{
  "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": 0.5693548387096774,
      "eval_loss": 0.19532112777233124,
      "eval_runtime": 1.9247,
      "eval_samples_per_second": 1610.663,
      "eval_steps_per_second": 33.772,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.6180300116539001,
      "learning_rate": 1.4758909853249476e-05,
      "loss": 0.3105,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8048387096774193,
      "eval_loss": 0.10033391416072845,
      "eval_runtime": 2.1516,
      "eval_samples_per_second": 1440.792,
      "eval_steps_per_second": 30.21,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8658064516129033,
      "eval_loss": 0.0706540197134018,
      "eval_runtime": 2.4126,
      "eval_samples_per_second": 1284.939,
      "eval_steps_per_second": 26.942,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5102590918540955,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.1155,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.057368192821741104,
      "eval_runtime": 2.1421,
      "eval_samples_per_second": 1447.207,
      "eval_steps_per_second": 30.345,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.3783129155635834,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0796,
      "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": 0.9715658576550433,
    "num_train_epochs": 6,
    "temperature": 19
  }
}