{
  "best_metric": 12.303981501169467,
  "best_model_checkpoint": "./checkpoint-4000",
  "epoch": 12.062,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 3.52e-07,
      "loss": 1.0655,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.52e-07,
      "loss": 0.9756,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.152e-06,
      "loss": 0.7902,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.552e-06,
      "loss": 0.4928,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9519999999999997e-06,
      "loss": 0.3652,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.3519999999999997e-06,
      "loss": 0.3222,
      "step": 150
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.7519999999999997e-06,
      "loss": 0.3031,
      "step": 175
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.152e-06,
      "loss": 0.3682,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.552e-06,
      "loss": 0.3873,
      "step": 225
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.952e-06,
      "loss": 0.3546,
      "step": 250
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.352e-06,
      "loss": 0.3597,
      "step": 275
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.7519999999999996e-06,
      "loss": 0.3844,
      "step": 300
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.152e-06,
      "loss": 0.3962,
      "step": 325
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.5519999999999995e-06,
      "loss": 0.3593,
      "step": 350
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.952e-06,
      "loss": 0.3424,
      "step": 375
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.352e-06,
      "loss": 0.348,
      "step": 400
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.751999999999999e-06,
      "loss": 0.3214,
      "step": 425
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.1519999999999995e-06,
      "loss": 0.2422,
      "step": 450
    },
    {
      "epoch": 1.0,
      "learning_rate": 7.551999999999999e-06,
      "loss": 0.2384,
      "step": 475
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.952e-06,
      "loss": 0.2678,
      "step": 500
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.967999999999999e-06,
      "loss": 0.2878,
      "step": 525
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.931636363636364e-06,
      "loss": 0.2682,
      "step": 550
    },
    {
      "epoch": 1.02,
      "learning_rate": 7.895272727272728e-06,
      "loss": 0.2454,
      "step": 575
    },
    {
      "epoch": 1.02,
      "learning_rate": 7.85890909090909e-06,
      "loss": 0.2459,
      "step": 600
    },
    {
      "epoch": 1.03,
      "learning_rate": 7.822545454545454e-06,
      "loss": 0.2372,
      "step": 625
    },
    {
      "epoch": 1.03,
      "learning_rate": 7.786181818181817e-06,
      "loss": 0.2726,
      "step": 650
    },
    {
      "epoch": 1.03,
      "learning_rate": 7.749818181818181e-06,
      "loss": 0.1999,
      "step": 675
    },
    {
      "epoch": 1.04,
      "learning_rate": 7.713454545454545e-06,
      "loss": 0.2111,
      "step": 700
    },
    {
      "epoch": 1.04,
      "learning_rate": 7.677090909090908e-06,
      "loss": 0.2177,
      "step": 725
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.640727272727272e-06,
      "loss": 0.2232,
      "step": 750
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.6043636363636365e-06,
      "loss": 0.2485,
      "step": 775
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.567999999999999e-06,
      "loss": 0.2419,
      "step": 800
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.531636363636363e-06,
      "loss": 0.2476,
      "step": 825
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.495272727272727e-06,
      "loss": 0.2193,
