trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.99527559055118,
  "global_step": 790,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.26,
      "learning_rate": 8.734177215189874e-05,
      "loss": 1.9271,
      "step": 100
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.468354430379747e-05,
      "loss": 1.5723,
      "step": 200
    },
    {
      "epoch": 3.79,
      "learning_rate": 6.20253164556962e-05,
      "loss": 1.4655,
      "step": 300
    },
    {
      "epoch": 5.06,
      "learning_rate": 4.936708860759494e-05,
      "loss": 1.4053,
      "step": 400
    },
    {
      "epoch": 6.33,
      "learning_rate": 3.670886075949367e-05,
      "loss": 1.3442,
      "step": 500
    },
    {
      "epoch": 7.59,
      "learning_rate": 2.4050632911392405e-05,
      "loss": 1.2954,
      "step": 600
    },
    {
      "epoch": 8.86,
      "learning_rate": 1.139240506329114e-05,
      "loss": 1.28,
      "step": 700
    },
    {
      "epoch": 10.0,
      "step": 790,
      "total_flos": 1.388887369371648e+16,
      "train_loss": 1.4465530395507813,
      "train_runtime": 7665.0914,
      "train_samples_per_second": 3.314,
      "train_steps_per_second": 0.103
    }
  ],
  "max_steps": 790,
  "num_train_epochs": 10,
  "total_flos": 1.388887369371648e+16,
  "trial_name": null,
  "trial_params": null
}