{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 29.12621359223301,
  "global_step": 15000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.97,
      "learning_rate": 5e-05,
      "loss": 8.8221,
      "step": 500
    },
    {
      "epoch": 0.97,
      "eval_loss": 3.6322524547576904,
      "eval_runtime": 46.46,
      "eval_samples_per_second": 21.739,
      "eval_steps_per_second": 2.734,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001,
      "loss": 3.4315,
      "step": 1000
    },
    {
      "epoch": 1.94,
      "eval_loss": 3.614928722381592,
      "eval_runtime": 46.1252,
      "eval_samples_per_second": 21.897,
      "eval_steps_per_second": 2.753,
      "eval_wer": 1.001953125,
      "step": 1000
    },
    {
      "epoch": 2.91,
      "learning_rate": 9.653979238754325e-05,
      "loss": 3.2825,
      "step": 1500
    },
    {
      "epoch": 2.91,
      "eval_loss": 4.486703395843506,
      "eval_runtime": 46.2406,
      "eval_samples_per_second": 21.842,
      "eval_steps_per_second": 2.747,
      "eval_wer": 1.0,
      "step": 1500
    },
    {
      "epoch": 3.88,
      "learning_rate": 9.307958477508652e-05,
      "loss": 3.0082,
      "step": 2000
    },
    {
      "epoch": 3.88,
      "eval_loss": 4.054150104522705,
      "eval_runtime": 46.7848,
      "eval_samples_per_second": 21.588,
      "eval_steps_per_second": 2.715,
      "eval_wer": 1.069921875,
      "step": 2000
    },
    {
      "epoch": 4.85,
      "learning_rate": 8.961937716262977e-05,
      "loss": 2.9323,
      "step": 2500
    },
    {
      "epoch": 4.85,
      "eval_loss": 3.1641268730163574,
      "eval_runtime": 46.5642,
      "eval_samples_per_second": 21.69,
      "eval_steps_per_second": 2.727,
      "eval_wer": 1.29140625,
      "step": 2500
    },
    {
      "epoch": 5.83,
      "learning_rate": 8.615916955017301e-05,
      "loss": 2.9131,
      "step": 3000
    },
    {
      "epoch": 5.83,
      "eval_loss": 3.072361707687378,
      "eval_runtime": 46.5467,
      "eval_samples_per_second": 21.699,
      "eval_steps_per_second": 2.728,
      "eval_wer": 1.29140625,
      "step": 3000
    },
    {
      "epoch": 6.8,
      "learning_rate": 8.269896193771626e-05,
      "loss": 2.8623,
      "step": 3500
    },
    {
      "epoch": 6.8,
      "eval_loss": 3.455599308013916,
      "eval_runtime": 46.5142,
      "eval_samples_per_second": 21.714,
      "eval_steps_per_second": 2.73,
      "eval_wer": 1.29140625,
      "step": 3500
    },
    {
      "epoch": 7.77,
      "learning_rate": 7.923875432525953e-05,
      "loss": 2.8501,
      "step": 4000
    },
    {
      "epoch": 7.77,
      "eval_loss": 3.1545495986938477,
      "eval_runtime": 46.4787,
      "eval_samples_per_second": 21.73,
      "eval_steps_per_second": 2.732,
      "eval_wer": 1.29140625,
      "step": 4000
    },
    {
      "epoch": 8.74,
      "learning_rate": 7.577854671280276e-05,
      "loss": 2.8556,
      "step": 4500
    },
    {
      "epoch": 8.74,
      "eval_loss": 3.0713276863098145,
      "eval_runtime": 46.5645,
      "eval_samples_per_second": 21.69,
      "eval_steps_per_second": 2.727,
      "eval_wer": 1.29140625,
      "step": 4500
    },
    {
      "epoch": 9.71,
      "learning_rate": 7.231833910034602e-05,
      "loss": 3.216,
      "step": 5000
    },
    {
      "epoch": 9.71,
      "eval_loss": 3.619907855987549,
      "eval_runtime": 46.8857,
      "eval_samples_per_second": 21.542,
      "eval_steps_per_second": 2.709,
      "eval_wer": 1.290234375,
      "step": 5000
    },
    {
      "epoch": 10.68,