      "step": 850
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.45890909090909e-06,
      "loss": 0.1901,
      "step": 875
    },
    {
      "epoch": 1.07,
      "learning_rate": 7.422545454545454e-06,
      "loss": 0.1994,
      "step": 900
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.386181818181818e-06,
      "loss": 0.2006,
      "step": 925
    },
    {
      "epoch": 2.0,
      "learning_rate": 7.3498181818181814e-06,
      "loss": 0.2218,
      "step": 950
    },
    {
      "epoch": 2.01,
      "learning_rate": 7.313454545454545e-06,
      "loss": 0.2398,
      "step": 975
    },
    {
      "epoch": 2.01,
      "learning_rate": 7.277090909090909e-06,
      "loss": 0.2737,
      "step": 1000
    },
    {
      "epoch": 2.01,
      "eval_loss": 0.27280035614967346,
      "eval_runtime": 5409.8244,
      "eval_samples_per_second": 2.773,
      "eval_steps_per_second": 0.087,
      "eval_wer": 13.409658728471188,
      "step": 1000
    },
    {
      "epoch": 2.01,
      "learning_rate": 7.2407272727272725e-06,
      "loss": 0.2345,
      "step": 1025
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.204363636363636e-06,
      "loss": 0.2426,
      "step": 1050
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.168e-06,
      "loss": 0.2022,
      "step": 1075
    },
    {
      "epoch": 2.03,
      "learning_rate": 7.1316363636363635e-06,
      "loss": 0.2105,
      "step": 1100
    },
    {
      "epoch": 2.03,
      "learning_rate": 7.095272727272726e-06,
      "loss": 0.1926,
      "step": 1125
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.058909090909091e-06,
      "loss": 0.1891,
      "step": 1150
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.0225454545454545e-06,
      "loss": 0.1971,
      "step": 1175
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.986181818181818e-06,
      "loss": 0.2074,
      "step": 1200
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.949818181818181e-06,
      "loss": 0.2283,
      "step": 1225
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.913454545454545e-06,
      "loss": 0.2194,
      "step": 1250
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.877090909090909e-06,
      "loss": 0.1934,
      "step": 1275
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.840727272727273e-06,
      "loss": 0.1857,
      "step": 1300
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.804363636363636e-06,
      "loss": 0.1681,
      "step": 1325
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.767999999999999e-06,
      "loss": 0.1534,
      "step": 1350
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.731636363636363e-06,
      "loss": 0.1522,
      "step": 1375
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.695272727272728e-06,
      "loss": 0.1874,
      "step": 1400
    },
    {
      "epoch": 3.0,
      "learning_rate": 6.658909090909091e-06,
      "loss": 0.2014,
      "step": 1425
    },
    {
      "epoch": 3.01,
      "learning_rate": 6.622545454545454e-06,
      "loss": 0.1817,
      "step": 1450
    },
    {
      "epoch": 3.01,
      "learning_rate": 6.586181818181818e-06,
      "loss": 0.1792,
      "step": 1475
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.5498181818181815e-06,
      "loss": 0.1578,
      "step": 1500
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.513454545454545e-06,
      "loss": 0.1522,
      "step": 1525
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.477090909090909e-06,
      "loss": 0.1457,
      "step": 1550
    },
    {
      "epoch": 3.03,
      "learning_rate": 6.4407272727272725e-06,
      "loss": 0.1482,
      "step": 1575
    },
    {
      "epoch": 3.03,
      "learning_rate": 6.404363636363636e-06,
      "loss": 0.1669,
      "step": 1600
    },
    {
