{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.862593783494106,
  "global_step": 11500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 3.7125e-06,
      "loss": 16.3142,
      "step": 100
    },
    {
      "epoch": 0.17,
      "learning_rate": 7.4625e-06,
      "loss": 9.7017,
      "step": 200
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.1212499999999998e-05,
      "loss": 6.7433,
      "step": 300
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.49625e-05,
      "loss": 5.7032,
      "step": 400
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.8712499999999997e-05,
      "loss": 4.7972,
      "step": 500
    },
    {
      "epoch": 0.43,
      "eval_loss": 5.140112400054932,
      "eval_runtime": 323.0627,
      "eval_samples_per_second": 31.833,
      "eval_steps_per_second": 3.981,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.2462499999999997e-05,
      "loss": 4.1097,
      "step": 600
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.6212499999999997e-05,
      "loss": 3.6756,
      "step": 700
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.99625e-05,
      "loss": 3.5057,
      "step": 800
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.37125e-05,
      "loss": 3.4116,
      "step": 900
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.7462499999999996e-05,
      "loss": 3.3241,
      "step": 1000
    },
    {
      "epoch": 0.86,
      "eval_loss": 3.3219833374023438,
      "eval_runtime": 316.4143,
      "eval_samples_per_second": 32.502,
      "eval_steps_per_second": 4.064,
      "eval_wer": 1.0,
      "step": 1000
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.12125e-05,
      "loss": 3.2968,
      "step": 1100
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.4962499999999995e-05,
      "loss": 3.2757,
      "step": 1200
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.871249999999999e-05,
      "loss": 3.2343,
      "step": 1300
    },
    {
      "epoch": 1.2,
      "learning_rate": 5.2462499999999994e-05,
      "loss": 3.1872,
      "step": 1400
    },
    {
      "epoch": 1.29,
      "learning_rate": 5.62125e-05,
      "loss": 3.1432,
      "step": 1500
    },
    {
      "epoch": 1.29,
      "eval_loss": 3.080551862716675,
      "eval_runtime": 318.2154,
      "eval_samples_per_second": 32.318,
      "eval_steps_per_second": 4.041,
      "eval_wer": 0.9998891086181085,
      "step": 1500
    },
    {
      "epoch": 1.37,
      "learning_rate": 5.9962499999999994e-05,
      "loss": 3.0996,
      "step": 1600
    },
    {
      "epoch": 1.46,
      "learning_rate": 6.37125e-05,
      "loss": 3.0582,
      "step": 1700
    },
    {
      "epoch": 1.54,
      "learning_rate": 6.746249999999999e-05,
      "loss": 3.0313,
      "step": 1800
    },
    {
      "epoch": 1.63,
      "learning_rate": 7.121249999999999e-05,
      "loss": 2.9824,
      "step": 1900
    },
    {
      "epoch": 1.72,
      "learning_rate": 7.49625e-05,
      "loss": 2.9297,
      "step": 2000
    },
    {
      "epoch": 1.72,
      "eval_loss": 2.567805528640747,
      "eval_runtime": 317.0688,
      "eval_samples_per_second": 32.435,
      "eval_steps_per_second": 4.056,
      "eval_wer": 1.005729388064391,
      "step": 2000
    },
    {
      "epoch": 1.8,
      "learning_rate": 7.423136645962732e-05,
      "loss": 2.8167,
      "step": 2100
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.345496894409937e-05,
      "loss": 2.6315,
      "step": 2200
    },
    {
      "epoch": 1.97,
      "learning_rate": 7.26863354037267e-05,
      "loss": 2.475,
      "step": 2300
    },
    {
      "epoch": 2.06,
      "learning_rate": 7.190993788819875e-05,
      "loss": 2.3571,
      "step": 2400
    },
    {
      "epoch": 2.14,
      "learning_rate": 7.11335403726708e-05,
      "loss": 2.2593,
      "step": 2500
    },
    {
      "epoch": 2.14,
      "eval_loss": 1.1067590713500977,
