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": 0.8966,
      "eval_gen_len": 19.970909090909092,
      "eval_loss": 1.5708835124969482,
      "eval_precision": 0.9093,
      "eval_recall": 0.8846,
      "eval_rouge1": 0.4119,
      "eval_rouge2": 0.2002,
      "eval_rougeL": 0.3529,
      "eval_rougeLsum": 0.3527,
      "eval_runtime": 231.9368,
      "eval_samples_per_second": 11.857,
      "eval_steps_per_second": 0.742,
      "step": 390
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.3589743589743592e-05,
      "loss": 1.8155,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.9003,
      "eval_gen_len": 19.918545454545455,
      "eval_loss": 1.5360783338546753,
      "eval_precision": 0.9123,
      "eval_recall": 0.8889,
      "eval_rouge1": 0.4331,
      "eval_rouge2": 0.2157,
      "eval_rougeL": 0.3717,
      "eval_rougeLsum": 0.3717,
      "eval_runtime": 228.1491,
      "eval_samples_per_second": 12.054,
      "eval_steps_per_second": 0.754,
      "step": 781
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.17948717948718e-06,
      "loss": 1.5875,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.899,
      "eval_gen_len": 19.954545454545453,
      "eval_loss": 1.5030012130737305,
      "eval_precision": 0.9117,
      "eval_recall": 0.8871,
      "eval_rouge1": 0.4263,
      "eval_rouge2": 0.2129,
      "eval_rougeL": 0.3671,
      "eval_rougeLsum": 0.3673,
      "eval_runtime": 228.8181,
      "eval_samples_per_second": 12.018,
      "eval_steps_per_second": 0.752,
      "step": 1172
    },
    {
      "epoch": 3.84,
      "learning_rate": 7.692307692307694e-07,
      "loss": 1.4978,
      "step": 1500
    },
    {
      "epoch": 3.99,
      "eval_f1": 0.9002,
      "eval_gen_len": 19.925454545454546,
      "eval_loss": 1.499870777130127,
      "eval_precision": 0.9125,
      "eval_recall": 0.8885,
      "eval_rouge1": 0.4331,
      "eval_rouge2": 0.2164,
      "eval_rougeL": 0.3724,
      "eval_rougeLsum": 0.3725,
      "eval_runtime": 229.5833,
      "eval_samples_per_second": 11.978,
      "eval_steps_per_second": 0.749,
      "step": 1560
    },
    {
      "epoch": 3.99,
      "step": 1560,
      "total_flos": 2.163117129644114e+17,
      "train_loss": 1.6275942337818634,
      "train_runtime": 5933.9925,
      "train_samples_per_second": 33.704,
      "train_steps_per_second": 0.263
    }
  ],
  "logging_steps": 500,
  "max_steps": 1560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 2.163117129644114e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}