multi_class_model/trainer_state.json

{
  "best_metric": 1.0723919868469238,
  "best_model_checkpoint": "clf_trainer/checkpoint-6560",
  "epoch": 2.0,
  "global_step": 6560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.5e-05,
      "loss": 1.1784,
      "step": 3280
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6109756097560975,
      "eval_loss": 1.100429654121399,
      "eval_runtime": 43.9741,
      "eval_samples_per_second": 186.474,
      "eval_steps_per_second": 18.647,
      "step": 3280
    },
    {
      "epoch": 2.0,
      "learning_rate": 4e-05,
      "loss": 1.0022,
      "step": 6560
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6201219512195122,
      "eval_loss": 1.0723919868469238,
      "eval_runtime": 43.854,
      "eval_samples_per_second": 186.984,
      "eval_steps_per_second": 18.698,
      "step": 6560
    }
  ],
  "max_steps": 32800,
  "num_train_epochs": 10,
  "total_flos": 8690946151219200.0,
  "trial_name": null,
  "trial_params": null
}