trainer_state.json

{
  "best_metric": 0.026788420975208282,
  "best_model_checkpoint": "./roberta-large-condaqa-neg-tag-token-classifier/checkpoint-24",
  "epoch": 6.0,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9587528418317636,
      "eval_f1": 0.0,
      "eval_loss": 0.15255515277385712,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8569,
      "eval_samples_per_second": 66.777,
      "eval_steps_per_second": 2.154,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9587528418317636,
      "eval_f1": 0.0,
      "eval_loss": 0.0874876081943512,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 2.0477,
      "eval_samples_per_second": 60.555,
      "eval_steps_per_second": 1.953,
      "step": 8
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9876583306268268,
      "eval_f1": 0.0,
      "eval_loss": 0.03961525484919548,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 2.0055,
      "eval_samples_per_second": 61.83,
      "eval_steps_per_second": 1.995,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9899317960376746,
      "eval_f1": 0.0,
      "eval_loss": 0.03219746798276901,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8577,
      "eval_samples_per_second": 66.749,
      "eval_steps_per_second": 2.153,
      "step": 16
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9905813575836311,
      "eval_f1": 0.0,
      "eval_loss": 0.026981903240084648,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8937,
      "eval_samples_per_second": 65.479,
      "eval_steps_per_second": 2.112,
      "step": 20
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9899317960376746,
      "eval_f1": 0.0,
      "eval_loss": 0.026788420975208282,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8835,
      "eval_samples_per_second": 65.835,
      "eval_steps_per_second": 2.124,
      "step": 24
    },
    {
      "epoch": 6.0,
      "step": 24,
      "total_flos": 897265181578104.0,
      "train_loss": 0.13241314888000488,
      "train_runtime": 520.3829,
      "train_samples_per_second": 11.034,
      "train_steps_per_second": 0.046
    }
  ],
  "max_steps": 24,
  "num_train_epochs": 6,
  "total_flos": 897265181578104.0,
  "trial_name": null,
  "trial_params": null
}