{
  "best_metric": 0.8090737462043762,
  "best_model_checkpoint": "/home/ccasimiro/ccasimiro/berta/src/finetuning/te/roberta-base-ca-cased-te/checkpoint-5290",
  "epoch": 9.999055712936732,
  "global_step": 5290,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.94,
      "learning_rate": 4.527410207939509e-05,
      "loss": 0.7295,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7693761587142944,
      "eval_loss": 0.5496639013290405,
      "eval_runtime": 15.769,
      "eval_samples_per_second": 134.188,
      "eval_steps_per_second": 4.249,
      "step": 529
    },
    {
      "epoch": 1.89,
      "learning_rate": 4.054820415879017e-05,
      "loss": 0.4162,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.782136082649231,
      "eval_loss": 0.5493781566619873,
      "eval_runtime": 15.7381,
      "eval_samples_per_second": 134.45,
      "eval_steps_per_second": 4.257,
      "step": 1058
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.582230623818525e-05,
      "loss": 0.237,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7948960065841675,
      "eval_loss": 0.6657664179801941,
      "eval_runtime": 15.8042,
      "eval_samples_per_second": 133.888,
      "eval_steps_per_second": 4.239,
      "step": 1587
    },
    {
      "epoch": 3.78,
      "learning_rate": 3.1096408317580344e-05,
      "loss": 0.1316,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.79347825050354,
      "eval_loss": 0.910639226436615,
      "eval_runtime": 15.913,
      "eval_samples_per_second": 132.973,
      "eval_steps_per_second": 4.21,
      "step": 2116
    },
    {
      "epoch": 4.73,
      "learning_rate": 2.637051039697543e-05,
      "loss": 0.0853,
      "step": 2500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8057655692100525,
      "eval_loss": 1.0349944829940796,
      "eval_runtime": 15.8236,
      "eval_samples_per_second": 133.724,
      "eval_steps_per_second": 4.234,
      "step": 2645
    },
    {
      "epoch": 5.67,
      "learning_rate": 2.1644612476370512e-05,
      "loss": 0.0593,
      "step": 3000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8010396957397461,
      "eval_loss": 1.2216821908950806,
      "eval_runtime": 15.8613,
      "eval_samples_per_second": 133.407,
      "eval_steps_per_second": 4.224,
      "step": 3174
    },
    {
      "epoch": 6.62,
      "learning_rate": 1.6918714555765597e-05,
      "loss": 0.0384,
      "step": 3500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8010396957397461,
      "eval_loss": 1.3548704385757446,
      "eval_runtime": 15.9719,
      "eval_samples_per_second": 132.483,
      "eval_steps_per_second": 4.195,
      "step": 3703
    },
    {
      "epoch": 7.56,
      "learning_rate": 1.2192816635160681e-05,
      "loss": 0.0231,
      "step": 4000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8024574518203735,
      "eval_loss": 1.385912537574768,
      "eval_runtime": 15.9537,
      "eval_samples_per_second": 132.633,
      "eval_steps_per_second": 4.2,
      "step": 4232
    },
    {
      "epoch": 8.51,
      "learning_rate": 7.466918714555767e-06,
      "loss": 0.0142,
      "step": 4500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8062381744384766,
      "eval_loss": 1.4750466346740723,
      "eval_runtime": 15.969,
      "eval_samples_per_second": 132.507,
      "eval_steps_per_second": 4.196,
      "step": 4761
    },
    {
      "epoch": 9.45,
      "learning_rate": 2.741020793950851e-06,
      "loss": 0.0118,
      "step": 5000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8090737462043762,
      "eval_loss": 1.4761372804641724,
      "eval_runtime": 15.9022,
      "eval_samples_per_second": 133.064,
      "eval_steps_per_second": 4.213,
      "step": 5290
    },
    {
      "epoch": 10.0,
      "step": 5290,
      "total_flos": 1.1136143848628736e+16,
      "train_loss": 0.1654055107068024,
      "train_runtime": 5360.0658,
      "train_samples_per_second": 31.585,
      "train_steps_per_second": 0.987
    }
  ],
  "max_steps": 5290,
  "num_train_epochs": 10,
  "total_flos": 1.1136143848628736e+16,
  "trial_name": null,
  "trial_params": null
}