
    },
    {
      "epoch": 0.22,
      "learning_rate": 8.826530612244899e-06,
      "loss": 0.8522,
      "step": 675
    },
    {
      "epoch": 0.23,
      "learning_rate": 8.775510204081633e-06,
      "loss": 0.8413,
      "step": 700
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.724489795918369e-06,
      "loss": 0.8276,
      "step": 725
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.673469387755103e-06,
      "loss": 0.8227,
      "step": 750
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.8335936069488525,
      "eval_runtime": 72.0249,
      "eval_samples_per_second": 13.773,
      "eval_steps_per_second": 0.111,
      "eval_wer": 0.4491304106239929,
      "step": 750
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.622448979591837e-06,
      "loss": 0.8161,
      "step": 775
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.8064,
      "step": 800
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.520408163265307e-06,
      "loss": 0.7929,
      "step": 825
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.469387755102042e-06,
      "loss": 0.8026,
      "step": 850
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.418367346938776e-06,
      "loss": 0.8115,
      "step": 875
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.36734693877551e-06,
      "loss": 0.7938,
      "step": 900
    },
    {
      "epoch": 0.31,
      "learning_rate": 8.316326530612246e-06,
      "loss": 0.7955,
      "step": 925
    },
    {
      "epoch": 0.32,
      "learning_rate": 8.26530612244898e-06,
      "loss": 0.7881,
      "step": 950
    },
    {
      "epoch": 0.32,
      "learning_rate": 8.214285714285714e-06,
      "loss": 0.791,
      "step": 975
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.16326530612245e-06,
      "loss": 0.777,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.7931004762649536,
      "eval_runtime": 72.0011,
      "eval_samples_per_second": 13.778,
      "eval_steps_per_second": 0.111,
      "eval_wer": 0.4314237558111537,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 8.112244897959184e-06,
      "loss": 0.7795,
      "step": 1025
    },
    {
      "epoch": 0.35,
      "learning_rate": 8.06122448979592e-06,
      "loss": 0.781,
      "step": 1050
    },
    {
      "epoch": 0.36,
      "learning_rate": 8.010204081632654e-06,
      "loss": 0.7721,
      "step": 1075
    },
    {
      "epoch": 0.37,
      "learning_rate": 7.959183673469388e-06,
      "loss": 0.7749,
      "step": 1100
    },
    {
      "epoch": 0.37,
      "learning_rate": 7.908163265306124e-06,
      "loss": 0.7652,
      "step": 1125
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.857142857142858e-06,
      "loss": 0.768,
      "step": 1150
    },
    {
      "epoch": 0.39,
      "learning_rate": 7.806122448979593e-06,
      "loss": 0.7626,
      "step": 1175
    },
    {
      "epoch": 0.4,
      "learning_rate": 7.755102040816327e-06,
      "loss": 0.7535,
      "step": 1200
    },
    {
      "epoch": 0.41,
      "learning_rate": 7.704081632653061e-06,
      "loss": 0.7615,
      "step": 1225
    },
    {
      "epoch": 0.42,
      "learning_rate": 7.653061224489796e-06,
      "loss": 0.7728,
      "step": 1250
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.7639850378036499,
      "eval_runtime": 71.9953,
      "eval_samples_per_second": 13.779,
      "eval_steps_per_second": 0.111,
      "eval_wer": 0.4216999129484544,
      "step": 1250
    },
    {
      "epoch": 0.42,
      "learning_rate": 7.602040816326531e-06,
      "loss": 0.7454,
      "step": 1275
    },
    {
      "epoch": 0.43,
      "learning_rate": 7.551020408163265e-06,
      "loss": 0.7385,
      "step": 1300
    },
    {
      "epoch": 0.44,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.7368,
      "step": 1325
    },
    {
      "epoch": 0.45,
      "learning_rate": 7.448979591836736e-06,
      "loss": 0.7366,
      "step": 1350
    },
    {
      "epoch": 0.46,
      "learning_rate": 7.39795918367347e-06,
      "loss": 0.7426,
      "step": 1375
    },
    {
      "epoch": 0.47,
      "learning_rate": 7.346938775510205e-06,
      "loss": 0.7244,
