checkpoint-1842/trainer_state.json

{
  "best_metric": 1.664982557296753,
  "best_model_checkpoint": "flan-t5-small-samsum/checkpoint-1842",
  "epoch": 1.0,
  "global_step": 1842,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 4.728555917480999e-05,
      "loss": 1.859,
      "step": 500
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.457111834961998e-05,
      "loss": 1.829,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.185667752442997e-05,
      "loss": 1.8116,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 16.714285714285715,
      "eval_loss": 1.664982557296753,
      "eval_rouge1": 42.222,
      "eval_rouge2": 18.4934,
      "eval_rougeL": 35.0441,
      "eval_rougeLsum": 38.4529,
      "eval_runtime": 232.334,
      "eval_samples_per_second": 3.525,
      "eval_steps_per_second": 0.443,
      "step": 1842
    }
  ],
  "max_steps": 9210,
  "num_train_epochs": 5,
  "total_flos": 2738538915299328.0,
  "trial_name": null,
  "trial_params": null
}