trainer_state.json

{
  "best_metric": 0.601811408996582,
  "best_model_checkpoint": "hBERTv1_new_pretrain_mrpc/checkpoint-116",
  "epoch": 9.0,
  "global_step": 261,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.6855,
      "step": 29
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6254739165306091,
      "eval_runtime": 0.7416,
      "eval_samples_per_second": 550.143,
      "eval_steps_per_second": 5.394,
      "step": 29
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.647,
      "step": 58
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6535907983779907,
      "eval_runtime": 0.7427,
      "eval_samples_per_second": 549.37,
      "eval_steps_per_second": 5.386,
      "step": 58
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6336,
      "step": 87
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6536518335342407,
      "eval_runtime": 0.7412,
      "eval_samples_per_second": 550.49,
      "eval_steps_per_second": 5.397,
      "step": 87
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6007,
      "step": 116
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.601811408996582,
      "eval_runtime": 0.7403,
      "eval_samples_per_second": 551.095,
      "eval_steps_per_second": 5.403,
      "step": 116
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.5196,
      "step": 145
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6544117647058824,
      "eval_combined_score": 0.6908422459893049,
      "eval_f1": 0.7272727272727273,
      "eval_loss": 0.6852147579193115,
      "eval_runtime": 0.7429,
      "eval_samples_per_second": 549.188,
      "eval_steps_per_second": 5.384,
      "step": 145
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.3703,
      "step": 174
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7273469334447811,
      "eval_f1": 0.7708703374777975,
      "eval_loss": 0.716745138168335,
      "eval_runtime": 0.7419,
      "eval_samples_per_second": 549.958,
      "eval_steps_per_second": 5.392,
      "step": 174
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.2697,
      "step": 203
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7009803921568627,
      "eval_combined_score": 0.7481412027898408,
      "eval_f1": 0.7953020134228188,
      "eval_loss": 0.9072341918945312,
      "eval_runtime": 0.7436,
      "eval_samples_per_second": 548.694,
      "eval_steps_per_second": 5.379,
      "step": 203
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.1997,
      "step": 232
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6764705882352942,
      "eval_combined_score": 0.7207975716977182,
      "eval_f1": 0.7651245551601423,
      "eval_loss": 1.046744465827942,
      "eval_runtime": 0.7418,
      "eval_samples_per_second": 550.012,
      "eval_steps_per_second": 5.392,
      "step": 232
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 0.1629,
      "step": 261
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6593137254901961,
      "eval_combined_score": 0.7044316375198727,
      "eval_f1": 0.7495495495495494,
      "eval_loss": 1.0808790922164917,
      "eval_runtime": 0.7458,
      "eval_samples_per_second": 547.093,
      "eval_steps_per_second": 5.364,
      "step": 261
    },
    {
      "epoch": 9.0,
      "step": 261,
      "total_flos": 4888970768941056.0,
      "train_loss": 0.4543412891841026,
      "train_runtime": 215.8989,
      "train_samples_per_second": 849.472,
      "train_steps_per_second": 6.716
    }
  ],
  "max_steps": 1450,
  "num_train_epochs": 50,
  "total_flos": 4888970768941056.0,
  "trial_name": null,
  "trial_params": null
}