      "eval_runtime": 319.3729,
      "eval_samples_per_second": 32.201,
      "eval_steps_per_second": 4.027,
      "eval_wer": 0.821834513094424,
      "step": 2500
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.035714285714285e-05,
      "loss": 2.1855,
      "step": 2600
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.95807453416149e-05,
      "loss": 2.1429,
      "step": 2700
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.880434782608696e-05,
      "loss": 2.1143,
      "step": 2800
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.802795031055901e-05,
      "loss": 2.081,
      "step": 2900
    },
    {
      "epoch": 2.57,
      "learning_rate": 6.725155279503105e-05,
      "loss": 2.0504,
      "step": 3000
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.7877965569496155,
      "eval_runtime": 316.6537,
      "eval_samples_per_second": 32.477,
      "eval_steps_per_second": 4.061,
      "eval_wer": 0.7113866967305524,
      "step": 3000
    },
    {
      "epoch": 2.66,
      "learning_rate": 6.64751552795031e-05,
      "loss": 2.014,
      "step": 3100
    },
    {
      "epoch": 2.74,
      "learning_rate": 6.569875776397515e-05,
      "loss": 1.9885,
      "step": 3200
    },
    {
      "epoch": 2.83,
      "learning_rate": 6.49223602484472e-05,
      "loss": 1.9532,
      "step": 3300
    },
    {
      "epoch": 2.92,
      "learning_rate": 6.414596273291925e-05,
      "loss": 1.936,
      "step": 3400
    },
    {
      "epoch": 3.0,
      "learning_rate": 6.33695652173913e-05,
      "loss": 1.937,
      "step": 3500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6955012679100037,
      "eval_runtime": 319.8543,
      "eval_samples_per_second": 32.152,
      "eval_steps_per_second": 4.021,
      "eval_wer": 0.6449812408745633,
      "step": 3500
    },
    {
      "epoch": 3.09,
      "learning_rate": 6.259316770186334e-05,
      "loss": 1.9074,
      "step": 3600
    },
    {
      "epoch": 3.17,
      "learning_rate": 6.18167701863354e-05,
      "loss": 1.8975,
      "step": 3700
    },
    {
      "epoch": 3.26,
      "learning_rate": 6.104037267080745e-05,
      "loss": 1.8628,
      "step": 3800
    },
    {
      "epoch": 3.34,
      "learning_rate": 6.02639751552795e-05,
      "loss": 1.8605,
      "step": 3900
    },
    {
      "epoch": 3.43,
      "learning_rate": 5.948757763975155e-05,
      "loss": 1.8491,
      "step": 4000
    },
    {
      "epoch": 3.43,
      "eval_loss": 0.645221471786499,
      "eval_runtime": 318.4935,
      "eval_samples_per_second": 32.29,
      "eval_steps_per_second": 4.038,
      "eval_wer": 0.6303620603618756,
      "step": 4000
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.87111801242236e-05,
      "loss": 1.85,
      "step": 4100
    },
    {
      "epoch": 3.6,
      "learning_rate": 5.7934782608695654e-05,
      "loss": 1.8298,
      "step": 4200
    },
    {
      "epoch": 3.69,
      "learning_rate": 5.715838509316769e-05,
      "loss": 1.8108,
      "step": 4300
    },
    {
      "epoch": 3.77,
      "learning_rate": 5.6381987577639744e-05,
      "loss": 1.8136,
      "step": 4400
    },
    {
      "epoch": 3.86,
      "learning_rate": 5.5605590062111795e-05,
      "loss": 1.803,
      "step": 4500
    },
    {
      "epoch": 3.86,
      "eval_loss": 0.5961059927940369,
      "eval_runtime": 322.5472,
      "eval_samples_per_second": 31.884,
      "eval_steps_per_second": 3.987,
      "eval_wer": 0.6041547304415326,
      "step": 4500
    },
    {
      "epoch": 3.94,
      "learning_rate": 5.482919254658385e-05,
      "loss": 1.7829,
      "step": 4600
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.40527950310559e-05,
      "loss": 1.7755,
      "step": 4700
    },
    {
      "epoch": 4.12,
      "learning_rate": 5.327639751552795e-05,
      "loss": 1.7805,
      "step": 4800
    },
    {
      "epoch": 4.2,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 1.7647,
