trainer_state.json

{
  "best_metric": 0.9403783584342401,
  "best_model_checkpoint": "model_saves/xlnet-large_lemon-spell_5k_1_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": 0.9399143550352754,
      "eval_loss": 0.4331456124782562,
      "eval_runtime": 4.9037,
      "eval_samples_per_second": 893.606,
      "eval_steps_per_second": 7.137,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9403783584342401,
      "eval_loss": 0.4501822292804718,
      "eval_runtime": 4.78,
      "eval_samples_per_second": 916.736,
      "eval_steps_per_second": 7.322,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9400205726808216,
      "eval_loss": 0.4826386570930481,
      "eval_runtime": 4.713,
      "eval_samples_per_second": 929.76,
      "eval_steps_per_second": 7.426,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.3276,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9399031742304811,
      "eval_loss": 0.5183663964271545,
      "eval_runtime": 4.721,
      "eval_samples_per_second": 928.191,
      "eval_steps_per_second": 7.414,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9392323259428214,
      "eval_loss": 0.560949444770813,
      "eval_runtime": 4.7734,
      "eval_samples_per_second": 918.011,
      "eval_steps_per_second": 7.332,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.12276846002176e+16,
      "train_loss": 0.29810903179111764,
      "train_runtime": 570.8717,
      "train_samples_per_second": 901.358,
      "train_steps_per_second": 7.042
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.12276846002176e+16,
  "trial_name": null,
  "trial_params": null
}