      "epoch": 3.04,
      "learning_rate": 6.368e-06,
      "loss": 0.2161,
      "step": 1625
    },
    {
      "epoch": 3.04,
      "learning_rate": 6.331636363636363e-06,
      "loss": 0.1928,
      "step": 1650
    },
    {
      "epoch": 3.04,
      "learning_rate": 6.295272727272727e-06,
      "loss": 0.1923,
      "step": 1675
    },
    {
      "epoch": 3.05,
      "learning_rate": 6.258909090909091e-06,
      "loss": 0.2077,
      "step": 1700
    },
    {
      "epoch": 3.05,
      "learning_rate": 6.2225454545454546e-06,
      "loss": 0.2226,
      "step": 1725
    },
    {
      "epoch": 3.06,
      "learning_rate": 6.186181818181817e-06,
      "loss": 0.2089,
      "step": 1750
    },
    {
      "epoch": 3.06,
      "learning_rate": 6.149818181818181e-06,
      "loss": 0.1904,
      "step": 1775
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.113454545454546e-06,
      "loss": 0.1825,
      "step": 1800
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.077090909090909e-06,
      "loss": 0.2006,
      "step": 1825
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.040727272727273e-06,
      "loss": 0.1328,
      "step": 1850
    },
    {
      "epoch": 3.08,
      "learning_rate": 6.004363636363636e-06,
      "loss": 0.1086,
      "step": 1875
    },
    {
      "epoch": 4.0,
      "learning_rate": 5.9679999999999994e-06,
      "loss": 0.1554,
      "step": 1900
    },
    {
      "epoch": 4.01,
      "learning_rate": 5.931636363636364e-06,
      "loss": 0.1507,
      "step": 1925
    },
    {
      "epoch": 4.01,
      "learning_rate": 5.895272727272728e-06,
      "loss": 0.1845,
      "step": 1950
    },
    {
      "epoch": 4.02,
      "learning_rate": 5.8589090909090905e-06,
      "loss": 0.1394,
      "step": 1975
    },
    {
      "epoch": 4.02,
      "learning_rate": 5.822545454545454e-06,
      "loss": 0.1536,
      "step": 2000
    },
    {
      "epoch": 4.02,
      "eval_loss": 0.2611413598060608,
      "eval_runtime": 5136.3747,
      "eval_samples_per_second": 2.921,
      "eval_steps_per_second": 0.091,
      "eval_wer": 12.989714012332554,
      "step": 2000
    },
    {
      "epoch": 4.02,
      "learning_rate": 5.786181818181818e-06,
      "loss": 0.1309,
      "step": 2025
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.7498181818181815e-06,
      "loss": 0.1297,
      "step": 2050
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.713454545454545e-06,
      "loss": 0.1217,
      "step": 2075
    },
    {
      "epoch": 4.04,
      "learning_rate": 5.677090909090909e-06,
      "loss": 0.1264,
      "step": 2100
    },
    {
      "epoch": 4.04,
      "learning_rate": 5.6407272727272725e-06,
      "loss": 0.1166,
      "step": 2125
    },
    {
      "epoch": 4.05,
      "learning_rate": 5.604363636363636e-06,
      "loss": 0.1529,
      "step": 2150
    },
    {
      "epoch": 4.05,
      "learning_rate": 5.567999999999999e-06,
      "loss": 0.1509,
      "step": 2175
    },
    {
      "epoch": 4.05,
      "learning_rate": 5.5316363636363636e-06,
      "loss": 0.1333,
      "step": 2200
    },
    {
      "epoch": 4.06,
      "learning_rate": 5.495272727272727e-06,
      "loss": 0.1253,
      "step": 2225
    },
    {
      "epoch": 4.06,
      "learning_rate": 5.458909090909091e-06,
      "loss": 0.1116,
      "step": 2250
    },
    {
      "epoch": 4.07,
      "learning_rate": 5.422545454545454e-06,
      "loss": 0.1084,
      "step": 2275
    },
    {
      "epoch": 4.07,
      "learning_rate": 5.386181818181817e-06,
      "loss": 0.0988,
      "step": 2300
    },
    {
      "epoch": 4.07,
      "learning_rate": 5.349818181818182e-06,
      "loss": 0.1122,
      "step": 2325
    },
    {
      "epoch": 5.0,
      "learning_rate": 5.313454545454546e-06,