      "step": 1400
    },
    {
      "epoch": 0.47,
      "learning_rate": 7.295918367346939e-06,
      "loss": 0.7403,
      "step": 1425
    },
    {
      "epoch": 0.48,
      "learning_rate": 7.244897959183675e-06,
      "loss": 0.7399,
      "step": 1450
    },
    {
      "epoch": 0.49,
      "learning_rate": 7.193877551020409e-06,
      "loss": 0.7253,
      "step": 1475
    },
    {
      "epoch": 0.5,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.7141,
      "step": 1500
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.7407031655311584,
      "eval_runtime": 72.8809,
      "eval_samples_per_second": 13.611,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.41338371210016484,
      "step": 1500
    },
    {
      "epoch": 0.51,
      "learning_rate": 7.091836734693878e-06,
      "loss": 0.7246,
      "step": 1525
    },
    {
      "epoch": 0.52,
      "learning_rate": 7.0408163265306125e-06,
      "loss": 0.7248,
      "step": 1550
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.989795918367348e-06,
      "loss": 0.719,
      "step": 1575
    },
    {
      "epoch": 0.53,
      "learning_rate": 6.938775510204082e-06,
      "loss": 0.7146,
      "step": 1600
    },
    {
      "epoch": 0.54,
      "learning_rate": 6.887755102040817e-06,
      "loss": 0.714,
      "step": 1625
    },
    {
      "epoch": 0.55,
      "learning_rate": 6.836734693877551e-06,
      "loss": 0.7094,
      "step": 1650
    },
    {
      "epoch": 0.56,
      "learning_rate": 6.785714285714287e-06,
      "loss": 0.7121,
      "step": 1675
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.734693877551021e-06,
      "loss": 0.709,
      "step": 1700
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.683673469387756e-06,
      "loss": 0.7112,
      "step": 1725
    },
    {
      "epoch": 0.58,
      "learning_rate": 6.63265306122449e-06,
      "loss": 0.7208,
      "step": 1750
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.7225156426429749,
      "eval_runtime": 73.303,
      "eval_samples_per_second": 13.533,
      "eval_steps_per_second": 0.109,
      "eval_wer": 0.4023448352503195,
      "step": 1750
    },
    {
      "epoch": 0.59,
      "learning_rate": 6.581632653061225e-06,
      "loss": 0.7153,
      "step": 1775
    },
    {
      "epoch": 0.6,
      "learning_rate": 6.530612244897959e-06,
      "loss": 0.7052,
      "step": 1800
    },
    {
      "epoch": 0.61,
      "learning_rate": 6.4795918367346946e-06,
      "loss": 0.7026,
      "step": 1825
    },
    {
      "epoch": 0.62,
      "learning_rate": 6.4285714285714295e-06,
      "loss": 0.7042,
      "step": 1850
    },
    {
      "epoch": 0.62,
      "learning_rate": 6.3775510204081635e-06,
      "loss": 0.7031,
      "step": 1875
    },
    {
      "epoch": 0.63,
      "learning_rate": 6.326530612244899e-06,
      "loss": 0.6939,
      "step": 1900
    },
    {
      "epoch": 0.64,
      "learning_rate": 6.275510204081633e-06,
      "loss": 0.6956,
      "step": 1925
    },
    {
      "epoch": 0.65,
      "learning_rate": 6.224489795918368e-06,
      "loss": 0.7066,
      "step": 1950
    },
    {
      "epoch": 0.66,
      "learning_rate": 6.173469387755102e-06,
      "loss": 0.7094,
      "step": 1975
    },
    {
      "epoch": 0.66,
      "learning_rate": 6.122448979591837e-06,
      "loss": 0.6911,
      "step": 2000
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.7082733511924744,
      "eval_runtime": 72.5463,
      "eval_samples_per_second": 13.674,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.3942138504565576,
      "step": 2000
    },
    {
      "epoch": 0.67,
      "learning_rate": 6.071428571428571e-06,
      "loss": 0.6901,
      "step": 2025
    },
    {
      "epoch": 0.68,
      "learning_rate": 6.020408163265307e-06,
      "loss": 0.6935,
      "step": 2050
    },
    {
      "epoch": 0.69,
      "learning_rate": 5.969387755102042e-06,
      "loss": 0.6961,
      "step": 2075
    },
    {
      "epoch": 0.7,
      "learning_rate": 5.918367346938776e-06,
      "loss": 0.6922,
      "step": 2100
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.867346938775511e-06,
