{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 500.0,
  "global_step": 11000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.55,
      "learning_rate": 3.675e-06,
      "loss": 15.8714,
      "step": 100
    },
    {
      "epoch": 9.09,
      "learning_rate": 7.425e-06,
      "loss": 8.905,
      "step": 200
    },
    {
      "epoch": 13.64,
      "learning_rate": 1.1174999999999999e-05,
      "loss": 5.8648,
      "step": 300
    },
    {
      "epoch": 18.18,
      "learning_rate": 1.4925e-05,
      "loss": 4.9504,
      "step": 400
    },
    {
      "epoch": 18.18,
      "eval_loss": 4.673031330108643,
      "eval_runtime": 13.8453,
      "eval_samples_per_second": 22.101,
      "eval_steps_per_second": 1.445,
      "eval_wer": 1.0,
      "step": 400
    },
    {
      "epoch": 22.73,
      "learning_rate": 1.8675e-05,
      "loss": 4.2429,
      "step": 500
    },
    {
      "epoch": 27.27,
      "learning_rate": 2.2424999999999996e-05,
      "loss": 3.72,
      "step": 600
    },
    {
      "epoch": 31.82,
      "learning_rate": 2.6174999999999996e-05,
      "loss": 3.4596,
      "step": 700
    },
    {
      "epoch": 36.36,
      "learning_rate": 2.9925e-05,
      "loss": 3.3766,
      "step": 800
    },
    {
      "epoch": 36.36,
      "eval_loss": 3.3464324474334717,
      "eval_runtime": 13.5196,
      "eval_samples_per_second": 22.634,
      "eval_steps_per_second": 1.479,
      "eval_wer": 1.0,
      "step": 800
    },
    {
      "epoch": 40.91,
      "learning_rate": 3.3675e-05,
      "loss": 3.3035,
      "step": 900
    },
    {
      "epoch": 45.45,
      "learning_rate": 3.7424999999999995e-05,
      "loss": 3.2559,
      "step": 1000
    },
    {
      "epoch": 50.0,
      "learning_rate": 4.1175e-05,
      "loss": 3.2202,
      "step": 1100
    },
    {
      "epoch": 54.55,
      "learning_rate": 4.4924999999999994e-05,
      "loss": 3.1128,
      "step": 1200
    },
    {
      "epoch": 54.55,
      "eval_loss": 3.0176751613616943,
      "eval_runtime": 13.6849,
      "eval_samples_per_second": 22.36,
      "eval_steps_per_second": 1.461,
      "eval_wer": 0.9979943842759728,
      "step": 1200
    },
    {
      "epoch": 59.09,
      "learning_rate": 4.8675e-05,
      "loss": 2.9121,
      "step": 1300
    },
    {
      "epoch": 63.64,
      "learning_rate": 5.2424999999999994e-05,
      "loss": 2.4741,
      "step": 1400
    },
    {
      "epoch": 68.18,
      "learning_rate": 5.6175e-05,
      "loss": 2.0461,
      "step": 1500
    },
    {
      "epoch": 72.73,
      "learning_rate": 5.9925e-05,
      "loss": 1.7966,
      "step": 1600
    },
    {
      "epoch": 72.73,
      "eval_loss": 0.8733049631118774,
      "eval_runtime": 13.6227,
      "eval_samples_per_second": 22.463,
      "eval_steps_per_second": 1.468,
      "eval_wer": 0.8038507821901324,
      "step": 1600
    },
    {
      "epoch": 77.27,
      "learning_rate": 6.367499999999999e-05,
      "loss": 1.6388,
      "step": 1700
    },
    {
      "epoch": 81.82,
      "learning_rate": 6.7425e-05,
      "loss": 1.5466,
      "step": 1800
    },
    {
      "epoch": 86.36,
      "learning_rate": 7.1175e-05,
      "loss": 1.461,
      "step": 1900
    },
    {
      "epoch": 90.91,
      "learning_rate": 7.48875e-05,
      "loss": 1.4085,
      "step": 2000
    },
    {
      "epoch": 90.91,
