training_checkpoints/checkpoint-643/trainer_state.json

{
  "best_metric": 0.7415140921427528,
  "best_model_checkpoint": "tiny-llama-lora-new/checkpoint-643",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 643,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8001549186676995,
      "eval_f1_macro": 0.6790360166213908,
      "eval_f1_micro": 0.8001549186676995,
      "eval_f1_weighted": 0.7959404181222782,
      "eval_loss": 0.6614832282066345,
      "eval_macro_fpr": 0.01824293967244035,
      "eval_macro_sensitivity": 0.6677834011062839,
      "eval_macro_specificity": 0.9848478627893485,
      "eval_precision": 0.8040294558027004,
      "eval_precision_macro": 0.7265846774862527,
      "eval_recall": 0.8001549186676995,
      "eval_recall_macro": 0.6677834011062839,
      "eval_runtime": 225.9328,
      "eval_samples_per_second": 5.714,
      "eval_steps_per_second": 0.717,
      "eval_weighted_fpr": 0.017527173913043478,
      "eval_weighted_sensitivity": 0.8001549186676995,
      "eval_weighted_specificity": 0.97256302317253,
      "step": 160
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8063516653756778,
      "eval_f1_macro": 0.723523893027765,
      "eval_f1_micro": 0.8063516653756778,
      "eval_f1_weighted": 0.8039099652383133,
      "eval_loss": 0.6995529532432556,
      "eval_macro_fpr": 0.01771710819775641,
      "eval_macro_sensitivity": 0.7207263437235008,
      "eval_macro_specificity": 0.9853387258716434,
      "eval_precision": 0.8110438944279544,
      "eval_precision_macro": 0.7448100337436813,
      "eval_recall": 0.8063516653756778,
      "eval_recall_macro": 0.7207263437235008,
      "eval_runtime": 230.7469,
      "eval_samples_per_second": 5.595,
      "eval_steps_per_second": 0.702,
      "eval_weighted_fpr": 0.01686454398273071,
      "eval_weighted_sensitivity": 0.8063516653756778,
      "eval_weighted_specificity": 0.9737292226989717,
      "step": 321
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8125484120836561,
      "eval_f1_macro": 0.7179663090811256,
      "eval_f1_micro": 0.8125484120836561,
      "eval_f1_weighted": 0.8085118934271095,
      "eval_loss": 0.8202366828918457,
      "eval_macro_fpr": 0.017097652587672894,
      "eval_macro_sensitivity": 0.7080107913825292,
      "eval_macro_specificity": 0.985578595986596,
      "eval_precision": 0.8118501609528374,
      "eval_precision_macro": 0.7576540712122397,
      "eval_recall": 0.8125484120836561,
      "eval_recall_macro": 0.7080107913825292,
      "eval_runtime": 231.655,
      "eval_samples_per_second": 5.573,
      "eval_steps_per_second": 0.699,
      "eval_weighted_fpr": 0.016211146838156484,
      "eval_weighted_sensitivity": 0.8125484120836561,
      "eval_weighted_specificity": 0.9711305277152844,
      "step": 482
    },
    {
      "epoch": 3.11,
      "learning_rate": 3.958333333333333e-05,
      "loss": 0.2932,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8140975987606507,
      "eval_f1_macro": 0.7415140921427528,
      "eval_f1_micro": 0.8140975987606507,
      "eval_f1_weighted": 0.8153735617531811,
      "eval_loss": 0.9493006467819214,
      "eval_macro_fpr": 0.016573672675540378,
      "eval_macro_sensitivity": 0.7326751453453961,
      "eval_macro_specificity": 0.9859001747376132,
      "eval_precision": 0.8204052204072106,
      "eval_precision_macro": 0.759251934556511,
      "eval_recall": 0.8140975987606507,
      "eval_recall_macro": 0.7326751453453961,
      "eval_runtime": 230.2371,
      "eval_samples_per_second": 5.607,
      "eval_steps_per_second": 0.704,
      "eval_weighted_fpr": 0.016049217600641968,
      "eval_weighted_sensitivity": 0.8140975987606507,
      "eval_weighted_specificity": 0.974405022303545,
      "step": 643
    }
  ],
  "logging_steps": 500,
  "max_steps": 2400,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 6.197742432524698e+16,
  "trial_name": null,
  "trial_params": null
}