trainer_state.json

{
  "best_metric": 54.0,
  "best_model_checkpoint": "outputs/bitfit/t5-base/superglue-copa/checkpoint-200",
  "epoch": 40.0,
  "global_step": 520,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.85,
      "eval_accuracy": 50.0,
      "eval_average_metrics": 50.0,
      "eval_loss": 0.2800115942955017,
      "eval_runtime": 0.3692,
      "eval_samples_per_second": 135.418,
      "step": 50
    },
    {
      "epoch": 7.69,
      "eval_accuracy": 42.0,
      "eval_average_metrics": 42.0,
      "eval_loss": 0.28102436661720276,
      "eval_runtime": 0.4452,
      "eval_samples_per_second": 112.304,
      "step": 100
    },
    {
      "epoch": 11.54,
      "eval_accuracy": 52.0,
      "eval_average_metrics": 52.0,
      "eval_loss": 0.3020976185798645,
      "eval_runtime": 0.3114,
      "eval_samples_per_second": 160.544,
      "step": 150
    },
    {
      "epoch": 15.38,
      "eval_accuracy": 54.0,
      "eval_average_metrics": 54.0,
      "eval_loss": 0.29293137788772583,
      "eval_runtime": 0.5099,
      "eval_samples_per_second": 98.067,
      "step": 200
    },
    {
      "epoch": 19.23,
      "eval_accuracy": 50.0,
      "eval_average_metrics": 50.0,
      "eval_loss": 0.4718337953090668,
      "eval_runtime": 0.2966,
      "eval_samples_per_second": 168.586,
      "step": 250
    },
    {
      "epoch": 23.08,
      "eval_accuracy": 48.0,
      "eval_average_metrics": 48.0,
      "eval_loss": 0.6855034828186035,
      "eval_runtime": 0.3659,
      "eval_samples_per_second": 136.664,
      "step": 300
    },
    {
      "epoch": 26.92,
      "eval_accuracy": 48.0,
      "eval_average_metrics": 48.0,
      "eval_loss": 0.9259137511253357,
      "eval_runtime": 0.5574,
      "eval_samples_per_second": 89.705,
      "step": 350
    },
    {
      "epoch": 30.77,
      "eval_accuracy": 44.0,
      "eval_average_metrics": 44.0,
      "eval_loss": 1.0988849401474,
      "eval_runtime": 0.3043,
      "eval_samples_per_second": 164.333,
      "step": 400
    },
    {
      "epoch": 34.62,
      "eval_accuracy": 50.0,
      "eval_average_metrics": 50.0,
      "eval_loss": 1.1635206937789917,
      "eval_runtime": 0.3906,
      "eval_samples_per_second": 128.015,
      "step": 450
    },
    {
      "epoch": 38.46,
      "learning_rate": 1.1538461538461538e-05,
      "loss": 0.19,
      "step": 500
    },
    {
      "epoch": 38.46,
      "eval_accuracy": 48.0,
      "eval_average_metrics": 48.0,
      "eval_loss": 1.2427139282226562,
      "eval_runtime": 0.3254,
      "eval_samples_per_second": 153.656,
      "step": 500
    },
    {
      "epoch": 40.0,
      "step": 520,
      "total_flos": 954277186351104.0,
      "train_loss": 0.18321549898156753,
      "train_runtime": 219.7861,
      "train_samples_per_second": 72.798,
      "train_steps_per_second": 2.366
    }
  ],
  "max_steps": 520,
  "num_train_epochs": 40,
  "total_flos": 954277186351104.0,
  "trial_name": null,
  "trial_params": null
}