trainer_state.json

{
  "best_metric": 17.2861,
  "best_model_checkpoint": "./out/finetuned_models/xsum_aligned_smallT5_full/checkpoint-10",
  "epoch": 0.0007842214641414735,
  "eval_steps": 5,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "eval_gen_len": 46.2436655778229,
      "eval_loss": 2.6444404125213623,
      "eval_rouge1": 22.3341,
      "eval_rouge2": 4.3395,
      "eval_rougeL": 16.2507,
      "eval_rougeLsum": 17.8303,
      "eval_runtime": 1003.3723,
      "eval_samples_per_second": 11.289,
      "eval_steps_per_second": 1.411,
      "step": 5
    },
    {
      "epoch": 0.0,
      "eval_gen_len": 33.63662046437715,
      "eval_loss": 2.4093151092529297,
      "eval_rouge1": 22.8498,
      "eval_rouge2": 4.7818,
      "eval_rougeL": 17.2861,
      "eval_rougeLsum": 18.0665,
      "eval_runtime": 845.1794,
      "eval_samples_per_second": 13.402,
      "eval_steps_per_second": 1.675,
      "step": 10
    },
    {
      "epoch": 0.0,
      "step": 10,
      "total_flos": 41941155446784.0,
      "train_loss": 3.090506744384766,
      "train_runtime": 1864.227,
      "train_samples_per_second": 0.086,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 41941155446784.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}