{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.716981132075472,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6835483870967742,
      "eval_loss": 0.40425631403923035,
      "eval_runtime": 2.0536,
      "eval_samples_per_second": 1509.522,
      "eval_steps_per_second": 31.651,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.9517033696174622,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.6417,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8487096774193549,
      "eval_loss": 0.13626283407211304,
      "eval_runtime": 2.0845,
      "eval_samples_per_second": 1487.158,
      "eval_steps_per_second": 31.182,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8980645161290323,
      "eval_loss": 0.06938854604959488,
      "eval_runtime": 2.0529,
      "eval_samples_per_second": 1510.085,
      "eval_steps_per_second": 31.663,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.7188680768013,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1602,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9170967741935484,
      "eval_loss": 0.049937449395656586,
      "eval_runtime": 2.0722,
      "eval_samples_per_second": 1496.005,
      "eval_steps_per_second": 31.368,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.5076115727424622,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0795,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 450371359983132.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.14941115644627356,
    "num_train_epochs": 10,
    "temperature": 2
  }
}