{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.289308176100629,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6380645161290323,
      "eval_loss": 0.2459673434495926,
      "eval_runtime": 2.0619,
      "eval_samples_per_second": 1503.494,
      "eval_steps_per_second": 31.525,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.665433406829834,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.397,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8367741935483871,
      "eval_loss": 0.1078028753399849,
      "eval_runtime": 2.0351,
      "eval_samples_per_second": 1523.287,
      "eval_steps_per_second": 31.94,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8829032258064516,
      "eval_loss": 0.0652221292257309,
      "eval_runtime": 2.1429,
      "eval_samples_per_second": 1446.633,
      "eval_steps_per_second": 30.333,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5546591877937317,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1259,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9070967741935484,
      "eval_loss": 0.04747674614191055,
      "eval_runtime": 2.0918,
      "eval_samples_per_second": 1481.987,
      "eval_steps_per_second": 31.074,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.41417187452316284,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0737,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.917741935483871,
      "eval_loss": 0.03854530677199364,
      "eval_runtime": 2.6012,
      "eval_samples_per_second": 1191.736,
      "eval_steps_per_second": 24.988,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9245161290322581,
      "eval_loss": 0.03373381868004799,
      "eval_runtime": 2.1107,
      "eval_samples_per_second": 1468.716,
      "eval_steps_per_second": 30.796,
      "step": 1908
    },
    {
      "epoch": 6.29,
      "grad_norm": 0.3389872610569,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0559,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 532459451005944.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.04832367034810514,
    "num_train_epochs": 10,
    "temperature": 4
  }
}