trainer_state.json

{
  "best_global_step": 1566,
  "best_metric": 0.5676447814653858,
  "best_model_checkpoint": "outputs/bartpho/checkpoint-1566",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 1566,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9942528735632183,
      "grad_norm": 8.253643035888672,
      "learning_rate": 6.88e-06,
      "loss": 1.7722,
      "step": 173
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.35422740524781343,
      "eval_loss": 1.6894474029541016,
      "eval_macro_f1": 0.16483983747511713,
      "eval_runtime": 4.6188,
      "eval_samples_per_second": 148.524,
      "eval_steps_per_second": 4.763,
      "step": 174
    },
    {
      "epoch": 1.9885057471264367,
      "grad_norm": 10.314130783081055,
      "learning_rate": 1.38e-05,
      "loss": 1.5474,
      "step": 346
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.49416909620991256,
      "eval_loss": 1.3798141479492188,
      "eval_macro_f1": 0.33564178199668815,
      "eval_runtime": 4.6167,
      "eval_samples_per_second": 148.592,
      "eval_steps_per_second": 4.765,
      "step": 348
    },
    {
      "epoch": 2.9827586206896552,
      "grad_norm": 11.580602645874023,
      "learning_rate": 1.999994401896218e-05,
      "loss": 1.3036,
      "step": 519
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5728862973760933,
      "eval_loss": 1.1752054691314697,
      "eval_macro_f1": 0.4355327116696494,
      "eval_runtime": 4.6311,
      "eval_samples_per_second": 148.13,
      "eval_steps_per_second": 4.751,
      "step": 522
    },
    {
      "epoch": 3.9770114942528734,
      "grad_norm": 14.440051078796387,
      "learning_rate": 1.9993697433304146e-05,
      "loss": 1.1727,
      "step": 692
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5787172011661808,
      "eval_loss": 1.1525901556015015,
      "eval_macro_f1": 0.4912122989755165,
      "eval_runtime": 4.6769,
      "eval_samples_per_second": 146.679,
      "eval_steps_per_second": 4.704,
      "step": 696
    },
    {
      "epoch": 4.971264367816092,
      "grad_norm": 13.43419361114502,
      "learning_rate": 1.9977115923137912e-05,
      "loss": 1.0254,
      "step": 865
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5932944606413995,
      "eval_loss": 1.1551982164382935,
      "eval_macro_f1": 0.5047627462427012,
      "eval_runtime": 4.7431,
      "eval_samples_per_second": 144.632,
      "eval_steps_per_second": 4.638,
      "step": 870
    },
    {
      "epoch": 5.9655172413793105,
      "grad_norm": 13.305834770202637,
      "learning_rate": 1.995021663613649e-05,
      "loss": 0.8855,
      "step": 1038
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5860058309037901,
      "eval_loss": 1.0997425317764282,
      "eval_macro_f1": 0.5274407663166836,
      "eval_runtime": 4.6958,
      "eval_samples_per_second": 146.087,
      "eval_steps_per_second": 4.685,
      "step": 1044
    },
    {
      "epoch": 6.959770114942529,
      "grad_norm": 11.853787422180176,
      "learning_rate": 1.991302739004225e-05,
      "loss": 0.7547,
      "step": 1211
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5816326530612245,
      "eval_loss": 1.2447866201400757,
      "eval_macro_f1": 0.5071616053303554,
      "eval_runtime": 4.6271,
      "eval_samples_per_second": 148.256,
      "eval_steps_per_second": 4.755,
      "step": 1218
    },
    {
      "epoch": 7.954022988505747,
      "grad_norm": 13.94935417175293,
      "learning_rate": 1.9865586643899352e-05,
      "loss": 0.6322,
      "step": 1384
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.575801749271137,
      "eval_loss": 1.263814091682434,
      "eval_macro_f1": 0.5216417103369302,
      "eval_runtime": 4.6197,
      "eval_samples_per_second": 148.494,
      "eval_steps_per_second": 4.762,
      "step": 1392
    },
    {
      "epoch": 8.948275862068966,
      "grad_norm": 13.111533164978027,
      "learning_rate": 1.9807943458281543e-05,
      "loss": 0.4984,
      "step": 1557
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6020408163265306,
      "eval_loss": 1.3265769481658936,
      "eval_macro_f1": 0.5676447814653858,
      "eval_runtime": 4.6357,
      "eval_samples_per_second": 147.98,
      "eval_steps_per_second": 4.746,
      "step": 1566
    }
  ],
  "logging_steps": 173,
  "max_steps": 17400,
  "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": 2.0879945994205656e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}