{
  "best_global_step": 3885,
  "best_metric": 0.07048454880714417,
  "best_model_checkpoint": "./results_albert_punctuation_casing/checkpoint-3885",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3885,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1287001287001287,
      "grad_norm": 4.2918195724487305,
      "learning_rate": 1.9143715143715144e-05,
      "loss": 0.2105482635498047,
      "step": 500
    },
    {
      "epoch": 0.2574002574002574,
      "grad_norm": 3.180814504623413,
      "learning_rate": 1.8285714285714288e-05,
      "loss": 0.09811102294921875,
      "step": 1000
    },
    {
      "epoch": 0.3861003861003861,
      "grad_norm": 4.069736480712891,
      "learning_rate": 1.742771342771343e-05,
      "loss": 0.08830223083496094,
      "step": 1500
    },
    {
      "epoch": 0.5148005148005148,
      "grad_norm": 2.3238089084625244,
      "learning_rate": 1.656971256971257e-05,
      "loss": 0.07262681579589844,
      "step": 2000
    },
    {
      "epoch": 0.6435006435006435,
      "grad_norm": 0.725167989730835,
      "learning_rate": 1.5711711711711713e-05,
      "loss": 0.07961682891845703,
      "step": 2500
    },
    {
      "epoch": 0.7722007722007722,
      "grad_norm": 13.671622276306152,
      "learning_rate": 1.4853710853710854e-05,
      "loss": 0.06993846893310547,
      "step": 3000
    },
    {
      "epoch": 0.9009009009009009,
      "grad_norm": 2.6098945140838623,
      "learning_rate": 1.3995709995709996e-05,
      "loss": 0.07217549133300781,
      "step": 3500
    },
    {
      "epoch": 1.0,
      "eval_casing_accuracy": 0.6387912059001499,
      "eval_loss": 0.07048454880714417,
      "eval_overall_accuracy": 0.6404223412931571,
      "eval_punctuation_accuracy": 0.6420534766861643,
      "eval_runtime": 86.2448,
      "eval_samples_per_second": 180.162,
      "eval_steps_per_second": 11.27,
      "step": 3885
    }
  ],
  "logging_steps": 500,
  "max_steps": 11655,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}