{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 135,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 4.7333,
      "step": 10
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 1.0511,
      "step": 20
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 0.2765,
      "step": 30
    },
    {
      "epoch": 0.89,
      "learning_rate": 5e-05,
      "loss": 0.1875,
      "step": 40
    },
    {
      "epoch": 1.11,
      "learning_rate": 5e-05,
      "loss": 0.2089,
      "step": 50
    },
    {
      "epoch": 1.33,
      "learning_rate": 5e-05,
      "loss": 0.1893,
      "step": 60
    },
    {
      "epoch": 1.56,
      "learning_rate": 5e-05,
      "loss": 0.1092,
      "step": 70
    },
    {
      "epoch": 1.78,
      "learning_rate": 5e-05,
      "loss": 0.1111,
      "step": 80
    },
    {
      "epoch": 2.0,
      "learning_rate": 5e-05,
      "loss": 0.0726,
      "step": 90
    },
    {
      "epoch": 2.22,
      "learning_rate": 5e-05,
      "loss": 0.048,
      "step": 100
    },
    {
      "epoch": 2.44,
      "learning_rate": 5e-05,
      "loss": 0.029,
      "step": 110
    },
    {
      "epoch": 2.67,
      "learning_rate": 5e-05,
      "loss": 0.037,
      "step": 120
    },
    {
      "epoch": 2.89,
      "learning_rate": 5e-05,
      "loss": 0.0554,
      "step": 130
    },
    {
      "epoch": 3.0,
      "step": 135,
      "total_flos": 26945073709056.0,
      "train_loss": 0.5275618712107341,
      "train_runtime": 2086.9604,
      "train_samples_per_second": 2.069,
      "train_steps_per_second": 0.065
    }
  ],
  "logging_steps": 10,
  "max_steps": 135,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 26945073709056.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}