trainer_state.json

{
  "best_metric": 0.005362513940781355,
  "best_model_checkpoint": "data/train-test///model/checkpoint-1676",
  "epoch": 1.0,
  "global_step": 1676,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3,
      "learning_rate": 1.7016706443914083e-05,
      "loss": 0.0394,
      "step": 500
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.4033412887828164e-05,
      "loss": 0.0059,
      "step": 1000
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.1050119331742244e-05,
      "loss": 0.0061,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": null,
      "eval_f1": 0.9930200093066543,
      "eval_loss": 0.005362513940781355,
      "eval_precision": 0.9888785912882299,
      "eval_recall": 0.997196261682243,
      "eval_runtime": 0.3843,
      "eval_samples_per_second": 2523.919,
      "eval_steps_per_second": 158.721,
      "step": 1676
    }
  ],
  "max_steps": 3352,
  "num_train_epochs": 2,
  "total_flos": 775012676912622.0,
  "trial_name": null,
  "trial_params": null
}