{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4506769825918762,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 4.8388136686009026e-05,
      "loss": 0.6868,
      "step": 100
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.677627337201805e-05,
      "loss": 0.5552,
      "step": 200
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.516441005802708e-05,
      "loss": 0.5657,
      "step": 300
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.355254674403611e-05,
      "loss": 0.4919,
      "step": 400
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.1940683430045136e-05,
      "loss": 0.4673,
      "step": 500
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.4286854565143585,
      "eval_runtime": 157.876,
      "eval_samples_per_second": 13.093,
      "eval_steps_per_second": 1.641,
      "step": 500
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.032882011605416e-05,
      "loss": 0.4705,
      "step": 600
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.8716956802063184e-05,
      "loss": 0.4717,
      "step": 700
    },
    {
      "epoch": 0.77,
      "learning_rate": 3.7105093488072215e-05,
      "loss": 0.4523,
      "step": 800
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.549323017408124e-05,
      "loss": 0.4333,
      "step": 900
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.388136686009026e-05,
      "loss": 0.4765,
      "step": 1000
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.3951832354068756,
      "eval_runtime": 154.675,
      "eval_samples_per_second": 13.364,
      "eval_steps_per_second": 1.674,
      "step": 1000
    },
    {
      "epoch": 1.06,
      "learning_rate": 3.226950354609929e-05,
      "loss": 0.3246,
      "step": 1100
    },
    {
      "epoch": 1.16,
      "learning_rate": 3.065764023210832e-05,
      "loss": 0.3084,
      "step": 1200
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.9045776918117345e-05,
      "loss": 0.3059,
      "step": 1300
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.7433913604126372e-05,
      "loss": 0.3234,
      "step": 1400
    },
    {
      "epoch": 1.45,
      "learning_rate": 2.5822050290135396e-05,
      "loss": 0.3353,
      "step": 1500
    },
    {
      "epoch": 1.45,
      "eval_loss": 0.5979052186012268,
      "eval_runtime": 152.068,
      "eval_samples_per_second": 13.593,
      "eval_steps_per_second": 1.703,
      "step": 1500
    }
  ],
  "logging_steps": 100,
  "max_steps": 3102,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 597867456650040.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}