      "step": 4900
    },
    {
      "epoch": 4.29,
      "learning_rate": 5.173136645962733e-05,
      "loss": 1.7545,
      "step": 5000
    },
    {
      "epoch": 4.29,
      "eval_loss": 0.5550380945205688,
      "eval_runtime": 317.6329,
      "eval_samples_per_second": 32.377,
      "eval_steps_per_second": 4.049,
      "eval_wer": 0.5747500323433197,
      "step": 5000
    },
    {
      "epoch": 4.37,
      "learning_rate": 5.095496894409938e-05,
      "loss": 1.735,
      "step": 5100
    },
    {
      "epoch": 4.46,
      "learning_rate": 5.017857142857142e-05,
      "loss": 1.7365,
      "step": 5200
    },
    {
      "epoch": 4.55,
      "learning_rate": 4.940217391304347e-05,
      "loss": 1.734,
      "step": 5300
    },
    {
      "epoch": 4.63,
      "learning_rate": 4.862577639751552e-05,
      "loss": 1.7254,
      "step": 5400
    },
    {
      "epoch": 4.72,
      "learning_rate": 4.7849378881987574e-05,
      "loss": 1.7045,
      "step": 5500
    },
    {
      "epoch": 4.72,
      "eval_loss": 0.5373523831367493,
      "eval_runtime": 318.8226,
      "eval_samples_per_second": 32.256,
      "eval_steps_per_second": 4.034,
      "eval_wer": 0.5743064668157539,
      "step": 5500
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.7072981366459626e-05,
      "loss": 1.7031,
      "step": 5600
    },
    {
      "epoch": 4.89,
      "learning_rate": 4.629658385093168e-05,
      "loss": 1.7135,
      "step": 5700
    },
    {
      "epoch": 4.97,
      "learning_rate": 4.5520186335403715e-05,
      "loss": 1.7053,
      "step": 5800
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.474378881987577e-05,
      "loss": 1.7078,
      "step": 5900
    },
    {
      "epoch": 5.15,
      "learning_rate": 4.396739130434782e-05,
      "loss": 1.6733,
      "step": 6000
    },
    {
      "epoch": 5.15,
      "eval_loss": 0.5336768627166748,
      "eval_runtime": 321.7447,
      "eval_samples_per_second": 31.963,
      "eval_steps_per_second": 3.997,
      "eval_wer": 0.5403737039569741,
      "step": 6000
    },
    {
      "epoch": 5.23,
      "learning_rate": 4.319099378881987e-05,
      "loss": 1.6737,
      "step": 6100
    },
    {
      "epoch": 5.32,
      "learning_rate": 4.241459627329192e-05,
      "loss": 1.6939,
      "step": 6200
    },
    {
      "epoch": 5.4,
      "learning_rate": 4.1638198757763974e-05,
      "loss": 1.674,
      "step": 6300
    },
    {
      "epoch": 5.49,
      "learning_rate": 4.0861801242236026e-05,
      "loss": 1.6685,
      "step": 6400
    },
    {
      "epoch": 5.57,
      "learning_rate": 4.008540372670807e-05,
      "loss": 1.6761,
      "step": 6500
    },
    {
      "epoch": 5.57,
      "eval_loss": 0.5054484009742737,
      "eval_runtime": 316.5106,
      "eval_samples_per_second": 32.492,
      "eval_steps_per_second": 4.063,
      "eval_wer": 0.5265862088084721,
      "step": 6500
    },
    {
      "epoch": 5.66,
      "learning_rate": 3.930900621118012e-05,
      "loss": 1.6729,
      "step": 6600
    },
    {
      "epoch": 5.75,
      "learning_rate": 3.853260869565217e-05,
      "loss": 1.6521,
      "step": 6700
    },
    {
      "epoch": 5.83,
      "learning_rate": 3.775621118012422e-05,
      "loss": 1.6876,
      "step": 6800
    },
    {
      "epoch": 5.92,
      "learning_rate": 3.697981366459627e-05,
      "loss": 1.6499,
      "step": 6900
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.620341614906832e-05,
      "loss": 1.655,
      "step": 7000
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.492550253868103,
      "eval_runtime": 321.8948,
      "eval_samples_per_second": 31.948,
      "eval_steps_per_second": 3.995,
      "eval_wer": 0.5243314173766795,
      "step": 7000
    },
    {
      "epoch": 6.09,
      "learning_rate": 3.543478260869565e-05,
      "loss": 1.6501,
      "step": 7100
    },
    {
      "epoch": 6.17,