      "loss": 0.1112,
      "step": 2350
    },
    {
      "epoch": 5.0,
      "learning_rate": 5.277090909090909e-06,
      "loss": 0.1192,
      "step": 2375
    },
    {
      "epoch": 5.01,
      "learning_rate": 5.240727272727272e-06,
      "loss": 0.1304,
      "step": 2400
    },
    {
      "epoch": 5.01,
      "learning_rate": 5.204363636363636e-06,
      "loss": 0.1241,
      "step": 2425
    },
    {
      "epoch": 5.02,
      "learning_rate": 5.168e-06,
      "loss": 0.1102,
      "step": 2450
    },
    {
      "epoch": 5.02,
      "learning_rate": 5.131636363636364e-06,
      "loss": 0.1181,
      "step": 2475
    },
    {
      "epoch": 5.03,
      "learning_rate": 5.095272727272727e-06,
      "loss": 0.1136,
      "step": 2500
    },
    {
      "epoch": 5.03,
      "learning_rate": 5.0589090909090905e-06,
      "loss": 0.1396,
      "step": 2525
    },
    {
      "epoch": 5.03,
      "learning_rate": 5.022545454545454e-06,
      "loss": 0.0885,
      "step": 2550
    },
    {
      "epoch": 5.04,
      "learning_rate": 4.986181818181818e-06,
      "loss": 0.1082,
      "step": 2575
    },
    {
      "epoch": 5.04,
      "learning_rate": 4.9498181818181815e-06,
      "loss": 0.1056,
      "step": 2600
    },
    {
      "epoch": 5.05,
      "learning_rate": 4.913454545454545e-06,
      "loss": 0.1102,
      "step": 2625
    },
    {
      "epoch": 5.05,
      "learning_rate": 4.877090909090909e-06,
      "loss": 0.123,
      "step": 2650
    },
    {
      "epoch": 5.05,
      "learning_rate": 4.8407272727272726e-06,
      "loss": 0.1202,
      "step": 2675
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.804363636363635e-06,
      "loss": 0.1317,
      "step": 2700
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.768e-06,
      "loss": 0.1099,
      "step": 2725
    },
    {
      "epoch": 5.07,
      "learning_rate": 4.731636363636364e-06,
      "loss": 0.0966,
      "step": 2750
    },
    {
      "epoch": 5.07,
      "learning_rate": 4.695272727272727e-06,
      "loss": 0.0976,
      "step": 2775
    },
    {
      "epoch": 5.08,
      "learning_rate": 4.658909090909091e-06,
      "loss": 0.1015,
      "step": 2800
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.622545454545454e-06,
      "loss": 0.0906,
      "step": 2825
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.586181818181818e-06,
      "loss": 0.0913,
      "step": 2850
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.549818181818182e-06,
      "loss": 0.0917,
      "step": 2875
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.513454545454546e-06,
      "loss": 0.0857,
      "step": 2900
    },
    {
      "epoch": 6.02,
      "learning_rate": 4.4770909090909085e-06,
      "loss": 0.0799,
      "step": 2925
    },
    {
      "epoch": 6.02,
      "learning_rate": 4.440727272727272e-06,
      "loss": 0.0794,
      "step": 2950
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.404363636363637e-06,
      "loss": 0.0874,
      "step": 2975
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.368e-06,
      "loss": 0.0905,
      "step": 3000
    },
    {
      "epoch": 6.03,
      "eval_loss": 0.26857462525367737,
      "eval_runtime": 5158.1008,
      "eval_samples_per_second": 2.909,
      "eval_steps_per_second": 0.091,
      "eval_wer": 12.927253880501807,
      "step": 3000
    },
    {
      "epoch": 6.04,
      "learning_rate": 4.331636363636363e-06,
      "loss": 0.0671,
      "step": 3025
    },
    {
      "epoch": 6.04,
      "learning_rate": 4.295272727272727e-06,
      "loss": 0.0776,
      "step": 3050
    },
    {
      "epoch": 6.04,
      "learning_rate": 4.2589090909090905e-06,
      "loss": 0.0743,
      "step": 3075
    },
    {
      "epoch": 6.05,
      "learning_rate": 4.222545454545454e-06,
