trainer_state.json

{
  "best_metric": 32.4749,
  "best_model_checkpoint": "Models/t5-base-medium-title-generation/checkpoint-2000",
  "epoch": 0.18705574261129818,
  "eval_steps": 2000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18705574261129818,
      "grad_norm": 4.753298282623291,
      "learning_rate": 3.2555181444070336e-05,
      "loss": 2.591,
      "step": 2000
    },
    {
      "epoch": 0.18705574261129818,
      "eval_gen_len": 11.6682,
      "eval_loss": 2.3164875507354736,
      "eval_rouge1": 32.4749,
      "eval_rouge2": 16.8238,
      "eval_rougeL": 29.9745,
      "eval_rougeLsum": 30.1279,
      "eval_runtime": 68.6794,
      "eval_samples_per_second": 12.507,
      "eval_steps_per_second": 1.573,
      "step": 2000
    }
  ],
  "logging_steps": 2000,
  "max_steps": 10692,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.946608113917952e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}