trainer_state.json

{
  "best_metric": 0.9425083017475598,
  "best_model_checkpoint": "model_saves/deberta-v3-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": 0.9421505159941412,
      "eval_loss": 0.42014917731285095,
      "eval_runtime": 4.637,
      "eval_samples_per_second": 945.015,
      "eval_steps_per_second": 7.548,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9425083017475598,
      "eval_loss": 0.4309523403644562,
      "eval_runtime": 4.6317,
      "eval_samples_per_second": 946.082,
      "eval_steps_per_second": 7.557,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.941121881953063,
      "eval_loss": 0.4580814242362976,
      "eval_runtime": 4.6382,
      "eval_samples_per_second": 944.765,
      "eval_steps_per_second": 7.546,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2884,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9403559968246514,
      "eval_loss": 0.4939667880535126,
      "eval_runtime": 4.6359,
      "eval_samples_per_second": 945.231,
      "eval_steps_per_second": 7.55,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9406187457373182,
      "eval_loss": 0.5340861678123474,
      "eval_runtime": 4.642,
      "eval_samples_per_second": 943.988,
      "eval_steps_per_second": 7.54,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 1.7256130287763456e+16,
      "train_loss": 0.25945684233708166,
      "train_runtime": 598.3689,
      "train_samples_per_second": 859.938,
      "train_steps_per_second": 6.718
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 1.7256130287763456e+16,
  "trial_name": null,
  "trial_params": null
}