trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.030034022916585194,
  "global_step": 42,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 7.61904761904762e-09,
      "loss": 0.1312,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.238095238095238e-09,
      "loss": 0.1106,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.857142857142857e-09,
      "loss": 0.0951,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.761904761904762e-10,
      "loss": 0.0701,
      "step": 40
    },
    {
      "epoch": 0.03,
      "step": 42,
      "total_flos": 1320357672517632.0,
      "train_loss": 0.09978416970088369,
      "train_runtime": 613.5462,
      "train_samples_per_second": 17.504,
      "train_steps_per_second": 0.068
    }
  ],
  "max_steps": 42,
  "num_train_epochs": 1,
  "total_flos": 1320357672517632.0,
  "trial_name": null,
  "trial_params": null
}