{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.922374429223744,
  "eval_steps": 500,
  "global_step": 16000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 1.9391171993911722e-05,
      "loss": 2.9417,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8782343987823442e-05,
      "loss": 1.7699,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8173515981735163e-05,
      "loss": 1.5303,
      "step": 1500
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.756468797564688e-05,
      "loss": 1.46,
      "step": 2000
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.69558599695586e-05,
      "loss": 1.393,
      "step": 2500
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.634703196347032e-05,
      "loss": 1.3692,
      "step": 3000
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.573820395738204e-05,
      "loss": 1.3134,
      "step": 3500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5129375951293761e-05,
      "loss": 1.2416,
      "step": 4000
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4520547945205482e-05,
      "loss": 1.2574,
      "step": 4500
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.39117199391172e-05,
      "loss": 1.2039,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1474684476852417,
      "eval_runtime": 59.0937,
      "eval_samples_per_second": 178.868,
      "eval_steps_per_second": 11.186,
      "step": 5475
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.330289193302892e-05,
      "loss": 1.1717,
      "step": 5500
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2694063926940641e-05,
      "loss": 0.9552,
      "step": 6000
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2085235920852361e-05,
      "loss": 0.9877,
      "step": 6500
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.147640791476408e-05,
      "loss": 0.9877,
      "step": 7000
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.08675799086758e-05,
      "loss": 0.9727,
      "step": 7500
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.025875190258752e-05,
      "loss": 0.9713,
      "step": 8000
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.64992389649924e-06,
      "loss": 0.9441,
      "step": 8500
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.04109589041096e-06,
      "loss": 0.9527,
      "step": 9000
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.432267884322679e-06,
      "loss": 0.9311,
      "step": 9500
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.823439878234399e-06,
      "loss": 0.9318,
      "step": 10000
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.214611872146119e-06,
      "loss": 0.9639,
      "step": 10500
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.0952799320220947,
      "eval_runtime": 59.2147,
      "eval_samples_per_second": 178.503,
      "eval_steps_per_second": 11.163,
      "step": 10950
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.605783866057839e-06,
      "loss": 0.9442,
      "step": 11000
    },
    {
      "epoch": 2.1,
      "learning_rate": 5.996955859969558e-06,
      "loss": 0.7817,
      "step": 11500
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.388127853881279e-06,
      "loss": 0.7787,
      "step": 12000
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.779299847792998e-06,
      "loss": 0.7367,
      "step": 12500
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.170471841704719e-06,
      "loss": 0.7513,
      "step": 13000
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5616438356164386e-06,
      "loss": 0.7739,
      "step": 13500
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9528158295281586e-06,
      "loss": 0.7591,
      "step": 14000
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.343987823439878e-06,
      "loss": 0.7491,
      "step": 14500
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7351598173515982e-06,
      "loss": 0.7394,
      "step": 15000
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1263318112633182e-06,
      "loss": 0.7548,
      "step": 15500
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.17503805175038e-07,
      "loss": 0.7728,
      "step": 16000
    }
  ],
  "logging_steps": 500,
  "max_steps": 16425,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.508519922649395e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}