trainer_state.json

{
  "best_metric": 29.5518,
  "best_model_checkpoint": "bin/liputan6-pt-pl5/checkpoint-315",
  "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": 0.8308019638061523,
      "learning_rate": 0.0008,
      "loss": 4.309,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 41.853,
      "eval_loss": 3.4071803092956543,
      "eval_rouge1": 20.861,
      "eval_rouge2": 8.8494,
      "eval_rougeL": 18.3118,
      "eval_rougeLsum": 19.3333,
      "eval_runtime": 253.5381,
      "eval_samples_per_second": 3.944,
      "eval_steps_per_second": 0.126,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9169403910636902,
      "learning_rate": 0.0006,
      "loss": 3.6184,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 51.44,
      "eval_loss": 3.1459953784942627,
      "eval_rouge1": 25.8149,
      "eval_rouge2": 9.9065,
      "eval_rougeL": 21.2242,
      "eval_rougeLsum": 23.6534,
      "eval_runtime": 238.2583,
      "eval_samples_per_second": 4.197,
      "eval_steps_per_second": 0.134,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7842546105384827,
      "learning_rate": 0.0004,
      "loss": 3.3773,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 55.215,
      "eval_loss": 2.9716238975524902,
      "eval_rouge1": 27.0263,
      "eval_rouge2": 11.563,
      "eval_rougeL": 22.6018,
      "eval_rougeLsum": 25.0075,
      "eval_runtime": 260.5928,
      "eval_samples_per_second": 3.837,
      "eval_steps_per_second": 0.123,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.033412218093872,
      "learning_rate": 0.0002,
      "loss": 3.1793,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 52.119,
      "eval_loss": 2.826566219329834,
      "eval_rouge1": 28.6871,
      "eval_rouge2": 12.9802,
      "eval_rougeL": 24.038,
      "eval_rougeLsum": 26.6558,
      "eval_runtime": 245.0999,
      "eval_samples_per_second": 4.08,
      "eval_steps_per_second": 0.131,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9423825740814209,
      "learning_rate": 0.0,
      "loss": 3.0027,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 51.956,
      "eval_loss": 2.752587080001831,
      "eval_rouge1": 29.5518,
      "eval_rouge2": 13.5408,
      "eval_rougeL": 24.834,
      "eval_rougeLsum": 27.6096,
      "eval_runtime": 228.7726,
      "eval_samples_per_second": 4.371,
      "eval_steps_per_second": 0.14,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3877644533760000.0,
      "train_loss": 3.4973431299603175,
      "train_runtime": 1609.8772,
      "train_samples_per_second": 3.106,
      "train_steps_per_second": 0.196
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3877644533760000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}