{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 9990,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 4.74974974974975e-05,
      "loss": 0.6173,
      "step": 500
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.4994994994994996e-05,
      "loss": 0.4542,
      "step": 1000
    },
    {
      "epoch": 1.5,
      "learning_rate": 4.24924924924925e-05,
      "loss": 0.2895,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.998998998998999e-05,
      "loss": 0.2248,
      "step": 2000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.7487487487487486e-05,
      "loss": 0.1654,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.498498498498499e-05,
      "loss": 0.1603,
      "step": 3000
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.2482482482482486e-05,
      "loss": 0.1201,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.9979979979979983e-05,
      "loss": 0.1236,
      "step": 4000
    },
    {
      "epoch": 4.5,
      "learning_rate": 2.7477477477477483e-05,
      "loss": 0.094,
      "step": 4500
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.4974974974974976e-05,
      "loss": 0.0905,
      "step": 5000
    },
    {
      "epoch": 5.51,
      "learning_rate": 2.2472472472472473e-05,
      "loss": 0.081,
      "step": 5500
    },
    {
      "epoch": 6.01,
      "learning_rate": 1.996996996996997e-05,
      "loss": 0.0672,
      "step": 6000
    },
    {
      "epoch": 6.51,
      "learning_rate": 1.746746746746747e-05,
      "loss": 0.0576,
      "step": 6500
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.4964964964964966e-05,
      "loss": 0.0536,
      "step": 7000
    },
    {
      "epoch": 7.51,
      "learning_rate": 1.2462462462462463e-05,
      "loss": 0.0356,
      "step": 7500
    },
    {
      "epoch": 8.01,
      "learning_rate": 9.959959959959962e-06,
      "loss": 0.0344,
      "step": 8000
    },
    {
      "epoch": 8.51,
      "learning_rate": 7.457457457457457e-06,
      "loss": 0.0231,
      "step": 8500
    },
    {
      "epoch": 9.01,
      "learning_rate": 4.954954954954955e-06,
      "loss": 0.0246,
      "step": 9000
    },
    {
      "epoch": 9.51,
      "learning_rate": 2.4524524524524525e-06,
      "loss": 0.0143,
      "step": 9500
    },
    {
      "epoch": 10.0,
      "step": 9990,
      "total_flos": 3.1793177520881664e+16,
      "train_runtime": 5930.4108,
      "train_samples_per_second": 1.685
    }
  ],
  "max_steps": 9990,
  "num_train_epochs": 10,
  "total_flos": 3.1793177520881664e+16,
  "trial_name": null,
  "trial_params": null
}