      "loss": 0.6832,
      "step": 2125
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.816326530612246e-06,
      "loss": 0.6816,
      "step": 2150
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.7653061224489805e-06,
      "loss": 0.6792,
      "step": 2175
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.6771,
      "step": 2200
    },
    {
      "epoch": 0.74,
      "learning_rate": 5.663265306122449e-06,
      "loss": 0.6978,
      "step": 2225
    },
    {
      "epoch": 0.75,
      "learning_rate": 5.6122448979591834e-06,
      "loss": 0.6924,
      "step": 2250
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.6948310732841492,
      "eval_runtime": 72.8197,
      "eval_samples_per_second": 13.623,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.39108369913504104,
      "step": 2250
    },
    {
      "epoch": 0.76,
      "learning_rate": 5.561224489795919e-06,
      "loss": 0.6889,
      "step": 2275
    },
    {
      "epoch": 0.76,
      "learning_rate": 5.510204081632653e-06,
      "loss": 0.6738,
      "step": 2300
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.459183673469388e-06,
      "loss": 0.6795,
      "step": 2325
    },
    {
      "epoch": 0.78,
      "learning_rate": 5.408163265306123e-06,
      "loss": 0.6903,
      "step": 2350
    },
    {
      "epoch": 0.79,
      "learning_rate": 5.357142857142857e-06,
      "loss": 0.682,
      "step": 2375
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.306122448979593e-06,
      "loss": 0.6776,
      "step": 2400
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.255102040816327e-06,
      "loss": 0.6952,
      "step": 2425
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.204081632653062e-06,
      "loss": 0.6678,
      "step": 2450
    },
    {
      "epoch": 0.82,
      "learning_rate": 5.153061224489796e-06,
      "loss": 0.6738,
      "step": 2475
    },
    {
      "epoch": 0.83,
      "learning_rate": 5.1020408163265315e-06,
      "loss": 0.6702,
      "step": 2500
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.6848832964897156,
      "eval_runtime": 72.6772,
      "eval_samples_per_second": 13.649,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.3883610231334852,
      "step": 2500
    },
    {
      "epoch": 0.84,
      "learning_rate": 5.0510204081632655e-06,
      "loss": 0.6636,
      "step": 2525
    },
    {
      "epoch": 0.85,
      "learning_rate": 5e-06,
      "loss": 0.6719,
      "step": 2550
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.948979591836735e-06,
      "loss": 0.6636,
      "step": 2575
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.897959183673469e-06,
      "loss": 0.6656,
      "step": 2600
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.846938775510204e-06,
      "loss": 0.6732,
      "step": 2625
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.795918367346939e-06,
      "loss": 0.6658,
      "step": 2650
    },
    {
      "epoch": 0.89,
      "learning_rate": 4.744897959183674e-06,
      "loss": 0.6662,
      "step": 2675
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.693877551020409e-06,
      "loss": 0.665,
      "step": 2700
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.642857142857144e-06,
      "loss": 0.6603,
      "step": 2725
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.591836734693878e-06,
      "loss": 0.663,
      "step": 2750
    },
    {
      "epoch": 0.91,
      "eval_loss": 0.6765531897544861,
      "eval_runtime": 72.8274,
      "eval_samples_per_second": 13.621,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.3769331925691319,
      "step": 2750
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.540816326530613e-06,
      "loss": 0.68,
      "step": 2775
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.489795918367348e-06,
      "loss": 0.6704,
      "step": 2800
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.438775510204082e-06,
      "loss": 0.6731,
      "step": 2825
    },
    {
      "epoch": 0.95,
      "learning_rate": 4.3877551020408165e-06,
      "loss": 0.6576,
      "step": 2850
