{
  "best_global_step": 16,
  "best_metric": 0.9664335664335665,
  "best_model_checkpoint": "/content/longformer_vietnamese_policy_model/checkpoint-16",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 16,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3870967741935484,
      "eval_f1_macro": 0.22135076252723312,
      "eval_loss": 1.569871187210083,
      "eval_runtime": 1.6168,
      "eval_samples_per_second": 19.174,
      "eval_steps_per_second": 2.474,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3870967741935484,
      "eval_f1_macro": 0.2541666666666667,
      "eval_loss": 1.5102384090423584,
      "eval_runtime": 1.6207,
      "eval_samples_per_second": 19.127,
      "eval_steps_per_second": 2.468,
      "step": 8
    },
    {
      "epoch": 2.533333333333333,
      "grad_norm": 8.057486534118652,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 5.993505477905273,
      "step": 10
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8387096774193549,
      "eval_f1_macro": 0.8164335664335665,
      "eval_loss": 1.4431389570236206,
      "eval_runtime": 1.6485,
      "eval_samples_per_second": 18.805,
      "eval_steps_per_second": 2.426,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.967741935483871,
      "eval_f1_macro": 0.9664335664335665,
      "eval_loss": 1.371219515800476,
      "eval_runtime": 1.6491,
      "eval_samples_per_second": 18.798,
      "eval_steps_per_second": 2.426,
      "step": 16
    }
  ],
  "logging_steps": 10,
  "max_steps": 20,
  "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": 1204028619915264.0,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}