last-checkpoint/trainer_state.json

{
  "best_metric": 1.7246909141540527,
  "best_model_checkpoint": "/home3/s5431786/nlp-final-project/results/google-flan-t5-small-e-snli-generation-explanation_use_prompt_label-selected-b64/checkpoint-6000",
  "epoch": 0.6989748369058714,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 0.0004659832246039143,
      "loss": 1.6051,
      "step": 2000
    },
    {
      "epoch": 0.23,
      "eval_accuracy": null,
      "eval_bertscore_f1": 0.9312730214807033,
      "eval_bleu": 0.39782836950988876,
      "eval_f1": null,
      "eval_loss": 1.7780297994613647,
      "eval_rouge1": 0.5911787258201253,
      "eval_rouge2": 0.38303308637957223,
      "eval_rougeL": 0.5377264257708202,
      "eval_rougeLsum": 0.5395006217014338,
      "eval_runtime": 168.0702,
      "eval_samples_per_second": 58.559,
      "eval_steps_per_second": 0.916,
      "step": 2000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0009319664492078286,
      "loss": 1.4342,
      "step": 4000
    },
    {
      "epoch": 0.47,
      "eval_accuracy": null,
      "eval_bertscore_f1": 0.933153417701039,
      "eval_bleu": 0.4002469468404885,
      "eval_f1": null,
      "eval_loss": 1.7909314632415771,
      "eval_rouge1": 0.5975097321291585,
      "eval_rouge2": 0.3847152683676083,
      "eval_rougeL": 0.5420448625219438,
      "eval_rougeLsum": 0.5439086899907004,
      "eval_runtime": 163.5422,
      "eval_samples_per_second": 60.18,
      "eval_steps_per_second": 0.942,
      "step": 4000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0009790552803256977,
      "loss": 1.3804,
      "step": 6000
    },
    {
      "epoch": 0.7,
      "eval_accuracy": null,
      "eval_bertscore_f1": 0.9348857560079773,
      "eval_bleu": 0.41900709929386093,
      "eval_f1": null,
      "eval_loss": 1.7246909141540527,
      "eval_rouge1": 0.616316027086431,
      "eval_rouge2": 0.4107303410412238,
      "eval_rougeL": 0.5610382702035865,
      "eval_rougeLsum": 0.5630504299315114,
      "eval_runtime": 164.6404,
      "eval_samples_per_second": 59.779,
      "eval_steps_per_second": 0.935,
      "step": 6000
    }
  ],
  "max_steps": 85840,
  "num_train_epochs": 10,
  "total_flos": 9074198514499584.0,
  "trial_name": null,
  "trial_params": null
}