{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.063868392403532,
  "eval_steps": 800,
  "global_step": 16800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1935405830410064,
      "grad_norm": NaN,
      "learning_rate": 0.0002979720793534166,
      "loss": 5.9034,
      "step": 800
    },
    {
      "epoch": 0.1935405830410064,
      "eval_loss": Infinity,
      "eval_runtime": 557.9284,
      "eval_samples_per_second": 41.837,
      "eval_steps_per_second": 5.23,
      "eval_wer": 1.0,
      "step": 800
    },
    {
      "epoch": 0.3870811660820128,
      "grad_norm": 0.0,
      "learning_rate": 0.00029226304188096985,
      "loss": 4.005,
      "step": 1600
    },
    {
      "epoch": 0.3870811660820128,
      "eval_loss": Infinity,
      "eval_runtime": 524.4916,
      "eval_samples_per_second": 44.504,
      "eval_steps_per_second": 5.563,
      "eval_wer": 1.0,
      "step": 1600
    },
    {
      "epoch": 0.5806217491230192,
      "grad_norm": 0.0,
      "learning_rate": 0.0002864364437913299,
      "loss": 4.0676,
      "step": 2400
    },
    {
      "epoch": 0.5806217491230192,
      "eval_loss": Infinity,
      "eval_runtime": 522.6766,
      "eval_samples_per_second": 44.659,
      "eval_steps_per_second": 5.583,
      "eval_wer": 1.0,
      "step": 2400
    },
    {
      "epoch": 0.7741623321640256,
      "grad_norm": 0.0,
      "learning_rate": 0.0002806833210874357,
      "loss": 4.0978,
      "step": 3200
    },
    {
      "epoch": 0.7741623321640256,
      "eval_loss": Infinity,
      "eval_runtime": 524.1686,
      "eval_samples_per_second": 44.531,
      "eval_steps_per_second": 5.567,
      "eval_wer": 1.0,
      "step": 3200
    },
    {
      "epoch": 0.967702915205032,
      "grad_norm": 0.0,
      "learning_rate": 0.0002749595885378398,
      "loss": 4.1059,
      "step": 4000
    },
    {
      "epoch": 0.967702915205032,
      "eval_loss": Infinity,
      "eval_runtime": 525.0456,
      "eval_samples_per_second": 44.457,
      "eval_steps_per_second": 5.558,
      "eval_wer": 1.0,
      "step": 4000
    },
    {
      "epoch": 1.161122535381638,
      "grad_norm": 0.0,
      "learning_rate": 0.0002692064658339456,
      "loss": 4.0961,
      "step": 4800
    },
    {
      "epoch": 1.161122535381638,
      "eval_loss": Infinity,
      "eval_runtime": 525.4326,
      "eval_samples_per_second": 44.424,
      "eval_steps_per_second": 5.554,
      "eval_wer": 1.0,
      "step": 4800
    },
    {
      "epoch": 1.3546631184226443,
      "grad_norm": 0.0,
      "learning_rate": 0.00026350477590007343,
      "loss": 4.1137,
      "step": 5600
    },
    {
      "epoch": 1.3546631184226443,
      "eval_loss": Infinity,
      "eval_runtime": 524.7239,
      "eval_samples_per_second": 44.484,
      "eval_steps_per_second": 5.561,
      "eval_wer": 1.0,
      "step": 5600
    },
    {
      "epoch": 1.5482037014636507,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 7.5668,
      "step": 6400
    },
    {
      "epoch": 1.5482037014636507,
      "eval_loss": NaN,
      "eval_runtime": 525.2952,
      "eval_samples_per_second": 44.436,
      "eval_steps_per_second": 5.555,
      "eval_wer": 1.0,
      "step": 6400
    },
    {
      "epoch": 1.7417442845046571,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 7200
    },
    {
      "epoch": 1.7417442845046571,
      "eval_loss": NaN,
      "eval_runtime": 524.3577,
      "eval_samples_per_second": 44.515,
      "eval_steps_per_second": 5.565,
      "eval_wer": 1.0,
      "step": 7200
    },
    {
      "epoch": 1.9352848675456635,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 8000
    },
    {
      "epoch": 1.9352848675456635,
      "eval_loss": NaN,
      "eval_runtime": 524.5203,
      "eval_samples_per_second": 44.502,
      "eval_steps_per_second": 5.563,
      "eval_wer": 1.0,
      "step": 8000
    },
    {
