trainer_state.json

{
  "best_metric": 6.5087,
  "best_model_checkpoint": "bin/liputan6-unipelt/checkpoint-63",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 315,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.6662813425064087,
      "learning_rate": 0.0008,
      "loss": 4.4282,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 127.0,
      "eval_loss": 3.9181036949157715,
      "eval_rouge1": 6.5087,
      "eval_rouge2": 1.1507,
      "eval_rougeL": 5.6509,
      "eval_rougeLsum": 5.8646,
      "eval_runtime": 719.7468,
      "eval_samples_per_second": 1.389,
      "eval_steps_per_second": 0.044,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9859250783920288,
      "learning_rate": 0.0006,
      "loss": 3.5887,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 127.0,
      "eval_loss": 3.7177910804748535,
      "eval_rouge1": 2.446,
      "eval_rouge2": 0.2311,
      "eval_rougeL": 2.2699,
      "eval_rougeLsum": 2.3165,
      "eval_runtime": 730.7553,
      "eval_samples_per_second": 1.368,
      "eval_steps_per_second": 0.044,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.097254991531372,
      "learning_rate": 0.0004,
      "loss": 3.2648,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 127.0,
      "eval_loss": 3.648098945617676,
      "eval_rouge1": 2.5164,
      "eval_rouge2": 0.1545,
      "eval_rougeL": 2.3984,
      "eval_rougeLsum": 2.4483,
      "eval_runtime": 742.5951,
      "eval_samples_per_second": 1.347,
      "eval_steps_per_second": 0.043,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.1174938678741455,
      "learning_rate": 0.0002,
      "loss": 3.0219,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 127.0,
      "eval_loss": 3.6383612155914307,
      "eval_rouge1": 1.8155,
      "eval_rouge2": 0.0676,
      "eval_rougeL": 1.7586,
      "eval_rougeLsum": 1.7921,
      "eval_runtime": 741.1853,
      "eval_samples_per_second": 1.349,
      "eval_steps_per_second": 0.043,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.148689031600952,
      "learning_rate": 0.0,
      "loss": 2.8422,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 127.0,
      "eval_loss": 3.6315574645996094,
      "eval_rouge1": 1.8166,
      "eval_rouge2": 0.0438,
      "eval_rougeL": 1.7867,
      "eval_rougeLsum": 1.8115,
      "eval_runtime": 723.9069,
      "eval_samples_per_second": 1.381,
      "eval_steps_per_second": 0.044,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3920273141760000.0,
      "train_loss": 3.429138425796751,
      "train_runtime": 4104.3672,
      "train_samples_per_second": 1.218,
      "train_steps_per_second": 0.077
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3920273141760000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}