{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.30752967180817836,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "ep_loss": 4.4238,
      "epoch": 0.04,
      "learning_rate": 2.5e-05,
      "loss": 11.1467,
      "mlm_loss": 6.7229,
      "step": 500
    },
    {
      "ep_loss": 0.4305,
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 2.9841,
      "mlm_loss": 2.5537,
      "step": 1000
    },
    {
      "ep_loss": 0.3724,
      "epoch": 0.12,
      "learning_rate": 7.5e-05,
      "loss": 2.2371,
      "mlm_loss": 1.8647,
      "step": 1500
    },
    {
      "ep_loss": 0.364,
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 1.8477,
      "mlm_loss": 1.4837,
      "step": 2000
    },
    {
      "ep_loss": 0.3678,
      "epoch": 0.19,
      "learning_rate": 0.00012495,
      "loss": 1.5215,
      "mlm_loss": 1.1538,
      "step": 2500
    },
    {
      "ep_loss": 0.3617,
      "epoch": 0.23,
      "learning_rate": 0.00014995,
      "loss": 1.4119,
      "mlm_loss": 1.0501,
      "step": 3000
    },
    {
      "ep_loss": 0.3336,
      "epoch": 0.27,
      "learning_rate": 0.0001749,
      "loss": 1.3027,
      "mlm_loss": 0.9691,
      "step": 3500
    },
    {
      "ep_loss": 0.3348,
      "epoch": 0.31,
      "learning_rate": 0.0001999,
      "loss": 1.2441,
      "mlm_loss": 0.9093,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 520240,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 3.8192690878008525e+18,
  "trial_name": null,
  "trial_params": null
}