      "eval_loss": 0.5555232763290405,
      "eval_runtime": 13.4681,
      "eval_samples_per_second": 22.72,
      "eval_steps_per_second": 1.485,
      "eval_wer": 0.645808263136783,
      "step": 2000
    },
    {
      "epoch": 95.45,
      "learning_rate": 7.419166666666666e-05,
      "loss": 1.3233,
      "step": 2100
    },
    {
      "epoch": 100.0,
      "learning_rate": 7.335833333333333e-05,
      "loss": 1.2821,
      "step": 2200
    },
    {
      "epoch": 104.55,
      "learning_rate": 7.252499999999999e-05,
      "loss": 1.2194,
      "step": 2300
    },
    {
      "epoch": 109.09,
      "learning_rate": 7.169166666666666e-05,
      "loss": 1.1731,
      "step": 2400
    },
    {
      "epoch": 109.09,
      "eval_loss": 0.4929651618003845,
      "eval_runtime": 13.4716,
      "eval_samples_per_second": 22.715,
      "eval_steps_per_second": 1.485,
      "eval_wer": 0.6438026474127557,
      "step": 2400
    },
    {
      "epoch": 113.64,
      "learning_rate": 7.085833333333333e-05,
      "loss": 1.135,
      "step": 2500
    },
    {
      "epoch": 118.18,
      "learning_rate": 7.0025e-05,
      "loss": 1.1009,
      "step": 2600
    },
    {
      "epoch": 122.73,
      "learning_rate": 6.919166666666666e-05,
      "loss": 1.0712,
      "step": 2700
    },
    {
      "epoch": 127.27,
      "learning_rate": 6.835833333333332e-05,
      "loss": 1.0271,
      "step": 2800
    },
    {
      "epoch": 127.27,
      "eval_loss": 0.47801950573921204,
      "eval_runtime": 13.4955,
      "eval_samples_per_second": 22.674,
      "eval_steps_per_second": 1.482,
      "eval_wer": 0.6093060569594866,
      "step": 2800
    },
    {
      "epoch": 131.82,
      "learning_rate": 6.753333333333333e-05,
      "loss": 1.0016,
      "step": 2900
    },
    {
      "epoch": 136.36,
      "learning_rate": 6.67e-05,
      "loss": 0.9638,
      "step": 3000
    },
    {
      "epoch": 140.91,
      "learning_rate": 6.586666666666666e-05,
      "loss": 0.9327,
      "step": 3100
    },
    {
      "epoch": 145.45,
      "learning_rate": 6.503333333333333e-05,
      "loss": 0.9045,
      "step": 3200
    },
    {
      "epoch": 145.45,
      "eval_loss": 0.46472442150115967,
      "eval_runtime": 13.3444,
      "eval_samples_per_second": 22.931,
      "eval_steps_per_second": 1.499,
      "eval_wer": 0.6578419574809466,
      "step": 3200
    },
    {
      "epoch": 150.0,
      "learning_rate": 6.419999999999999e-05,
      "loss": 0.8865,
      "step": 3300
    },
    {
      "epoch": 154.55,
      "learning_rate": 6.336666666666666e-05,
      "loss": 0.8703,
      "step": 3400
    },
    {
      "epoch": 159.09,
      "learning_rate": 6.253333333333333e-05,
      "loss": 0.8379,
      "step": 3500
    },
    {
      "epoch": 163.64,
      "learning_rate": 6.17e-05,
      "loss": 0.807,
      "step": 3600
    },
    {
      "epoch": 163.64,
      "eval_loss": 0.4505322575569153,
      "eval_runtime": 13.3663,
      "eval_samples_per_second": 22.893,
      "eval_steps_per_second": 1.496,
      "eval_wer": 0.5924588848776574,
      "step": 3600
    },
    {
      "epoch": 168.18,
      "learning_rate": 6.0866666666666664e-05,
      "loss": 0.805,
      "step": 3700
    },
    {
      "epoch": 172.73,
      "learning_rate": 6.003333333333333e-05,
      "loss": 0.7889,
      "step": 3800
    },
    {
