trainer_state.json

{
  "best_metric": 0.9399702590592471,
  "best_model_checkpoint": "model_saves/xlnet-large_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": 0.939696329341786,
      "eval_loss": 0.4382520616054535,
      "eval_runtime": 4.7794,
      "eval_samples_per_second": 916.855,
      "eval_steps_per_second": 7.323,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9399702590592471,
      "eval_loss": 0.4678247570991516,
      "eval_runtime": 4.7915,
      "eval_samples_per_second": 914.541,
      "eval_steps_per_second": 7.305,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.939696329341786,
      "eval_loss": 0.491964191198349,
      "eval_runtime": 4.7989,
      "eval_samples_per_second": 913.133,
      "eval_steps_per_second": 7.293,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2974,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9389528058229631,
      "eval_loss": 0.5350586175918579,
      "eval_runtime": 4.749,
      "eval_samples_per_second": 922.725,
      "eval_steps_per_second": 7.37,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9388242265678284,
      "eval_loss": 0.5907213687896729,
      "eval_runtime": 4.7376,
      "eval_samples_per_second": 924.934,
      "eval_steps_per_second": 7.388,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.124454503120896e+16,
      "train_loss": 0.269802509137054,
      "train_runtime": 552.5128,
      "train_samples_per_second": 931.309,
      "train_steps_per_second": 7.276
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.124454503120896e+16,
  "trial_name": null,
  "trial_params": null
}