trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 517094,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 2e-05,
      "loss": 1.5646,
      "step": 51710
    },
    {
      "epoch": 0.2,
      "learning_rate": 2e-05,
      "loss": 1.4615,
      "step": 103420
    },
    {
      "epoch": 0.3,
      "learning_rate": 2e-05,
      "loss": 1.4251,
      "step": 155130
    },
    {
      "epoch": 0.4,
      "learning_rate": 2e-05,
      "loss": 1.4013,
      "step": 206840
    },
    {
      "epoch": 0.5,
      "learning_rate": 2e-05,
      "loss": 1.3836,
      "step": 258550
    },
    {
      "epoch": 0.6,
      "learning_rate": 2e-05,
      "loss": 1.3704,
      "step": 310260
    },
    {
      "epoch": 0.7,
      "learning_rate": 2e-05,
      "loss": 1.3598,
      "step": 361970
    },
    {
      "epoch": 0.8,
      "learning_rate": 2e-05,
      "loss": 1.3506,
      "step": 413680
    },
    {
      "epoch": 0.9,
      "learning_rate": 2e-05,
      "loss": 1.3429,
      "step": 465390
    },
    {
      "epoch": 1.0,
      "step": 517094,
      "total_flos": 8.710545978622679e+18,
      "train_loss": 1.3995336800465679,
      "train_runtime": 507280.6591,
      "train_samples_per_second": 65.238,
      "train_steps_per_second": 1.019
    }
  ],
  "max_steps": 517094,
  "num_train_epochs": 1,
  "total_flos": 8.710545978622679e+18,
  "trial_name": null,
  "trial_params": null
}