      "epoch": 177.27,
      "learning_rate": 5.9199999999999996e-05,
      "loss": 0.764,
      "step": 3900
    },
    {
      "epoch": 181.82,
      "learning_rate": 5.8366666666666665e-05,
      "loss": 0.741,
      "step": 4000
    },
    {
      "epoch": 181.82,
      "eval_loss": 0.4745561182498932,
      "eval_runtime": 13.2681,
      "eval_samples_per_second": 23.063,
      "eval_steps_per_second": 1.507,
      "eval_wer": 0.6024869634977938,
      "step": 4000
    },
    {
      "epoch": 186.36,
      "learning_rate": 5.753333333333333e-05,
      "loss": 0.7192,
      "step": 4100
    },
    {
      "epoch": 190.91,
      "learning_rate": 5.6699999999999996e-05,
      "loss": 0.6977,
      "step": 4200
    },
    {
      "epoch": 195.45,
      "learning_rate": 5.5866666666666665e-05,
      "loss": 0.6882,
      "step": 4300
    },
    {
      "epoch": 200.0,
      "learning_rate": 5.503333333333333e-05,
      "loss": 0.6706,
      "step": 4400
    },
    {
      "epoch": 200.0,
      "eval_loss": 0.5003653168678284,
      "eval_runtime": 13.3324,
      "eval_samples_per_second": 22.952,
      "eval_steps_per_second": 1.5,
      "eval_wer": 0.5844364219815483,
      "step": 4400
    },
    {
      "epoch": 204.55,
      "learning_rate": 5.4199999999999996e-05,
      "loss": 0.6627,
      "step": 4500
    },
    {
      "epoch": 209.09,
      "learning_rate": 5.3366666666666665e-05,
      "loss": 0.6453,
      "step": 4600
    },
    {
      "epoch": 213.64,
      "learning_rate": 5.253333333333333e-05,
      "loss": 0.6311,
      "step": 4700
    },
    {
      "epoch": 218.18,
      "learning_rate": 5.1699999999999996e-05,
      "loss": 0.6186,
      "step": 4800
    },
    {
      "epoch": 218.18,
      "eval_loss": 0.4983522891998291,
      "eval_runtime": 13.4916,
      "eval_samples_per_second": 22.681,
      "eval_steps_per_second": 1.482,
      "eval_wer": 0.5996791014841556,
      "step": 4800
    },
    {
      "epoch": 222.73,
      "learning_rate": 5.086666666666666e-05,
      "loss": 0.5976,
      "step": 4900
    },
    {
      "epoch": 227.27,
      "learning_rate": 5.003333333333333e-05,
      "loss": 0.5832,
      "step": 5000
    },
    {
      "epoch": 231.82,
      "learning_rate": 4.9199999999999997e-05,
      "loss": 0.5663,
      "step": 5100
    },
    {
      "epoch": 236.36,
      "learning_rate": 4.836666666666666e-05,
      "loss": 0.5508,
      "step": 5200
    },
    {
      "epoch": 236.36,
      "eval_loss": 0.5298363566398621,
      "eval_runtime": 13.4008,
      "eval_samples_per_second": 22.834,
      "eval_steps_per_second": 1.492,
      "eval_wer": 0.5635780184516647,
      "step": 5200
    },
    {
      "epoch": 240.91,
      "learning_rate": 4.753333333333333e-05,
      "loss": 0.554,
      "step": 5300
    },
    {
      "epoch": 245.45,
      "learning_rate": 4.67e-05,
      "loss": 0.543,
      "step": 5400
    },
    {
      "epoch": 250.0,
      "learning_rate": 4.586666666666666e-05,
      "loss": 0.5221,
      "step": 5500
    },
    {
      "epoch": 254.55,
      "learning_rate": 4.503333333333333e-05,
      "loss": 0.5123,
      "step": 5600
    },
    {
      "epoch": 254.55,
      "eval_loss": 0.5410219430923462,
      "eval_runtime": 13.3891,
      "eval_samples_per_second": 22.854,
      "eval_steps_per_second": 1.494,
      "eval_wer": 0.51103088648215,