      "loss": 0.0817,
      "step": 3100
    },
    {
      "epoch": 6.05,
      "learning_rate": 4.186181818181818e-06,
      "loss": 0.088,
      "step": 3125
    },
    {
      "epoch": 6.06,
      "learning_rate": 4.1498181818181816e-06,
      "loss": 0.0901,
      "step": 3150
    },
    {
      "epoch": 6.06,
      "learning_rate": 4.113454545454545e-06,
      "loss": 0.094,
      "step": 3175
    },
    {
      "epoch": 6.06,
      "learning_rate": 4.077090909090909e-06,
      "loss": 0.0753,
      "step": 3200
    },
    {
      "epoch": 6.07,
      "learning_rate": 4.040727272727273e-06,
      "loss": 0.0676,
      "step": 3225
    },
    {
      "epoch": 6.07,
      "learning_rate": 4.004363636363636e-06,
      "loss": 0.0714,
      "step": 3250
    },
    {
      "epoch": 6.08,
      "learning_rate": 3.968e-06,
      "loss": 0.0729,
      "step": 3275
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.931636363636364e-06,
      "loss": 0.1062,
      "step": 3300
    },
    {
      "epoch": 7.01,
      "learning_rate": 3.895272727272727e-06,
      "loss": 0.0978,
      "step": 3325
    },
    {
      "epoch": 7.01,
      "learning_rate": 3.858909090909091e-06,
      "loss": 0.0966,
      "step": 3350
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.822545454545455e-06,
      "loss": 0.0864,
      "step": 3375
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.7861818181818183e-06,
      "loss": 0.0805,
      "step": 3400
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.7498181818181816e-06,
      "loss": 0.0788,
      "step": 3425
    },
    {
      "epoch": 7.03,
      "learning_rate": 3.7134545454545453e-06,
      "loss": 0.0825,
      "step": 3450
    },
    {
      "epoch": 7.03,
      "learning_rate": 3.677090909090909e-06,
      "loss": 0.0943,
      "step": 3475
    },
    {
      "epoch": 7.04,
      "learning_rate": 3.6407272727272726e-06,
      "loss": 0.1333,
      "step": 3500
    },
    {
      "epoch": 7.04,
      "learning_rate": 3.604363636363636e-06,
      "loss": 0.1137,
      "step": 3525
    },
    {
      "epoch": 7.04,
      "learning_rate": 3.568e-06,
      "loss": 0.1184,
      "step": 3550
    },
    {
      "epoch": 7.05,
      "learning_rate": 3.5316363636363632e-06,
      "loss": 0.1253,
      "step": 3575
    },
    {
      "epoch": 7.05,
      "learning_rate": 3.4952727272727273e-06,
      "loss": 0.1369,
      "step": 3600
    },
    {
      "epoch": 7.06,
      "learning_rate": 3.4589090909090906e-06,
      "loss": 0.1325,
      "step": 3625
    },
    {
      "epoch": 7.06,
      "learning_rate": 3.4225454545454543e-06,
      "loss": 0.1182,
      "step": 3650
    },
    {
      "epoch": 7.07,
      "learning_rate": 3.3861818181818184e-06,
      "loss": 0.112,
      "step": 3675
    },
    {
      "epoch": 7.07,
      "learning_rate": 3.3498181818181816e-06,
      "loss": 0.1322,
      "step": 3700
    },
    {
      "epoch": 7.07,
      "learning_rate": 3.3134545454545457e-06,
      "loss": 0.0821,
      "step": 3725
    },
    {
      "epoch": 7.08,
      "learning_rate": 3.277090909090909e-06,
      "loss": 0.0599,
      "step": 3750
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.2407272727272726e-06,
      "loss": 0.1378,
      "step": 3775
    },
    {
      "epoch": 8.01,
      "learning_rate": 3.2043636363636363e-06,
      "loss": 0.1425,
      "step": 3800
    },
    {
      "epoch": 8.01,
      "learning_rate": 3.168e-06,
      "loss": 0.1534,
      "step": 3825
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.1316363636363632e-06,
      "loss": 0.139,
      "step": 3850
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.0952727272727273e-06,
      "loss": 0.1286,
      "step": 3875
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.0589090909090906e-06,
