trainer_state.json

{
  "best_metric": 15.5769,
  "best_model_checkpoint": "saved/tobyoki-pairwise-wo_space/bart-base-japanese/BaseModel/checkpoint-2868",
  "epoch": 5.0,
  "global_step": 3585,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7,
      "learning_rate": 4.302649930264993e-05,
      "loss": 2.1701,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 47.7375,
      "eval_loss": 1.9507354497909546,
      "eval_rouge1": 12.6467,
      "eval_rouge2": 2.901,
      "eval_rougeL": 10.0035,
      "eval_rougeLsum": 11.1471,
      "eval_runtime": 40.6217,
      "eval_samples_per_second": 1.969,
      "eval_steps_per_second": 1.969,
      "step": 717
    },
    {
      "epoch": 1.39,
      "learning_rate": 3.6052998605299864e-05,
      "loss": 1.4042,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 21.1375,
      "eval_loss": 1.9519145488739014,
      "eval_rouge1": 11.9515,
      "eval_rouge2": 3.096,
      "eval_rougeL": 10.2259,
      "eval_rougeLsum": 10.8478,
      "eval_runtime": 26.1922,
      "eval_samples_per_second": 3.054,
      "eval_steps_per_second": 3.054,
      "step": 1434
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.9079497907949792e-05,
      "loss": 1.1417,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.2105997210599723e-05,
      "loss": 0.8952,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 76.35,
      "eval_loss": 2.0322935581207275,
      "eval_rouge1": 15.5721,
      "eval_rouge2": 3.5875,
      "eval_rougeL": 10.6382,
      "eval_rougeLsum": 12.9346,
      "eval_runtime": 74.4084,
      "eval_samples_per_second": 1.075,
      "eval_steps_per_second": 1.075,
      "step": 2151
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.5132496513249652e-05,
      "loss": 0.7489,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 63.7625,
      "eval_loss": 2.0723605155944824,
      "eval_rouge1": 15.5769,
      "eval_rouge2": 3.3042,
      "eval_rougeL": 11.0176,
      "eval_rougeLsum": 12.8107,
      "eval_runtime": 64.4381,
      "eval_samples_per_second": 1.242,
      "eval_steps_per_second": 1.242,
      "step": 2868
    },
    {
      "epoch": 4.18,
      "learning_rate": 8.158995815899583e-06,
      "loss": 0.6757,
      "step": 3000
    },
    {
      "epoch": 4.88,
      "learning_rate": 1.185495118549512e-06,
      "loss": 0.5941,
      "step": 3500
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 66.075,
      "eval_loss": 2.0849809646606445,
      "eval_rouge1": 15.3454,
      "eval_rouge2": 2.9489,
      "eval_rougeL": 10.7691,
      "eval_rougeLsum": 12.7028,
      "eval_runtime": 57.1206,
      "eval_samples_per_second": 1.401,
      "eval_steps_per_second": 1.401,
      "step": 3585
    },
    {
      "epoch": 5.0,
      "step": 3585,
      "total_flos": 682678854881280.0,
      "train_loss": 1.0774688316355856,
      "train_runtime": 782.0878,
      "train_samples_per_second": 4.584,
      "train_steps_per_second": 4.584
    }
  ],
  "max_steps": 3585,
  "num_train_epochs": 5,
  "total_flos": 682678854881280.0,
  "trial_name": null,
  "trial_params": null
}