trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.2833,
      "step": 1
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018314696123025454,
      "loss": 1.2373,
      "step": 5
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.168,
      "step": 10
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.6853038769745467e-05,
      "loss": 1.1117,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.0914130210876465,
      "eval_runtime": 15.7258,
      "eval_samples_per_second": 7.885,
      "eval_steps_per_second": 0.509,
      "step": 18
    },
    {
      "epoch": 1.0,
      "step": 18,
      "total_flos": 1.0560439065981747e+17,
      "train_loss": 1.161470651626587,
      "train_runtime": 443.1723,
      "train_samples_per_second": 2.579,
      "train_steps_per_second": 0.041
    }
  ],
  "logging_steps": 5,
  "max_steps": 18,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.0560439065981747e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}