{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.643835616438356,
  "global_step": 9000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 1.9391171993911722e-05,
      "loss": 1.0449,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8782343987823442e-05,
      "loss": 0.821,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8173515981735163e-05,
      "loss": 0.8326,
      "step": 1500
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.756468797564688e-05,
      "loss": 0.8143,
      "step": 2000
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.69558599695586e-05,
      "loss": 0.7836,
      "step": 2500
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.634703196347032e-05,
      "loss": 0.7796,
      "step": 3000
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.573820395738204e-05,
      "loss": 0.7676,
      "step": 3500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5129375951293761e-05,
      "loss": 0.7507,
      "step": 4000
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4520547945205482e-05,
      "loss": 0.7615,
      "step": 4500
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.39117199391172e-05,
      "loss": 0.7582,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8703365921974182,
      "eval_runtime": 270.4723,
      "eval_samples_per_second": 39.08,
      "eval_steps_per_second": 2.444,
      "step": 5475
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.330289193302892e-05,
      "loss": 0.7364,
      "step": 5500
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2694063926940641e-05,
      "loss": 0.5517,
      "step": 6000
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2085235920852361e-05,
      "loss": 0.5915,
      "step": 6500
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.147640791476408e-05,
      "loss": 0.5903,
      "step": 7000
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.08675799086758e-05,
      "loss": 0.5782,
      "step": 7500
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.025875190258752e-05,
      "loss": 0.5823,
      "step": 8000
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.64992389649924e-06,
      "loss": 0.5812,
      "step": 8500
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.04109589041096e-06,
      "loss": 0.5604,
      "step": 9000
    }
  ],
  "max_steps": 16425,
  "num_train_epochs": 3,
  "total_flos": 2.821985375492045e+16,
  "trial_name": null,
  "trial_params": null
}