training_checkpoints/checkpoint-643/trainer_state.json

{
  "best_metric": 0.6806109083601769,
  "best_model_checkpoint": "tiny-llama-lora-no-grad/checkpoint-643",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 643,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.78,
      "learning_rate": 4.740798341109383e-05,
      "loss": 1.1276,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8086754453911696,
      "eval_f1_macro": 0.6806109083601769,
      "eval_f1_micro": 0.8086754453911695,
      "eval_f1_weighted": 0.8034368965438116,
      "eval_loss": 0.6704964637756348,
      "eval_macro_fpr": 0.017184987145855975,
      "eval_macro_sensitivity": 0.6852607939619662,
      "eval_macro_specificity": 0.985525532858602,
      "eval_precision": 0.8055191978891799,
      "eval_precision_macro": 0.7052877886582329,
      "eval_recall": 0.8086754453911696,
      "eval_recall_macro": 0.6852607939619662,
      "eval_runtime": 225.0756,
      "eval_samples_per_second": 5.736,
      "eval_steps_per_second": 0.72,
      "eval_weighted_fpr": 0.016618448496265894,
      "eval_weighted_sensitivity": 0.8086754453911696,
      "eval_weighted_specificity": 0.9742075474878604,
      "step": 643
    }
  ],
  "logging_steps": 500,
  "max_steps": 9645,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 1.5494356081311744e+16,
  "trial_name": null,
  "trial_params": null
}