trainer_state.json

{
  "best_metric": 0.9396683773298002,
  "best_model_checkpoint": "model_saves/xlnet-large_lemon-spell_10k_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": 0.93925468755241,
      "eval_loss": 0.4631763696670532,
      "eval_runtime": 4.8353,
      "eval_samples_per_second": 906.259,
      "eval_steps_per_second": 7.238,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9396683773298002,
      "eval_loss": 0.48520079255104065,
      "eval_runtime": 4.8367,
      "eval_samples_per_second": 905.981,
      "eval_steps_per_second": 7.236,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9395565692818569,
      "eval_loss": 0.5397723913192749,
      "eval_runtime": 4.8298,
      "eval_samples_per_second": 907.277,
      "eval_steps_per_second": 7.247,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2685,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.938913673006183,
      "eval_loss": 0.5819151997566223,
      "eval_runtime": 4.8167,
      "eval_samples_per_second": 909.744,
      "eval_steps_per_second": 7.266,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9380527510370197,
      "eval_loss": 0.6364547610282898,
      "eval_runtime": 4.8139,
      "eval_samples_per_second": 910.288,
      "eval_steps_per_second": 7.271,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.1554750788993024e+16,
      "train_loss": 0.2428875709647563,
      "train_runtime": 555.5784,
      "train_samples_per_second": 926.17,
      "train_steps_per_second": 7.236
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.1554750788993024e+16,
  "trial_name": null,
  "trial_params": null
}