{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 2000.0,
  "global_step": 4719,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "learning_rate": 2.5e-05,
      "loss": 1.5255,
      "step": 500
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 1.1342,
      "step": 1000
    },
    {
      "epoch": 0.95,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0615,
      "step": 1500
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.021,
      "step": 2000
    },
    {
      "epoch": 1.59,
      "learning_rate": 8.161088635527768e-05,
      "loss": 0.9927,
      "step": 2500
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.322177271055535e-05,
      "loss": 0.9672,
      "step": 3000
    },
    {
      "epoch": 2.23,
      "learning_rate": 4.483265906583303e-05,
      "loss": 0.9361,
      "step": 3500
    },
    {
      "epoch": 2.54,
      "learning_rate": 2.6443545421110705e-05,
      "loss": 0.9202,
      "step": 4000
    },
    {
      "epoch": 2.86,
      "learning_rate": 8.054431776388379e-06,
      "loss": 0.9096,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "step": 4719,
      "total_flos": 2.30068809105408e+16,
      "train_loss": 1.0451076868909677,
      "train_runtime": 3967.7657,
      "train_samples_per_second": 19.02,
      "train_steps_per_second": 1.189
    }
  ],
  "logging_steps": 500,
  "max_steps": 4719,
  "num_train_epochs": 3,
  "save_steps": 2000,
  "total_flos": 2.30068809105408e+16,
  "trial_name": null,
  "trial_params": null
}