{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 250.0,
  "global_step": 134000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 25.0,
      "learning_rate": 9.474312647289867e-05,
      "loss": 20.2416,
      "step": 13400
    },
    {
      "epoch": 25.0,
      "eval_cer": 0.9767676767676767,
      "eval_loss": 3.248976230621338,
      "eval_runtime": 0.8563,
      "eval_samples_per_second": 210.199,
      "eval_steps_per_second": 26.859,
      "eval_wer": 1.0,
      "step": 13400
    },
    {
      "epoch": 50.0,
      "learning_rate": 8.422073841319718e-05,
      "loss": 2.7136,
      "step": 26800
    },
    {
      "epoch": 50.0,
      "eval_cer": 0.9222222222222223,
      "eval_loss": 3.162766218185425,
      "eval_runtime": 0.8572,
      "eval_samples_per_second": 209.98,
      "eval_steps_per_second": 26.831,
      "eval_wer": 1.0,
      "step": 26800
    },
    {
      "epoch": 75.0,
      "learning_rate": 7.369835035349568e-05,
      "loss": 2.574,
      "step": 40200
    },
    {
      "epoch": 75.0,
      "eval_cer": 0.9141414141414141,
      "eval_loss": 3.141190528869629,
      "eval_runtime": 0.8427,
      "eval_samples_per_second": 213.593,
      "eval_steps_per_second": 27.292,
      "eval_wer": 1.0,
      "step": 40200
    },
    {
      "epoch": 100.0,
      "learning_rate": 6.317674783974863e-05,
      "loss": 2.5132,
      "step": 53600
    },
    {
      "epoch": 100.0,
      "eval_cer": 0.9101010101010101,
      "eval_loss": 3.159379720687866,
      "eval_runtime": 0.8438,
      "eval_samples_per_second": 213.316,
      "eval_steps_per_second": 27.257,
      "eval_wer": 1.0,
      "step": 53600
    },
    {
      "epoch": 125.0,
      "learning_rate": 5.2655145326001575e-05,
      "loss": 2.4764,
      "step": 67000
    },
    {
      "epoch": 125.0,
      "eval_cer": 0.9050505050505051,
      "eval_loss": 3.132108449935913,
      "eval_runtime": 0.8431,
      "eval_samples_per_second": 213.503,
      "eval_steps_per_second": 27.281,
      "eval_wer": 1.0,
      "step": 67000
    },
    {
      "epoch": 150.0,
      "learning_rate": 4.2132757266300084e-05,
      "loss": 2.4522,
      "step": 80400
    },
    {
      "epoch": 150.0,
      "eval_cer": 0.9030303030303031,
      "eval_loss": 3.149836301803589,
      "eval_runtime": 0.842,
      "eval_samples_per_second": 213.776,
      "eval_steps_per_second": 27.316,
      "eval_wer": 1.0,
      "step": 80400
    },
    {
      "epoch": 175.0,
      "learning_rate": 3.1610369206598586e-05,
      "loss": 2.4366,
      "step": 93800
    },
    {
      "epoch": 175.0,
      "eval_cer": 0.901010101010101,
      "eval_loss": 3.154324769973755,
      "eval_runtime": 0.8423,
      "eval_samples_per_second": 213.693,
      "eval_steps_per_second": 27.305,
      "eval_wer": 1.0,
      "step": 93800
    },
    {
      "epoch": 200.0,
      "learning_rate": 2.1088766692851532e-05,
      "loss": 2.426,
      "step": 107200
    },
    {
      "epoch": 200.0,
      "eval_cer": 0.901010101010101,
      "eval_loss": 3.1591243743896484,
      "eval_runtime": 0.8434,
      "eval_samples_per_second": 213.431,
      "eval_steps_per_second": 27.272,
      "eval_wer": 1.0,
      "step": 107200
    },
    {
      "epoch": 225.0,
      "learning_rate": 1.056637863315004e-05,
      "loss": 2.4199,
      "step": 120600
    },
    {
      "epoch": 225.0,
      "eval_cer": 0.898989898989899,
      "eval_loss": 3.1549627780914307,
      "eval_runtime": 0.8433,
      "eval_samples_per_second": 213.458,
      "eval_steps_per_second": 27.275,
      "eval_wer": 1.0,
      "step": 120600
    },
    {
      "epoch": 250.0,
      "learning_rate": 4.4776119402985075e-08,
      "loss": 2.4161,
      "step": 134000
    },
    {
      "epoch": 250.0,
      "eval_cer": 0.898989898989899,
      "eval_loss": 3.1552622318267822,
      "eval_runtime": 0.843,
      "eval_samples_per_second": 213.521,
      "eval_steps_per_second": 27.283,
      "eval_wer": 1.0,
      "step": 134000
    }
  ],
  "max_steps": 134000,
  "num_train_epochs": 250,
  "total_flos": 1.2448746848771059e+20,
  "trial_name": null,
  "trial_params": null
}