run-8/checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.1446540880503147,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5864516129032258,
      "eval_loss": 0.1933201551437378,
      "eval_runtime": 5.533,
      "eval_samples_per_second": 560.278,
      "eval_steps_per_second": 11.748,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.3095,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.09366016089916229,
      "eval_runtime": 5.65,
      "eval_samples_per_second": 548.672,
      "eval_steps_per_second": 11.504,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8912903225806451,
      "eval_loss": 0.06257741898298264,
      "eval_runtime": 5.4549,
      "eval_samples_per_second": 568.294,
      "eval_steps_per_second": 11.916,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1078,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 377730567005652.0,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9205924553180014,
    "num_train_epochs": 10,
    "temperature": 18
  }
}