{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0204081632653061,
  "eval_steps": 30,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10204081632653061,
      "grad_norm": 32.964595794677734,
      "learning_rate": 3e-06,
      "loss": 3.9694,
      "step": 30
    },
    {
      "epoch": 0.10204081632653061,
      "eval_cer": 1.08866660375578,
      "eval_loss": 3.778170347213745,
      "eval_runtime": 309.6504,
      "eval_samples_per_second": 3.669,
      "eval_steps_per_second": 0.229,
      "eval_wer": 1.8747932306909276,
      "step": 30
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 17.289554595947266,
      "learning_rate": 6e-06,
      "loss": 3.3735,
      "step": 60
    },
    {
      "epoch": 0.20408163265306123,
      "eval_cer": 0.42536566952911203,
      "eval_loss": 2.9597644805908203,
      "eval_runtime": 203.9425,
      "eval_samples_per_second": 5.57,
      "eval_steps_per_second": 0.348,
      "eval_wer": 1.0019086397760528,
      "step": 60
    },
    {
      "epoch": 0.30612244897959184,
      "grad_norm": 14.627169609069824,
      "learning_rate": 9e-06,
      "loss": 2.5449,
      "step": 90
    },
    {
      "epoch": 0.30612244897959184,
      "eval_cer": 0.3221289044069076,
      "eval_loss": 2.198906421661377,
      "eval_runtime": 188.0543,
      "eval_samples_per_second": 6.041,
      "eval_steps_per_second": 0.378,
      "eval_wer": 0.8820460618399287,
      "step": 90
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 17.809280395507812,
      "learning_rate": 9e-06,
      "loss": 1.9987,
      "step": 120
    },
    {
      "epoch": 0.40816326530612246,
      "eval_cer": 0.2606398037180334,
      "eval_loss": 1.8648453950881958,
      "eval_runtime": 173.0054,
      "eval_samples_per_second": 6.566,
      "eval_steps_per_second": 0.41,
      "eval_wer": 0.8003562794248632,
      "step": 120
    },
    {
      "epoch": 0.5102040816326531,
      "grad_norm": 12.8538236618042,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.7671,
      "step": 150
    },
    {
      "epoch": 0.5102040816326531,
      "eval_cer": 0.2312352552609229,
      "eval_loss": 1.6909141540527344,
      "eval_runtime": 166.0118,
      "eval_samples_per_second": 6.843,
      "eval_steps_per_second": 0.428,
      "eval_wer": 0.7619289986003308,
      "step": 150
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 1594495.5,
      "learning_rate": 6e-06,
      "loss": 1.6285,
      "step": 180
    },
    {
      "epoch": 0.6122448979591837,
      "eval_cer": 0.22453524582428988,
      "eval_loss": 1.5862839221954346,
      "eval_runtime": 170.3953,
      "eval_samples_per_second": 6.667,
      "eval_steps_per_second": 0.417,
      "eval_wer": 0.7335538872630105,
      "step": 180
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 13.195433616638184,
      "learning_rate": 4.5e-06,
      "loss": 1.5475,
      "step": 210
    },
    {
      "epoch": 0.7142857142857143,
      "eval_cer": 0.2212512975370388,
      "eval_loss": 1.525095820426941,
      "eval_runtime": 176.0288,
      "eval_samples_per_second": 6.453,
      "eval_steps_per_second": 0.403,
      "eval_wer": 0.7215930779997455,
      "step": 210
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 11.149357795715332,
      "learning_rate": 3e-06,
      "loss": 1.4793,
      "step": 240
    },
    {
      "epoch": 0.8163265306122449,
      "eval_cer": 0.20349155421345663,
      "eval_loss": 1.4806641340255737,
      "eval_runtime": 165.379,
      "eval_samples_per_second": 6.869,
      "eval_steps_per_second": 0.429,
      "eval_wer": 0.6942359078763202,
      "step": 240
    },
    {
      "epoch": 0.9183673469387755,
      "grad_norm": 11.996837615966797,
      "learning_rate": 1.5e-06,
      "loss": 1.5013,
      "step": 270
    },
    {
      "epoch": 0.9183673469387755,
      "eval_cer": 0.2057374728696801,
      "eval_loss": 1.4582278728485107,
      "eval_runtime": 171.0364,
      "eval_samples_per_second": 6.642,
      "eval_steps_per_second": 0.415,
      "eval_wer": 0.6904186283242143,
      "step": 270
    },
    {
      "epoch": 1.0204081632653061,
      "grad_norm": 13.957674980163574,
      "learning_rate": 0.0,
      "loss": 1.4438,
      "step": 300
    },
    {
      "epoch": 1.0204081632653061,
      "eval_cer": 0.20496366896291404,
      "eval_loss": 1.4505608081817627,
      "eval_runtime": 170.7406,
      "eval_samples_per_second": 6.653,
      "eval_steps_per_second": 0.416,
      "eval_wer": 0.6883827458964245,
      "step": 300
    },
    {
      "epoch": 1.0204081632653061,
      "step": 300,
      "total_flos": 2.3614434607104e+17,
      "train_loss": 2.125396842956543,
      "train_runtime": 3027.4927,
      "train_samples_per_second": 3.171,
      "train_steps_per_second": 0.099
    }
  ],
  "logging_steps": 30,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3614434607104e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}