trainer_state.json

{
  "best_metric": 2.515381097793579,
  "best_model_checkpoint": "flan-t5-base-Amalq/shared_taskA/checkpoint-302",
  "epoch": 2.0,
  "global_step": 302,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 18.97,
      "eval_loss": 2.5547688007354736,
      "eval_rouge1": 27.7971,
      "eval_rouge2": 14.471,
      "eval_rougeL": 27.4683,
      "eval_rougeLsum": 27.617,
      "eval_runtime": 86.9289,
      "eval_samples_per_second": 1.15,
      "eval_steps_per_second": 0.15,
      "step": 151
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 18.85,
      "eval_loss": 2.515381097793579,
      "eval_rouge1": 28.1748,
      "eval_rouge2": 14.384,
      "eval_rougeL": 27.6673,
      "eval_rougeLsum": 27.8465,
      "eval_runtime": 64.4505,
      "eval_samples_per_second": 1.552,
      "eval_steps_per_second": 0.202,
      "step": 302
    }
  ],
  "max_steps": 755,
  "num_train_epochs": 5,
  "total_flos": 1644787160580096.0,
  "trial_name": null,
  "trial_params": null
}