trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 4080,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.23,
      "learning_rate": 4.387254901960784e-05,
      "loss": 1.0399,
      "step": 500
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.774509803921569e-05,
      "loss": 0.7758,
      "step": 1000
    },
    {
      "epoch": 3.68,
      "learning_rate": 3.161764705882353e-05,
      "loss": 0.7063,
      "step": 1500
    },
    {
      "epoch": 4.9,
      "learning_rate": 2.5490196078431373e-05,
      "loss": 0.6595,
      "step": 2000
    },
    {
      "epoch": 6.13,
      "learning_rate": 1.936274509803922e-05,
      "loss": 0.6224,
      "step": 2500
    },
    {
      "epoch": 7.35,
      "learning_rate": 1.323529411764706e-05,
      "loss": 0.6014,
      "step": 3000
    },
    {
      "epoch": 8.58,
      "learning_rate": 7.107843137254902e-06,
      "loss": 0.5836,
      "step": 3500
    },
    {
      "epoch": 9.8,
      "learning_rate": 9.80392156862745e-07,
      "loss": 0.5712,
      "step": 4000
    },
    {
      "epoch": 10.0,
      "step": 4080,
      "total_flos": 5.413466748248064e+16,
      "train_loss": 0.6925308077943091,
      "train_runtime": 3096.2997,
      "train_samples_per_second": 13.161,
      "train_steps_per_second": 1.318
    }
  ],
  "max_steps": 4080,
  "num_train_epochs": 10,
  "total_flos": 5.413466748248064e+16,
  "trial_name": null,
  "trial_params": null
}