trainer_state.json

{
  "best_metric": 0.9395845212938427,
  "best_model_checkpoint": "model_saves/xlnet-large_lemon-spell_5k_3_p3/checkpoint-536",
  "epoch": 5.0,
  "global_step": 1340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9392379163452186,
      "eval_loss": 0.44053345918655396,
      "eval_runtime": 4.7501,
      "eval_samples_per_second": 922.513,
      "eval_steps_per_second": 7.368,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9395845212938427,
      "eval_loss": 0.4633668065071106,
      "eval_runtime": 4.7229,
      "eval_samples_per_second": 927.829,
      "eval_steps_per_second": 7.411,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9390925658828922,
      "eval_loss": 0.498786985874176,
      "eval_runtime": 4.7484,
      "eval_samples_per_second": 922.834,
      "eval_steps_per_second": 7.371,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2992,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9394335804291193,
      "eval_loss": 0.5408743023872375,
      "eval_runtime": 4.7473,
      "eval_samples_per_second": 923.047,
      "eval_steps_per_second": 7.373,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9392602779548072,
      "eval_loss": 0.5943452715873718,
      "eval_runtime": 4.7987,
      "eval_samples_per_second": 913.169,
      "eval_steps_per_second": 7.294,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.127979812277453e+16,
      "train_loss": 0.2714570799870277,
      "train_runtime": 550.3629,
      "train_samples_per_second": 934.947,
      "train_steps_per_second": 7.304
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.127979812277453e+16,
  "trial_name": null,
  "trial_params": null
}