trainer_state.json

{
  "best_metric": 0.9413902212681269,
  "best_model_checkpoint": "model_saves/deberta-large_lemon-spell_5k_5_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": 0.9413063652321694,
      "eval_loss": 0.4176751971244812,
      "eval_runtime": 4.6415,
      "eval_samples_per_second": 944.087,
      "eval_steps_per_second": 7.541,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9413902212681269,
      "eval_loss": 0.44149360060691833,
      "eval_runtime": 4.6565,
      "eval_samples_per_second": 941.053,
      "eval_steps_per_second": 7.516,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.940546070506155,
      "eval_loss": 0.47912463545799255,
      "eval_runtime": 4.6556,
      "eval_samples_per_second": 941.237,
      "eval_steps_per_second": 7.518,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2552,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9403000928006798,
      "eval_loss": 0.5206338167190552,
      "eval_runtime": 4.6448,
      "eval_samples_per_second": 943.417,
      "eval_steps_per_second": 7.535,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9397075101465804,
      "eval_loss": 0.5826528072357178,
      "eval_runtime": 4.6432,
      "eval_samples_per_second": 943.748,
      "eval_steps_per_second": 7.538,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.009276016177971e+16,
      "train_loss": 0.2226787766413902,
      "train_runtime": 573.2437,
      "train_samples_per_second": 897.629,
      "train_steps_per_second": 7.013
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.009276016177971e+16,
  "trial_name": null,
  "trial_params": null
}