trainer_state.json

{
  "best_metric": 0.8927897563753328,
  "best_model_checkpoint": "/mnt/d/models_iberlef22/exist22/task1/models_es/retrained_bertin-project/checkpoint-483",
  "epoch": 3.0,
  "global_step": 483,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.0993153594484644e-05,
      "loss": 0.4374,
      "step": 161
    },
    {
      "epoch": 1.0,
      "eval_f1-score": 0.8591688247626095,
      "eval_loss": 0.33084869384765625,
      "eval_precision": 0.8609666394234341,
      "eval_recall": 0.8584189946881762,
      "eval_runtime": 0.6951,
      "eval_samples_per_second": 821.415,
      "eval_steps_per_second": 12.947,
      "eval_support": 571,
      "step": 161
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.5687892560171237e-05,
      "loss": 0.2522,
      "step": 322
    },
    {
      "epoch": 2.0,
      "eval_f1-score": 0.8802314677721843,
      "eval_loss": 0.3257339298725128,
      "eval_precision": 0.8825520833333333,
      "eval_recall": 0.8793158567774936,
      "eval_runtime": 0.7147,
      "eval_samples_per_second": 798.967,
      "eval_steps_per_second": 12.593,
      "eval_support": 571,
      "step": 322
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.869736443933763e-07,
      "loss": 0.1442,
      "step": 483
    },
    {
      "epoch": 3.0,
      "eval_f1-score": 0.8927897563753328,
      "eval_loss": 0.3625585436820984,
      "eval_precision": 0.8935010482180293,
      "eval_recall": 0.8923494983277591,
      "eval_runtime": 0.6958,
      "eval_samples_per_second": 820.621,
      "eval_steps_per_second": 12.934,
      "eval_support": 571,
      "step": 483
    }
  ],
  "max_steps": 483,
  "num_train_epochs": 3,
  "total_flos": 1012319785497600.0,
  "trial_name": null,
  "trial_params": null
}