{
  "best_global_step": 392,
  "best_metric": 1.9849951267242432,
  "best_model_checkpoint": "./bert-resume/checkpoint-392",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 588,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 2.013901472091675,
      "eval_runtime": 32.6312,
      "eval_samples_per_second": 53.905,
      "eval_steps_per_second": 1.686,
      "step": 196
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.9849951267242432,
      "eval_runtime": 32.6699,
      "eval_samples_per_second": 53.842,
      "eval_steps_per_second": 1.684,
      "step": 392
    },
    {
      "epoch": 2.5510204081632653,
      "grad_norm": 7.228590965270996,
      "learning_rate": 1.7454081632653063e-05,
      "loss": 1.8102947998046874,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.102543830871582,
      "eval_runtime": 33.1614,
      "eval_samples_per_second": 53.044,
      "eval_steps_per_second": 1.659,
      "step": 588
    }
  ],
  "logging_steps": 500,
  "max_steps": 3920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 4926272520121344.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}