      "step": 5600
    },
    {
      "epoch": 259.09,
      "learning_rate": 4.42e-05,
      "loss": 0.5023,
      "step": 5700
    },
    {
      "epoch": 263.64,
      "learning_rate": 4.336666666666666e-05,
      "loss": 0.4854,
      "step": 5800
    },
    {
      "epoch": 268.18,
      "learning_rate": 4.253333333333333e-05,
      "loss": 0.4795,
      "step": 5900
    },
    {
      "epoch": 272.73,
      "learning_rate": 4.17e-05,
      "loss": 0.4623,
      "step": 6000
    },
    {
      "epoch": 272.73,
      "eval_loss": 0.5591160655021667,
      "eval_runtime": 13.3234,
      "eval_samples_per_second": 22.967,
      "eval_steps_per_second": 1.501,
      "eval_wer": 0.538307260328921,
      "step": 6000
    },
    {
      "epoch": 277.27,
      "learning_rate": 4.086666666666666e-05,
      "loss": 0.4593,
      "step": 6100
    },
    {
      "epoch": 281.82,
      "learning_rate": 4.003333333333333e-05,
      "loss": 0.4435,
      "step": 6200
    },
    {
      "epoch": 286.36,
      "learning_rate": 3.919999999999999e-05,
      "loss": 0.4432,
      "step": 6300
    },
    {
      "epoch": 290.91,
      "learning_rate": 3.836666666666666e-05,
      "loss": 0.4281,
      "step": 6400
    },
    {
      "epoch": 290.91,
      "eval_loss": 0.5775493383407593,
      "eval_runtime": 13.3759,
      "eval_samples_per_second": 22.877,
      "eval_steps_per_second": 1.495,
      "eval_wer": 0.5599679101484155,
      "step": 6400
    },
    {
      "epoch": 295.45,
      "learning_rate": 3.753333333333333e-05,
      "loss": 0.4264,
      "step": 6500
    },
    {
      "epoch": 300.0,
      "learning_rate": 3.67e-05,
      "loss": 0.4187,
      "step": 6600
    },
    {
      "epoch": 304.55,
      "learning_rate": 3.586666666666666e-05,
      "loss": 0.3965,
      "step": 6700
    },
    {
      "epoch": 309.09,
      "learning_rate": 3.503333333333333e-05,
      "loss": 0.4045,
      "step": 6800
    },
    {
      "epoch": 309.09,
      "eval_loss": 0.5923808217048645,
      "eval_runtime": 13.6647,
      "eval_samples_per_second": 22.394,
      "eval_steps_per_second": 1.464,
      "eval_wer": 0.5579622944243883,
      "step": 6800
    },
    {
      "epoch": 313.64,
      "learning_rate": 3.42e-05,
      "loss": 0.3857,
      "step": 6900
    },
    {
      "epoch": 318.18,
      "learning_rate": 3.336666666666667e-05,
      "loss": 0.373,
      "step": 7000
    },
    {
      "epoch": 322.73,
      "learning_rate": 3.253333333333333e-05,
      "loss": 0.3752,
      "step": 7100
    },
    {
      "epoch": 327.27,
      "learning_rate": 3.17e-05,
      "loss": 0.3651,
      "step": 7200
    },
    {
      "epoch": 327.27,
      "eval_loss": 0.567070484161377,
      "eval_runtime": 13.4998,
      "eval_samples_per_second": 22.667,
      "eval_steps_per_second": 1.482,
      "eval_wer": 0.5683914961893302,
      "step": 7200
    },
    {
      "epoch": 331.82,
      "learning_rate": 3.086666666666667e-05,
      "loss": 0.3567,
      "step": 7300
    },
    {
      "epoch": 336.36,
      "learning_rate": 3.003333333333333e-05,
      "loss": 0.3551,
      "step": 7400
    },
    {
      "epoch": 340.91,
      "learning_rate": 2.9199999999999995e-05,
      "loss": 0.351,
      "step": 7500
    },
    {
      "epoch": 345.45,
      "learning_rate": 2.8366666666666664e-05,
      "loss": 0.343,