      "loss": 0.1312,
      "step": 3900
    },
    {
      "epoch": 8.03,
      "learning_rate": 3.0225454545454547e-06,
      "loss": 0.1612,
      "step": 3925
    },
    {
      "epoch": 8.03,
      "learning_rate": 2.986181818181818e-06,
      "loss": 0.127,
      "step": 3950
    },
    {
      "epoch": 8.04,
      "learning_rate": 2.9498181818181816e-06,
      "loss": 0.1268,
      "step": 3975
    },
    {
      "epoch": 8.04,
      "learning_rate": 2.9134545454545453e-06,
      "loss": 0.1301,
      "step": 4000
    },
    {
      "epoch": 8.04,
      "eval_loss": 0.25342342257499695,
      "eval_runtime": 5076.796,
      "eval_samples_per_second": 2.955,
      "eval_steps_per_second": 0.092,
      "eval_wer": 12.303981501169467,
      "step": 4000
    },
    {
      "epoch": 8.05,
      "learning_rate": 2.877090909090909e-06,
      "loss": 0.133,
      "step": 4025
    },
    {
      "epoch": 8.05,
      "learning_rate": 2.8407272727272722e-06,
      "loss": 0.1366,
      "step": 4050
    },
    {
      "epoch": 8.05,
      "learning_rate": 2.8043636363636363e-06,
      "loss": 0.1338,
      "step": 4075
    },
    {
      "epoch": 8.06,
      "learning_rate": 2.7679999999999996e-06,
      "loss": 0.1319,
      "step": 4100
    },
    {
      "epoch": 8.06,
      "learning_rate": 2.7316363636363637e-06,
      "loss": 0.1254,
      "step": 4125
    },
    {
      "epoch": 8.07,
      "learning_rate": 2.6952727272727274e-06,
      "loss": 0.1219,
      "step": 4150
    },
    {
      "epoch": 8.07,
      "learning_rate": 2.6589090909090906e-06,
      "loss": 0.1427,
      "step": 4175
    },
    {
      "epoch": 8.07,
      "learning_rate": 2.6225454545454547e-06,
      "loss": 0.1301,
      "step": 4200
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.586181818181818e-06,
      "loss": 0.1234,
      "step": 4225
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.549818181818182e-06,
      "loss": 0.0937,
      "step": 4250
    },
    {
      "epoch": 9.01,
      "learning_rate": 2.5134545454545453e-06,
      "loss": 0.0843,
      "step": 4275
    },
    {
      "epoch": 9.01,
      "learning_rate": 2.477090909090909e-06,
      "loss": 0.1116,
      "step": 4300
    },
    {
      "epoch": 9.02,
      "learning_rate": 2.4407272727272727e-06,
      "loss": 0.0894,
      "step": 4325
    },
    {
      "epoch": 9.02,
      "learning_rate": 2.4043636363636364e-06,
      "loss": 0.0813,
      "step": 4350
    },
    {
      "epoch": 9.03,
      "learning_rate": 2.3679999999999996e-06,
      "loss": 0.0782,
      "step": 4375
    },
    {
      "epoch": 9.03,
      "learning_rate": 2.3316363636363637e-06,
      "loss": 0.0706,
      "step": 4400
    },
    {
      "epoch": 9.03,
      "learning_rate": 2.295272727272727e-06,
      "loss": 0.0674,
      "step": 4425
    },
    {
      "epoch": 9.04,
      "learning_rate": 2.258909090909091e-06,
      "loss": 0.0769,
      "step": 4450
    },
    {
      "epoch": 9.04,
      "learning_rate": 2.2225454545454543e-06,
      "loss": 0.0688,
      "step": 4475
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.186181818181818e-06,
      "loss": 0.0904,
      "step": 4500
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.1498181818181817e-06,
      "loss": 0.0896,
      "step": 4525
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.1134545454545453e-06,
      "loss": 0.0699,
      "step": 4550
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.0770909090909086e-06,
      "loss": 0.071,
      "step": 4575
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.0407272727272727e-06,
      "loss": 0.0629,
      "step": 4600
    },
    {
      "epoch": 9.07,
      "learning_rate": 2.0043636363636364e-06,
      "loss": 0.0588,
      "step": 4625
