trainer_state.json

{
  "best_metric": 17.698,
  "best_model_checkpoint": "bin/liputan6-base/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": 5.591057300567627,
      "learning_rate": 0.0008,
      "loss": 3.8271,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 47.473,
      "eval_loss": 3.9787089824676514,
      "eval_rouge1": 14.5233,
      "eval_rouge2": 4.127,
      "eval_rougeL": 12.7611,
      "eval_rougeLsum": 13.5205,
      "eval_runtime": 251.9921,
      "eval_samples_per_second": 3.968,
      "eval_steps_per_second": 0.127,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.016976833343506,
      "learning_rate": 0.0006,
      "loss": 2.2739,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 44.229,
      "eval_loss": 4.131580352783203,
      "eval_rouge1": 15.9563,
      "eval_rouge2": 4.7752,
      "eval_rougeL": 13.8242,
      "eval_rougeLsum": 14.8005,
      "eval_runtime": 252.4735,
      "eval_samples_per_second": 3.961,
      "eval_steps_per_second": 0.127,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.183892726898193,
      "learning_rate": 0.0004,
      "loss": 1.2999,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 33.112,
      "eval_loss": 4.484961032867432,
      "eval_rouge1": 17.2932,
      "eval_rouge2": 4.6352,
      "eval_rougeL": 14.8582,
      "eval_rougeLsum": 16.1555,
      "eval_runtime": 221.3655,
      "eval_samples_per_second": 4.517,
      "eval_steps_per_second": 0.145,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.1966776847839355,
      "learning_rate": 0.0002,
      "loss": 0.6423,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 36.399,
      "eval_loss": 4.919987678527832,
      "eval_rouge1": 17.5707,
      "eval_rouge2": 4.9772,
      "eval_rougeL": 14.949,
      "eval_rougeLsum": 16.1838,
      "eval_runtime": 217.3583,
      "eval_samples_per_second": 4.601,
      "eval_steps_per_second": 0.147,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.510921001434326,
      "learning_rate": 0.0,
      "loss": 0.2536,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 31.108,
      "eval_loss": 5.426603317260742,
      "eval_rouge1": 17.698,
      "eval_rouge2": 4.7021,
      "eval_rougeL": 14.8138,
      "eval_rougeLsum": 16.3595,
      "eval_runtime": 159.266,
      "eval_samples_per_second": 6.279,
      "eval_steps_per_second": 0.201,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3423786762240000.0,
      "train_loss": 1.6593605313982283,
      "train_runtime": 1412.1276,
      "train_samples_per_second": 3.541,
      "train_steps_per_second": 0.223
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3423786762240000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}