      "learning_rate": 6.885813148788927e-05,
      "loss": 3.0364,
      "step": 5500
    },
    {
      "epoch": 10.68,
      "eval_loss": 3.2613348960876465,
      "eval_runtime": 46.689,
      "eval_samples_per_second": 21.633,
      "eval_steps_per_second": 2.72,
      "eval_wer": 1.290625,
      "step": 5500
    },
    {
      "epoch": 11.65,
      "learning_rate": 6.539792387543253e-05,
      "loss": 3.0891,
      "step": 6000
    },
    {
      "epoch": 11.65,
      "eval_loss": 8.374248504638672,
      "eval_runtime": 46.7038,
      "eval_samples_per_second": 21.626,
      "eval_steps_per_second": 2.719,
      "eval_wer": 1.19453125,
      "step": 6000
    },
    {
      "epoch": 12.62,
      "learning_rate": 6.193771626297578e-05,
      "loss": 3.2234,
      "step": 6500
    },
    {
      "epoch": 12.62,
      "eval_loss": 3.33834171295166,
      "eval_runtime": 46.7037,
      "eval_samples_per_second": 21.626,
      "eval_steps_per_second": 2.719,
      "eval_wer": 1.290234375,
      "step": 6500
    },
    {
      "epoch": 13.59,
      "learning_rate": 5.847750865051903e-05,
      "loss": 3.0869,
      "step": 7000
    },
    {
      "epoch": 13.59,
      "eval_loss": 3.750485420227051,
      "eval_runtime": 46.6392,
      "eval_samples_per_second": 21.656,
      "eval_steps_per_second": 2.723,
      "eval_wer": 1.29140625,
      "step": 7000
    },
    {
      "epoch": 14.56,
      "learning_rate": 5.501730103806229e-05,
      "loss": 3.1178,
      "step": 7500
    },
    {
      "epoch": 14.56,
      "eval_loss": 3.240990400314331,
      "eval_runtime": 46.8161,
      "eval_samples_per_second": 21.574,
      "eval_steps_per_second": 2.713,
      "eval_wer": 1.29140625,
      "step": 7500
    },
    {
      "epoch": 15.53,
      "learning_rate": 5.1557093425605544e-05,
      "loss": 3.119,
      "step": 8000
    },
    {
      "epoch": 15.53,
      "eval_loss": 3.2141737937927246,
      "eval_runtime": 46.8628,
      "eval_samples_per_second": 21.552,
      "eval_steps_per_second": 2.71,
      "eval_wer": 1.01015625,
      "step": 8000
    },
    {
      "epoch": 16.5,
      "learning_rate": 4.809688581314879e-05,
      "loss": 3.2652,
      "step": 8500
    },
    {
      "epoch": 16.5,
      "eval_loss": 3.2599737644195557,
      "eval_runtime": 46.771,
      "eval_samples_per_second": 21.595,
      "eval_steps_per_second": 2.715,
      "eval_wer": 1.29140625,
      "step": 8500
    },
    {
      "epoch": 17.48,
      "learning_rate": 4.463667820069204e-05,
      "loss": 2.8827,
      "step": 9000
    },
    {
      "epoch": 17.48,
      "eval_loss": 3.091489791870117,
      "eval_runtime": 47.0421,
      "eval_samples_per_second": 21.47,
      "eval_steps_per_second": 2.7,
      "eval_wer": 1.29140625,
      "step": 9000
    },
    {
      "epoch": 18.45,
      "learning_rate": 4.11764705882353e-05,
      "loss": 2.8601,
      "step": 9500
    },
    {
      "epoch": 18.45,
      "eval_loss": 3.061061143875122,
      "eval_runtime": 46.9034,
      "eval_samples_per_second": 21.534,
      "eval_steps_per_second": 2.708,
      "eval_wer": 1.29140625,
      "step": 9500
    },
    {
      "epoch": 19.42,
      "learning_rate": 3.7716262975778546e-05,
      "loss": 2.8083,
      "step": 10000
    },
    {
      "epoch": 19.42,
      "eval_loss": 3.1701200008392334,
      "eval_runtime": 47.1199,
      "eval_samples_per_second": 21.435,
      "eval_steps_per_second": 2.695,
      "eval_wer": 1.281640625,
      "step": 10000
    },
    {
      "epoch": 20.39,
      "learning_rate": 3.4256055363321795e-05,
      "loss": 2.771,
      "step": 10500
