{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9872386641325006,
  "global_step": 5500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 1.8203874705775847e-05,
      "loss": 1.5351,
      "step": 500
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.639326453014666e-05,
      "loss": 1.418,
      "step": 1000
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4586275574868732e-05,
      "loss": 1.3833,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 25.9756,
      "eval_loss": 1.4744428396224976,
      "eval_rouge1": 53.2728,
      "eval_rouge2": 28.5907,
      "eval_rougeL": 43.9891,
      "eval_rougeLsum": 48.6797,
      "eval_runtime": 208.0144,
      "eval_samples_per_second": 3.932,
      "eval_steps_per_second": 0.986,
      "step": 1841
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2775665399239545e-05,
      "loss": 1.2657,
      "step": 2000
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0965055223610357e-05,
      "loss": 1.0604,
      "step": 2500
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.15444504798117e-06,
      "loss": 1.0653,
      "step": 3000
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.343834872351984e-06,
      "loss": 1.0762,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 30.9328,
      "eval_loss": 1.4248533248901367,
      "eval_rouge1": 54.4293,
      "eval_rouge2": 29.3125,
      "eval_rougeL": 44.6626,
      "eval_rougeLsum": 50.0031,
      "eval_runtime": 237.2771,
      "eval_samples_per_second": 3.447,
      "eval_steps_per_second": 0.864,
      "step": 3682
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5332246967227965e-06,
      "loss": 0.9395,
      "step": 4000
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7262357414448676e-06,
      "loss": 0.8509,
      "step": 4500
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.91562556581568e-06,
      "loss": 0.8581,
      "step": 5000
    },
    {
      "epoch": 2.99,
      "learning_rate": 1.0501539018649285e-07,
      "loss": 0.8555,
      "step": 5500
    }
  ],
  "max_steps": 5523,
  "num_train_epochs": 3,
  "total_flos": 2.597261321227469e+16,
  "trial_name": null,
  "trial_params": null
}