run-2/checkpoint-1468/trainer_state.json

{
  "best_metric": 0.861878453038674,
  "best_model_checkpoint": "BERTForDetectingDepression-Twitter2015/run-2/checkpoint-734",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 1468,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8207489257213014,
      "eval_loss": 0.41005730628967285,
      "eval_runtime": 1.5429,
      "eval_samples_per_second": 1055.797,
      "eval_steps_per_second": 66.109,
      "step": 367
    },
    {
      "epoch": 1.3623978201634879,
      "grad_norm": 9.666861534118652,
      "learning_rate": 1.831949194907621e-05,
      "loss": 0.5034,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.861878453038674,
      "eval_loss": 0.5616857409477234,
      "eval_runtime": 1.528,
      "eval_samples_per_second": 1066.087,
      "eval_steps_per_second": 66.753,
      "step": 734
    },
    {
      "epoch": 2.7247956403269757,
      "grad_norm": 0.05537036433815956,
      "learning_rate": 1.1458259009347293e-05,
      "loss": 0.3057,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8575813382443217,
      "eval_loss": 0.6872314810752869,
      "eval_runtime": 1.521,
      "eval_samples_per_second": 1070.983,
      "eval_steps_per_second": 67.06,
      "step": 1101
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.858195211786372,
      "eval_loss": 0.8370489478111267,
      "eval_runtime": 1.8437,
      "eval_samples_per_second": 883.566,
      "eval_steps_per_second": 55.325,
      "step": 1468
    }
  ],
  "logging_steps": 500,
  "max_steps": 1835,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 81016621722120.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.518072488880513e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 4,
    "seed": 4
  }
}