{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.02139037433155,
  "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": 0.6878980994224548,
      "eval_loss": 0.5814566016197205,
      "eval_runtime": 8.9565,
      "eval_samples_per_second": 333.054,
      "eval_steps_per_second": 20.879,
      "step": 187
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7194100022315979,
      "eval_loss": 0.5218587517738342,
      "eval_runtime": 9.0661,
      "eval_samples_per_second": 329.027,
      "eval_steps_per_second": 20.626,
      "step": 374
    },
    {
      "epoch": 2.6737967914438503,
      "grad_norm": 7.46340274810791,
      "learning_rate": 1.5373335026225161e-06,
      "loss": 0.5847,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7344954609870911,
      "eval_loss": 0.5096723437309265,
      "eval_runtime": 9.0234,
      "eval_samples_per_second": 330.583,
      "eval_steps_per_second": 20.724,
      "step": 561
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7398592233657837,
      "eval_loss": 0.5069338083267212,
      "eval_runtime": 9.0796,
      "eval_samples_per_second": 328.54,
      "eval_steps_per_second": 20.596,
      "step": 748
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7475695610046387,
      "eval_loss": 0.5030502676963806,
      "eval_runtime": 9.111,
      "eval_samples_per_second": 327.405,
      "eval_steps_per_second": 20.525,
      "step": 935
    },
    {
      "epoch": 5.347593582887701,
      "grad_norm": 5.598100662231445,
      "learning_rate": 9.762628812274372e-07,
      "loss": 0.4789,
      "step": 1000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7499161958694458,
      "eval_loss": 0.5032511949539185,
      "eval_runtime": 9.0706,
      "eval_samples_per_second": 328.864,
      "eval_steps_per_second": 20.616,
      "step": 1122
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7512571215629578,
      "eval_loss": 0.5065814852714539,
      "eval_runtime": 9.1058,
      "eval_samples_per_second": 327.593,
      "eval_steps_per_second": 20.536,
      "step": 1309
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7515923380851746,
      "eval_loss": 0.5092657804489136,
      "eval_runtime": 9.0837,
      "eval_samples_per_second": 328.392,
      "eval_steps_per_second": 20.586,
      "step": 1496
    },
    {
      "epoch": 8.02139037433155,
      "grad_norm": 8.30627727508545,
      "learning_rate": 4.1519225983235836e-07,
      "loss": 0.4364,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1870,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.6202496657094632e+16,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.098404124017595e-06,
    "per_device_train_batch_size": 64
  }
}