trainer_state.json

{
  "best_metric": 0.6239516139030457,
  "best_model_checkpoint": "hBERTv2_data_aug_mrpc/checkpoint-1960",
  "epoch": 7.0,
  "global_step": 6860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 0.6319,
      "step": 980
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6244884133338928,
      "eval_runtime": 0.489,
      "eval_samples_per_second": 834.271,
      "eval_steps_per_second": 4.09,
      "step": 980
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.6305,
      "step": 1960
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6239516139030457,
      "eval_runtime": 0.4884,
      "eval_samples_per_second": 835.412,
      "eval_steps_per_second": 4.095,
      "step": 1960
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.6303,
      "step": 2940
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6259029507637024,
      "eval_runtime": 0.489,
      "eval_samples_per_second": 834.359,
      "eval_steps_per_second": 4.09,
      "step": 2940
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6302,
      "step": 3920
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6251693964004517,
      "eval_runtime": 0.4902,
      "eval_samples_per_second": 832.302,
      "eval_steps_per_second": 4.08,
      "step": 3920
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.500102040816327e-05,
      "loss": 0.6302,
      "step": 4900
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6241449117660522,
      "eval_runtime": 0.4885,
      "eval_samples_per_second": 835.178,
      "eval_steps_per_second": 4.094,
      "step": 4900
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4001020408163265e-05,
      "loss": 0.6302,
      "step": 5880
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6241024136543274,
      "eval_runtime": 0.4883,
      "eval_samples_per_second": 835.632,
      "eval_steps_per_second": 4.096,
      "step": 5880
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.300102040816327e-05,
      "loss": 0.6301,
      "step": 6860
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6838235294117647,
      "eval_combined_score": 0.7480253018237863,
      "eval_f1": 0.8122270742358079,
      "eval_loss": 0.6241885423660278,
      "eval_runtime": 0.4876,
      "eval_samples_per_second": 836.76,
      "eval_steps_per_second": 4.102,
      "step": 6860
    },
    {
      "epoch": 7.0,
      "step": 6860,
      "total_flos": 2.1861021041400218e+17,
      "train_loss": 0.6304825585368076,
      "train_runtime": 6112.7657,
      "train_samples_per_second": 2050.921,
      "train_steps_per_second": 8.016
    }
  ],
  "max_steps": 49000,
  "num_train_epochs": 50,
  "total_flos": 2.1861021041400218e+17,
  "trial_name": null,
  "trial_params": null
}