{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 11721,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 4.7867076188038565e-05,
      "loss": 0.3418,
      "step": 500
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.573415237607713e-05,
      "loss": 0.2892,
      "step": 1000
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.360122856411569e-05,
      "loss": 0.2876,
      "step": 1500
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.146830475215426e-05,
      "loss": 0.2511,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.933538094019282e-05,
      "loss": 0.2528,
      "step": 2500
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.720245712823138e-05,
      "loss": 0.2545,
      "step": 3000
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.506953331626994e-05,
      "loss": 0.2496,
      "step": 3500
    },
    {
      "epoch": 1.02,
      "learning_rate": 3.2936609504308506e-05,
      "loss": 0.2534,
      "step": 4000
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.0803685692347076e-05,
      "loss": 0.2498,
      "step": 4500
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.8670761880385632e-05,
      "loss": 0.2333,
      "step": 5000
    },
    {
      "epoch": 1.41,
      "learning_rate": 2.6537838068424198e-05,
      "loss": 0.2448,
      "step": 5500
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.440491425646276e-05,
      "loss": 0.2426,
      "step": 6000
    },
    {
      "epoch": 1.66,
      "learning_rate": 2.2271990444501324e-05,
      "loss": 0.2371,
      "step": 6500
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.0139066632539887e-05,
      "loss": 0.2307,
      "step": 7000
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.800614282057845e-05,
      "loss": 0.2255,
      "step": 7500
    },
    {
      "epoch": 2.05,
      "learning_rate": 1.5873219008617013e-05,
      "loss": 0.2384,
      "step": 8000
    },
    {
      "epoch": 2.18,
      "learning_rate": 1.3740295196655576e-05,
      "loss": 0.226,
      "step": 8500
    },
    {
      "epoch": 2.3,
      "learning_rate": 1.1607371384694139e-05,
      "loss": 0.2295,
      "step": 9000
    },
    {
      "epoch": 2.43,
      "learning_rate": 9.474447572732703e-06,
      "loss": 0.2324,
      "step": 9500
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.3415237607712656e-06,
      "loss": 0.236,
      "step": 10000
    },
    {
      "epoch": 2.69,
      "learning_rate": 5.2085999488098285e-06,
      "loss": 0.2278,
      "step": 10500
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.075676136848392e-06,
      "loss": 0.2199,
      "step": 11000
    },
    {
      "epoch": 2.94,
      "learning_rate": 9.42752324886955e-07,
      "loss": 0.2283,
      "step": 11500
    },
    {
      "epoch": 3.0,
      "step": 11721,
      "total_flos": 298502687293440.0,
      "train_loss": 0.24679102432645433,
      "train_runtime": 13355.3232,
      "train_samples_per_second": 3.51,
      "train_steps_per_second": 0.878
    }
  ],
  "logging_steps": 500,
  "max_steps": 11721,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 298502687293440.0,
  "trial_name": null,
  "trial_params": null
}