      "learning_rate": 3.4658385093167694e-05,
      "loss": 1.653,
      "step": 7200
    },
    {
      "epoch": 6.26,
      "learning_rate": 3.3881987577639746e-05,
      "loss": 1.6381,
      "step": 7300
    },
    {
      "epoch": 6.35,
      "learning_rate": 3.31055900621118e-05,
      "loss": 1.6231,
      "step": 7400
    },
    {
      "epoch": 6.43,
      "learning_rate": 3.232919254658385e-05,
      "loss": 1.6252,
      "step": 7500
    },
    {
      "epoch": 6.43,
      "eval_loss": 0.49458181858062744,
      "eval_runtime": 317.6079,
      "eval_samples_per_second": 32.38,
      "eval_steps_per_second": 4.049,
      "eval_wer": 0.5182508732696324,
      "step": 7500
    },
    {
      "epoch": 6.52,
      "learning_rate": 3.1552795031055894e-05,
      "loss": 1.6409,
      "step": 7600
    },
    {
      "epoch": 6.6,
      "learning_rate": 3.0776397515527946e-05,
      "loss": 1.6256,
      "step": 7700
    },
    {
      "epoch": 6.69,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 1.6113,
      "step": 7800
    },
    {
      "epoch": 6.78,
      "learning_rate": 2.9223602484472046e-05,
      "loss": 1.6096,
      "step": 7900
    },
    {
      "epoch": 6.86,
      "learning_rate": 2.8447204968944097e-05,
      "loss": 1.6209,
      "step": 8000
    },
    {
      "epoch": 6.86,
      "eval_loss": 0.491542249917984,
      "eval_runtime": 319.2369,
      "eval_samples_per_second": 32.214,
      "eval_steps_per_second": 4.028,
      "eval_wer": 0.5193597870885468,
      "step": 8000
    },
    {
      "epoch": 6.95,
      "learning_rate": 2.767080745341615e-05,
      "loss": 1.5917,
      "step": 8100
    },
    {
      "epoch": 7.03,
      "learning_rate": 2.6894409937888194e-05,
      "loss": 1.6049,
      "step": 8200
    },
    {
      "epoch": 7.12,
      "learning_rate": 2.6118012422360246e-05,
      "loss": 1.5891,
      "step": 8300
    },
    {
      "epoch": 7.2,
      "learning_rate": 2.5341614906832297e-05,
      "loss": 1.5915,
      "step": 8400
    },
    {
      "epoch": 7.29,
      "learning_rate": 2.4565217391304346e-05,
      "loss": 1.5772,
      "step": 8500
    },
    {
      "epoch": 7.29,
      "eval_loss": 0.47245877981185913,
      "eval_runtime": 314.2017,
      "eval_samples_per_second": 32.731,
      "eval_steps_per_second": 4.093,
      "eval_wer": 0.5103775851553404,
      "step": 8500
    },
    {
      "epoch": 7.38,
      "learning_rate": 2.3788819875776394e-05,
      "loss": 1.5932,
      "step": 8600
    },
    {
      "epoch": 7.46,
      "learning_rate": 2.3012422360248446e-05,
      "loss": 1.574,
      "step": 8700
    },
    {
      "epoch": 7.55,
      "learning_rate": 2.2236024844720494e-05,
      "loss": 1.5818,
      "step": 8800
    },
    {
      "epoch": 7.63,
      "learning_rate": 2.1459627329192546e-05,
      "loss": 1.5852,
      "step": 8900
    },
    {
      "epoch": 7.72,
      "learning_rate": 2.0683229813664594e-05,
      "loss": 1.5602,
      "step": 9000
    },
    {
      "epoch": 7.72,
      "eval_loss": 0.47263726592063904,
      "eval_runtime": 323.393,
      "eval_samples_per_second": 31.8,
      "eval_steps_per_second": 3.977,
      "eval_wer": 0.509656791173046,
      "step": 9000
    },
    {
      "epoch": 7.8,
      "learning_rate": 1.9906832298136646e-05,
      "loss": 1.5737,
      "step": 9100
    },
    {
      "epoch": 7.89,
      "learning_rate": 1.9130434782608694e-05,
      "loss": 1.5894,
      "step": 9200
    },
    {
      "epoch": 7.98,
      "learning_rate": 1.8361801242236024e-05,
      "loss": 1.5694,
      "step": 9300
    },
    {
      "epoch": 8.06,
      "learning_rate": 1.7585403726708073e-05,
      "loss": 1.537,
      "step": 9400
    },
    {
      "epoch": 8.15,
      "learning_rate": 1.680900621118012e-05,
      "loss": 1.5783,
      "step": 9500
    },
    {
      "epoch": 8.15,
      "eval_loss": 0.4666772186756134,
      "eval_runtime": 318.3833,
