{
  "best_metric": 0.739513099193573,
  "best_model_checkpoint": "LLM-bartlarge-1/checkpoint-1681",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1681,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 30.921918869018555,
      "learning_rate": 1.6336633663366337e-05,
      "loss": 1.7556,
      "step": 170
    },
    {
      "epoch": 0.2,
      "grad_norm": 12.515706062316895,
      "learning_rate": 3.306930693069307e-05,
      "loss": 0.1761,
      "step": 340
    },
    {
      "epoch": 0.3,
      "grad_norm": 20.075252532958984,
      "learning_rate": 4.9900990099009906e-05,
      "loss": 0.2054,
      "step": 510
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.381262302398682,
      "learning_rate": 4.813794623182018e-05,
      "loss": 0.2198,
      "step": 680
    },
    {
      "epoch": 0.51,
      "grad_norm": 16.09966468811035,
      "learning_rate": 4.626487439400617e-05,
      "loss": 0.2034,
      "step": 850
    },
    {
      "epoch": 0.61,
      "grad_norm": 17.01612663269043,
      "learning_rate": 4.439180255619216e-05,
      "loss": 0.1417,
      "step": 1020
    },
    {
      "epoch": 0.71,
      "grad_norm": 19.581165313720703,
      "learning_rate": 4.251873071837814e-05,
      "loss": 0.1227,
      "step": 1190
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.1578723192214966,
      "learning_rate": 4.064565888056412e-05,
      "loss": 0.1216,
      "step": 1360
    },
    {
      "epoch": 0.91,
      "grad_norm": 25.563976287841797,
      "learning_rate": 3.877258704275011e-05,
      "loss": 0.0994,
      "step": 1530
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 57.763,
      "eval_loss": 0.739513099193573,
      "eval_rouge1": 16.5172,
      "eval_rouge2": 4.4155,
      "eval_rougeL": 16.5116,
      "eval_rougeLsum": 16.5175,
      "eval_runtime": 917.8133,
      "eval_samples_per_second": 7.448,
      "eval_steps_per_second": 0.466,
      "step": 1681
    }
  ],
  "logging_steps": 170,
  "max_steps": 5043,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3641006619623424.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}