{
  "best_metric": 8.2013,
  "best_model_checkpoint": "bin/indosum-unipelt-0/checkpoint-1784",
  "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": 2.078096628189087,
      "learning_rate": 0.0008,
      "loss": 2.4307,
      "step": 892
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 127.0,
      "eval_loss": 1.2643963098526,
      "eval_rouge1": 5.9923,
      "eval_rouge2": 1.617,
      "eval_rougeL": 5.7169,
      "eval_rougeLsum": 5.8449,
      "eval_runtime": 588.1575,
      "eval_samples_per_second": 1.275,
      "eval_steps_per_second": 0.041,
      "step": 892
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9211084842681885,
      "learning_rate": 0.0006,
      "loss": 1.5043,
      "step": 1784
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 127.0,
      "eval_loss": 0.9409825205802917,
      "eval_rouge1": 8.2013,
      "eval_rouge2": 1.997,
      "eval_rougeL": 7.4285,
      "eval_rougeLsum": 7.8145,
      "eval_runtime": 606.2591,
      "eval_samples_per_second": 1.237,
      "eval_steps_per_second": 0.04,
      "step": 1784
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.439053773880005,
      "learning_rate": 0.0004,
      "loss": 1.2064,
      "step": 2676
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 127.0,
      "eval_loss": 0.7985809445381165,
      "eval_rouge1": 6.9869,
      "eval_rouge2": 1.3612,
      "eval_rougeL": 6.3365,
      "eval_rougeLsum": 6.7002,
      "eval_runtime": 569.5911,
      "eval_samples_per_second": 1.317,
      "eval_steps_per_second": 0.042,
      "step": 2676
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.8025060892105103,
      "learning_rate": 0.0002,
      "loss": 1.0352,
      "step": 3568
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 127.0,
      "eval_loss": 0.7307025790214539,
      "eval_rouge1": 5.9136,
      "eval_rouge2": 0.995,
      "eval_rougeL": 5.4713,
      "eval_rougeLsum": 5.7309,
      "eval_runtime": 553.251,
      "eval_samples_per_second": 1.356,
      "eval_steps_per_second": 0.043,
      "step": 3568
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0839836597442627,
      "learning_rate": 0.0,
      "loss": 0.9265,
      "step": 4460
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 127.0,
      "eval_loss": 0.6944215893745422,
      "eval_rouge1": 5.7938,
      "eval_rouge2": 0.9075,
      "eval_rougeL": 5.2885,
      "eval_rougeLsum": 5.5754,
      "eval_runtime": 611.2108,
      "eval_samples_per_second": 1.227,
      "eval_steps_per_second": 0.039,
      "step": 4460
    },
    {
      "epoch": 5.0,
      "step": 4460,
      "total_flos": 5.591093554778112e+16,
      "train_loss": 1.420613091302025,
      "train_runtime": 9240.9223,
      "train_samples_per_second": 7.717,
      "train_steps_per_second": 0.483
    }
  ],
  "logging_steps": 500,
  "max_steps": 4460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5.591093554778112e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}