trainer_state.json

{
  "best_metric": 0.26231876015663147,
  "best_model_checkpoint": "spa_mBERT/checkpoint-4480",
  "epoch": 6.0,
  "global_step": 5376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.56,
      "learning_rate": 1.988839285714286e-05,
      "loss": 1.0204,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8996683415089468,
      "eval_f1": 0.7103438993202774,
      "eval_loss": 0.3859018087387085,
      "eval_precision": 0.7230442526373476,
      "eval_recall": 0.6980820105820106,
      "eval_runtime": 5.926,
      "eval_samples_per_second": 279.11,
      "eval_steps_per_second": 17.55,
      "step": 896
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.9776785714285716e-05,
      "loss": 0.4747,
      "step": 1000
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.9665178571428575e-05,
      "loss": 0.3496,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.917184676186126,
      "eval_f1": 0.764059043027441,
      "eval_loss": 0.30577903985977173,
      "eval_precision": 0.7629756644463497,
      "eval_recall": 0.7651455026455026,
      "eval_runtime": 5.9393,
      "eval_samples_per_second": 278.485,
      "eval_steps_per_second": 17.511,
      "step": 1792
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.955357142857143e-05,
      "loss": 0.3017,
      "step": 2000
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.944196428571429e-05,
      "loss": 0.2568,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9256648064663476,
      "eval_f1": 0.7892974370876299,
      "eval_loss": 0.28278017044067383,
      "eval_precision": 0.7913702546373247,
      "eval_recall": 0.7872354497354498,
      "eval_runtime": 6.0079,
      "eval_samples_per_second": 275.304,
      "eval_steps_per_second": 17.311,
      "step": 2688
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.9330357142857144e-05,
      "loss": 0.22,
      "step": 3000
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.9218750000000003e-05,
      "loss": 0.1981,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9319405000695092,
      "eval_f1": 0.810671256454389,
      "eval_loss": 0.26689934730529785,
      "eval_precision": 0.8114232706069441,
      "eval_recall": 0.8099206349206349,
      "eval_runtime": 6.0528,
      "eval_samples_per_second": 273.261,
      "eval_steps_per_second": 17.182,
      "step": 3584
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.910714285714286e-05,
      "loss": 0.1757,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9351975056103906,
      "eval_f1": 0.8214273897362936,
      "eval_loss": 0.26231876015663147,
      "eval_precision": 0.8218896907899093,
      "eval_recall": 0.8209656084656085,
      "eval_runtime": 5.8306,
      "eval_samples_per_second": 283.675,
      "eval_steps_per_second": 17.837,
      "step": 4480
    },
    {
      "epoch": 5.02,
      "learning_rate": 1.8995535714285717e-05,
      "loss": 0.1648,
      "step": 4500
    },
    {
      "epoch": 5.58,
      "learning_rate": 1.8883928571428573e-05,
      "loss": 0.1388,
      "step": 5000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9327348916648461,
      "eval_f1": 0.8123542346904777,
      "eval_loss": 0.2874906361103058,
      "eval_precision": 0.8185175238351013,
      "eval_recall": 0.8062830687830688,
      "eval_runtime": 6.0062,
      "eval_samples_per_second": 275.382,
      "eval_steps_per_second": 17.315,
      "step": 5376
    }
  ],
  "max_steps": 89600,
  "num_train_epochs": 100,
  "total_flos": 3919265715030336.0,
  "trial_name": null,
  "trial_params": null
}