trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 5440,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.92,
      "learning_rate": 2.724264705882353e-05,
      "loss": 1.6219,
      "step": 500
    },
    {
      "epoch": 1.84,
      "learning_rate": 2.448529411764706e-05,
      "loss": 1.0069,
      "step": 1000
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.1727941176470588e-05,
      "loss": 0.8043,
      "step": 1500
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.8970588235294116e-05,
      "loss": 0.6756,
      "step": 2000
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.6213235294117647e-05,
      "loss": 0.5709,
      "step": 2500
    },
    {
      "epoch": 5.51,
      "learning_rate": 1.3455882352941177e-05,
      "loss": 0.4862,
      "step": 3000
    },
    {
      "epoch": 6.43,
      "learning_rate": 1.0698529411764707e-05,
      "loss": 0.4233,
      "step": 3500
    },
    {
      "epoch": 7.35,
      "learning_rate": 7.941176470588236e-06,
      "loss": 0.3724,
      "step": 4000
    },
    {
      "epoch": 8.27,
      "learning_rate": 5.183823529411765e-06,
      "loss": 0.3353,
      "step": 4500
    },
    {
      "epoch": 9.19,
      "learning_rate": 2.4264705882352943e-06,
      "loss": 0.3038,
      "step": 5000
    },
    {
      "epoch": 10.0,
      "step": 5440,
      "total_flos": 2.0576085260160205e+17,
      "train_loss": 0.6307362275965074,
      "train_runtime": 1688.4541,
      "train_samples_per_second": 773.157,
      "train_steps_per_second": 3.222
    }
  ],
  "max_steps": 5440,
  "num_train_epochs": 10,
  "total_flos": 2.0576085260160205e+17,
  "trial_name": null,
  "trial_params": null
}