last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1499652858134691,
  "global_step": 1296,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.9997685720897943e-05,
      "loss": 2.316,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9259430687340893e-05,
      "loss": 2.541,
      "step": 324
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.850960425827355e-05,
      "loss": 2.3336,
      "step": 648
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.7759777829206203e-05,
      "loss": 2.1896,
      "step": 972
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.7009951400138856e-05,
      "loss": 2.2036,
      "step": 1296
    }
  ],
  "max_steps": 8642,
  "num_train_epochs": 1,
  "total_flos": 2407208507670528.0,
  "trial_name": null,
  "trial_params": null
}