{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "global_step": 1620,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 1.9176954732510288e-05,
      "loss": 1.5253,
      "step": 100
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.835390946502058e-05,
      "loss": 0.5112,
      "step": 200
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7530864197530865e-05,
      "loss": 0.3811,
      "step": 300
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.670781893004115e-05,
      "loss": 0.2638,
      "step": 400
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.588477366255144e-05,
      "loss": 0.2136,
      "step": 500
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.506172839506173e-05,
      "loss": 0.1868,
      "step": 600
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4238683127572017e-05,
      "loss": 0.17,
      "step": 700
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3415637860082307e-05,
      "loss": 0.1193,
      "step": 800
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2592592592592593e-05,
      "loss": 0.108,
      "step": 900
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1769547325102882e-05,
      "loss": 0.0724,
      "step": 1000
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0946502057613168e-05,
      "loss": 0.072,
      "step": 1100
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0123456790123458e-05,
      "loss": 0.0732,
      "step": 1200
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.300411522633745e-06,
      "loss": 0.0483,
      "step": 1300
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.477366255144033e-06,
      "loss": 0.0586,
      "step": 1400
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.654320987654322e-06,
      "loss": 0.0432,
      "step": 1500
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.83127572016461e-06,
      "loss": 0.0275,
      "step": 1600
    }
  ],
  "max_steps": 2430,
  "num_train_epochs": 3,
  "total_flos": 5.698796483985408e+18,
  "trial_name": null,
  "trial_params": null
}