      "step": 7600
    },
    {
      "epoch": 345.45,
      "eval_loss": 0.6082874536514282,
      "eval_runtime": 13.5565,
      "eval_samples_per_second": 22.572,
      "eval_steps_per_second": 1.475,
      "eval_wer": 0.5944645006016848,
      "step": 7600
    },
    {
      "epoch": 350.0,
      "learning_rate": 2.7541666666666664e-05,
      "loss": 0.3379,
      "step": 7700
    },
    {
      "epoch": 354.55,
      "learning_rate": 2.6716666666666664e-05,
      "loss": 0.3257,
      "step": 7800
    },
    {
      "epoch": 359.09,
      "learning_rate": 2.588333333333333e-05,
      "loss": 0.3285,
      "step": 7900
    },
    {
      "epoch": 363.64,
      "learning_rate": 2.505e-05,
      "loss": 0.3085,
      "step": 8000
    },
    {
      "epoch": 363.64,
      "eval_loss": 0.6242865324020386,
      "eval_runtime": 13.4343,
      "eval_samples_per_second": 22.778,
      "eval_steps_per_second": 1.489,
      "eval_wer": 0.5728038507821901,
      "step": 8000
    },
    {
      "epoch": 368.18,
      "learning_rate": 2.4216666666666665e-05,
      "loss": 0.3127,
      "step": 8100
    },
    {
      "epoch": 372.73,
      "learning_rate": 2.338333333333333e-05,
      "loss": 0.3149,
      "step": 8200
    },
    {
      "epoch": 377.27,
      "learning_rate": 2.255e-05,
      "loss": 0.307,
      "step": 8300
    },
    {
      "epoch": 381.82,
      "learning_rate": 2.1716666666666665e-05,
      "loss": 0.2941,
      "step": 8400
    },
    {
      "epoch": 381.82,
      "eval_loss": 0.6244971752166748,
      "eval_runtime": 13.5184,
      "eval_samples_per_second": 22.636,
      "eval_steps_per_second": 1.479,
      "eval_wer": 0.5579622944243883,
      "step": 8400
    },
    {
      "epoch": 386.36,
      "learning_rate": 2.088333333333333e-05,
      "loss": 0.295,
      "step": 8500
    },
    {
      "epoch": 390.91,
      "learning_rate": 2.0049999999999996e-05,
      "loss": 0.2842,
      "step": 8600
    },
    {
      "epoch": 395.45,
      "learning_rate": 1.9216666666666665e-05,
      "loss": 0.2817,
      "step": 8700
    },
    {
      "epoch": 400.0,
      "learning_rate": 1.838333333333333e-05,
      "loss": 0.2735,
      "step": 8800
    },
    {
      "epoch": 400.0,
      "eval_loss": 0.6457676291465759,
      "eval_runtime": 13.5206,
      "eval_samples_per_second": 22.632,
      "eval_steps_per_second": 1.479,
      "eval_wer": 0.5804251905334937,
      "step": 8800
    },
    {
      "epoch": 404.55,
      "learning_rate": 1.755e-05,
      "loss": 0.2786,
      "step": 8900
    },
    {
      "epoch": 409.09,
      "learning_rate": 1.6716666666666665e-05,
      "loss": 0.2708,
      "step": 9000
    },
    {
      "epoch": 413.64,
      "learning_rate": 1.5883333333333334e-05,
      "loss": 0.2661,
      "step": 9100
    },
    {
      "epoch": 418.18,
      "learning_rate": 1.5049999999999998e-05,
      "loss": 0.262,
      "step": 9200
    },
    {
      "epoch": 418.18,
      "eval_loss": 0.656562328338623,
      "eval_runtime": 13.3117,
      "eval_samples_per_second": 22.987,
      "eval_steps_per_second": 1.502,
      "eval_wer": 0.5824308062575211,
      "step": 9200
    },
    {
      "epoch": 422.73,
      "learning_rate": 1.4216666666666666e-05,
      "loss": 0.2577,
      "step": 9300
    },
    {
      "epoch": 427.27,
      "learning_rate": 1.3383333333333331e-05,
