{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 9495,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.47,
      "learning_rate": 4.211690363349131e-05,
      "loss": 2.4319,
      "step": 500
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.4218009478672984e-05,
      "loss": 1.6405,
      "step": 1000
    },
    {
      "epoch": 1.42,
      "learning_rate": 2.631911532385466e-05,
      "loss": 1.3383,
      "step": 1500
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.8420221169036335e-05,
      "loss": 1.2262,
      "step": 2000
    },
    {
      "epoch": 2.37,
      "learning_rate": 1.052132701421801e-05,
      "loss": 1.0607,
      "step": 2500
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.622432859399684e-06,
      "loss": 0.9868,
      "step": 3000
    },
    {
      "epoch": 3.32,
      "learning_rate": 2.2361769352290678e-05,
      "loss": 0.9531,
      "step": 3500
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.8412322274881517e-05,
      "loss": 0.9439,
      "step": 4000
    },
    {
      "epoch": 4.27,
      "learning_rate": 1.4462875197472355e-05,
      "loss": 0.8432,
      "step": 4500
    },
    {
      "epoch": 4.74,
      "learning_rate": 1.0513428120063192e-05,
      "loss": 0.7916,
      "step": 5000
    },
    {
      "epoch": 5.21,
      "learning_rate": 6.563981042654029e-06,
      "loss": 0.7389,
      "step": 5500
    },
    {
      "epoch": 5.69,
      "learning_rate": 2.6145339652448658e-06,
      "loss": 0.6903,
      "step": 6000
    },
    {
      "epoch": 6.16,
      "learning_rate": 1.577672459189047e-05,
      "loss": 0.6838,
      "step": 6500
    },
    {
      "epoch": 6.64,
      "learning_rate": 1.3143759873617695e-05,
      "loss": 0.675,
      "step": 7000
    },
    {
      "epoch": 7.11,
      "learning_rate": 1.0510795155344918e-05,
      "loss": 0.6546,
      "step": 7500
    },
    {
      "epoch": 7.58,
      "learning_rate": 7.877830437072143e-06,
      "loss": 0.5931,
      "step": 8000
    },
    {
      "epoch": 8.06,
      "learning_rate": 5.244865718799368e-06,
      "loss": 0.5853,
      "step": 8500
    },
    {
      "epoch": 8.53,
      "learning_rate": 2.611901000526593e-06,
      "loss": 0.5335,
      "step": 9000
    },
    {
      "epoch": 9.0,
      "step": 9495,
      "total_flos": 4.383681832019558e+16,
      "train_loss": 0.22391470878484565,
      "train_runtime": 758.6651,
      "train_samples_per_second": 600.467,
      "train_steps_per_second": 12.515
    }
  ],
  "logging_steps": 500,
  "max_steps": 9495,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 4.383681832019558e+16,
  "trial_name": null,
  "trial_params": null
}