run-5/checkpoint-500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.25,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 1.0626,
      "eval_samples_per_second": 2917.34,
      "eval_steps_per_second": 15.998,
      "step": 80
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 1.0855,
      "eval_samples_per_second": 2855.753,
      "eval_steps_per_second": 15.661,
      "step": 160
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 1.0841,
      "eval_samples_per_second": 2859.551,
      "eval_steps_per_second": 15.681,
      "step": 240
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 1.0886,
      "eval_samples_per_second": 2847.783,
      "eval_steps_per_second": 15.617,
      "step": 320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 1.0641,
      "eval_samples_per_second": 2913.324,
      "eval_steps_per_second": 15.976,
      "step": 400
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 1.0801,
      "eval_samples_per_second": 2870.155,
      "eval_steps_per_second": 15.74,
      "step": 480
    },
    {
      "epoch": 6.25,
      "learning_rate": 4.3750000000000005e-06,
      "loss": 0.0,
      "step": 500
    }
  ],
  "logging_steps": 500,
  "max_steps": 640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 591309336775836.0,
  "train_batch_size": 192,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6606443487477446,
    "num_train_epochs": 8,
    "temperature": 12
  }
}