trainer_state.json

{
  "best_metric": 43.7351,
  "best_model_checkpoint": "bin/liputan6-seq_bn-rf64/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": 1.1267616748809814,
      "learning_rate": 0.0008,
      "loss": 1.1116,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 49.406,
      "eval_loss": 0.3952869176864624,
      "eval_rouge1": 39.3374,
      "eval_rouge2": 31.1269,
      "eval_rougeL": 36.447,
      "eval_rougeLsum": 37.9075,
      "eval_runtime": 223.3788,
      "eval_samples_per_second": 4.477,
      "eval_steps_per_second": 0.143,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5356050729751587,
      "learning_rate": 0.0006,
      "loss": 0.5774,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 55.7,
      "eval_loss": 0.33090007305145264,
      "eval_rouge1": 43.6615,
      "eval_rouge2": 34.9524,
      "eval_rougeL": 40.0101,
      "eval_rougeLsum": 42.2478,
      "eval_runtime": 200.7713,
      "eval_samples_per_second": 4.981,
      "eval_steps_per_second": 0.159,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5113601684570312,
      "learning_rate": 0.0004,
      "loss": 0.4729,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 52.999,
      "eval_loss": 0.2864004969596863,
      "eval_rouge1": 41.4092,
      "eval_rouge2": 32.5453,
      "eval_rougeL": 37.9684,
      "eval_rougeLsum": 39.9099,
      "eval_runtime": 178.2036,
      "eval_samples_per_second": 5.612,
      "eval_steps_per_second": 0.18,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9492444396018982,
      "learning_rate": 0.0002,
      "loss": 0.4209,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 56.612,
      "eval_loss": 0.2844858467578888,
      "eval_rouge1": 43.3834,
      "eval_rouge2": 34.6436,
      "eval_rougeL": 39.8104,
      "eval_rougeLsum": 41.9569,
      "eval_runtime": 205.9714,
      "eval_samples_per_second": 4.855,
      "eval_steps_per_second": 0.155,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0183289051055908,
      "learning_rate": 0.0,
      "loss": 0.3931,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 58.611,
      "eval_loss": 0.2748122215270996,
      "eval_rouge1": 43.7351,
      "eval_rouge2": 34.936,
      "eval_rougeL": 40.0408,
      "eval_rougeLsum": 42.3437,
      "eval_runtime": 204.4602,
      "eval_samples_per_second": 4.891,
      "eval_steps_per_second": 0.157,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3430869073920000.0,
      "train_loss": 0.5951776110936725,
      "train_runtime": 1265.3043,
      "train_samples_per_second": 3.952,
      "train_steps_per_second": 0.249
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3430869073920000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}