{
  "best_metric": 0.6075,
  "best_model_checkpoint": "BERTForDetectingDepression-Twitter2020/run-1/checkpoint-452",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 565,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5945,
      "eval_loss": 0.6746140718460083,
      "eval_runtime": 3.0608,
      "eval_samples_per_second": 653.432,
      "eval_steps_per_second": 40.84,
      "step": 113
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5985,
      "eval_loss": 0.7479358315467834,
      "eval_runtime": 3.0745,
      "eval_samples_per_second": 650.511,
      "eval_steps_per_second": 40.657,
      "step": 226
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.602,
      "eval_loss": 1.183725118637085,
      "eval_runtime": 3.1173,
      "eval_samples_per_second": 641.587,
      "eval_steps_per_second": 40.099,
      "step": 339
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6075,
      "eval_loss": 1.88389253616333,
      "eval_runtime": 3.1299,
      "eval_samples_per_second": 638.99,
      "eval_steps_per_second": 39.937,
      "step": 452
    },
    {
      "epoch": 4.424778761061947,
      "grad_norm": 0.07030410319566727,
      "learning_rate": 5.8869146368766855e-06,
      "loss": 0.3456,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6035,
      "eval_loss": 2.064941167831421,
      "eval_runtime": 3.266,
      "eval_samples_per_second": 612.362,
      "eval_steps_per_second": 38.273,
      "step": 565
    }
  ],
  "logging_steps": 500,
  "max_steps": 565,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 360416923630560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.117087338208196e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 16,
    "seed": 26
  }
}