{
  "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": 0.5735483870967742,
      "eval_loss": 0.19572289288043976,
      "eval_runtime": 6.6194,
      "eval_samples_per_second": 468.32,
      "eval_steps_per_second": 9.82,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.6219620108604431,
      "learning_rate": 1.4758909853249476e-05,
      "loss": 0.3113,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8058064516129032,
      "eval_loss": 0.10050024837255478,
      "eval_runtime": 4.6485,
      "eval_samples_per_second": 666.879,
      "eval_steps_per_second": 13.983,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8651612903225806,
      "eval_loss": 0.07076359540224075,
      "eval_runtime": 2.889,
      "eval_samples_per_second": 1073.036,
      "eval_steps_per_second": 22.499,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5108679533004761,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.1157,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8858064516129033,
      "eval_loss": 0.0574265792965889,
      "eval_runtime": 3.3052,
      "eval_samples_per_second": 937.905,
      "eval_steps_per_second": 19.666,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.3781363070011139,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0797,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1908,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 391368939443328.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.07810074874529571,
    "num_train_epochs": 6,
    "temperature": 18
  }
}