{
  "best_metric": 0.4457,
  "best_model_checkpoint": "bin/summarization-base-0/checkpoint-3566",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 17830,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.3418121337890625,
      "learning_rate": 0.0008,
      "loss": 1.5349,
      "step": 3566
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.9113389849662781,
      "eval_rouge1": 0.4457,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.4447,
      "eval_rougeLsum": 0.4422,
      "eval_runtime": 633.9025,
      "eval_samples_per_second": 1.183,
      "eval_steps_per_second": 0.148,
      "step": 3566
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0075907707214355,
      "learning_rate": 0.0006,
      "loss": 0.8222,
      "step": 7132
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.7657843828201294,
      "eval_rouge1": 0.4293,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.4263,
      "eval_rougeLsum": 0.4253,
      "eval_runtime": 631.9317,
      "eval_samples_per_second": 1.187,
      "eval_steps_per_second": 0.149,
      "step": 7132
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6747327446937561,
      "learning_rate": 0.0004,
      "loss": 0.6231,
      "step": 10698
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.6905989050865173,
      "eval_rouge1": 0.4208,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.417,
      "eval_rougeLsum": 0.415,
      "eval_runtime": 620.388,
      "eval_samples_per_second": 1.209,
      "eval_steps_per_second": 0.152,
      "step": 10698
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5738873481750488,
      "learning_rate": 0.0002,
      "loss": 0.4701,
      "step": 14264
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.6816205978393555,
      "eval_rouge1": 0.4158,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.4113,
      "eval_rougeLsum": 0.41,
      "eval_runtime": 613.0796,
      "eval_samples_per_second": 1.223,
      "eval_steps_per_second": 0.153,
      "step": 14264
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.3324648141860962,
      "learning_rate": 0.0,
      "loss": 0.3217,
      "step": 17830
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 1.0,
      "eval_loss": 0.7147426009178162,
      "eval_rouge1": 0.4109,
      "eval_rouge2": 0.0,
      "eval_rougeL": 0.4081,
      "eval_rougeLsum": 0.4066,
      "eval_runtime": 610.9308,
      "eval_samples_per_second": 1.228,
      "eval_steps_per_second": 0.154,
      "step": 17830
    },
    {
      "epoch": 5.0,
      "step": 17830,
      "total_flos": 4.883004680306688e+16,
      "train_loss": 0.7543908821486784,
      "train_runtime": 12843.575,
      "train_samples_per_second": 5.552,
      "train_steps_per_second": 1.388
    }
  ],
  "logging_steps": 500,
  "max_steps": 17830,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 4.883004680306688e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}