      "eval_samples_per_second": 32.301,
      "eval_steps_per_second": 4.039,
      "eval_wer": 0.49557358567283344,
      "step": 9500
    },
    {
      "epoch": 8.23,
      "learning_rate": 1.604037267080745e-05,
      "loss": 1.5432,
      "step": 9600
    },
    {
      "epoch": 8.32,
      "learning_rate": 1.5263975155279503e-05,
      "loss": 1.5439,
      "step": 9700
    },
    {
      "epoch": 8.4,
      "learning_rate": 1.4487577639751551e-05,
      "loss": 1.5589,
      "step": 9800
    },
    {
      "epoch": 8.49,
      "learning_rate": 1.3711180124223601e-05,
      "loss": 1.5573,
      "step": 9900
    },
    {
      "epoch": 8.58,
      "learning_rate": 1.2934782608695651e-05,
      "loss": 1.5442,
      "step": 10000
    },
    {
      "epoch": 8.58,
      "eval_loss": 0.46852901577949524,
      "eval_runtime": 317.9433,
      "eval_samples_per_second": 32.345,
      "eval_steps_per_second": 4.045,
      "eval_wer": 0.49374387787162477,
      "step": 10000
    },
    {
      "epoch": 8.66,
      "learning_rate": 1.2158385093167701e-05,
      "loss": 1.5679,
      "step": 10100
    },
    {
      "epoch": 8.75,
      "learning_rate": 1.1381987577639751e-05,
      "loss": 1.5604,
      "step": 10200
    },
    {
      "epoch": 8.83,
      "learning_rate": 1.06055900621118e-05,
      "loss": 1.5458,
      "step": 10300
    },
    {
      "epoch": 8.92,
      "learning_rate": 9.829192546583851e-06,
      "loss": 1.5501,
      "step": 10400
    },
    {
      "epoch": 9.01,
      "learning_rate": 9.0527950310559e-06,
      "loss": 1.5597,
      "step": 10500
    },
    {
      "epoch": 9.01,
      "eval_loss": 0.4707895815372467,
      "eval_runtime": 323.6683,
      "eval_samples_per_second": 31.773,
      "eval_steps_per_second": 3.973,
      "eval_wer": 0.495665995157743,
      "step": 10500
    },
    {
      "epoch": 9.09,
      "learning_rate": 8.27639751552795e-06,
      "loss": 1.5189,
      "step": 10600
    },
    {
      "epoch": 9.18,
      "learning_rate": 7.499999999999999e-06,
      "loss": 1.5272,
      "step": 10700
    },
    {
      "epoch": 9.26,
      "learning_rate": 6.7236024844720485e-06,
      "loss": 1.5319,
      "step": 10800
    },
    {
      "epoch": 9.35,
      "learning_rate": 5.9472049689440985e-06,
      "loss": 1.5175,
      "step": 10900
    },
    {
      "epoch": 9.43,
      "learning_rate": 5.1708074534161485e-06,
      "loss": 1.5406,
      "step": 11000
    },
    {
      "epoch": 9.43,
      "eval_loss": 0.45390617847442627,
      "eval_runtime": 320.1014,
      "eval_samples_per_second": 32.127,
      "eval_steps_per_second": 4.017,
      "eval_wer": 0.48104681464505517,
      "step": 11000
    },
    {
      "epoch": 9.52,
      "learning_rate": 4.3944099378881985e-06,
      "loss": 1.5326,
      "step": 11100
    },
    {
      "epoch": 9.61,
      "learning_rate": 3.6180124223602485e-06,
      "loss": 1.5142,
      "step": 11200
    },
    {
      "epoch": 9.69,
      "learning_rate": 2.8416149068322976e-06,
      "loss": 1.5195,
      "step": 11300
    },
    {
      "epoch": 9.78,
      "learning_rate": 2.0652173913043476e-06,
      "loss": 1.5147,
      "step": 11400
    },
    {
      "epoch": 9.86,
      "learning_rate": 1.2888198757763974e-06,
      "loss": 1.5274,
      "step": 11500
    },
    {
      "epoch": 9.86,
      "eval_loss": 0.45016008615493774,
      "eval_runtime": 317.8491,
      "eval_samples_per_second": 32.355,
      "eval_steps_per_second": 4.046,
      "eval_wer": 0.4782560482007873,
      "step": 11500
    },
    {
      "epoch": 9.86,
      "step": 11500,
      "total_flos": 3.154318018894781e+19,
      "train_loss": 0.0,
      "train_runtime": 3.4176,
      "train_samples_per_second": 54596.703,
      "train_steps_per_second": 1705.873
    }
  ],
  "max_steps": 5830,
  "num_train_epochs": 5,
  "total_flos": 3.154318018894781e+19,
  "trial_name": null,
  "trial_params": null
}