training_checkpoints/checkpoint-160/trainer_state.json

{
  "best_metric": 0.602571869042318,
  "best_model_checkpoint": "tiny-llama/checkpoint-160",
  "epoch": 0.995334370139969,
  "eval_steps": 500,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7869868319132456,
      "eval_f1_macro": 0.602571869042318,
      "eval_f1_micro": 0.7869868319132455,
      "eval_f1_weighted": 0.7752313093692699,
      "eval_loss": 0.6672300100326538,
      "eval_macro_fpr": 0.019618845690595806,
      "eval_macro_sensitivity": 0.6223339304749645,
      "eval_macro_specificity": 0.9839698388016963,
      "eval_precision": 0.7799565431301605,
      "eval_precision_macro": 0.6206625059288113,
      "eval_recall": 0.7869868319132456,
      "eval_recall_macro": 0.6223339304749645,
      "eval_runtime": 1156.4947,
      "eval_samples_per_second": 1.116,
      "eval_steps_per_second": 0.14,
      "eval_weighted_fpr": 0.018966825298296436,
      "eval_weighted_sensitivity": 0.7869868319132456,
      "eval_weighted_specificity": 0.9725607501121987,
      "step": 160
    }
  ],
  "logging_steps": 500,
  "max_steps": 1600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 6.197742432524698e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}