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.336734693877551e-06,
      "loss": 0.6503,
      "step": 2875
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.6514,
      "step": 2900
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.234693877551021e-06,
      "loss": 0.6608,
      "step": 2925
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.183673469387755e-06,
      "loss": 0.659,
      "step": 2950
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.13265306122449e-06,
      "loss": 0.6662,
      "step": 2975
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.081632653061225e-06,
      "loss": 0.6548,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.668616533279419,
      "eval_runtime": 72.9383,
      "eval_samples_per_second": 13.601,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.3758959826637773,
      "step": 3000
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.03061224489796e-06,
      "loss": 0.6534,
      "step": 3025
    },
    {
      "epoch": 1.01,
      "learning_rate": 3.979591836734694e-06,
      "loss": 0.6375,
      "step": 3050
    },
    {
      "epoch": 1.02,
      "learning_rate": 3.928571428571429e-06,
      "loss": 0.6438,
      "step": 3075
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.877551020408164e-06,
      "loss": 0.6323,
      "step": 3100
    },
    {
      "epoch": 1.04,
      "learning_rate": 3.826530612244898e-06,
      "loss": 0.644,
      "step": 3125
    },
    {
      "epoch": 1.05,
      "learning_rate": 3.7755102040816327e-06,
      "loss": 0.6333,
      "step": 3150
    },
    {
      "epoch": 1.06,
      "learning_rate": 3.724489795918368e-06,
      "loss": 0.6469,
      "step": 3175
    },
    {
      "epoch": 1.06,
      "learning_rate": 3.6734693877551024e-06,
      "loss": 0.6489,
      "step": 3200
    },
    {
      "epoch": 1.07,
      "learning_rate": 3.6224489795918373e-06,
      "loss": 0.6414,
      "step": 3225
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.5714285714285718e-06,
      "loss": 0.638,
      "step": 3250
    },
    {
      "epoch": 1.08,
      "eval_loss": 0.6626881957054138,
      "eval_runtime": 73.1024,
      "eval_samples_per_second": 13.57,
      "eval_steps_per_second": 0.109,
      "eval_wer": 0.37280287455316624,
      "step": 3250
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.5204081632653062e-06,
      "loss": 0.6471,
      "step": 3275
    },
    {
      "epoch": 1.1,
      "learning_rate": 3.469387755102041e-06,
      "loss": 0.643,
      "step": 3300
    },
    {
      "epoch": 1.11,
      "learning_rate": 3.4183673469387756e-06,
      "loss": 0.6386,
      "step": 3325
    },
    {
      "epoch": 1.11,
      "learning_rate": 3.3673469387755105e-06,
      "loss": 0.6287,
      "step": 3350
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.316326530612245e-06,
      "loss": 0.6432,
      "step": 3375
    },
    {
      "epoch": 1.13,
      "learning_rate": 3.2653061224489794e-06,
      "loss": 0.6338,
      "step": 3400
    },
    {
      "epoch": 1.14,
      "learning_rate": 3.2142857142857147e-06,
      "loss": 0.6443,
      "step": 3425
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.1632653061224496e-06,
      "loss": 0.6367,
      "step": 3450
    },
    {
      "epoch": 1.16,
      "learning_rate": 3.112244897959184e-06,
      "loss": 0.6495,
      "step": 3475
    },
    {
      "epoch": 1.16,
      "learning_rate": 3.0612244897959185e-06,
      "loss": 0.6222,
      "step": 3500
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.6573545336723328,
      "eval_runtime": 73.0325,
      "eval_samples_per_second": 13.583,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.37334000111129634,
      "step": 3500
    },
    {
      "epoch": 1.17,
      "learning_rate": 3.0102040816326534e-06,
      "loss": 0.6375,
      "step": 3525
    },
    {
      "epoch": 1.18,
      "learning_rate": 2.959183673469388e-06,
      "loss": 0.6291,
      "step": 3550
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.908163265306123e-06,
      "loss": 0.6246,
      "step": 3575
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.6314,
      "step": 3600