      "loss": 0.2576,
      "step": 9400
    },
    {
      "epoch": 431.82,
      "learning_rate": 1.2549999999999998e-05,
      "loss": 0.2573,
      "step": 9500
    },
    {
      "epoch": 436.36,
      "learning_rate": 1.1716666666666666e-05,
      "loss": 0.2578,
      "step": 9600
    },
    {
      "epoch": 436.36,
      "eval_loss": 0.6557896733283997,
      "eval_runtime": 13.599,
      "eval_samples_per_second": 22.502,
      "eval_steps_per_second": 1.471,
      "eval_wer": 0.596470116325712,
      "step": 9600
    },
    {
      "epoch": 440.91,
      "learning_rate": 1.0883333333333331e-05,
      "loss": 0.2445,
      "step": 9700
    },
    {
      "epoch": 445.45,
      "learning_rate": 1.005e-05,
      "loss": 0.2422,
      "step": 9800
    },
    {
      "epoch": 450.0,
      "learning_rate": 9.216666666666666e-06,
      "loss": 0.2481,
      "step": 9900
    },
    {
      "epoch": 454.55,
      "learning_rate": 8.383333333333333e-06,
      "loss": 0.2388,
      "step": 10000
    },
    {
      "epoch": 454.55,
      "eval_loss": 0.659763514995575,
      "eval_runtime": 13.4458,
      "eval_samples_per_second": 22.758,
      "eval_steps_per_second": 1.487,
      "eval_wer": 0.5992779783393501,
      "step": 10000
    },
    {
      "epoch": 459.09,
      "learning_rate": 7.55e-06,
      "loss": 0.2362,
      "step": 10100
    },
    {
      "epoch": 463.64,
      "learning_rate": 6.716666666666666e-06,
      "loss": 0.2374,
      "step": 10200
    },
    {
      "epoch": 468.18,
      "learning_rate": 5.883333333333333e-06,
      "loss": 0.2352,
      "step": 10300
    },
    {
      "epoch": 472.73,
      "learning_rate": 5.049999999999999e-06,
      "loss": 0.2328,
      "step": 10400
    },
    {
      "epoch": 472.73,
      "eval_loss": 0.6699528694152832,
      "eval_runtime": 13.3932,
      "eval_samples_per_second": 22.847,
      "eval_steps_per_second": 1.493,
      "eval_wer": 0.6040914560770156,
      "step": 10400
    },
    {
      "epoch": 477.27,
      "learning_rate": 4.216666666666666e-06,
      "loss": 0.2286,
      "step": 10500
    },
    {
      "epoch": 481.82,
      "learning_rate": 3.3916666666666662e-06,
      "loss": 0.232,
      "step": 10600
    },
    {
      "epoch": 486.36,
      "learning_rate": 2.558333333333333e-06,
      "loss": 0.2323,
      "step": 10700
    },
    {
      "epoch": 490.91,
      "learning_rate": 1.7249999999999998e-06,
      "loss": 0.2286,
      "step": 10800
    },
    {
      "epoch": 490.91,
      "eval_loss": 0.6684302091598511,
      "eval_runtime": 13.5254,
      "eval_samples_per_second": 22.624,
      "eval_steps_per_second": 1.479,
      "eval_wer": 0.5956678700361011,
      "step": 10800
    },
    {
      "epoch": 495.45,
      "learning_rate": 8.916666666666666e-07,
      "loss": 0.229,
      "step": 10900
    },
    {
      "epoch": 500.0,
      "learning_rate": 5.8333333333333326e-08,
      "loss": 0.2255,
      "step": 11000
    },
    {
      "epoch": 500.0,
      "step": 11000,
      "total_flos": 6.534167693112984e+19,
      "train_loss": 1.1389288659529253,
      "train_runtime": 23734.473,
      "train_samples_per_second": 14.704,
      "train_steps_per_second": 0.463
    }
  ],
  "max_steps": 11000,
  "num_train_epochs": 500,
  "total_flos": 6.534167693112984e+19,
  "trial_name": null,
  "trial_params": null
}