trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9923224568138194,
  "eval_steps": 500,
  "global_step": 1560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_f1": 0.9106,
      "eval_gen_len": 38.35490909090909,
      "eval_loss": 1.621419906616211,
      "eval_precision": 0.9049,
      "eval_recall": 0.9166,
      "eval_rouge1": 0.4804,
      "eval_rouge2": 0.2218,
      "eval_rougeL": 0.3873,
      "eval_rougeLsum": 0.3873,
      "eval_runtime": 654.035,
      "eval_samples_per_second": 4.205,
      "eval_steps_per_second": 0.263,
      "step": 390
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3589743589743592e-05,
      "loss": 1.5842,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.9113,
      "eval_gen_len": 37.86036363636364,
      "eval_loss": 1.5547642707824707,
      "eval_precision": 0.9059,
      "eval_recall": 0.9171,
      "eval_rouge1": 0.4874,
      "eval_rouge2": 0.2283,
      "eval_rougeL": 0.3945,
      "eval_rougeLsum": 0.3945,
      "eval_runtime": 634.7852,
      "eval_samples_per_second": 4.332,
      "eval_steps_per_second": 0.271,
      "step": 781
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.17948717948718e-06,
      "loss": 1.3014,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.912,
      "eval_gen_len": 37.75636363636364,
      "eval_loss": 1.5460655689239502,
      "eval_precision": 0.9064,
      "eval_recall": 0.918,
      "eval_rouge1": 0.49,
      "eval_rouge2": 0.2294,
      "eval_rougeL": 0.3975,
      "eval_rougeLsum": 0.3974,
      "eval_runtime": 629.8341,
      "eval_samples_per_second": 4.366,
      "eval_steps_per_second": 0.273,
      "step": 1172
    },
    {
      "epoch": 3.84,
      "learning_rate": 7.692307692307694e-07,
      "loss": 1.18,
      "step": 1500
    },
    {
      "epoch": 3.99,
      "eval_f1": 0.9123,
      "eval_gen_len": 38.22872727272727,
      "eval_loss": 1.559003472328186,
      "eval_precision": 0.9063,
      "eval_recall": 0.9187,
      "eval_rouge1": 0.4909,
      "eval_rouge2": 0.2303,
      "eval_rougeL": 0.3967,
      "eval_rougeLsum": 0.3965,
      "eval_runtime": 638.1793,
      "eval_samples_per_second": 4.309,
      "eval_steps_per_second": 0.27,
      "step": 1560
    },
    {
      "epoch": 3.99,
      "step": 1560,
      "total_flos": 4.304969545380004e+17,
      "train_loss": 1.3475837952051408,
      "train_runtime": 10536.0779,
      "train_samples_per_second": 18.982,
      "train_steps_per_second": 0.148
    }
  ],
  "logging_steps": 500,
  "max_steps": 1560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 4.304969545380004e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}