{
  "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": 0.5916129032258064,
      "eval_loss": 0.19515082240104675,
      "eval_runtime": 1.8956,
      "eval_samples_per_second": 1635.33,
      "eval_steps_per_second": 34.289,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.6071546077728271,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.315,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8209677419354838,
      "eval_loss": 0.09566613286733627,
      "eval_runtime": 2.0569,
      "eval_samples_per_second": 1507.132,
      "eval_steps_per_second": 31.601,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8745161290322581,
      "eval_loss": 0.06388552486896515,
      "eval_runtime": 3.4293,
      "eval_samples_per_second": 903.986,
      "eval_steps_per_second": 18.955,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.4936941862106323,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1104,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.049174223095178604,
      "eval_runtime": 3.2604,
      "eval_samples_per_second": 950.797,
      "eval_steps_per_second": 19.936,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.36201611161231995,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0706,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9058064516129032,
      "eval_loss": 0.040870048105716705,
      "eval_runtime": 2.5708,
      "eval_samples_per_second": 1205.872,
      "eval_steps_per_second": 25.284,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9112903225806451,
      "eval_loss": 0.035991765558719635,
      "eval_runtime": 3.5579,
      "eval_samples_per_second": 871.299,
      "eval_steps_per_second": 18.269,
      "step": 1908
    },
    {
      "epoch": 6.29,
      "grad_norm": 0.304591566324234,
      "learning_rate": 6.02375960866527e-06,
      "loss": 0.0557,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9180645161290323,
      "eval_loss": 0.033146169036626816,
      "eval_runtime": 1.8785,
      "eval_samples_per_second": 1650.226,
      "eval_steps_per_second": 34.602,
      "step": 2226
    },
    {
      "epoch": 7.86,
      "grad_norm": 0.20553433895111084,
      "learning_rate": 2.5296995108315863e-06,
      "loss": 0.0489,
      "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": 0.44158831361956363,
    "num_train_epochs": 9,
    "temperature": 13
  }
}