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.968e-06,
      "loss": 0.0579,
      "step": 4650
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.9316363636363633e-06,
      "loss": 0.0671,
      "step": 4675
    },
    {
      "epoch": 10.0,
      "learning_rate": 1.8952727272727272e-06,
      "loss": 0.0735,
      "step": 4700
    },
    {
      "epoch": 10.01,
      "learning_rate": 1.8589090909090909e-06,
      "loss": 0.0715,
      "step": 4725
    },
    {
      "epoch": 10.01,
      "learning_rate": 1.8225454545454543e-06,
      "loss": 0.0705,
      "step": 4750
    },
    {
      "epoch": 10.01,
      "learning_rate": 1.786181818181818e-06,
      "loss": 0.0586,
      "step": 4775
    },
    {
      "epoch": 10.02,
      "learning_rate": 1.7498181818181817e-06,
      "loss": 0.0556,
      "step": 4800
    },
    {
      "epoch": 10.02,
      "learning_rate": 1.7134545454545456e-06,
      "loss": 0.0543,
      "step": 4825
    },
    {
      "epoch": 10.03,
      "learning_rate": 1.677090909090909e-06,
      "loss": 0.053,
      "step": 4850
    },
    {
      "epoch": 10.03,
      "learning_rate": 1.6407272727272727e-06,
      "loss": 0.0578,
      "step": 4875
    },
    {
      "epoch": 10.04,
      "learning_rate": 1.6043636363636364e-06,
      "loss": 0.0944,
      "step": 4900
    },
    {
      "epoch": 10.04,
      "learning_rate": 1.568e-06,
      "loss": 0.0829,
      "step": 4925
    },
    {
      "epoch": 10.04,
      "learning_rate": 1.5316363636363635e-06,
      "loss": 0.0856,
      "step": 4950
    },
    {
      "epoch": 10.05,
      "learning_rate": 1.4952727272727272e-06,
      "loss": 0.0906,
      "step": 4975
    },
    {
      "epoch": 10.05,
      "learning_rate": 1.4589090909090909e-06,
      "loss": 0.096,
      "step": 5000
    },
    {
      "epoch": 10.05,
      "eval_loss": 0.27268335223197937,
      "eval_runtime": 5022.2596,
      "eval_samples_per_second": 2.987,
      "eval_steps_per_second": 0.093,
      "eval_wer": 12.612959812885391,
      "step": 5000
    },
    {
      "epoch": 10.06,
      "learning_rate": 1.4225454545454546e-06,
      "loss": 0.1049,
      "step": 5025
    },
    {
      "epoch": 10.06,
      "learning_rate": 1.386181818181818e-06,
      "loss": 0.09,
      "step": 5050
    },
    {
      "epoch": 10.06,
      "learning_rate": 1.3498181818181817e-06,
      "loss": 0.0809,
      "step": 5075
    },
    {
      "epoch": 10.07,
      "learning_rate": 1.3134545454545454e-06,
      "loss": 0.1007,
      "step": 5100
    },
    {
      "epoch": 10.07,
      "learning_rate": 1.277090909090909e-06,
      "loss": 0.0755,
      "step": 5125
    },
    {
      "epoch": 10.08,
      "learning_rate": 1.2407272727272725e-06,
      "loss": 0.0448,
      "step": 5150
    },
    {
      "epoch": 11.0,
      "learning_rate": 1.2043636363636362e-06,
      "loss": 0.0836,
      "step": 5175
    },
    {
      "epoch": 11.01,
      "learning_rate": 1.1679999999999999e-06,
      "loss": 0.1083,
      "step": 5200
    },
    {
      "epoch": 11.01,
      "learning_rate": 1.1316363636363635e-06,
      "loss": 0.1112,
      "step": 5225
    },
    {
      "epoch": 11.02,
      "learning_rate": 1.0952727272727272e-06,
      "loss": 0.1122,
      "step": 5250
    },
    {
      "epoch": 11.02,
      "learning_rate": 1.0589090909090907e-06,
      "loss": 0.0921,
      "step": 5275
    },
    {
      "epoch": 11.02,
      "learning_rate": 1.0225454545454546e-06,
      "loss": 0.0977,
      "step": 5300
    },
    {
      "epoch": 11.03,
      "learning_rate": 9.86181818181818e-07,
      "loss": 0.1235,
      "step": 5325
    },
    {
      "epoch": 11.03,
      "learning_rate": 9.498181818181817e-07,
      "loss": 0.1045,
      "step": 5350
    },
    {
