{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.861635220125786,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6535483870967742,
      "eval_loss": 0.28503966331481934,
      "eval_runtime": 1.889,
      "eval_samples_per_second": 1641.078,
      "eval_steps_per_second": 34.41,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.7276282906532288,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.4585,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8422580645161291,
      "eval_loss": 0.1152644231915474,
      "eval_runtime": 2.104,
      "eval_samples_per_second": 1473.356,
      "eval_steps_per_second": 30.893,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8938709677419355,
      "eval_loss": 0.06576532125473022,
      "eval_runtime": 3.2907,
      "eval_samples_per_second": 942.06,
      "eval_steps_per_second": 19.753,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.6005321741104126,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1354,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9119354838709678,
      "eval_loss": 0.04726028814911842,
      "eval_runtime": 2.7258,
      "eval_samples_per_second": 1137.262,
      "eval_steps_per_second": 23.846,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.449545294046402,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0753,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.038489848375320435,
      "eval_runtime": 3.2654,
      "eval_samples_per_second": 949.346,
      "eval_steps_per_second": 19.906,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9261290322580645,
      "eval_loss": 0.03405739739537239,
      "eval_runtime": 2.7619,
      "eval_samples_per_second": 1122.42,
      "eval_steps_per_second": 23.535,
      "step": 1908
    },
    {
      "epoch": 6.29,
      "grad_norm": 0.3650922179222107,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0567,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9296774193548387,
      "eval_loss": 0.030985437333583832,
      "eval_runtime": 7.9603,
      "eval_samples_per_second": 389.434,
      "eval_steps_per_second": 8.166,
      "step": 2226
    },
    {
      "epoch": 7.86,
      "grad_norm": 0.22622260451316833,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0488,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 651155886807636.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.45486992782879154,
    "num_train_epochs": 10,
    "temperature": 3
  }
}