trainer_state.json

{
  "best_metric": 0.9399646686568499,
  "best_model_checkpoint": "model_saves/xlnet-large_lemon-spell_10k_2_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": 0.9395230268674739,
      "eval_loss": 0.4556009769439697,
      "eval_runtime": 4.816,
      "eval_samples_per_second": 909.878,
      "eval_steps_per_second": 7.267,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9399646686568499,
      "eval_loss": 0.47372329235076904,
      "eval_runtime": 4.8123,
      "eval_samples_per_second": 910.589,
      "eval_steps_per_second": 7.273,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9398137277921265,
      "eval_loss": 0.5032979846000671,
      "eval_runtime": 4.8023,
      "eval_samples_per_second": 912.478,
      "eval_steps_per_second": 7.288,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2938,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9392994107715873,
      "eval_loss": 0.5553809404373169,
      "eval_runtime": 4.7923,
      "eval_samples_per_second": 914.384,
      "eval_steps_per_second": 7.303,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9390981562852894,
      "eval_loss": 0.6073898077011108,
      "eval_runtime": 4.7991,
      "eval_samples_per_second": 913.094,
      "eval_steps_per_second": 7.293,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.1579393029636096e+16,
      "train_loss": 0.2661068048050155,
      "train_runtime": 557.8515,
      "train_samples_per_second": 922.396,
      "train_steps_per_second": 7.206
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.1579393029636096e+16,
  "trial_name": null,
  "trial_params": null
}