{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 8728,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 3.597898028912073e-05,
      "loss": 2.1559,
      "step": 500
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.379737828274668e-05,
      "loss": 1.5255,
      "step": 1000
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.1624502684398136e-05,
      "loss": 1.4008,
      "step": 1500
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.9442900678024096e-05,
      "loss": 1.3043,
      "step": 2000
    },
    {
      "epoch": 1.15,
      "learning_rate": 2.7261298671650053e-05,
      "loss": 1.1264,
      "step": 2500
    },
    {
      "epoch": 1.37,
      "learning_rate": 2.5079696665276006e-05,
      "loss": 1.0354,
      "step": 3000
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.290245786291471e-05,
      "loss": 1.0141,
      "step": 3500
    },
    {
      "epoch": 1.83,
      "learning_rate": 2.0720855856540665e-05,
      "loss": 1.0036,
      "step": 4000
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.854361705417937e-05,
      "loss": 0.9137,
      "step": 4500
    },
    {
      "epoch": 2.29,
      "learning_rate": 1.6362015047805328e-05,
      "loss": 0.756,
      "step": 5000
    },
    {
      "epoch": 2.52,
      "learning_rate": 1.4180413041431285e-05,
      "loss": 0.741,
      "step": 5500
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.199881103505724e-05,
      "loss": 0.7376,
      "step": 6000
    },
    {
      "epoch": 2.98,
      "learning_rate": 9.817209028683196e-06,
      "loss": 0.7175,
      "step": 6500
    },
    {
      "epoch": 3.21,
      "learning_rate": 7.639970226321902e-06,
      "loss": 0.5542,
      "step": 7000
    },
    {
      "epoch": 3.44,
      "learning_rate": 5.458368219947858e-06,
      "loss": 0.5413,
      "step": 7500
    },
    {
      "epoch": 3.67,
      "learning_rate": 3.276766213573814e-06,
      "loss": 0.5365,
      "step": 8000
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.09516420719977e-06,
      "loss": 0.5279,
      "step": 8500
    },
    {
      "epoch": 4.0,
      "step": 8728,
      "total_flos": 136961485963264.0,
      "train_loss": 0.9642123808891373,
      "train_runtime": 1310.5941,
      "train_samples_per_second": 399.41,
      "train_steps_per_second": 6.66
    }
  ],
  "max_steps": 8728,
  "num_train_epochs": 4,
  "total_flos": 136961485963264.0,
  "trial_name": null,
  "trial_params": null
}