trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.875968992248062,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8262548262548263,
      "eval_f1": 0.8262548262548263,
      "eval_loss": 0.6026379466056824,
      "eval_precision": 0.8262548262548263,
      "eval_recall": 0.8262548262548263,
      "eval_runtime": 2.6881,
      "eval_samples_per_second": 96.35,
      "eval_steps_per_second": 12.276,
      "step": 129
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8455598455598455,
      "eval_f1": 0.8455598455598455,
      "eval_loss": 0.5269076228141785,
      "eval_precision": 0.8455598455598455,
      "eval_recall": 0.8455598455598455,
      "eval_runtime": 2.3725,
      "eval_samples_per_second": 109.17,
      "eval_steps_per_second": 13.91,
      "step": 258
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8301158301158301,
      "eval_f1": 0.8301158301158301,
      "eval_loss": 0.6534943580627441,
      "eval_precision": 0.8301158301158301,
      "eval_recall": 0.8301158301158301,
      "eval_runtime": 2.3717,
      "eval_samples_per_second": 109.205,
      "eval_steps_per_second": 13.914,
      "step": 387
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.1240310077519382e-05,
      "loss": 0.3772,
      "step": 500
    }
  ],
  "max_steps": 645,
  "num_train_epochs": 5,
  "total_flos": 205568431200000.0,
  "trial_name": null,
  "trial_params": null
}