{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 7900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.33,
      "learning_rate": 4.683544303797468e-05,
      "loss": 0.8991,
      "step": 500
    },
    {
      "epoch": 12.66,
      "learning_rate": 4.367088607594937e-05,
      "loss": 0.1014,
      "step": 1000
    },
    {
      "epoch": 18.99,
      "learning_rate": 4.050632911392405e-05,
      "loss": 0.0476,
      "step": 1500
    },
    {
      "epoch": 25.32,
      "learning_rate": 3.7341772151898736e-05,
      "loss": 0.0275,
      "step": 2000
    },
    {
      "epoch": 31.65,
      "learning_rate": 3.4177215189873416e-05,
      "loss": 0.0243,
      "step": 2500
    },
    {
      "epoch": 37.97,
      "learning_rate": 3.10126582278481e-05,
      "loss": 0.0165,
      "step": 3000
    },
    {
      "epoch": 44.3,
      "learning_rate": 2.7848101265822786e-05,
      "loss": 0.0252,
      "step": 3500
    },
    {
      "epoch": 50.63,
      "learning_rate": 2.468354430379747e-05,
      "loss": 0.034,
      "step": 4000
    },
    {
      "epoch": 56.96,
      "learning_rate": 2.1518987341772153e-05,
      "loss": 0.0469,
      "step": 4500
    },
    {
      "epoch": 63.29,
      "learning_rate": 1.8354430379746836e-05,
      "loss": 0.091,
      "step": 5000
    },
    {
      "epoch": 69.62,
      "learning_rate": 1.5189873417721521e-05,
      "loss": 0.2735,
      "step": 5500
    },
    {
      "epoch": 75.95,
      "learning_rate": 1.2025316455696203e-05,
      "loss": 0.4183,
      "step": 6000
    },
    {
      "epoch": 82.28,
      "learning_rate": 8.860759493670886e-06,
      "loss": 0.5516,
      "step": 6500
    },
    {
      "epoch": 88.61,
      "learning_rate": 5.69620253164557e-06,
      "loss": 0.6946,
      "step": 7000
    },
    {
      "epoch": 94.94,
      "learning_rate": 2.531645569620253e-06,
      "loss": 1.0278,
      "step": 7500
    },
    {
      "epoch": 100.0,
      "step": 7900,
      "total_flos": 9.7986283776e+16,
      "train_loss": 0.33136190450644193,
      "train_runtime": 3703.1966,
      "train_samples_per_second": 135.018,
      "train_steps_per_second": 2.133
    }
  ],
  "logging_steps": 500,
  "max_steps": 7900,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 9.7986283776e+16,
  "trial_name": null,
  "trial_params": null
}