    },
    {
      "epoch": 20.39,
      "eval_loss": 3.059804916381836,
      "eval_runtime": 47.1428,
      "eval_samples_per_second": 21.424,
      "eval_steps_per_second": 2.694,
      "eval_wer": 1.29140625,
      "step": 10500
    },
    {
      "epoch": 21.36,
      "learning_rate": 3.079584775086505e-05,
      "loss": 2.7693,
      "step": 11000
    },
    {
      "epoch": 21.36,
      "eval_loss": 3.0247151851654053,
      "eval_runtime": 46.8824,
      "eval_samples_per_second": 21.543,
      "eval_steps_per_second": 2.709,
      "eval_wer": 1.29140625,
      "step": 11000
    },
    {
      "epoch": 22.33,
      "learning_rate": 2.733564013840831e-05,
      "loss": 2.7861,
      "step": 11500
    },
    {
      "epoch": 22.33,
      "eval_loss": 3.054908514022827,
      "eval_runtime": 47.1813,
      "eval_samples_per_second": 21.407,
      "eval_steps_per_second": 2.692,
      "eval_wer": 1.29140625,
      "step": 11500
    },
    {
      "epoch": 23.3,
      "learning_rate": 2.387543252595156e-05,
      "loss": 2.9903,
      "step": 12000
    },
    {
      "epoch": 23.3,
      "eval_loss": 3.235363245010376,
      "eval_runtime": 47.3429,
      "eval_samples_per_second": 21.334,
      "eval_steps_per_second": 2.683,
      "eval_wer": 1.29140625,
      "step": 12000
    },
    {
      "epoch": 24.27,
      "learning_rate": 2.041522491349481e-05,
      "loss": 2.8883,
      "step": 12500
    },
    {
      "epoch": 24.27,
      "eval_loss": 3.0525033473968506,
      "eval_runtime": 46.4887,
      "eval_samples_per_second": 21.726,
      "eval_steps_per_second": 2.732,
      "eval_wer": 1.29140625,
      "step": 12500
    },
    {
      "epoch": 25.24,
      "learning_rate": 1.6955017301038063e-05,
      "loss": 2.743,
      "step": 13000
    },
    {
      "epoch": 25.24,
      "eval_loss": 3.0735745429992676,
      "eval_runtime": 46.8938,
      "eval_samples_per_second": 21.538,
      "eval_steps_per_second": 2.708,
      "eval_wer": 1.29140625,
      "step": 13000
    },
    {
      "epoch": 26.21,
      "learning_rate": 1.3494809688581317e-05,
      "loss": 2.7344,
      "step": 13500
    },
    {
      "epoch": 26.21,
      "eval_loss": 3.0599749088287354,
      "eval_runtime": 46.6796,
      "eval_samples_per_second": 21.637,
      "eval_steps_per_second": 2.721,
      "eval_wer": 1.29140625,
      "step": 13500
    },
    {
      "epoch": 27.18,
      "learning_rate": 1.0034602076124568e-05,
      "loss": 2.697,
      "step": 14000
    },
    {
      "epoch": 27.18,
      "eval_loss": 3.055352210998535,
      "eval_runtime": 47.0781,
      "eval_samples_per_second": 21.454,
      "eval_steps_per_second": 2.698,
      "eval_wer": 1.29140625,
      "step": 14000
    },
    {
      "epoch": 28.16,
      "learning_rate": 6.5743944636678194e-06,
      "loss": 2.7185,
      "step": 14500
    },
    {
      "epoch": 28.16,
      "eval_loss": 3.130082607269287,
      "eval_runtime": 46.9873,
      "eval_samples_per_second": 21.495,
      "eval_steps_per_second": 2.703,
      "eval_wer": 1.29140625,
      "step": 14500
    },
    {
      "epoch": 29.13,
      "learning_rate": 3.1141868512110726e-06,
      "loss": 2.6788,
      "step": 15000
    },
    {
      "epoch": 29.13,
      "eval_loss": 3.2066450119018555,
      "eval_runtime": 46.7382,
      "eval_samples_per_second": 21.61,
      "eval_steps_per_second": 2.717,
      "eval_wer": 1.29140625,
      "step": 15000
    }
  ],
  "max_steps": 15450,
  "num_train_epochs": 30,
  "total_flos": 3.5108035253147873e+18,
  "trial_name": null,
  "trial_params": null
}