      "epoch": 11.04,
      "learning_rate": 9.134545454545454e-07,
      "loss": 0.0944,
      "step": 5375
    },
    {
      "epoch": 11.04,
      "learning_rate": 8.770909090909091e-07,
      "loss": 0.0946,
      "step": 5400
    },
    {
      "epoch": 11.04,
      "learning_rate": 8.407272727272727e-07,
      "loss": 0.1045,
      "step": 5425
    },
    {
      "epoch": 11.05,
      "learning_rate": 8.043636363636363e-07,
      "loss": 0.1078,
      "step": 5450
    },
    {
      "epoch": 11.05,
      "learning_rate": 7.68e-07,
      "loss": 0.1053,
      "step": 5475
    },
    {
      "epoch": 11.06,
      "learning_rate": 7.316363636363636e-07,
      "loss": 0.1013,
      "step": 5500
    },
    {
      "epoch": 11.06,
      "learning_rate": 6.952727272727272e-07,
      "loss": 0.098,
      "step": 5525
    },
    {
      "epoch": 11.07,
      "learning_rate": 6.589090909090908e-07,
      "loss": 0.0974,
      "step": 5550
    },
    {
      "epoch": 11.07,
      "learning_rate": 6.225454545454545e-07,
      "loss": 0.1084,
      "step": 5575
    },
    {
      "epoch": 11.07,
      "learning_rate": 5.861818181818181e-07,
      "loss": 0.1079,
      "step": 5600
    },
    {
      "epoch": 11.08,
      "learning_rate": 5.498181818181818e-07,
      "loss": 0.0986,
      "step": 5625
    },
    {
      "epoch": 12.0,
      "learning_rate": 5.134545454545455e-07,
      "loss": 0.0563,
      "step": 5650
    },
    {
      "epoch": 12.01,
      "learning_rate": 4.770909090909091e-07,
      "loss": 0.0572,
      "step": 5675
    },
    {
      "epoch": 12.01,
      "learning_rate": 4.407272727272727e-07,
      "loss": 0.0561,
      "step": 5700
    },
    {
      "epoch": 12.02,
      "learning_rate": 4.0436363636363634e-07,
      "loss": 0.0505,
      "step": 5725
    },
    {
      "epoch": 12.02,
      "learning_rate": 3.6799999999999996e-07,
      "loss": 0.0479,
      "step": 5750
    },
    {
      "epoch": 12.02,
      "learning_rate": 3.316363636363636e-07,
      "loss": 0.0522,
      "step": 5775
    },
    {
      "epoch": 12.03,
      "learning_rate": 2.9527272727272726e-07,
      "loss": 0.0617,
      "step": 5800
    },
    {
      "epoch": 12.03,
      "learning_rate": 2.5890909090909094e-07,
      "loss": 0.0468,
      "step": 5825
    },
    {
      "epoch": 12.04,
      "learning_rate": 2.2254545454545453e-07,
      "loss": 0.0468,
      "step": 5850
    },
    {
      "epoch": 12.04,
      "learning_rate": 1.8618181818181816e-07,
      "loss": 0.0472,
      "step": 5875
    },
    {
      "epoch": 12.05,
      "learning_rate": 1.498181818181818e-07,
      "loss": 0.0495,
      "step": 5900
    },
    {
      "epoch": 12.05,
      "learning_rate": 1.1345454545454546e-07,
      "loss": 0.0546,
      "step": 5925
    },
    {
      "epoch": 12.05,
      "learning_rate": 7.709090909090908e-08,
      "loss": 0.0579,
      "step": 5950
    },
    {
      "epoch": 12.06,
      "learning_rate": 4.072727272727273e-08,
      "loss": 0.0623,
      "step": 5975
    },
    {
      "epoch": 12.06,
      "learning_rate": 4.363636363636364e-09,
      "loss": 0.0604,
      "step": 6000
    },
    {
      "epoch": 12.06,
      "eval_loss": 0.26977524161338806,
      "eval_runtime": 5061.9846,
      "eval_samples_per_second": 2.964,
      "eval_steps_per_second": 0.093,
      "eval_wer": 12.502657877950243,
      "step": 6000
    },
    {
      "epoch": 12.06,
      "step": 6000,
      "total_flos": 1.1074060717719552e+20,
      "train_loss": 0.1507516227165858,
      "train_runtime": 84190.0984,
      "train_samples_per_second": 4.561,
      "train_steps_per_second": 0.071
    }
  ],
  "max_steps": 6000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.1074060717719552e+20,
  "trial_name": null,
  "trial_params": null
}