trainer_state.json

{
  "best_global_step": 3568,
  "best_metric": 0.8764965343415249,
  "best_model_checkpoint": "outputs/bartpho-spam-binary/checkpoint-3568",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 3568,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 5.2079925537109375,
      "learning_rate": 4.998771926094749e-05,
      "loss": 0.4627,
      "step": 446
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8084436042848141,
      "eval_loss": 0.4561808407306671,
      "eval_runtime": 19.4902,
      "eval_samples_per_second": 81.426,
      "eval_steps_per_second": 2.565,
      "step": 446
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.896441459655762,
      "learning_rate": 4.995077872180951e-05,
      "loss": 0.3735,
      "step": 892
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8418399495904222,
      "eval_loss": 0.3941121995449066,
      "eval_runtime": 19.0372,
      "eval_samples_per_second": 83.363,
      "eval_steps_per_second": 2.626,
      "step": 892
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.555406332015991,
      "learning_rate": 4.988921477641713e-05,
      "loss": 0.3147,
      "step": 1338
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8613736609955892,
      "eval_loss": 0.3282012343406677,
      "eval_runtime": 19.0715,
      "eval_samples_per_second": 83.213,
      "eval_steps_per_second": 2.622,
      "step": 1338
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.6473870277404785,
      "learning_rate": 4.980308818095173e-05,
      "loss": 0.2677,
      "step": 1784
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8714555765595463,
      "eval_loss": 0.29463911056518555,
      "eval_runtime": 19.1426,
      "eval_samples_per_second": 82.904,
      "eval_steps_per_second": 2.612,
      "step": 1784
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.139180660247803,
      "learning_rate": 4.9692483931964865e-05,
      "loss": 0.2441,
      "step": 2230
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.868935097668557,
      "eval_loss": 0.32677021622657776,
      "eval_runtime": 19.0711,
      "eval_samples_per_second": 83.215,
      "eval_steps_per_second": 2.622,
      "step": 2230
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.7669497728347778,
      "learning_rate": 4.9557511182496865e-05,
      "loss": 0.2137,
      "step": 2676
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8752362948960303,
      "eval_loss": 0.37486177682876587,
      "eval_runtime": 19.1107,
      "eval_samples_per_second": 83.042,
      "eval_steps_per_second": 2.616,
      "step": 2676
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.3637311458587646,
      "learning_rate": 4.939830313435599e-05,
      "loss": 0.1957,
      "step": 3122
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8720856962822936,
      "eval_loss": 0.43288952112197876,
      "eval_runtime": 19.1972,
      "eval_samples_per_second": 82.668,
      "eval_steps_per_second": 2.605,
      "step": 3122
    },
    {
      "epoch": 8.0,
      "grad_norm": 7.402363300323486,
      "learning_rate": 4.9215016906664344e-05,
      "loss": 0.1769,
      "step": 3568
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8764965343415249,
      "eval_loss": 0.4071377217769623,
      "eval_runtime": 19.1521,
      "eval_samples_per_second": 82.863,
      "eval_steps_per_second": 2.611,
      "step": 3568
    }
  ],
  "logging_steps": 446,
  "max_steps": 44600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.101365327850419e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}