checkpoint-2345/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.25769230769230766,
  "eval_steps": 500,
  "global_step": 2345,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 0.00010967032967032969,
      "loss": 1.2182,
      "step": 500
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019782661782661784,
      "loss": 0.147,
      "step": 1000
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00018561660561660563,
      "loss": 0.384,
      "step": 1500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00017340659340659342,
      "loss": 1.1442,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 9100,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 5.1255665246601216e+17,
  "trial_name": null,
  "trial_params": null
}