      "epoch": 2.128704487722269,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 8800
    },
    {
      "epoch": 2.128704487722269,
      "eval_loss": NaN,
      "eval_runtime": 525.5034,
      "eval_samples_per_second": 44.418,
      "eval_steps_per_second": 5.553,
      "eval_wer": 1.0,
      "step": 8800
    },
    {
      "epoch": 2.322245070763276,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 9600
    },
    {
      "epoch": 2.322245070763276,
      "eval_loss": NaN,
      "eval_runtime": 524.5899,
      "eval_samples_per_second": 44.496,
      "eval_steps_per_second": 5.562,
      "eval_wer": 1.0,
      "step": 9600
    },
    {
      "epoch": 2.515785653804282,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 10400
    },
    {
      "epoch": 2.515785653804282,
      "eval_loss": NaN,
      "eval_runtime": 524.8508,
      "eval_samples_per_second": 44.474,
      "eval_steps_per_second": 5.56,
      "eval_wer": 1.0,
      "step": 10400
    },
    {
      "epoch": 2.7093262368452886,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 11200
    },
    {
      "epoch": 2.7093262368452886,
      "eval_loss": NaN,
      "eval_runtime": 524.5261,
      "eval_samples_per_second": 44.501,
      "eval_steps_per_second": 5.563,
      "eval_wer": 1.0,
      "step": 11200
    },
    {
      "epoch": 2.902866819886295,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 12000
    },
    {
      "epoch": 2.902866819886295,
      "eval_loss": NaN,
      "eval_runtime": 523.894,
      "eval_samples_per_second": 44.555,
      "eval_steps_per_second": 5.57,
      "eval_wer": 1.0,
      "step": 12000
    },
    {
      "epoch": 3.096286440062901,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 12800
    },
    {
      "epoch": 3.096286440062901,
      "eval_loss": NaN,
      "eval_runtime": 524.8229,
      "eval_samples_per_second": 44.476,
      "eval_steps_per_second": 5.56,
      "eval_wer": 1.0,
      "step": 12800
    },
    {
      "epoch": 3.289827023103907,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 13600
    },
    {
      "epoch": 3.289827023103907,
      "eval_loss": NaN,
      "eval_runtime": 523.5675,
      "eval_samples_per_second": 44.583,
      "eval_steps_per_second": 5.573,
      "eval_wer": 1.0,
      "step": 13600
    },
    {
      "epoch": 3.4833676061449137,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 14400
    },
    {
      "epoch": 3.4833676061449137,
      "eval_loss": NaN,
      "eval_runtime": 531.6114,
      "eval_samples_per_second": 43.908,
      "eval_steps_per_second": 5.489,
      "eval_wer": 1.0,
      "step": 14400
    },
    {
      "epoch": 3.67690818918592,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 15200
    },
    {
      "epoch": 3.67690818918592,
      "eval_loss": NaN,
      "eval_runtime": 544.415,
      "eval_samples_per_second": 42.875,
      "eval_steps_per_second": 5.36,
      "eval_wer": 1.0,
      "step": 15200
    },
    {
      "epoch": 3.8704487722269265,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 16000
    },
    {
      "epoch": 3.8704487722269265,
      "eval_loss": NaN,
      "eval_runtime": 540.5292,
      "eval_samples_per_second": 43.184,
      "eval_steps_per_second": 5.398,
      "eval_wer": 1.0,
      "step": 16000
    },
    {
      "epoch": 4.063868392403532,
      "grad_norm": NaN,
      "learning_rate": 0.0002604114621601763,
      "loss": 0.0,
      "step": 16800
    },
    {
      "epoch": 4.063868392403532,
      "eval_loss": NaN,
      "eval_runtime": 527.7152,
      "eval_samples_per_second": 44.232,
      "eval_steps_per_second": 5.529,
      "eval_wer": 1.0,
      "step": 16800
    }
  ],
  "logging_steps": 800,
  "max_steps": 41330,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 800,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.7959210547118785e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}