checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.873563218390805,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.1485695297263177,
      "eval_runtime": 4.6066,
      "eval_samples_per_second": 134.156,
      "eval_steps_per_second": 8.466,
      "step": 348
    },
    {
      "epoch": 1.44,
      "learning_rate": 3.2040229885057474e-05,
      "loss": 0.18,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.12702645080563052,
      "eval_runtime": 4.6083,
      "eval_samples_per_second": 134.105,
      "eval_steps_per_second": 8.463,
      "step": 696
    },
    {
      "epoch": 2.87,
      "learning_rate": 1.4080459770114942e-05,
      "loss": 0.1044,
      "step": 1000
    }
  ],
  "max_steps": 1392,
  "num_train_epochs": 4,
  "total_flos": 1050809472513024.0,
  "trial_name": null,
  "trial_params": null
}