{
  "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": 0.6806451612903226,
      "eval_loss": 0.40702202916145325,
      "eval_runtime": 1.8726,
      "eval_samples_per_second": 1655.484,
      "eval_steps_per_second": 34.712,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.9639246463775635,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.6436,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8480645161290322,
      "eval_loss": 0.13934274017810822,
      "eval_runtime": 2.1147,
      "eval_samples_per_second": 1465.91,
      "eval_steps_per_second": 30.737,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.895483870967742,
      "eval_loss": 0.07144162803888321,
      "eval_runtime": 3.2019,
      "eval_samples_per_second": 968.173,
      "eval_steps_per_second": 20.3,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.7319316267967224,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1636,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9151612903225806,
      "eval_loss": 0.0515269860625267,
      "eval_runtime": 2.0391,
      "eval_samples_per_second": 1520.288,
      "eval_steps_per_second": 31.877,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.5300735831260681,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0819,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9229032258064516,
      "eval_loss": 0.04307929426431656,
      "eval_runtime": 2.0917,
      "eval_samples_per_second": 1482.018,
      "eval_steps_per_second": 31.075,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9287096774193548,
      "eval_loss": 0.038923561573028564,
      "eval_runtime": 2.117,
      "eval_samples_per_second": 1464.33,
      "eval_steps_per_second": 30.704,
      "step": 1908
    },
    {
      "epoch": 6.29,
      "grad_norm": 0.4357357323169708,
      "learning_rate": 6.02375960866527e-06,
      "loss": 0.0618,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9290322580645162,
      "eval_loss": 0.03594091534614563,
      "eval_runtime": 2.0849,
      "eval_samples_per_second": 1486.855,
      "eval_steps_per_second": 31.176,
      "step": 2226
    },
    {
      "epoch": 7.86,
      "grad_norm": 0.26694580912590027,
      "learning_rate": 2.5296995108315863e-06,
      "loss": 0.0541,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 651155886807636.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.2693940258587564,
    "num_train_epochs": 9,
    "temperature": 2
  }
}