{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4460,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9366385340690613,
      "learning_rate": 0.0008,
      "loss": 0.7738,
      "step": 892
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.5310206413269043,
      "eval_rouge1": 0.704,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.6981,
      "eval_rougeLsum": 0.7001,
      "eval_runtime": 38.4784,
      "eval_samples_per_second": 19.491,
      "eval_steps_per_second": 0.624,
      "step": 892
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.2285586595535278,
      "learning_rate": 0.0006,
      "loss": 0.5926,
      "step": 1784
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.5095808506011963,
      "eval_rouge1": 0.6968,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.6921,
      "eval_rougeLsum": 0.6922,
      "eval_runtime": 38.0389,
      "eval_samples_per_second": 19.717,
      "eval_steps_per_second": 0.631,
      "step": 1784
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.663176953792572,
      "learning_rate": 0.0004,
      "loss": 0.5492,
      "step": 2676
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.4948200285434723,
      "eval_rouge1": 0.6619,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.659,
      "eval_rougeLsum": 0.6573,
      "eval_runtime": 37.9852,
      "eval_samples_per_second": 19.745,
      "eval_steps_per_second": 0.632,
      "step": 2676
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5773711800575256,
      "learning_rate": 0.0002,
      "loss": 0.5199,
      "step": 3568
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.49569639563560486,
      "eval_rouge1": 0.7098,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.7065,
      "eval_rougeLsum": 0.7048,
      "eval_runtime": 38.4364,
      "eval_samples_per_second": 19.513,
      "eval_steps_per_second": 0.624,
      "step": 3568
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6949478983879089,
      "learning_rate": 0.0,
      "loss": 0.4984,
      "step": 4460
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.48721930384635925,
      "eval_rouge1": 0.6882,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.6851,
      "eval_rougeLsum": 0.6829,
      "eval_runtime": 37.7454,
      "eval_samples_per_second": 19.87,
      "eval_steps_per_second": 0.636,
      "step": 4460
    },
    {
      "epoch": 5.0,
      "step": 4460,
      "total_flos": 4.921767498350592e+16,
      "train_loss": 0.5867589565670543,
      "train_runtime": 3933.0288,
      "train_samples_per_second": 18.131,
      "train_steps_per_second": 1.134
    }
  ],
  "logging_steps": 500,
  "max_steps": 4460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 4.921767498350592e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}