{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.433682161109759,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 1.919201752251156e-05,
      "loss": 2.3883,
      "step": 500
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8380790135474976e-05,
      "loss": 1.4529,
      "step": 1000
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7569562748438388e-05,
      "loss": 1.2698,
      "step": 1500
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6758335361401804e-05,
      "loss": 1.1845,
      "step": 2000
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.594873042913929e-05,
      "loss": 1.1239,
      "step": 2500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5137503042102703e-05,
      "loss": 1.0576,
      "step": 3000
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4326275655066115e-05,
      "loss": 1.0744,
      "step": 3500
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3515048268029529e-05,
      "loss": 1.0113,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8904165625572205,
      "eval_runtime": 91.887,
      "eval_samples_per_second": 131.694,
      "eval_steps_per_second": 2.068,
      "step": 4109
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2703820880992943e-05,
      "loss": 0.9452,
      "step": 4500
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1892593493956356e-05,
      "loss": 0.9205,
      "step": 5000
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.108136610691977e-05,
      "loss": 0.8961,
      "step": 5500
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0271761174657257e-05,
      "loss": 0.906,
      "step": 6000
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.460533787620671e-06,
      "loss": 0.8696,
      "step": 6500
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.649306400584085e-06,
      "loss": 0.8745,
      "step": 7000
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.838079013547499e-06,
      "loss": 0.8528,
      "step": 7500
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.028474081284985e-06,
      "loss": 0.8645,
      "step": 8000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8758471012115479,
      "eval_runtime": 91.8658,
      "eval_samples_per_second": 131.725,
      "eval_steps_per_second": 2.068,
      "step": 8218
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.217246694248399e-06,
      "loss": 0.8209,
      "step": 8500
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.406019307211812e-06,
      "loss": 0.798,
      "step": 9000
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.594791920175226e-06,
      "loss": 0.7944,
      "step": 9500
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.783564533138639e-06,
      "loss": 0.7893,
      "step": 10000
    }
  ],
  "max_steps": 12327,
  "num_train_epochs": 3,
  "total_flos": 1.5698961223584768e+16,
  "trial_name": null,
  "trial_params": null
}