    },
    {
      "epoch": 1.21,
      "learning_rate": 2.8061224489795917e-06,
      "loss": 0.6339,
      "step": 3625
    },
    {
      "epoch": 1.21,
      "learning_rate": 2.7551020408163266e-06,
      "loss": 0.6433,
      "step": 3650
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.7040816326530615e-06,
      "loss": 0.6319,
      "step": 3675
    },
    {
      "epoch": 1.23,
      "learning_rate": 2.6530612244897964e-06,
      "loss": 0.6334,
      "step": 3700
    },
    {
      "epoch": 1.24,
      "learning_rate": 2.602040816326531e-06,
      "loss": 0.6307,
      "step": 3725
    },
    {
      "epoch": 1.25,
      "learning_rate": 2.5510204081632657e-06,
      "loss": 0.6323,
      "step": 3750
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.652826189994812,
      "eval_runtime": 73.0176,
      "eval_samples_per_second": 13.586,
      "eval_steps_per_second": 0.11,
      "eval_wer": 0.3690800318571614,
      "step": 3750
    },
    {
      "epoch": 1.25,
      "learning_rate": 2.5e-06,
      "loss": 0.6203,
      "step": 3775
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.4489795918367347e-06,
      "loss": 0.6357,
      "step": 3800
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.3979591836734696e-06,
      "loss": 0.6332,
      "step": 3825
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.3469387755102044e-06,
      "loss": 0.6336,
      "step": 3850
    },
    {
      "epoch": 1.29,
      "learning_rate": 2.295918367346939e-06,
      "loss": 0.6349,
      "step": 3875
    },
    {
      "epoch": 1.3,
      "learning_rate": 2.244897959183674e-06,
      "loss": 0.6358,
      "step": 3900
    },
    {
      "epoch": 1.3,
      "learning_rate": 2.1938775510204083e-06,
      "loss": 0.6247,
      "step": 3925
    },
    {
      "epoch": 1.31,
      "learning_rate": 2.1428571428571427e-06,
      "loss": 0.631,
      "step": 3950
    },
    {
      "epoch": 1.32,
      "learning_rate": 2.0918367346938776e-06,
      "loss": 0.64,
      "step": 3975
    },
    {
      "epoch": 1.33,
      "learning_rate": 2.0408163265306125e-06,
      "loss": 0.6192,
      "step": 4000
    },
    {
      "epoch": 1.33,
      "eval_loss": 0.6498121619224548,
      "eval_runtime": 73.6303,
      "eval_samples_per_second": 13.473,
      "eval_steps_per_second": 0.109,
      "eval_wer": 0.36880220777537,
      "step": 4000
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.989795918367347e-06,
      "loss": 0.6346,
      "step": 4025
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.938775510204082e-06,
      "loss": 0.6234,
      "step": 4050
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.8877551020408163e-06,
      "loss": 0.6271,
      "step": 4075
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.8367346938775512e-06,
      "loss": 0.6217,
      "step": 4100
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.7857142857142859e-06,
      "loss": 0.6215,
      "step": 4125
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.7346938775510206e-06,
      "loss": 0.6218,
      "step": 4150
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.6836734693877552e-06,
      "loss": 0.6337,
      "step": 4175
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.6326530612244897e-06,
      "loss": 0.63,
      "step": 4200
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.5816326530612248e-06,
      "loss": 0.6203,
      "step": 4225
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.5306122448979593e-06,
      "loss": 0.633,
      "step": 4250
    },
    {
      "epoch": 1.41,
      "eval_loss": 0.6468986868858337,
      "eval_runtime": 73.531,
      "eval_samples_per_second": 13.491,
      "eval_steps_per_second": 0.109,
      "eval_wer": 0.36772795465910985,
      "step": 4250
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.479591836734694e-06,
      "loss": 0.6229,
      "step": 4275
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.6218,
      "step": 4300
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.3775510204081633e-06,
      "loss": 0.6203,
      "step": 4325
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.3265306122448982e-06,
