trainer_state.json

{
  "best_metric": 0.7328244274809161,
  "best_model_checkpoint": "outputs/indobert-large-p1-twitter-indonesia-sarcastic/checkpoint-236",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 413,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.997777306367356e-06,
      "loss": 0.5836,
      "step": 59
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8059701492537313,
      "eval_f1": 0.5737704918032788,
      "eval_loss": 0.41532421112060547,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.5223880597014925,
      "eval_runtime": 3.3886,
      "eval_samples_per_second": 79.088,
      "eval_steps_per_second": 1.476,
      "step": 59
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.990628789297524e-06,
      "loss": 0.3766,
      "step": 118
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8432835820895522,
      "eval_f1": 0.5961538461538461,
      "eval_loss": 0.33532315492630005,
      "eval_precision": 0.8378378378378378,
      "eval_recall": 0.4626865671641791,
      "eval_runtime": 3.3898,
      "eval_samples_per_second": 79.062,
      "eval_steps_per_second": 1.475,
      "step": 118
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.978800734821829e-06,
      "loss": 0.2476,
      "step": 177
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8619402985074627,
      "eval_f1": 0.694214876033058,
      "eval_loss": 0.3113671541213989,
      "eval_precision": 0.7777777777777778,
      "eval_recall": 0.6268656716417911,
      "eval_runtime": 3.3876,
      "eval_samples_per_second": 79.111,
      "eval_steps_per_second": 1.476,
      "step": 177
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.961896985969137e-06,
      "loss": 0.1356,
      "step": 236
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8694029850746269,
      "eval_f1": 0.7328244274809161,
      "eval_loss": 0.327924907207489,
      "eval_precision": 0.75,
      "eval_recall": 0.7164179104477612,
      "eval_runtime": 3.3876,
      "eval_samples_per_second": 79.113,
      "eval_steps_per_second": 1.476,
      "step": 236
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.940096443848375e-06,
      "loss": 0.0536,
      "step": 295
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8582089552238806,
      "eval_f1": 0.7164179104477613,
      "eval_loss": 0.42652806639671326,
      "eval_precision": 0.7164179104477612,
      "eval_recall": 0.7164179104477612,
      "eval_runtime": 3.3976,
      "eval_samples_per_second": 78.88,
      "eval_steps_per_second": 1.472,
      "step": 295
    },
    {
      "epoch": 6.0,
      "learning_rate": 9.913420622962606e-06,
      "loss": 0.0157,
      "step": 354
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8619402985074627,
      "eval_f1": 0.6666666666666667,
      "eval_loss": 0.644801914691925,
      "eval_precision": 0.8409090909090909,
      "eval_recall": 0.5522388059701493,
      "eval_runtime": 3.3901,
      "eval_samples_per_second": 79.054,
      "eval_steps_per_second": 1.475,
      "step": 354
    },
    {
      "epoch": 7.0,
      "learning_rate": 9.881895849126432e-06,
      "loss": 0.0076,
      "step": 413
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8619402985074627,
      "eval_f1": 0.7218045112781953,
      "eval_loss": 0.5739138722419739,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.7164179104477612,
      "eval_runtime": 3.3841,
      "eval_samples_per_second": 79.193,
      "eval_steps_per_second": 1.477,
      "step": 413
    },
    {
      "epoch": 7.0,
      "step": 413,
      "total_flos": 3062792425577472.0,
      "train_loss": 0.20290469026450095,
      "train_runtime": 601.0116,
      "train_samples_per_second": 312.473,
      "train_steps_per_second": 9.817
    }
  ],
  "logging_steps": 500,
  "max_steps": 5900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 3062792425577472.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}