{
  "best_metric": 42.3835,
  "best_model_checkpoint": "text_gen/train/checkpoint/t5-large/samsum_42/checkpoint-1842",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4605,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 29.073349633251834,
      "eval_loss": 1.7106927633285522,
      "eval_rouge1": 49.4859,
      "eval_rouge2": 25.6708,
      "eval_rougeL": 40.6695,
      "eval_rougeLsum": 45.4657,
      "eval_runtime": 1388.4704,
      "eval_samples_per_second": 0.589,
      "eval_steps_per_second": 0.009,
      "step": 921
    },
    {
      "epoch": 1.09,
      "grad_norm": 2.411003589630127,
      "learning_rate": 3.914223669923996e-05,
      "loss": 3.053,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 26.381418092909534,
      "eval_loss": 1.459146499633789,
      "eval_rouge1": 50.8056,
      "eval_rouge2": 27.2825,
      "eval_rougeL": 42.3835,
      "eval_rougeLsum": 47.2015,
      "eval_runtime": 1162.1896,
      "eval_samples_per_second": 0.704,
      "eval_steps_per_second": 0.011,
      "step": 1842
    },
    {
      "epoch": 2.17,
      "grad_norm": 2.016871690750122,
      "learning_rate": 2.8284473398479917e-05,
      "loss": 1.8078,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 24.509779951100246,
      "eval_loss": 1.4260163307189941,
      "eval_rouge1": 47.4424,
      "eval_rouge2": 24.5737,
      "eval_rougeL": 38.5731,
      "eval_rougeLsum": 43.77,
      "eval_runtime": 880.791,
      "eval_samples_per_second": 0.929,
      "eval_steps_per_second": 0.015,
      "step": 2763
    },
    {
      "epoch": 3.26,
      "grad_norm": 2.286728858947754,
      "learning_rate": 1.742671009771987e-05,
      "loss": 1.6115,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 24.721271393643033,
      "eval_loss": 1.4194447994232178,
      "eval_rouge1": 47.6099,
      "eval_rouge2": 24.4291,
      "eval_rougeL": 38.6691,
      "eval_rougeLsum": 43.651,
      "eval_runtime": 1053.9525,
      "eval_samples_per_second": 0.776,
      "eval_steps_per_second": 0.012,
      "step": 3684
    },
    {
      "epoch": 4.34,
      "grad_norm": 1.7063580751419067,
      "learning_rate": 6.568946796959827e-06,
      "loss": 1.5189,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 25.162591687041566,
      "eval_loss": 1.418218731880188,
      "eval_rouge1": 47.9053,
      "eval_rouge2": 25.0788,
      "eval_rougeL": 39.0674,
      "eval_rougeLsum": 43.9306,
      "eval_runtime": 969.9541,
      "eval_samples_per_second": 0.843,
      "eval_steps_per_second": 0.013,
      "step": 4605
    },
    {
      "epoch": 5.0,
      "step": 4605,
      "total_flos": 1.3401835354139443e+17,
      "train_loss": 1.930547320337948,
      "train_runtime": 10277.6026,
      "train_samples_per_second": 7.167,
      "train_steps_per_second": 0.448
    }
  ],
  "logging_steps": 1000,
  "max_steps": 4605,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.3401835354139443e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}