trainer_state.json

{
  "best_metric": 0.9402609599838997,
  "best_model_checkpoint": "model_saves/xlnet-large_spell_10k_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": 0.9399423070472612,
      "eval_loss": 0.4410298764705658,
      "eval_runtime": 4.8733,
      "eval_samples_per_second": 899.177,
      "eval_steps_per_second": 7.182,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9402609599838997,
      "eval_loss": 0.44978341460227966,
      "eval_runtime": 5.0594,
      "eval_samples_per_second": 866.102,
      "eval_steps_per_second": 6.918,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9393385435883674,
      "eval_loss": 0.4808991849422455,
      "eval_runtime": 4.8831,
      "eval_samples_per_second": 897.377,
      "eval_steps_per_second": 7.168,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.323,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9394279900267222,
      "eval_loss": 0.5203689932823181,
      "eval_runtime": 4.8771,
      "eval_samples_per_second": 898.49,
      "eval_steps_per_second": 7.176,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9399255358400698,
      "eval_loss": 0.5856708288192749,
      "eval_runtime": 4.8903,
      "eval_samples_per_second": 896.057,
      "eval_steps_per_second": 7.157,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.1552156360310784e+16,
      "train_loss": 0.2938878016685372,
      "train_runtime": 558.6733,
      "train_samples_per_second": 921.039,
      "train_steps_per_second": 7.196
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.1552156360310784e+16,
  "trial_name": null,
  "trial_params": null
}