trainer_state.json

{
  "best_metric": 0.9413231364393609,
  "best_model_checkpoint": "model_saves/deberta-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": 0.9407696866020416,
      "eval_loss": 0.41556528210639954,
      "eval_runtime": 4.6348,
      "eval_samples_per_second": 945.458,
      "eval_steps_per_second": 7.552,
      "step": 268
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9413231364393609,
      "eval_loss": 0.4426611661911011,
      "eval_runtime": 4.6378,
      "eval_samples_per_second": 944.852,
      "eval_steps_per_second": 7.547,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9406522881517012,
      "eval_loss": 0.47103145718574524,
      "eval_runtime": 4.6337,
      "eval_samples_per_second": 945.689,
      "eval_steps_per_second": 7.553,
      "step": 804
    },
    {
      "epoch": 3.73,
      "learning_rate": 1e-05,
      "loss": 0.2543,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9397242813537718,
      "eval_loss": 0.5292877554893494,
      "eval_runtime": 4.6416,
      "eval_samples_per_second": 944.076,
      "eval_steps_per_second": 7.541,
      "step": 1072
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9391428795044667,
      "eval_loss": 0.5923100709915161,
      "eval_runtime": 4.6379,
      "eval_samples_per_second": 944.827,
      "eval_steps_per_second": 7.547,
      "step": 1340
    },
    {
      "epoch": 5.0,
      "step": 1340,
      "total_flos": 2.0108696705040384e+16,
      "train_loss": 0.22194259700490468,
      "train_runtime": 572.4463,
      "train_samples_per_second": 898.879,
      "train_steps_per_second": 7.022
    }
  ],
  "max_steps": 4020,
  "num_train_epochs": 15,
  "total_flos": 2.0108696705040384e+16,
  "trial_name": null,
  "trial_params": null
}