checkpoint-1132/trainer_state.json

{
  "best_metric": 1.3724207878112793,
  "best_model_checkpoint": "./LongT5-Large-NSPCC/checkpoint-1132",
  "epoch": 5.997350993377483,
  "eval_steps": 500,
  "global_step": 1132,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9960264900662251,
      "grad_norm": 1.2397888898849487,
      "learning_rate": 0.00029303322414305795,
      "loss": 6.0401,
      "step": 188
    },
    {
      "epoch": 0.9960264900662251,
      "eval_gen_len": 151.7021,
      "eval_loss": 2.708937406539917,
      "eval_rouge1": 0.2766,
      "eval_rouge2": 0.0617,
      "eval_rougeL": 0.1655,
      "eval_rougeLsum": 0.1657,
      "eval_runtime": 1372.78,
      "eval_samples_per_second": 0.068,
      "eval_steps_per_second": 0.068,
      "step": 188
    },
    {
      "epoch": 1.9973509933774833,
      "grad_norm": 0.9210834503173828,
      "learning_rate": 0.00026343936751493783,
      "loss": 2.4805,
      "step": 377
    },
    {
      "epoch": 1.9973509933774833,
      "eval_gen_len": 211.1809,
      "eval_loss": 1.8809192180633545,
      "eval_rouge1": 0.382,
      "eval_rouge2": 0.1178,
      "eval_rougeL": 0.2092,
      "eval_rougeLsum": 0.2092,
      "eval_runtime": 1618.944,
      "eval_samples_per_second": 0.058,
      "eval_steps_per_second": 0.058,
      "step": 377
    },
    {
      "epoch": 2.9986754966887417,
      "grad_norm": 1.2795201539993286,
      "learning_rate": 0.0002152904846699957,
      "loss": 1.8093,
      "step": 566
    },
    {
      "epoch": 2.9986754966887417,
      "eval_gen_len": 246.1277,
      "eval_loss": 1.576943039894104,
      "eval_rouge1": 0.4356,
      "eval_rouge2": 0.1527,
      "eval_rougeL": 0.2409,
      "eval_rougeLsum": 0.2409,
      "eval_runtime": 1917.7123,
      "eval_samples_per_second": 0.049,
      "eval_steps_per_second": 0.049,
      "step": 566
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6965745091438293,
      "learning_rate": 0.00015646218178508154,
      "loss": 1.4653,
      "step": 755
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 245.0851,
      "eval_loss": 1.435921311378479,
      "eval_rouge1": 0.4661,
      "eval_rouge2": 0.1722,
      "eval_rougeL": 0.26,
      "eval_rougeLsum": 0.2603,
      "eval_runtime": 1747.772,
      "eval_samples_per_second": 0.054,
      "eval_steps_per_second": 0.054,
      "step": 755
    },
    {
      "epoch": 4.9960264900662255,
      "grad_norm": 1.0618396997451782,
      "learning_rate": 9.687901331136185e-05,
      "loss": 1.2626,
      "step": 943
    },
    {
      "epoch": 4.9960264900662255,
      "eval_gen_len": 239.8617,
      "eval_loss": 1.390848994255066,
      "eval_rouge1": 0.4829,
      "eval_rouge2": 0.1931,
      "eval_rougeL": 0.2717,
      "eval_rougeLsum": 0.2717,
      "eval_runtime": 1739.5487,
      "eval_samples_per_second": 0.054,
      "eval_steps_per_second": 0.054,
      "step": 943
    },
    {
      "epoch": 5.997350993377483,
      "grad_norm": 0.8346843123435974,
      "learning_rate": 4.566183323963369e-05,
      "loss": 1.117,
      "step": 1132
    },
    {
      "epoch": 5.997350993377483,
      "eval_gen_len": 244.4255,
      "eval_loss": 1.3724207878112793,
      "eval_rouge1": 0.4864,
      "eval_rouge2": 0.1988,
      "eval_rougeL": 0.2804,
      "eval_rougeLsum": 0.2804,
      "eval_runtime": 1739.9152,
      "eval_samples_per_second": 0.054,
      "eval_steps_per_second": 0.054,
      "step": 1132
    }
  ],
  "logging_steps": 500,
  "max_steps": 1504,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 3.341101062684672e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}