run-0/checkpoint-920/trainer_state.json

{
  "best_metric": 0.8594229588704727,
  "best_model_checkpoint": "BERTForDetectingDepression-Twitter2015/run-0/checkpoint-736",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 920,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7925107427869859,
      "eval_loss": 0.46051156520843506,
      "eval_runtime": 1.5937,
      "eval_samples_per_second": 1022.13,
      "eval_steps_per_second": 64.001,
      "step": 184
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8465316144874155,
      "eval_loss": 0.43229982256889343,
      "eval_runtime": 1.6666,
      "eval_samples_per_second": 977.463,
      "eval_steps_per_second": 61.204,
      "step": 368
    },
    {
      "epoch": 2.717391304347826,
      "grad_norm": 0.23171758651733398,
      "learning_rate": 7.102447094738944e-06,
      "loss": 0.3446,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8397790055248618,
      "eval_loss": 0.6237844824790955,
      "eval_runtime": 1.5569,
      "eval_samples_per_second": 1046.301,
      "eval_steps_per_second": 65.514,
      "step": 552
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8594229588704727,
      "eval_loss": 0.6028990745544434,
      "eval_runtime": 1.5442,
      "eval_samples_per_second": 1054.886,
      "eval_steps_per_second": 66.052,
      "step": 736
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.852670349907919,
      "eval_loss": 0.6441431045532227,
      "eval_runtime": 1.5731,
      "eval_samples_per_second": 1035.524,
      "eval_steps_per_second": 64.839,
      "step": 920
    }
  ],
  "logging_steps": 500,
  "max_steps": 920,
  "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": 91478369505360.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.555774125514245e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 8,
    "seed": 16
  }
}