      "loss": 0.6238,
      "step": 4350
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.2755102040816329e-06,
      "loss": 0.6249,
      "step": 4375
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.2244897959183673e-06,
      "loss": 0.6159,
      "step": 4400
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.1734693877551022e-06,
      "loss": 0.6258,
      "step": 4425
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.122448979591837e-06,
      "loss": 0.6258,
      "step": 4450
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0714285714285714e-06,
      "loss": 0.6236,
      "step": 4475
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0204081632653063e-06,
      "loss": 0.6229,
      "step": 4500
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.6451022028923035,
      "eval_runtime": 73.6892,
      "eval_samples_per_second": 13.462,
      "eval_steps_per_second": 0.109,
      "eval_wer": 0.3680983867681651,
      "step": 4500
    },
    {
      "epoch": 1.5,
      "learning_rate": 9.69387755102041e-07,
      "loss": 0.6218,
      "step": 4525
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.183673469387756e-07,
      "loss": 0.6307,
      "step": 4550
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.673469387755103e-07,
      "loss": 0.6185,
      "step": 4575
    },
    {
      "epoch": 1.53,
      "learning_rate": 8.163265306122449e-07,
      "loss": 0.6174,
      "step": 4600
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.653061224489796e-07,
      "loss": 0.6314,
      "step": 4625
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.6168,
      "step": 4650
    },
    {
      "epoch": 1.55,
      "learning_rate": 6.632653061224491e-07,
      "loss": 0.6246,
      "step": 4675
    },
    {
      "epoch": 1.56,
      "learning_rate": 6.122448979591837e-07,
      "loss": 0.6162,
      "step": 4700
    },
    {
      "epoch": 1.57,
      "learning_rate": 5.612244897959184e-07,
      "loss": 0.6192,
      "step": 4725
    },
    {
      "epoch": 1.58,
      "learning_rate": 5.102040816326531e-07,
      "loss": 0.6246,
      "step": 4750
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.6439383625984192,
      "eval_runtime": 74.0251,
      "eval_samples_per_second": 13.401,
      "eval_steps_per_second": 0.108,
      "eval_wer": 0.37059880350428775,
      "step": 4750
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.591836734693878e-07,
      "loss": 0.6128,
      "step": 4775
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.0816326530612243e-07,
      "loss": 0.6108,
      "step": 4800
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.6218,
      "step": 4825
    },
    {
      "epoch": 1.61,
      "learning_rate": 3.0612244897959183e-07,
      "loss": 0.6298,
      "step": 4850
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.5510204081632656e-07,
      "loss": 0.6229,
      "step": 4875
    },
    {
      "epoch": 1.63,
      "learning_rate": 2.0408163265306121e-07,
      "loss": 0.6067,
      "step": 4900
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.5306122448979592e-07,
      "loss": 0.6201,
      "step": 4925
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.0204081632653061e-07,
      "loss": 0.6221,
      "step": 4950
    },
    {
      "epoch": 1.65,
      "learning_rate": 5.1020408163265303e-08,
      "loss": 0.6209,
      "step": 4975
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0,
      "loss": 0.6214,
      "step": 5000
    },
    {
      "epoch": 1.66,
      "eval_loss": 0.6434767842292786,
      "eval_runtime": 73.9972,
      "eval_samples_per_second": 13.406,
      "eval_steps_per_second": 0.108,
      "eval_wer": 0.3700987201570632,
      "step": 5000
    },
    {
      "epoch": 1.66,
      "step": 5000,
      "total_flos": 1.5756085807389082e+19,
      "train_loss": 0.7225225172042846,
      "train_runtime": 18771.4877,
      "train_samples_per_second": 34.094,
      "train_steps_per_second": 0.266
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 2,
  "total_flos": 1.5756085807389082e+19,
  "trial_name": null,
  "trial_params": null
}