trainer_state.json

{
  "best_metric": 42.8202,
  "best_model_checkpoint": "bin/liputan6-lora-8/checkpoint-252",
  "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": 1.214747428894043,
      "learning_rate": 0.0008,
      "loss": 1.4611,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 26.342,
      "eval_loss": 0.4168391823768616,
      "eval_rouge1": 20.2294,
      "eval_rouge2": 15.1672,
      "eval_rougeL": 18.3462,
      "eval_rougeLsum": 19.5551,
      "eval_runtime": 361.7096,
      "eval_samples_per_second": 2.765,
      "eval_steps_per_second": 0.088,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.2018976211547852,
      "learning_rate": 0.0006,
      "loss": 0.6123,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 54.709,
      "eval_loss": 0.3029869794845581,
      "eval_rouge1": 42.4943,
      "eval_rouge2": 33.9837,
      "eval_rougeL": 39.042,
      "eval_rougeLsum": 40.9843,
      "eval_runtime": 392.3812,
      "eval_samples_per_second": 2.549,
      "eval_steps_per_second": 0.082,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.4286189079284668,
      "learning_rate": 0.0004,
      "loss": 0.4836,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 48.258,
      "eval_loss": 0.2516452670097351,
      "eval_rouge1": 39.2983,
      "eval_rouge2": 30.3972,
      "eval_rougeL": 36.1696,
      "eval_rougeLsum": 37.7681,
      "eval_runtime": 320.2113,
      "eval_samples_per_second": 3.123,
      "eval_steps_per_second": 0.1,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9979777932167053,
      "learning_rate": 0.0002,
      "loss": 0.4256,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 54.56,
      "eval_loss": 0.2482166737318039,
      "eval_rouge1": 42.8202,
      "eval_rouge2": 33.9496,
      "eval_rougeL": 39.2137,
      "eval_rougeLsum": 41.343,
      "eval_runtime": 345.5791,
      "eval_samples_per_second": 2.894,
      "eval_steps_per_second": 0.093,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.1315256357192993,
      "learning_rate": 0.0,
      "loss": 0.4014,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 55.005,
      "eval_loss": 0.23890838027000427,
      "eval_rouge1": 42.5107,
      "eval_rouge2": 33.6017,
      "eval_rougeL": 39.0118,
      "eval_rougeLsum": 41.0601,
      "eval_runtime": 343.6619,
      "eval_samples_per_second": 2.91,
      "eval_steps_per_second": 0.093,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3437376307200000.0,
      "train_loss": 0.6767798650832403,
      "train_runtime": 2125.3106,
      "train_samples_per_second": 2.353,
      "train_steps_per_second": 0.148
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3437376307200000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}